Hybrid: Input -> x264/x265/Xvid/VP8/VP9

[Selur]

n other GUIs encoder that I have used like, staxrip. These steps are not necessary. Just the enter the parameters and it would stick during encoding
Ah, yes. StaxRip basically has only 'Custom' as audio handling option.

Is it possible to have the dependencies updated too?
Whenever you want to update the dependencies tell Hybrid by 'Config->Tools->Clear all cached tool infos'
I know they are a lot that needs to be check from time to time
Which is hey the are not checked each time, but only on new releases (revision changes).
It seems changes are made in the dependencies like, x265 almost daily
Not really, see: https://bitbucket.org/multicoreware/x265/commits/all
Just because folks release new builds doesn't mean anything has changed aside from some different build option. ;)

I normally release a new version when one of the following things occurs:

I'm aware of a change in one of the dependencies that I would call important (+ I was able to adjust Hybrid to it)
I fixed some bug which I deem important + my list of 'bugs to fix' doesn't include any important bugs and I got feedback from all the folks that reported problems whether the problem was fixed
I added some feature that seems really important

something that stops me from releasing a new version is typically:

waiting for feedback regarding bugs
problems building some of the required tools
working on some major bugs
I'm in the middle of some code refactoring/cleanup to keep the code maintainable adding workarounds and quick-fixes during the week require some major clean-up from time to time
simply swamped with real world occurrences


Is it also possible to have 2 updates a month?
short:
Sorry, but my free time doesn't offer that option.
longer:
During the week I have 1-1.5 hrs travel time to work, 8 hrs work, 1-1.5 hrs travel time back home, ~7 hrs sleep that's 17-18hrs a day gone then there is that matter of getting ready for work, shopping for food&co, making food, eating, cleaning the apartment, occasionally meeting friends&co, reading up on stuff (system maintenance etc.). So after answering questions and working on bug fixes *, changes etc. releases and larger adjustments (requiring more testing or coding) are normally planned for the weekend which, like for most working folks is my main relaxing and coding time. Also note that it takes quite some time to build all the tools required for an update. (on windows for example I use https://github.com/jb-alvarado/media-autobuild_suite to build new versions of the tools which sometimes quite takes a while to complete,..)
Don't see me releasing a regular two releases a month unless I win the lottery, change workplace or similar.
=> That's the downside of hobby projects :D.
They usually simply don't pay the bills and if you have an intact social life with real people interaction and time for coding&co is limited.

Cu Selur

* Which both (bug fixes and general user support) can be quite challenging.
Especially if you got google translator in the middle. Especially when Arabic users use it I sometimes spend a few hours just trying to figure out if the folk on the other side of the translation is trying to describe a bug or whether they simply tried to use Hybrid in an (for me) unexpected way.

[Magik Mark]

Maybe I can offer you beer or coffee to ease you with your hectic life? ;)

[Magik Mark]

Selur where can I find documentation for nlmeanscl2? Thanks


Sent from my iPhone using Tapatalk

[Selur]

nlmeanscl2 evolved to knlmeanscl which is why the nlmeanscl2 thread was renamed
http://forum.doom9.org/showthread.php?p=1718490#post1718490 <- there is the last version named NLMeansCL2
http://forum.doom9.org/showthread.php?p=1719416#post1719416 <- here's the first version named KNLMeansCL
I just kept NLMeansCL2 due to the request of a user, don't know if there is still a working download for the NLMeansCL2 download somewhere.

[Magik Mark]

Thanks!

You may want to take a look at fluxsmooth. Cleans the video well with minimal degradation on the image itself. Most importantly - it's fast

[Magik Mark]

I could add an option to specify the 'Cores' value of Interfame. Still 27 seems a rather high value.

1. I have been toying around the test built with "Core" parameter. Is there really any difference if I just use Avisynth -> Misc ->"MT Thread" setting? Right now I don't see any difference. Maybe the right combination of "Core" and MT Threads" must be right? Would you suggest some value? I have a 28 threads Xeon Processor. Maybe the right amount of memory setting?

2. Can you recommend an optimum memory setting for the ff filters:

f3dk - > Undot -> VagueDenoiser -> Interframe

[Selur]

Sorry, no clue. Due to my experience on high core count systems avisynth simply sucks in regard of cpu core count vs speed scaling. Vapoursynth seems to scale better, but is way slower.

[Magik Mark]

Selur

Why is it that qaac -> low complexity can only resample up to 48khz? Is this a limitation of qaac?

[Selur]

You can see the supported formats<>sample rate<>bit rate qaac supports by calling:
qaac.exe --formats
which gives you:
LC 8000Hz Mono -- 8,12,16,20,24
LC 8000Hz Stereo -- 16,20,24,28,32,40,48
LC 8000Hz 3.0 (C L R) -- 24,28,32,40,48,56,64,72
LC 8000Hz 4.0 (L R Ls Rs) -- 32,40,48,56,64,72,80,96
LC 8000Hz 4.0 (C L R Cs) -- 32,40,48,56,64,72,80,96
LC 8000Hz 5.0 (C L R Ls Rs) -- 40,48,56,64,72,80,96,112
LC 8000Hz 5.1 (C L R Ls Rs LFE) -- 40,48,56,64,72,80,96,112
LC 8000Hz 6.0 (C L R Ls Rs Cs) -- 48,56,64,72,80,96,112,128,144
LC 8000Hz 6.1 (C L R Ls Rs Cs LFE) -- 48,56,64,72,80,96,112,128,144
LC 8000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 56,64,72,80,96,112,128,144,160
LC 8000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 56,64,72,80,96,112,128,144,160
LC 8000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 64,72,80,96,112,128,144,160,192
LC 11025Hz Mono -- 8,12,16,20,24,28,32
LC 11025Hz Stereo -- 16,20,24,28,32,40,48,56,64
LC 11025Hz 3.0 (C L R) -- 40,48,56,64,72,80,96
LC 11025Hz 4.0 (L R Ls Rs) -- 48,56,64,72,80,96,112,128
LC 11025Hz 4.0 (C L R Cs) -- 48,56,64,72,80,96,112,128
LC 11025Hz 5.0 (C L R Ls Rs) -- 64,72,80,96,112,128,144,160
LC 11025Hz 5.1 (C L R Ls Rs LFE) -- 64,72,80,96,112,128,144,160
LC 11025Hz 6.0 (C L R Ls Rs Cs) -- 72,80,96,112,128,144,160,192
LC 11025Hz 6.1 (C L R Ls Rs Cs LFE) -- 72,80,96,112,128,144,160,192
LC 11025Hz 7.0 (C L R Ls Rs Rls Rrs) -- 96,112,128,144,160,192,224
LC 11025Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 96,112,128,144,160,192,224
LC 11025Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 96,112,128,144,160,192,224,256
LC 12000Hz Mono -- 12,16,20,24,28,32
LC 12000Hz Stereo -- 24,28,32,40,48,56,64
LC 12000Hz 3.0 (C L R) -- 40,48,56,64,72,80,96
LC 12000Hz 4.0 (L R Ls Rs) -- 48,56,64,72,80,96,112,128
LC 12000Hz 4.0 (C L R Cs) -- 48,56,64,72,80,96,112,128
LC 12000Hz 5.0 (C L R Ls Rs) -- 64,72,80,96,112,128,144,160
LC 12000Hz 5.1 (C L R Ls Rs LFE) -- 64,72,80,96,112,128,144,160
LC 12000Hz 6.0 (C L R Ls Rs Cs) -- 72,80,96,112,128,144,160,192
LC 12000Hz 6.1 (C L R Ls Rs Cs LFE) -- 72,80,96,112,128,144,160,192
LC 12000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 96,112,128,144,160,192,224
LC 12000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 96,112,128,144,160,192,224
LC 12000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 96,112,128,144,160,192,224,256
LC 16000Hz Mono -- 12,16,20,24,28,32,40,48
LC 16000Hz Stereo -- 24,28,32,40,48,56,64,72,80,96
LC 16000Hz 3.0 (C L R) -- 40,48,56,64,72,80,96,112,128,144
LC 16000Hz 4.0 (L R Ls Rs) -- 48,56,64,72,80,96,112,128,144,160,192
LC 16000Hz 4.0 (C L R Cs) -- 48,56,64,72,80,96,112,128,144,160,192
LC 16000Hz 5.0 (C L R Ls Rs) -- 64,72,80,96,112,128,144,160,192,224
LC 16000Hz 5.1 (C L R Ls Rs LFE) -- 64,72,80,96,112,128,144,160,192,224
LC 16000Hz 6.0 (C L R Ls Rs Cs) -- 72,80,96,112,128,144,160,192,224,256,288
LC 16000Hz 6.1 (C L R Ls Rs Cs LFE) -- 72,80,96,112,128,144,160,192,224,256,288
LC 16000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 96,112,128,144,160,192,224,256,288,320
LC 16000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 96,112,128,144,160,192,224,256,288,320
LC 16000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 96,112,128,144,160,192,224,256,288,320,384
LC 22050Hz Mono -- 16,20,24,28,32,40,48,56,64
LC 22050Hz Stereo -- 32,40,48,56,64,72,80,96,112,128
LC 22050Hz 3.0 (C L R) -- 48,56,64,72,80,96,112,128,144,160,192
LC 22050Hz 4.0 (L R Ls Rs) -- 64,72,80,96,112,128,144,160,192,224,256
LC 22050Hz 4.0 (C L R Cs) -- 64,72,80,96,112,128,144,160,192,224,256
LC 22050Hz 5.0 (C L R Ls Rs) -- 80,96,112,128,144,160,192,224,256,288,320
LC 22050Hz 5.1 (C L R Ls Rs LFE) -- 80,96,112,128,144,160,192,224,256,288,320
LC 22050Hz 6.0 (C L R Ls Rs Cs) -- 96,112,128,144,160,192,224,256,288,320,384
LC 22050Hz 6.1 (C L R Ls Rs Cs LFE) -- 96,112,128,144,160,192,224,256,288,320,384
LC 22050Hz 7.0 (C L R Ls Rs Rls Rrs) -- 112,128,144,160,192,224,256,288,320,384,448
LC 22050Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 112,128,144,160,192,224,256,288,320,384,448
LC 22050Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 128,144,160,192,224,256,288,320,384,448,512
LC 24000Hz Mono -- 16,20,24,28,32,40,48,56,64
LC 24000Hz Stereo -- 32,40,48,56,64,72,80,96,112,128
LC 24000Hz 3.0 (C L R) -- 48,56,64,72,80,96,112,128,144,160,192
LC 24000Hz 4.0 (L R Ls Rs) -- 64,72,80,96,112,128,144,160,192,224,256
LC 24000Hz 4.0 (C L R Cs) -- 64,72,80,96,112,128,144,160,192,224,256
LC 24000Hz 5.0 (C L R Ls Rs) -- 80,96,112,128,144,160,192,224,256,288,320
LC 24000Hz 5.1 (C L R Ls Rs LFE) -- 80,96,112,128,144,160,192,224,256,288,320
LC 24000Hz 6.0 (C L R Ls Rs Cs) -- 96,112,128,144,160,192,224,256,288,320,384
LC 24000Hz 6.1 (C L R Ls Rs Cs LFE) -- 96,112,128,144,160,192,224,256,288,320,384
LC 24000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 112,128,144,160,192,224,256,288,320,384,448
LC 24000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 112,128,144,160,192,224,256,288,320,384,448
LC 24000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 128,144,160,192,224,256,288,320,384,448,512
LC 32000Hz Mono -- 24,28,32,40,48,56,64,72,80,96
LC 32000Hz Stereo -- 48,56,64,72,80,96,112,128,144,160,192
LC 32000Hz 3.0 (C L R) -- 72,80,96,112,128,144,160,192,224,256,288
LC 32000Hz 4.0 (L R Ls Rs) -- 96,112,128,144,160,192,224,256,288,320,384
LC 32000Hz 4.0 (C L R Cs) -- 96,112,128,144,160,192,224,256,288,320,384
LC 32000Hz 5.0 (C L R Ls Rs) -- 128,144,160,192,224,256,288,320,384,448
LC 32000Hz 5.1 (C L R Ls Rs LFE) -- 128,144,160,192,224,256,288,320,384,448
LC 32000Hz 6.0 (C L R Ls Rs Cs) -- 144,160,192,224,256,288,320,384,448,512,576
LC 32000Hz 6.1 (C L R Ls Rs Cs LFE) -- 144,160,192,224,256,288,320,384,448,512,576
LC 32000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 192,224,256,288,320,384,448,512,576,640
LC 32000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 192,224,256,288,320,384,448,512,576,640
LC 32000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 192,224,256,288,320,384,448,512,576,640,768
LC 44100Hz Mono -- 32,40,48,56,64,72,80,96,112,128,144,160,192,224,256
LC 44100Hz Stereo -- 64,72,80,96,112,128,144,160,192,224,256,288,320
LC 44100Hz 3.0 (C L R) -- 96,112,128,144,160,192,224,256,288,320,384,448
LC 44100Hz 4.0 (L R Ls Rs) -- 128,144,160,192,224,256,288,320,384,448,512,576,640
LC 44100Hz 4.0 (C L R Cs) -- 128,144,160,192,224,256,288,320,384,448,512,576,640
LC 44100Hz 5.0 (C L R Ls Rs) -- 160,192,224,256,288,320,384,448,512,576,640,768
LC 44100Hz 5.1 (C L R Ls Rs LFE) -- 160,192,224,256,288,320,384,448,512,576,640,768
LC 44100Hz 6.0 (C L R Ls Rs Cs) -- 192,224,256,288,320,384,448,512,576,640,768,960
LC 44100Hz 6.1 (C L R Ls Rs Cs LFE) -- 192,224,256,288,320,384,448,512,576,640,768,960
LC 44100Hz 7.0 (C L R Ls Rs Rls Rrs) -- 224,256,288,320,384,448,512,576,640,768,960
LC 44100Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 224,256,288,320,384,448,512,576,640,768,960
LC 44100Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 256,288,320,384,448,512,576,640,768,960,1280
LC 48000Hz Mono -- 32,40,48,56,64,72,80,96,112,128,144,160,192,224,256
LC 48000Hz Stereo -- 64,72,80,96,112,128,144,160,192,224,256,288,320
LC 48000Hz 3.0 (C L R) -- 96,112,128,144,160,192,224,256,288,320,384,448
LC 48000Hz 4.0 (L R Ls Rs) -- 128,144,160,192,224,256,288,320,384,448,512,576,640
LC 48000Hz 4.0 (C L R Cs) -- 128,144,160,192,224,256,288,320,384,448,512,576,640
LC 48000Hz 5.0 (C L R Ls Rs) -- 160,192,224,256,288,320,384,448,512,576,640,768
LC 48000Hz 5.1 (C L R Ls Rs LFE) -- 160,192,224,256,288,320,384,448,512,576,640,768
LC 48000Hz 6.0 (C L R Ls Rs Cs) -- 192,224,256,288,320,384,448,512,576,640,768,960
LC 48000Hz 6.1 (C L R Ls Rs Cs LFE) -- 192,224,256,288,320,384,448,512,576,640,768,960
LC 48000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 224,256,288,320,384,448,512,576,640,768,960
LC 48000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 224,256,288,320,384,448,512,576,640,768,960
LC 48000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 256,288,320,384,448,512,576,640,768,960,1280
HE 32000Hz Mono -- 12,16,24,32,40
HE 32000Hz Stereo -- 24,32,40,48,56,64,80
HE 32000Hz 4.0 (L R Ls Rs) -- 48,56,64,80,96,112,128,160
HE 32000Hz 5.1 (C L R Ls Rs LFE) -- 64,80,96,112,128,160,192
HE 32000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 96,112,128,160,192,224,256
HE 44100Hz Mono -- 16,24,32,40
HE 44100Hz Stereo -- 32,40,48,56,64,80
HE 44100Hz 4.0 (L R Ls Rs) -- 64,80,96,112,128,160
HE 44100Hz 5.1 (C L R Ls Rs LFE) -- 80,96,112,128,160,192
HE 44100Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 112,128,160,192,224,256
HE 48000Hz Mono -- 16,24,32,40
HE 48000Hz Stereo -- 32,40,48,56,64,80
HE 48000Hz 4.0 (L R Ls Rs) -- 64,80,96,112,128,160
HE 48000Hz 5.1 (C L R Ls Rs LFE) -- 80,96,112,128,160,192
HE 48000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 112,128,160,192,224,256
HE 88200Hz Mono -- 32,40,48,56,64,80,96,112,128,160
HE 88200Hz Stereo -- 64,80,96,112,128,160,192,224,256,320
HE 88200Hz 4.0 (L R Ls Rs) -- 128,160,192,224,256,320,448,640
HE 88200Hz 5.1 (C L R Ls Rs LFE) -- 160,192,224,256,320,448,640
HE 88200Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 224,256,320,448,640,1120
HE 96000Hz Mono -- 32,40,48,56,64,80,96,112,128,160
HE 96000Hz Stereo -- 64,80,96,112,128,160,192,224,256,320
HE 96000Hz 4.0 (L R Ls Rs) -- 128,160,192,224,256,320,448,640
HE 96000Hz 5.1 (C L R Ls Rs LFE) -- 160,192,224,256,320,448,640
-> yes, it's a restriction of qaac

[Magik Mark]

Thanks


Sent from my iPhone using Tapatalk

[Magik Mark]

Selur,

The latest x265 has new switches:

--qpmin <integer> sets a hard lower limit on QP allowed to ratecontrol. Default 0
--qpmax <integer> sets a hard upper limit on QP allowed to ratecontrol. Default 69
...
--log2-max-poc-lsb <integer> Maximum of the picture order count
--[no-]sei-dump Emit SEI packets in bitstream. Default enabled

[Selur]

--log2-max-poc-lsb <integer> Maximum of the picture order count
--[no-]sei-dump Emit SEI packets in bitstream. Default enabled
-> don't seem really interesting. Any inside how these could help with encoding?

--qpmin <integer> sets a hard lower limit on QP allowed to ratecontrol. Default 0
--qpmax <integer> sets a hard upper limit on QP allowed to ratecontrol. Default 69
'x265->Quantization->General Quantization->Quantizer (min/max)' (wrote code for that a while ago, but it's not in the current released version)

[Gravitator]

С наступающим Новым годом!
General
Unique ID : 216635845488522259592977839275463268625 (0xA2FA8BD2CE6AE64B88C5F06F0289D911)
Complete name : Z:\1.mkv
Format : Matroska
Format version : Version 4 / Version 2
File size : 1.39 MiB
Duration : 2s 100ms
Overall bit rate : 5 552 Kbps
Encoded date : UTC 2016-09-04 06:35:35
Writing application : mkvmerge v9.0.0 ('Power to progress') 64bit
Writing library : libebml v1.3.3 + libmatroska v1.4.4
Encoding Gui : Hybrid 2016.08.27.1
DURATION : 00:00:02.100000000
NUMBER_OF_FRAMES : 105
NUMBER_OF_BYTES : 1450771
_STATISTICS_WRITING_APP : mkvmerge v9.0.0 ('Power to progress') 64bit
_STATISTICS_WRITING_DATE_UTC : 2016-09-04 06:35:35
_STATISTICS_TAGS : BPS DURATION NUMBER_OF_FRAMES NUMBER_OF_BYTES

Video
ID : 1
Format : VP9
Codec ID : V_VP9
Duration : 2s 100ms
Bit rate : 5 442 Kbps
Width : 1 920 pixels
Height : 1 088 pixels
Display aspect ratio : 16:9
Frame rate mode : Constant
Frame rate : 50.000 fps
Bits/(Pixel*Frame) : 0.052
Stream size : 1.36 MiB (98%)
Language : English
Default : Yes
Forced : Yes




VP9 refused to adhere to encoding with a given 1500 kbit/s.

[Selur]

will look at it once the attachment is approved, which might take a while,... (so don't blame me if I forgot about it until then)
In case the encoding command line is as it should, the only way to fix it is to report it to the webm-team and probably wait 1/2 year till they look into it.

[Gravitator]

AVS preview naughty when trying to view ProRes. :p

[Selur]

I probably fixed that a few days ago,.. (see: http://forum.selur.de/topic1287-avisynth-preview-error.html)
-> will be fixed in the next release (avisynth filter preview didn't behave properly when handling 10bit input; normal avisynth preview should work)
Might look at the debug output in case it gets approved before I forget about it.

[Gravitator]

Is there a way to control the smoothing I-frames for VP9? Starting frames at the test got very smeared (high bitrate does not help).

[Selur]

maybe the undershoot percentage, but I doubt it.

[Gravitator]

Unable encoding VP9 through constant guantizer.

[Gravitator]

x264 do not like the combination of settings.

[Selur]

still can't look at the attachments,...

[Magik Mark]

Selur

x265 is now on build 45. Does this warrant for new version of hybrid?

[Havokdan]

Strange, @Selur not posted the latest release on the topic, has this changelog:

rev 2016.08.27.1:
*fixed*
avisynth: srt subtitle styling
avisynth: dither mod4
input: ffmpeg timecode extraction
subtitle: tagging subtitle streams which get converted form sup->idx/sub
nvenc: ui glitch with cabac
some unicode file handling
decoder: mencoder->ffmpeg yuv444 handling
preview: color handling during filter preview and yuv444 content
*changed*
some file size checks during subtitle extraction
avisynth: filters updated
prores: interlaced signaling
cuda: 1060&1080 cards should be supported
*added*
avisynth: support save settings for each filter
nvenc: set target size (note: this is just an estimation, target size can still differ a lot since nvenc doesn't have 2pass encoding)
nvenc: support 10bit and lookahead feature
subtitle: 'add to all' option
x265: added support for --log2-max-poc-lsb, --discard-sei,--qpmin, --qpm

[Selur]

yeah, busy the last couple of weeks :)

[luigizaninoni]

any news re vapoursynth support ?

[Selur]

working on it when I have the time, see: http://forum.selur.de/topic1267-yes-im-working-on-vapoursynth-support.html (mainly working on the windows side atm.)

[luigizaninoni]

working on it when I have the time, see: http://forum.selur.de/topic1267-yes-im-working-on-vapoursynth-support.html (mainly working on the windows side atm.)

Great, looking forward to it, especially on linux, where nothing comparable exists

[Selur]

One of the road bumps atm. is that mplayer, mencoder don't support vpy scripts as input thus there's no av-preview. In my current dev versions I use a modified Vapoursynth Editor version (https://github.com/Selur/vapoursynth-editor) which adds IPC support that is needed to control the editor via Hybrid to use it for preview.

[Vesdaris]

Hmm..I dunno why but I never seem to make preview working. I've been ripping a dvd and I couldn't make preview work at all. I was getting just a blank screen. Are there any settings I can try changing specifically for preview\cropping preview?

PS Where can I set(and see) acceptable aspect rario error?

[Selur]

Hybrid doesn't calculate the aspect ratio error.
http://forum.selur.de/topic33-hidden-hybrid-options.html -> changing the vomode might help,...

[Vesdaris]

http://forum.selur.de/topic33-hidden-hybrid-options.html -> changing the vomode might help,...

I've changed it to directx and it fixed it! Thanx a lot ;)


Hybrid doesn't calculate the aspect ratio error.


How am I supposed to check it? It is kinda hard for me to determine whether the image gets stretched or not yet another person spots it the moment he sees it. That's why I always checked in megui so that AR error wasn't higher than 1% to be on the safe side.

PS Oh and is there something like meguis suggested resolution that changes depending on your height and mod?

[Magik Mark]

Selur

I new milestone has been released for x265

[Selur]

@Vesdaris: Hybrid will not hint at any aspect ratio error since I think it's silly. You specify a width or height and a modulo and Hybrid will select the other height/width to match both as good as possible. Hybrid will not start to suggest resolutions based on some wild guess work.
@MagikMark: only thing that isn't supported atm is ' --limit-TU' iirc. -> so no hurry :)

[Gravitator]

AVS preview does not give information about the frames I-P-B for x264/x265-10bit.

[Selur]

That should depend on the input filter used.

[Gravitator]

Also experiencing the problem with decoding > hotfuzz.vc1 (https://drive.google.com/open?id=0BzeF_1syecQwakhsX3RuZGhWcjA) (before playing it is desirable to pack the file through mkvtoolnix to .mkv).
AVS preview and the output is the result of a dirty after encoding.

[Selur]

Decoding you vc1 file with:
- DGDecNV, I have no problem and all frame types are shown
- using LibavSource decoding is fine, but no frame types are shown
- using FFVideoSource decoding is fine, but no frame types are shown (signals always I frame)
- FRIM doesn't work
=> whether Hybrid can show the frame type depends on the 'info'-option of the decoder used. Nothing I can do about it.

[Selur]

Hybrid rev 2016.12.23.1:

*fixed*

a bunch of small bugs

*added*

support for VCEEnc
NVEncC custom command line

*changed*

adjusted to latest x265 changes
updated detected NVIDIA cards

-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: Merry Christmas to you all!
For those that want to try the experimental Vapoursynth support on:
- Windows 64bit: download the Addon-File from the download page and extract it into your Hybrid folder
- Linux: use the repository, add the Hybrid Vapoursynth packages and add vapoursynth="true" to your misc.ini file.
(no Vapoursynth for 32bit Windows or Mac atm.)

General goals for the next release are:
a. further Vapoursynth stuff
b. updating VP9 encoding (just noticed that vp9 got some new features Hybrid doesn't support atm. :))

PPs.: This and future version will probably not work under Windows XP, since latest ffmpeg, mencoder, mplayer,.. do not support it.
Hybrid itself should still work.

[Jamaika]

b. updating VP9 encoding (just noticed that vp9 got some new features Hybrid doesn't support atm. :)
So, you failed to connect the codec VPX with AOM.;)

[Selur]

Didn't look into aom any more. :)
Will look into it once again when libav supports it and thus decoding in a normal player is possible.

[Magik Mark]

Selur,

The audio tab is missing "configuration save and apply" buttons

[Selur]

Not here,... the current Hybrid release has a bug which will remove all buttons when a fontSize is configured inside the misc.ini file, may be that is the problem for you.

[Havokdan]

Previously Hybrid let me use my GTX 670MX to convert videos to my notebook. Are you no longer compatible?

[Selur]

The "GeForce GTX 670MX" should still be detected as compatible card for nvenc, it that is not the case for you, what does the Hybrid log mention. My guess is for some reason the name inside the registry changed.
-> need details.

[Havokdan]

Well, it just only detects my Intel HD 4000 (Quick sync), it is not that version only, it has not detected since the last version, I use a Clevo P151 notebook, with Intel HD 4000 and Nvidia GTX 670MX. Older versions still work, much like StaxRip works, but the Hybrid does not, nor does it appear for selection.

Hybrid: homepage
Forum: public forum
Regarding problems please read: needed infos
Donate: via PayPal

Detected the following cards which seem to lack NVIDIA PureVideo support: Intel(R) HD Graphics 4000
Finished initialization, finished after 2.804s

[Selur]

Hybrid checks the registry 'Device Description' entries under "HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Video" to find the video cards available.
My guess is that for some reason your NVIDIA card isn't listed there or it is, but under a name that Hybrid doesn't recognize.
-> What entries are there on your system?

[Havokdan]

It has 3 entries. All Intel HD 4000. Maybe reinstall nvidia driver help me.
Well, I removed all the intel and nvidia drivers I had installed and with the native Windows 10 drivers, the Nvenc option reappeared.

[Selur]

Okay, so it seems like something messed up you registry entries for the video drivers. -> nothing to fix in Hybrid :D

[Magik Mark]

Not here,... the current Hybrid release has a bug which will remove all buttons when a fontSize is configured inside the misc.ini file, may be that is the problem for you.

You are right! My misc.ini setting are as ff:

fontSize=15
toolTipFontSize=15

However, I don't have this problem in previous version. Is there another way to enlarge the window of hybrid? I have 60" panel and the default window size is really small. It's really hard to see

[Selur]

Fixed the issue in my dev version, will send you a link for testing.

[Magik Mark]

Thanks that did it!

I was just wondering about the ff:

1. Why queuing entries are deleted when you restart hybrid?
2. Why can't the user change the settings / preset once the video has been moved job queue ?

Is this a limitation of the software used or your mere preference?

Thanks a lot to all your work!

[Havokdan]

Is it possible to include in the first tab, somewhere near the box in which we chose the input file, an option to choose the input method? Something like Avisynth and its various options (L-smash source, ffmpeg2source and others) As well as those who do not use avisynth.

[Selur]

Why queuing entries are deleted when you restart hybrid?
for me the entries inside the queue do not get removed.

Why can't the user change the settings / preset once the video has been moved job queue ?
Simply because the entries inside the queue do not contain enough data to reconstruct the gui state and all the settings.

Is this a limitation of the software used or your mere preference?
This is how I decided to design the software,...

Is it possible to include in the first tab, somewhere near the box in which we chose the input file, an option to choose the input method? Something like Avisynth and its various options (L-smash source, ffmpeg2source and others) As well as those who do not use avisynth.
No.
a. Hybrid selects the method to handle the input based on a best guess and restriction model, this will not change. Especially since I don't see any useful case for this.
b. You want to select from mencoder, mplayer, ffmpeg, all avisynth source filters and all vapoursynth source filters and then complain if you selected a filter that isn't fit for your input?
c. since avisynth and vapoursynth are optional adding them to the main tab would just bloat the size of the tab.

Cu Selur

[Havokdan]

No.
a. Hybrid selects the method to handle the input based on a best guess and restriction model, this will not change. Especially since I don't see any useful case for this.
b. You want to select from mencoder, mplayer, ffmpeg, all avisynth source filters and all vapoursynth source filters and then complain if you selected a filter that isn't fit for your input?
c. since avisynth and vapoursynth are optional adding them to the main tab would just bloat the size of the tab.

The problem is that sometimes for the conversion to be successful, I need to use a certain combination to open the files, as I see it is possible with the Hybrid, but the options are bad, so much so that I only discovered them after you answered a Comment here from forum, in StaxRip for example is easier. But thanks for the answer.

[Selur]

Then use StaxRip or if you have some criteria which allows to identify those sources that require a specific input method share that info and I can look into using that selection method in Hybrid.

Cu Selur

[Havokdan]

Then use StaxRip or if you have some criteria which allows to identify those sources that require a specific input method share that info and I can look into using that selection method in Hybrid.

Cu Selur

I do not have any files now with the problem, after much breaking my head, discover that it was a problem with colors (colorspaces), in the opportunity StaxRip offers when opening a file a certain combination of "source filters", in which the The only one that opened the video was this combination highlighted in the image below:

https://i.imgur.com/jT3tQlK.png

If the image did not appear, it was using the source filter to use such a LWLibAVVideoSource, either with avisynth+ or vapoursynth in StaxRip, at the time did not know where to make that change in the Hybrid.

[Selur]

Something like that will probably not happen, sorry. But like I said if you come up with a way to identify problematic files that require a specific filter, let me know.

[alter4]

Hi Selur, downloaded recent version, open video file and got error "x265Model doSet->found no field: optionalHrdSignalling" when i tried to apply x265 preset. OS windows 10 64bit.

--Thanks and happy new year.

[Selur]

you can simply ignore those errors, deleted two parameters that weren't used anymore.
(will probably release a new version on Saturday or Sunday, which will fix this and some other minor stuff, waiting for feedback from a few users regarding some bug fixes)

[alter4]

>>will probably release a new version on Saturday or Sunday
You probably really are one of the best Santa I know :-)

>>you can simply ignore those errors,
Got it, thanks

[Emulgator]

Hi Selur, first off: Many thanks for your efforts !
Starting from Hybrid 2014, then 2015, early version, now recent version 2016.12.23.0
I sometimes try to run your big toolbox on a Win7U64, which all updates applied.
Never got past this Win7 crash screen:
"Hybrid Encoder GUI funktioniert nicht mehr"

Problemsignatur:
Problemereignisname: APPCRASH
Anwendungsname: Hybrid.exe
Anwendungsversion: 2016.12.23.0
Anwendungszeitstempel: 585d6abc
Fehlermodulname: Qt5Core.dll
Fehlermodulversion: 5.7.0.0
Fehlermodulzeitstempel: 575a6ab1
Ausnahmecode: c00000fd
Ausnahmeoffset: 00000000000aedfa
Betriebsystemversion: 6.1.7601.2.1.0.256.1
Gebietsschema-ID: 1031
Zusatzinformation 1: f88f
Zusatzinformation 2: f88ff1916e94783332ccfa2e559abd45
Zusatzinformation 3: efc5
Zusatzinformation 4: efc599c078a28f4164978e345401d462
...

The mentioned previous Hybrid version I tried also crashed on Qt5core.dll, IIRC.

I wonder that after installing Hybrid I see 2 locations of msvcp/msvcr.dlls and Qt.dlls,
one is "Hybrid" folder and the other is "dynamic" within "Hybrid".
Additionally the Qt.dlls have different versions (2015 in "dynamic" vs.2016 in "Hybrid").
Since other users happily report success with Hybrid I am curious to find out where it borks on my side.

I started with my chosen install path C:/_PROG.
Startup.Crash.Uninstall..Checked that the "Hybrid" folder is completely empty.
Reinstall with suggested install path C:/Programme,
leaving everything at default setting.
Startup.Crash.Uninstall.Checked that the "Hybrid" folder is completely empty.
Reinstall with suggested install path C:/Programme,
letting reset everything to default on install
but keep getting the same behaviour: Startup.Crash.

[digitall.h]

Thank you Selur for an updated Hybrid version:
I hope you had Merry Christmas and wish you a very good New Year 2017
:thanks:

Using Linux64bit 2016.12.23.1 version, when Hybrid starts up it does not show the image (Hybrid logo) it used, just a white square showing files verified.
And icons for Bluray input and DVD input, in main tab, are also lost. Is this supposed to be?

When you say:
Hybrid rev 2016.12.23.1:
...
For those that want to try the experimental Vapoursynth support on:
- Windows 64bit: download the Addon-File from the download page and extract it into your Hybrid folder
- Linux: use the repository, add the Hybrid Vapoursynth packages and add vapoursynth="true" to your misc.ini file.
(no Vapoursynth for 32bit Windows or Mac atm.)
...
are you referring to Linux repository?.
My OS is Debian and sorrily there are neither Hybrid nor Vapoursynth packages...
Or do you refer to another repository?

[Selur]

@digtal.h:
about the missing splashscreen at start up:
- I know tested some stuff and forgot to reenable it :)
about the missing icons inside Hybrid:
- this happens due to a bug (which I fixed locally), but it causes Hybrid to kill the icons if a custom font size was specified inside the misc.ini (removing the font size setting from your misc.ini should fix the problem, till the next release)
about the repository:
- I'm referring to the Ubuntu Linux repository that is linked on the download page (under Hybrid packages by others) maintained by djcj

@Emulgator:
about the different Qt versions:
- the 32bit qt binaries often use a different version (dynamic = 32bit versions)
about the crash:
- make sure to uninstall Hyrid and it's config files.

Cu Selur

Ps.: Planned to release a new version today, but this isn't going to happen since I wait for some feedback from users that reported a bug. (Main problem atm. is that Hybrid seems to crash on Macbook Pro with the new touch bar; can't reproduce the issue here, since I don't own such a device)

[digitall.h]

@digtal.h:
...
about the repository:
- I'm referring to the Ubuntu Linux repository that is linked on the download page (under Hybrid packages by others) maintained by djcj


Thank you.
Ubuntu ppa... in Debian it may work, or not. I may take a look.
It may also help to know which packages we need to have installed in the distribution, and some configurations or adjustments needed, so everybody can have a system working with Hybrid and Vapoursynth independently of the distro.

[Selur]

Best ask djcj about what packages he uses etc. (personally I lost track of all the changes :))

[Magik Mark]

Selur

How do I make sure that only the ff are encoded:

1. Video
2. Audio
3. Chapters

After running mkvtoolnix I always see the word "tags". I want that remove. Which in the settings I could do that?

[Selur]

In relation to tags Hybrid by default:
a. passes through the Audio/Video language and title tags. (can be disabled under: Config->Automation->Passthrough Tags)
b. always sets the global 'Encoding Gui' tag to the Hybrid version info in example 'Hybrid 2016.12.23.1' (can't be disabled)

Since I'm not sure what Hybrid does that isn't to your liking can you be more specific about what you don't like that Hybrid does?

Cu Selur

[Selur]

Hybrid rev 2017.01.01.1:

*fixed*

x265: optionalhrdSignaling and optionalHrdListSignaling popup removed
gui: font pixel size scaling via misc.ini
avisynth: aWarpSharp problem
VP9: support for --color-space, --min-gf-interval, --max-gf-interval, --gf-cbr-boost
DivX265: not shown on Linux system

*added*

x265: --[no-]multi-pass-opt-distortion, --[no-]multi-pass-opt-analysis, --[no-]opt-cu-delta-qp, --[no-]ssim-rd, --[no-]aq-motion support
ffnvenc: exprimental support for nvenc through ffmpeg

*changed*

switched to Qt 5.7.1 for Windows 64bit and Mac OS X builds

-> downloads: http://www.selur.de/downloads

Cu Selur

[Magik Mark]

Selur,

Have you tried the new x265 option? It crashes!

[Selur]

the new x265 option? Which there are multiple --[no-]multi-pass-opt-distortion, --[no-]multi-pass-opt-analysis, --[no-]opt-cu-delta-qp, --[no-]ssim-rd, --[no-]aq-motion support,....
Tested them yesterday and they seemed to work here fine.
-> a debug output would be nice :)

[Magik Mark]

This is the error I'm getting:


Creating job for current source
-> Creating jobs for source 1,...
-> Generating calls for: C:\Users\Magik Mark\Desktop\Reencoded x265\Modern Family S08E09.mkv
adding x265 calls for source: 1
createJobs for C:\Users\Magik Mark\Desktop\Reencoded x265\Modern Family S08E09.mkv
creating attachment extraction subjobs
creating video subjobs
creating muxing subjobs
optimizing the subJobs
-> finished creating subjobs for current source
added new job with id 13_29_41_3810
Starting Main@13:29:41.619: "C:\PROGRA~1\Hybrid\x265.exe" --input - --output-depth 10 --y4m --profile main422-10 --limit-modes --no-open-gop --lookahead-slices 0 --pass 1 --no-slow-firstpass --bitrate 1500 --cbqpoffs -2 --crqpoffs -2 --ssim-rd --psy-rd 2.50 --rdoq-level 2 --psy-rdoq 15.00 --aq-mode 3 --cu-lossless --range limited --colormatrix bt709 --stats "C:\Users\MAGIKM~1\Desktop\HYBRID~1\Modern Family S08E09_13_29_41_3810_02.stats" --multi-pass-opt-rps --multi-pass-opt-analysis --multi-pass-opt-distortion --output NUL
y4m [info]: 1920x1080 fps 24000/1001 i420p10 unknown frame count
raw [info]: output file: NUL
x265 [info]: HEVC encoder version 2.2+17-a2fe29ca5c6c
x265 [info]: build info [Windows][GCC 6.2.0][64 bit] 10bit
x265 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX AVX2 FMA3 LZCNT BMI2
x265 [warning]: --cu-lossless disabled, requires --rdlevel 3 or higher
x265 [info]: Main 10 profile, Level-4 (Main tier)
x265 [info]: Thread pool created using 28 threads
x265 [info]: Slices : 1
x265 [info]: frame threads / pool features : 5 / wpp(17 rows)
x265 [error]: Analysis 2 pass: failed to open file x265_analysis.dat.temp
x265 [error]: failed to open encoder
process finished with exitcode: 2 and exitStatue: 0
finished after 00:00:24.854
-> 13_29_41_3810_02_video crashed:ERROR: C:\Users\MAGIKM~1\Desktop\HYBRID~1\Modern Family S08E09_13_29_41_3810_02.stats was not created!
Job 13_29_41_3810 finished!

[Ma]

Starting Main@13:29:41.619: "C:\PROGRA~1\Hybrid\x265.exe" --input - --output-depth 10 --y4m --profile main422-10 --limit-modes --no-open-gop --lookahead-slices 0 --pass 1 --no-slow-firstpass --bitrate 1500 --cbqpoffs -2 --crqpoffs -2 --ssim-rd --psy-rd 2.50 --rdoq-level 2 --psy-rdoq 15.00 --aq-mode 3 --cu-lossless --range limited --colormatrix bt709 --stats "C:\Users\MAGIKM~1\Desktop\HYBRID~1\Modern Family S08E09_13_29_41_3810_02.stats" --multi-pass-opt-rps --multi-pass-opt-analysis --multi-pass-opt-distortion --output NUL

Maybe your default directory is without write permission. You can try to add option:
--analysis-file "C:\Users\MAGIKM~1\Desktop\HYBRID~1\Modern Family S08E09_13_29_41_3810_02.analysis"

[Magik Mark]

How am I suppose to that in Hybrid?

[Ma]

How am I suppose to that in Hybrid?

I'm not Hybrid user, but from screenshots (based on x264):
x265 -> Misc -> Command line addition:
------------------
After installation of Hybrid -- you can write any option(s) in the box:
x265 -> Misc -> Custom command line addition:

[Selur]

Seems like I missed '--analysis-mode <string|int>' and '--analysis-file <filename> :)

[alter4]

Does it means new build is necessary to fix this?

[Selur]

You could simply disable the 'Analysis refinement' and 'QP refinement' and thus not use the option, other than that a new build would be required. (fixed it locally, but busy with another bug atm.)

[DJ-1]

Hi Selur

Have been sitting on a older version, as it `works nicely`...

Ive been doing lots of QTGMC + lossless / sourcematch etc + Nnedi3 on interlaced pal stuff no probs

i get a progressive source, set input type to 1, and crash.
maybe a silly question, is lossless not meant to be used with progrssive? because when i disable lossless mode it converts `input type 1` fine.

...i tried the latest Hybrid installer on my whs,.. same job setup
crash
debug:https://www.dropbox.com/s/ljx14afi3h1alo0/HybridDebugOutput%20Lossless%202.txt?dl=0

QTGMC preset - custom
select `tried all 3 settings`
input type 1
shapness 0.2
lossless `before resharpening`
final temporal smoothing `1`
source matching `2x refined`
ezkeepgrain 1.00
noise preset - fast
edi threads -1


thanks.

[Selur]

Does the preview work?

LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\LoadDll.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\DGDecode.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\AddGrainC.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\dfttest.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\DirectShowSource.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\EEDI2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\eedi3.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\FFT3DFilter.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\mt_masktools-26.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\mvtools2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\nnedi.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\nnedi2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\SSE2Tools.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\TDeint.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\VerticalCleanerSSE2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\nnedi3.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\RepairSSE2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\RemoveGrainSSE2.dll")
LoadCPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\yadif.dll")
LoadDll("C:\PROGRA~1\Hybrid\AVISYN~1\fftw3.dll")
LoadDll("C:\PROGRA~1\Hybrid\AVISYN~1\LIBFFT~1.DLL")
Import("C:\Program Files\Hybrid\avisynthPlugins\QTGMC-3.33.avsi")
# input luminance scale tv
MPEG2Source(d2v="C:\Users\Administrator\AppData\Local\Temp\2\19_32_45_4910.d2v")
# deinterlacing
AssumeTFF()
QTGMC(Preset="Placebo", InputType=1, TR2=3, Sharpness=0.2, SourceMatch=3, Lossless=2, EZKeepGrain=1.00, NoisePreset="Fast", ediThreads=1)
# filtering
return last does seem fine to me, don't see anything that conflicts with http://avisynth.nl/index.php/QTGMC.
Maybe Avisynth is running out-of-memory,..

[DJ-1]

didnt think to check,
preview ok
preview (filter box checked) tries but window closes (no error)

avisynthprview (same if filter box checked or not)

1st hybrid pop-up saying
avisynth preview error
lossless modes incompatible with inputtype=1
couldnt import re incompatible with inputtype=1

plus errors referencing the script that gets used...
https://www.dropbox.com/s/kir2q2qsa294xta/HybridDebugOutput%20preview%20errro.txt?dl=0


ta

[Selur]

-> "lossless modes incompatible with inputtype=1"
=> I will adjust Hybrid accordingly to make sure this doesn't happen with the next release, but until then: don't use input type 1 with lossless :)

[DJ-1]

cheers, yeah did look at the avisynth page earlier couldnt see anything that says they cant be used together.

was only when i tried the preview, i see Hybrid thinks otherwise...

i was thinking it was my global profile initially...checking that.

cheers.

[Selur]

It's not Hybrid that thinks that way. The error message comes from the QTGMC script!
2017.01.02 - 20:20:00_Windows 7-64bit_2017.01.01.1 - level 4: Avisynth preview error: Lossless modes are incompatible with InputType=1
2017.01.02 - 20:20:02_Windows 7-64bit_2017.01.01.1 - level 4: Avisynth preview error: (C:\Program Files\Hybrid\avisynthPlugins\QTGMC-3.33.avsi, line 938)
2017.01.02 - 20:20:03_Windows 7-64bit_2017.01.01.1 - level 4: Avisynth preview error: (C:\Program Files\Hybrid\avisynthPlugins\QTGMC-3.33.avsi, line 609)
here's line 934+:
# Insert the source lines into the result to create a true lossless output. However, the other lines in the result have had considerable processing and won't
# exactly match source lines. There will be some slight residual combing. Use vertical medians to clean a little of this away
function QTGMC_MakeLossless( clip Input, clip Source, int InputType )
{
Assert( InputType != 1, "Lossless modes are incompatible with InputType=1" )

# >>>> YUY2: 'Input' is planar, 'Source' is interleaved (changed to planar here for processing) - returns planar result

# Weave the source fields and the "new" fields that have generated in the input
srcFields1 = (InputType == 0) ? Source.SeparateFields() : Source.SeparateFields().SelectEvery( 4, 0,3 )
srcFields = Source.IsYUY2() ? srcFields1.interleaved2planar() : srcFields1
newFields = Input.SeparateFields().SelectEvery( 4, 1,2 )
processed = Interleave( srcFields, newFields ).SelectEvery(4, 0,1,3,2 ).Weave()

# Clean some of the artefacts caused by the above - creating a second version of the "new" fields
vertMedian = processed.VerticalCleaner( mode=1, planar=true )
vertMedDiff = mt_makediff( Isyuy2(Source) ? processed.planar2interleaved().ConvertToYV16() : \
processed, Isyuy2(Source) ? vertMedian.planar2interleaved().ConvertToYV16() : vertMedian, U=3,V=3 )
vertMedDiff = Isyv16(vertMedDiff) && Isyuy2(Source) ? vertMedDiff.ConvertToYUY2().interleaved2planar() : vertMedDiff
vmNewDiff1 = vertMedDiff.SeparateFields().SelectEvery( 4, 1,2 )
vmNewDiff2 = vmNewDiff1.VerticalCleaner( mode=1, planar=true )
vmNewDiff2 = Isyuy2(vmNewDiff2) ? vmNewDiff2.planar2interleaved().ConvertToYV16().mt_lutxy( vmNewDiff1.planar2interleaved().ConvertToYV16(), \
"x 128 - y 128 - * 0 < 128 x 128 - abs y 128 - abs < x y ? ?", U=3,V=3 ) : \
vmNewDiff2.mt_lutxy( vmNewDiff1, "x 128 - y 128 - * 0 < 128 x 128 - abs y 128 - abs < x y ? ?", U=3,V=3 )
vmNewDiff2 = Isyv16(vmNewDiff2) && Isyuy2(Source) ? vmNewDiff2.ConvertToYUY2().interleaved2planar() : vmNewDiff2
vmNewDiff3 = vmNewDiff2.Repair( vmNewDiff2.RemoveGrain( 2, planar=true ), 1, planar=true )

# Reweave final result
newfd = Isyuy2(Source) ? newFields.planar2interleaved().ConvertToYV16().mt_makediff( vmNewDiff3.planar2interleaved().ConvertToYV16(), U=3,V=3 ).ConvertToYUY2().interleaved2planar() : \
newFields.mt_makediff( vmNewDiff3, U=3,V=3 )
return Interleave( srcFields, newfd).SelectEvery( 4, 0,1,3,2 ).Weave()
}

=> I simply wasn't aware of the restriction of QTGMC and thus didn't hinder the user from using inputtype=1 and lossless together :)

[DJ-1]

ahh ok, cheers

[Jamaika]

Problem with converter subtitles mp4 to ass

I have ffmpeg 2017 and can't convert files.
When I use alignment an4 in mov_text this I have information in ffmpeg:
<TextSampleDescription horizontalJustification="left" verticalJustification="center" backColor="0 0 0 0" verticalText="no" fillTextRegion="no" continuousKaraoke="no" scroll="None">

ffmpeg.exe -sub_charenc utf-8 -i input.mp4 -f ass -c:s ass output.ass
[mov_text @ 000001d506ec1660] Unable to recode subtitle event "?Đ..............."

[Selur]

and this is related to Hybrid: How? (don't think Hybrid would try to extract subtitle from an .mp4 file as ass,...)

[Jamaika]

I would not bother you, but ffmpeg begins to read some information from movtext. There is some progress.

convert file mov_text to
<TextSampleDescription horizontalJustification="left" verticalJustification="center" backColor="0 0 0 0" verticalText="no" fillTextRegion="no" continuousKaraoke="no" scroll="None">
<FontTable>
<FontTableEntry fontName="Courier New" fontID="1"/>
</FontTable>
<TextBox top="0" left="0" bottom="1080" right="1920"/>
<Style styles="Normal" fontID="1" fontSize="40" color="ff ff ff ff"/>

file ASS
[V4+ Styles]
Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
Style: Default,Courier New,40,&Hffffff,&Hffffff,&H0,&H0,0,0,0,0,100,100,0,0,1,1,0,4,10,10,10,0

Unfortunately adds some unnecessary characters into text.:(

[Selur]

a. don't see a problem with what you posted
b. still don't see why you posted this inside the Hybrid thread, since this has nothing to do with Hybrid (Hybrid doesn't use ffmpeg for subtitle conversions,..)

[Vesdaris]

Hey Selur.

I was hoping if you could help me out with something.

I've found out that when I try to screenrecord any video (youtube or other souces) playing on my screen (refresh rate 60) at their original fps, say, 23.976 or 25 ) then my recorded file gets some kind of micro-stutter as if frames are dropped.

Sometimes it starts happening from the very beginning(I can observe it real time via OBS preview), sometimes it happens after x minutes of recording. CPU usage, settings(except one , see below) and whatnot do not affect this issue.

However, recording at 60 fps (same as my refresh rate) fixes this issue.

I read somewhere that it has something to do with me trying to screenrecord 23.976\25 fps video on my 60 refresh rate monitor and that to complete avoid this from happening I need to set my refresh rate to, say, 72(seeing as you can't force 24 or 25 refresh rate) for instance(for 23.976 material). Sadly my monitor only works at 60 refresh rate and can't be overclocked.

I was thinking about a workaround..recording at 60 fps using lossless mode(yeah I get it, lossless for a screencapture lol) and then somehow converting it to the original fps during a final encoding.

How could I do it in hybrid ? I presume I need to use some avisynth script in "Frame" settings? What should I use? Would the final file be stuttery or something seeing as the original frame rate is 25 fps not 60?


Thanx a lot.

PS I read that virtualdub supposedly has some plgins that can do

analyzing sequential frames and removing duplicates, or blending frames / fields in order to reach the lower fps without audio/video desync.

but I have never worked with it. Is it possible to do something like this in hybrid "frame" settings?

[Vesdaris]

Hybrid crashes when I try to encode PCM audio from a certain video file to either AAC or AC3

It's a large,80Gb, file (video codec - MAGY, audio codec- PCM

Here is my log:

http://pastebin.com/pa2jV81s

Any ideas on what might be wrong?

Cheers.

[Selur]

@Vesdaris:
Regarding changing the frame rate from 60 to X: Try Avisynth->Frame->Restore->sRestore
Regarding the crash: I think I know why this is happening. (a small bug I fixed a few days ago, will send you a link to a dev version for testing)

[Vesdaris]

The dev version works :) Thank you.

[Selur]

Hybrid rev 2017.01.07.1:

*fixed*

ffnvenc: switching between H.264&H.265 doesn't switch the profiles (they are stuck at H.265s profiles)
x265: x265->Main->General Settings->Analysis Refinement and x265->Main->General Settings->QP Refinement require the setting of a analysis file otherwise they will cause a crash
sox: crash when running multiple sox instances in parallel (seems like sox is trying to create the same temp file, now creating, and later deleting, a separate sub-folder for each sox instance)
avisynth: removed 'disableDistributor', added 'addDistributor'
avisynth: changed Hybrid default QTGMC ediThread count from -1 to 2.
avisynth: nnedi3- and ResizeHQ-resizer call fixed

-> downloads: http://www.selur.de/downloads

Cu Selur

[Havokdan]

Just a question, I have a file that mediainfo informs that it has 120 fps, what option should I check to reduce fps?

[Selur]

Assuming you are on a Windows System your best option is probably Avisynth->Frame->Restore->sRestore.

[Havokdan]

Thanks, i will try.

[Magik Mark]

Hy Selur!

Thanks for the new build. We may Have some issues on the ff:

1. Multiple redundant x265 CLI entries on "Analysis & QP Refinement" and "Save RPS"

http://i.imgur.com/4Hz0GuZ.png

2. When "Motion based adaptive Quantization" is ticked it disables it. If unticked it activates it. Isn't suppose to be the other way around?

http://i.imgur.com/xDsVzX0.png

[Selur]

argh,...
- the double entry only happens when 'fast 1st pass is disabled' :)
- for aq-motion I accidentally added a 'not' thus enabling and disabling the option are switched :)
-> will fix (-> fixed locally, send you a link to a dev version)

[Magik Mark]

Hi Selur,

1. The default audio is greyed out. I can't set my audio as default

http://i.imgur.com/e58I0Ow.png

2. In order to use "--multi-pass-opt-analysis or --multi-pass-opt-distortion", "cu tree" has to be turned off. Where can I find "cu tree"

[Selur]

regarding 1.: you already set the audio stream that is inside the audio to be flagged as default for an mkv container,... you can't flag multiple streams as default
regarding 2.: Good to know. 'cu tree' = Coding unit tree -> disable "x265->Quantization->Coding unit tree for adaptive quantization' and Hybrid should add '--no-cutree' to the command line

[Magik Mark]

Thanks Selur!

I was even told not to activate "multi-pass-opt* option together with analysis load or save". So, the encoder has to choose any one of them. Otherwise, the app will just deactivate it

[Magik Mark]

Some mkv has fonts in them to be used by the subtitle. How do I view, change or even delete this?

[Selur]

I was even told not to activate "multi-pass-opt* option together with analysis load or save". So, the encoder has to choose any one of them. Otherwise, the app will just deactivate it
Any documentation for this available?

Some mkv has fonts in them to be used by the subtitle. How do I view, change or even delete this?
Hybrid by default automatically passes through all attached fonts, there is no option to manage the attachments aside form disabling 'Config->Automation->Matroska Settings->On load extract fonts from mkv'.
Hybrid extracts fonts to 'Config->Path->Infos->Font path'.

=> if you want to change fonts use a tool meant for that. :) (iirc. Hybrids font selection options only support switching the first font when hard coding the font)

Cu Selur

[Vesdaris]

For some reason I keep getting hex in my video file despite umh being set in x264 settings.

Here is the log

http://rgho.st/6qTQ6TN4P
Am I doing something wrong?

[Selur]

hex? I guess you are not speaking of hexagonal search,.. wild guess: might be related to the cutting,.... whatever the problem really is,.. may be enforcing Avisynth + AviSource decoding helps,...
-

[Vesdaris]

Yeah, I get hexagonal instead of multihexagonal for some reason(slow preset). I've tried setting always use avisynth but I stil get hex instead of umh.

Very Slow preset- fine
Slower- fine.

Slow preset- I always get hex instead of multi-hex.HYbrid shows that it should be multi-hexagonal.

I think the preset might be bugged.

[Selur]

Still can't really follow you.

Looking at the debug output:
... | "K:\Hybrid\x264-10bit.exe" --preset slow --crf 18.00 --profile high10 --level 4.1 --bframes 10 --rc-lookahead 60 --psy-rd 1.00:0.15 --vbv-maxrate 150000 --vbv-bufsize 187500 --sar 1:1 --deblock -1:-1 --non-deterministic --colormatrix bt709 --demuxer raw --input-depth 10 --input-csp i420 --fps 25 --input-res 1280x720 --output "O:\000000~1\00_36_09_0510_01.264" -
is used and '--slow' is the same as:
--b-adapt 2 --direct auto --me umh --rc-lookahead 50 --ref 5 --subme 8
where '--rc-lookahead 50' will be overwritten by '--rc-lookahead 60',...

Disabling the 'Minimize command line' option in Hybrid, I also see '--me umh' when I apply the 'slow' preset.

So, what makes you thing that you there is a bug and Hybrid uses 'hex' instead of 'umh' as '--me'-value?


Cu Selur

[Vesdaris]

Selur
So, what makes you thing that you there is a bug and Hybrid uses 'hex' instead of 'umh' as '--me'-value?

Mediainfo reports that me value is hex unlike very slow and slower preset.

[Selur]

Okay, than this isn't a bug in Hybrid. :)
It's either:
a. a bug in x264 reporting the wrong info or stating the wrong info inside their help output
or
b. MediaInfo misreading the info
-> you might want to contact the developers of MediaInfo and x264 :)

Cu Selur

[Vesdaris]

Any other encoding tool I used (handbrake,megui) created a file that had me=umh as specified.

So, I've copied x264 binaries from megui into hybrids folder and made some test runs.

slow preset(me=umh)

x264-10bit(32) - me=hex
x264-10bit(64) -me=hex

Hybrids original x264-10bit - me=hex (same with 8bit version)

Megui's encoding using the same x264 binaries results in a file being encoded at me=umh as specified.

So, I dunno what to think about all this..

[Selur]

No clue either. I showed you the call Hybrid uses.
The call looks fine to me, so I don't think it's Hybrids fault.

[Magik Mark]

Selur,

After encoding, I see an entry for "Global Tag". How do I disable this during encoding?

[Selur]

By default Hybrid passes through audio&video language and title tags, which can be disabled under 'Config->Automation->Passthrough tags'.
Those tags can be overwritten, through the tags tab.
For mkv there are additional tagging options under 'Config->Output->MKV Tags', which by default do nothing.
If that doesn't help, you need to share details, since it's just wild guessing for me otherwise.

Cu Selur

[Magik Mark]

By default Hybrid passes through audio&video language and title tags, which can be disabled under 'Config->Automation->Passthrough tags'.
Those tags can be overwritten, through the tags tab.
For mkv there are additional tagging options under 'Config->Output->MKV Tags', which by default do nothing.
If that doesn't help, you need to share details, since it's just wild guessing for me otherwise.

Cu Selur

This is what I'm talking about:

http://i.imgur.com/A03gXAP.png

Is there a way to automatically remove the "Global Entry Tag"? I can't find the right setting

[Magik Mark]

Selur,

When changing the size of the window through the "ini" file some area are truncated:

http://i.imgur.com/5Oc5pcP.png

Is there a way to automatically adjust this?

[Selur]

about the Global tag entry: If that file was created with Hybrid, it might be the 'Copyright Hybrid VERSIONINFO' data Hybrid always sets when outputting mkv.
about the progress information: you could remove some of the information inside the progress bar (Config->View->Progressbar Main), other than that it might be possible to adjust the font size of just the progress bar using a custom style sheet (http://forum.selur.de/topic255-hybrid-and-styling.html) other than that I would probably have to write additional code for it to allow to set a custom font size just for the progressbar. (will also look if there is a way to tell Hybrid to automatically adjust the size, but I doubt that is possible,..)

-> just had another idea, seeing your screenshot the icons/buttons seem to be too large decreasing their size might already help (send you a link to a dev version)

[Magik Mark]

Tried the file you sent.

1. Fonts became too small
2. x265 -> "Save Rps" is now greyed out. Can't be ticked

[Selur]

regarding 1: then change the setting in the misc.ini
regarding 2: you did disable cuTree (x265->Quantization->General Quantization->Coding tree for adaptive quantization, right? remember https://forum.doom9.org/showthread.php?p=1792735#post1792735)

[Selur]

Icons have the size they are meant to have.
about 'Save RPS' read my previous post.

[Magik Mark]

Thanks Selur!

[Magik Mark]

about the Global tag entry: If that file was created with Hybrid, it might be the 'Copyright Hybrid VERSIONINFO' data Hybrid always sets when outputting mkv.

Is there a way to turn this off?

[Selur]

No, there is not. :D Any reason why that is a problem?

[Magik Mark]

No, there is not. :D Any reason why that is a problem?



Not actually a problem. But I tinker my mkv files a lot using mkvtoolnix. I prefer to have video, audio, subtitles and chapters content only. I want it to be clean and free from other entries that are not needed in playing videos


Sent from my iPhone using Tapatalk

[Selur]

Only thing Hybrid does (just checked) is setting:
Encoding Gui : Hybrid VERSION_STRING
If you don't like it you could use mkvtoolnix or ffmpeg remove it, but I don't see a reason to remove this. :)

Cu Selur

[Magik Mark]

https://forum.doom9.org/showthread.php?p=1792735#post1792735[/url])

The way I understand it --no-cutree are for:

1. --multi-pass-opt-analysis
2. --multi-pass-opt-distortion

cutree can be turned ON for --multi-pass-opt-rps

Tried it in staxrip and it's working perfectly

[Selur]

not in Hybrid atm. :)

[Magik Mark]

1.Tried the test file. It doesn't add "--multi-pass-opt-rps" CLI even if I tick "Save RPS"
2. "--analysis file" is also missing

[Selur]

You are using 2pass encoding, right?

[Magik Mark]

Correct 2pass!

Tried the file same problem #3127

[Selur]

iirc analysis file should only be needed for--multi-pass-opt-analysis and --multi-pass-opt-distortion not for --multi-pass-opt-rps,..

[Magik Mark]

The latest test file still doesn't add "--multi-pass-opt-rps" CLI even if I tick "Save RPS"

[Magik Mark]

Thanks that did it!

[Magik Mark]

Selur, we have an error when "Analysis refinement is ticked":

http://i.imgur.com/Mo0K8Vp.png

cutree is off

[Selur]

seems like I accidentally removed the analysis file part,... argh,..

[Magik Mark]

Selur

Is hybrid's "limited sharpen" the same as "limited sharpen faster"?

Tried it. Encoding 50% slower compared to staxrip

[Selur]

Yes, limited sharpen in Hybrid uses 'LimitedSharpenFaster', if you compare the avisynth scripts and the encoding settings you can probably figure out why in your case StaxRip is faster,...

[Havokdan]

I am not an expert, but StaxRip uses Avisynth+ and The Vapoursynth, both 64-bit, would not be the case of the plugin run in 64-bit in StaxRip and in 32 bits in the Hybrid?

[Selur]

I my tests so far Vapoursynth (64bit) offers better multithreading than Avisynth MT (32bit), but is slower per thread, so it only delivered faster output when you use 2+ CPUs.
Couldn't find any speed gain in using Avisynth+.
-> If someone has different experiences please share.

[Magik Mark]

Yes, limited sharpen in Hybrid uses 'LimitedSharpenFaster', if you compare the avisynth scripts and the encoding settings you can probably figure out why in your case StaxRip is faster,...

The reason why it's a lot faster in staxrip is because it uses "Limited Sharpen Faster Mod" (LSFMod)

Maybe we can use this in hybrid as well?

[Selur]

I could probably add support for it, but that would take a while since it has tons or parameters:
LSFmod (clip input, int "strength", int "Smode", int "Smethod", int "kernel", string "preblur", bool "secure", clip "source", int "Szrp", int "Spwr", int "SdmpLo", int "SdmpHi", int "Lmode", int "overshoot", int "undershoot", int "overshoot2", int "undershoot2", int "soft", bool "soothe", int "keep", int "edgemode", bool "edgemaskHQ", float "ss_x", float "ss_y", int "dest_x", int "dest_y", bool "show", int "screenW", int "screenH", string "defaults")

Cu Selur

[Magik Mark]

Thanks for this.

There are preset for "old setting", "slow" and "fast". These helps a lot in choosing your preference. Is it still possible to integrate this?

[Magik Mark]

Selur,

Can you guide me on the ff:
1. Convert 2160p -> 1080p
2. What steps should I take in the resizer tab
3. Which among the resizer should I take bicubic? Your personal preference?
4. Convert color to bt709

[Selur]

There are preset for "old setting", "slow" and "fast". These helps a lot in choosing your preference. Is it still possible to integrate this?
if I find the time, I'll look into it.

There are preset for "old setting", "slow" and "fast". These helps a lot in choosing your preference. Is it still possible to integrate this?
1. enable Resize, set 'Target resolution' width to 1920. (Height should be automatically set to 1080)
2. select the Resizer you want (here, alternatively enable Filtering->Avisynth->Line->Resizer and select a resizer there)
Which among the resizer should I take bicubic? Your personal preference?
personally I like Bicubic and GaussResize

Convert color to bt709
What color matrix does your source use? Not sure whether Avisynth even offers a way to convert in example bt2020 properly to bt709. iirc dither supports "2020" (2020nc to bt709) conversions,..

[Ajvar]

[BugReport]
Hello again! Last time I used Hybrid was a 2015 version but now I deleted old version and installed new one however I found a frustration with this bug:
Whenever I launch Hybrid for the second time I have notebook completely freeze with an error message "Nvidia GPU 660M is not a removable device. You can't remove it". And then frozen screen with repeated sound.
If I delete everything rom AppData/hybrid it launches fine but next time... freeze.
I use Windows 7 and 375.95 drivers. Old Hybrid was doing fine.

[Selur]

No clue. First time I hear of that error.
May be it's a problem with NVEncC,.. try if it helps if you move NVEncC out of the Hybrid folder.

[Ajvar]

No clue. First time I hear of that error.
May be it's a problem with NVEncC,.. try if it helps if you move NVEncC out of the Hybrid folder.

Yes, works fine now. Interesting...

[Selur]

-> I guess some incompatibility with your graphic card drivers and NVEncC is the source of the issue then. :)

[Magik Mark]

Selur,

Can you help in figuring out why the lsfmod in staxrip is still faster? In staxrip i use prefetch 8 and in hybrid just 4. Anything higher doesn't make any improvements. I also use f3 debander and undot for both encoders.

I believe staxrip uses the "fast" default. Same encoder settings. x265 2pass 10bit.

I was thinking maybe the culprit is one of the parameter in lsfmod. It's really hard to find

[Selur]

Prefetch? No clue what you are referring to.
Without more details I can't help.
-> Send me a debug output of Hybrid and a log of what StaxRip is doing so that I can see the calls and scripts they use and I can probably see the differences.

Cu Selur

[Selur]

Hybrid rev 2017.02.12.2:
*fixed*

10bit encoding: don't use 'pixel_format' instead of 'pix_fmt' in ffmpeg since it breaks rawvideo output<


Hybrid rev 2017.02.12.1:

*fixed*

x265: analysis file handling
audio: wav handling

*changed*

x265: supporting --dynamic-rd <0..4>
x265: aq-motion allowed without adaptive quantization
x265: supporting --slices X (for better CPU usage and speed on systems with lots of cpu cores)
QSVEnc: adjusted to latest version
NVEnc: adjusted to latest version
VCEEnc: adjusted to latest version (H.265)
avisynth: updated some scripts and dlls
avisynth: high bit depth handling

*added*

audio: mono stream merge option
gui: scalingMode -1 to 3 to misc.ini options
NVEnvC: added HDR signaling
avisynth: prefer avs2yuv

-> downloads: http://www.selur.de/downloads

Cu Selur

[Jamaika]

Hybrid rev 2017.02.12.2:
*fixed*

10bit encoding: don't use 'pixel_format' instead of 'pix_fmt' in ffmpeg since it breaks rawvideo output<


Sorry, I shouldn't interfere.
It is not better to use function 'vf scale=W:H,format=yuv422p10le -strict -1'

Thx for AV1. ;)

[Selur]

It is not better to use function 'vf scale=W:H,format=yuv422p10le -strict -1'
shouldn't that do the same?
+ W&H don't seem to be constants, using:
ffmpeg -y -threads 8 -i "F:\TestClips&Co\vc-3\original_vc3.mxf" -map 0:0 -an -sn -vsync 0 -vf scale=W:H,format=yuv444p10le -strict -1 -pix_fmt yuv444p10le -f yuv4mpegpipe -
gives me:
[Parsed_scale_0 @ 0000000002546060] [Eval @ 00000000005fe490] Undefined constant or missing '(' in 'W'
[Parsed_scale_0 @ 0000000002546060] [Eval @ 00000000005fe490] Undefined constant or missing '(' in 'H'
[Parsed_scale_0 @ 0000000002546060] Error when evaluating the expression 'H'.
Maybe the expression for out_w:'W' or for out_h:'H' is self-referencing.
[Parsed_scale_0 @ 0000000002546060] Failed to configure output pad on Parsed_scale_0

btw. seems like ffmpeg can't properly handle yuv444p10le content.

[Jamaika]

You should specify the size of the frame(ex.1920x1080). It is an inconvenience.

[Selur]

Did that and the result seems to be the same. -> will stick with pix_fmt since it's shorter. :)

[Jamaika]

Will stick, but be aware that it has blocked the possibility of scaling colorrange and matrix. ;)

[Selur]

When color matrix is changed Hybrid already does use vf with "scale,format=TARGETFORMAT', you don't have to use the W:H part ;)

[Jamaika]

Frankly, I always writes all the possible parameters. I check what is the info 'loglevel trace'. I never trust the program. Sometimes compares image. It happened to me that the program default information avoided.
[Parsed_scale_0 @ 0000000000653c40] Setting 'in_color_matrix' to value 'bt2020_ncl'
[Parsed_scale_0 @ 0000000000653c40] Setting 'in_range' to value 'full'
[Parsed_scale_0 @ 0000000000653c40] Setting 'out_color_matrix' to value 'bt2020_ncl'
[Parsed_scale_0 @ 0000000000653c40] Setting 'out_range' to value 'full'
[Parsed_scale_0 @ 0000000000653c40] Setting 'flags' to value 'bicubic'
[Parsed_scale_0 @ 0000000000653c40] w:iw h:ih flags:'bicubic' interl:0
[Parsed_format_1 @ 00000000028e4260] compat: called with args=[yuv444p10le]
[Parsed_format_1 @ 00000000028e4260] Setting 'pix_fmts' to value 'yuv444p10le'

[Selur]

He, he, might be a bit overkill for a gui to present the 'trace' log level. ;)

[Jamaika]

A disadvantage of FFmpeg is that the info verobse doesn't contain this information. All the more codec vpx/aom this information isn't displayed.

[Dclose]

Anyone know if DPL and DPL2 in audio settings with QAAC has a minimum bitrate requirement? I can get it to do as low as 32 kbps when loading a single file, but it seems to want to default to 192 with a low setting option of 80 kbps if I input more than one file at a time. And then I can change the downmix to stereo to be able to click on 32 kbps, and then I can click on dpl or dpl2 to have that setting again, but with multiple files they're already in the queue with the previous audio settings.

Or in other words, if I have the downmix set to stereo, load the file, I can then change stereo downmix to dpl or dpl2 downmix.

But I can't input multiple files this way because when inputting multiple files the bitrate defaults to 192 and 80 kbps is the lowest setting it allows.

I think I could load multiple files with 32 kbps and DPL/DPL2 in the past, but if I could, I can't now.

Yes, this is with clicking on High Efficiency setting.

[Selur]

a. DPL and DPL2 are methods for downmixing 5.1 to stereo which have no influence on the required bit rate
b. no clue what you do with multiple files or not, the file count should not have any influence on whether a downmix is possible or not.
-> please describe in details what you do and what doesn't work since currently I have no clue. + create a debug output, if you don't know how to create a debug output read http://www.selur.de/support

There might be a bug in Hybrid, but unless I can reproduce and understand the issue I can't do anything about it.

Cu Selur

[Dclose]

b. no clue what you do with multiple files or not, the file count should not have any influence on whether a downmix is possible or not.
-> please describe in details what you do and what doesn't work since currently I have no clue.
It should be easy to duplicate:

Audio tab, set Filtering to Stereo, set audio Encoder tab to Qaac cvbr and High Efficiency, set audio bitrate to 32. Now drag and drop a file. The audio box to the right in the Audio tab then shows what the audio output will be. It shows 32 kbps and stereo.

Now do the same but instead of Stereo choose DPL or DPL2. When I drag and drop the file, the bitrate that was set at High Efficiency 32 kbps changes to Low Complexity and 192 kbps. If I change it to High Efficiency, the lowest bitrate I can choose is 80 kbps.

My assumption is DPL and DPL2 require a minimum bitrate to be set, since that's what it is doing. I can get around that by drag-dropping the file and then setting the audio to 32kbps and DPL and then pressing the minus button in the audio output box to delete the drag-dropped audio and pressing the + button to add the 32kbps DPL audio. And it does output in 32 kbps and in what sounds like DPL.

If I drag-drop more than one file at a time, I can't use that trick because pressing the + button only adds the audio for one of the drag-dropped files. If I press the audio Source button, it adds audio for both drag-dropped files, but puts them at 192 kbps instead of 32 kbps.

The main question is the first part of this post. If audio is set to QAAC, DPL, High Efficiency, 32 kbps, is drag-dropping a file supposed to change the audio to Low Complexity 192 kbps? And is DPL/DPL2 with High Efficiency supposed to have a lowest kbps possible setting of 80 kbps?

Even if DPL/DPL2 with High Efficiency has a lowest kbps setting of 80, maybe I'm missing a setting somewhere that forces Hybrid to keep the 80 kbps setting, since every time I drag-drop with DPL or DPL2 and High Efficiency QAAC, the bitrate changes to 192. (Though it does let me manually change it to as low as 80 kbps.)

FDK let's me drag-drop to DPL/DPL2 with 32 kbps, but if I had it set to High Efficiency or High Efficiency V2, it changes it to Low Complexity (but does keep it at 32 kbps.)

[Selur]

Reason isn't DPL/DPL2, but simply that Hybrid seems to ignore the downmix when you use drag&drop.
Since you provided no information about:
a. What audio handling method you use.
b. What characteristics your audio files have.
c. What OS you are on.
d. What other settings you use.
this will take a while.
btw. to see the restrictions regarding bit rates from the qaac encoder you can call 'qaac --formats':
LC 8000Hz Mono -- 8,12,16,20,24
LC 8000Hz Stereo -- 16,20,24,28,32,40,48
LC 8000Hz 3.0 (C L R) -- 24,28,32,40,48,56,64,72
LC 8000Hz 4.0 (L R Ls Rs) -- 32,40,48,56,64,72,80,96
LC 8000Hz 4.0 (C L R Cs) -- 32,40,48,56,64,72,80,96
LC 8000Hz 5.0 (C L R Ls Rs) -- 40,48,56,64,72,80,96,112
LC 8000Hz 5.1 (C L R Ls Rs LFE) -- 40,48,56,64,72,80,96,112
LC 8000Hz 6.0 (C L R Ls Rs Cs) -- 48,56,64,72,80,96,112,128,144
LC 8000Hz 6.1 (C L R Ls Rs Cs LFE) -- 48,56,64,72,80,96,112,128,144
LC 8000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 56,64,72,80,96,112,128,144,160
LC 8000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 56,64,72,80,96,112,128,144,160
LC 8000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 64,72,80,96,112,128,144,160,192
LC 11025Hz Mono -- 8,12,16,20,24,28,32
LC 11025Hz Stereo -- 16,20,24,28,32,40,48,56,64
LC 11025Hz 3.0 (C L R) -- 40,48,56,64,72,80,96
LC 11025Hz 4.0 (L R Ls Rs) -- 48,56,64,72,80,96,112,128
LC 11025Hz 4.0 (C L R Cs) -- 48,56,64,72,80,96,112,128
LC 11025Hz 5.0 (C L R Ls Rs) -- 64,72,80,96,112,128,144,160
LC 11025Hz 5.1 (C L R Ls Rs LFE) -- 64,72,80,96,112,128,144,160
LC 11025Hz 6.0 (C L R Ls Rs Cs) -- 72,80,96,112,128,144,160,192
LC 11025Hz 6.1 (C L R Ls Rs Cs LFE) -- 72,80,96,112,128,144,160,192
LC 11025Hz 7.0 (C L R Ls Rs Rls Rrs) -- 96,112,128,144,160,192,224
LC 11025Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 96,112,128,144,160,192,224
LC 11025Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 96,112,128,144,160,192,224,256
LC 12000Hz Mono -- 12,16,20,24,28,32
LC 12000Hz Stereo -- 24,28,32,40,48,56,64
LC 12000Hz 3.0 (C L R) -- 40,48,56,64,72,80,96
LC 12000Hz 4.0 (L R Ls Rs) -- 48,56,64,72,80,96,112,128
LC 12000Hz 4.0 (C L R Cs) -- 48,56,64,72,80,96,112,128
LC 12000Hz 5.0 (C L R Ls Rs) -- 64,72,80,96,112,128,144,160
LC 12000Hz 5.1 (C L R Ls Rs LFE) -- 64,72,80,96,112,128,144,160
LC 12000Hz 6.0 (C L R Ls Rs Cs) -- 72,80,96,112,128,144,160,192
LC 12000Hz 6.1 (C L R Ls Rs Cs LFE) -- 72,80,96,112,128,144,160,192
LC 12000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 96,112,128,144,160,192,224
LC 12000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 96,112,128,144,160,192,224
LC 12000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 96,112,128,144,160,192,224,256
LC 16000Hz Mono -- 12,16,20,24,28,32,40,48
LC 16000Hz Stereo -- 24,28,32,40,48,56,64,72,80,96
LC 16000Hz 3.0 (C L R) -- 40,48,56,64,72,80,96,112,128,144
LC 16000Hz 4.0 (L R Ls Rs) -- 48,56,64,72,80,96,112,128,144,160,192
LC 16000Hz 4.0 (C L R Cs) -- 48,56,64,72,80,96,112,128,144,160,192
LC 16000Hz 5.0 (C L R Ls Rs) -- 64,72,80,96,112,128,144,160,192,224
LC 16000Hz 5.1 (C L R Ls Rs LFE) -- 64,72,80,96,112,128,144,160,192,224
LC 16000Hz 6.0 (C L R Ls Rs Cs) -- 72,80,96,112,128,144,160,192,224,256,288
LC 16000Hz 6.1 (C L R Ls Rs Cs LFE) -- 72,80,96,112,128,144,160,192,224,256,288
LC 16000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 96,112,128,144,160,192,224,256,288,320
LC 16000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 96,112,128,144,160,192,224,256,288,320
LC 16000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 96,112,128,144,160,192,224,256,288,320,384
LC 22050Hz Mono -- 16,20,24,28,32,40,48,56,64
LC 22050Hz Stereo -- 32,40,48,56,64,72,80,96,112,128
LC 22050Hz 3.0 (C L R) -- 48,56,64,72,80,96,112,128,144,160,192
LC 22050Hz 4.0 (L R Ls Rs) -- 64,72,80,96,112,128,144,160,192,224,256
LC 22050Hz 4.0 (C L R Cs) -- 64,72,80,96,112,128,144,160,192,224,256
LC 22050Hz 5.0 (C L R Ls Rs) -- 80,96,112,128,144,160,192,224,256,288,320
LC 22050Hz 5.1 (C L R Ls Rs LFE) -- 80,96,112,128,144,160,192,224,256,288,320
LC 22050Hz 6.0 (C L R Ls Rs Cs) -- 96,112,128,144,160,192,224,256,288,320,384
LC 22050Hz 6.1 (C L R Ls Rs Cs LFE) -- 96,112,128,144,160,192,224,256,288,320,384
LC 22050Hz 7.0 (C L R Ls Rs Rls Rrs) -- 112,128,144,160,192,224,256,288,320,384,448
LC 22050Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 112,128,144,160,192,224,256,288,320,384,448
LC 22050Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 128,144,160,192,224,256,288,320,384,448,512
LC 24000Hz Mono -- 16,20,24,28,32,40,48,56,64
LC 24000Hz Stereo -- 32,40,48,56,64,72,80,96,112,128
LC 24000Hz 3.0 (C L R) -- 48,56,64,72,80,96,112,128,144,160,192
LC 24000Hz 4.0 (L R Ls Rs) -- 64,72,80,96,112,128,144,160,192,224,256
LC 24000Hz 4.0 (C L R Cs) -- 64,72,80,96,112,128,144,160,192,224,256
LC 24000Hz 5.0 (C L R Ls Rs) -- 80,96,112,128,144,160,192,224,256,288,320
LC 24000Hz 5.1 (C L R Ls Rs LFE) -- 80,96,112,128,144,160,192,224,256,288,320
LC 24000Hz 6.0 (C L R Ls Rs Cs) -- 96,112,128,144,160,192,224,256,288,320,384
LC 24000Hz 6.1 (C L R Ls Rs Cs LFE) -- 96,112,128,144,160,192,224,256,288,320,384
LC 24000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 112,128,144,160,192,224,256,288,320,384,448
LC 24000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 112,128,144,160,192,224,256,288,320,384,448
LC 24000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 128,144,160,192,224,256,288,320,384,448,512
LC 32000Hz Mono -- 24,28,32,40,48,56,64,72,80,96
LC 32000Hz Stereo -- 48,56,64,72,80,96,112,128,144,160,192
LC 32000Hz 3.0 (C L R) -- 72,80,96,112,128,144,160,192,224,256,288
LC 32000Hz 4.0 (L R Ls Rs) -- 96,112,128,144,160,192,224,256,288,320,384
LC 32000Hz 4.0 (C L R Cs) -- 96,112,128,144,160,192,224,256,288,320,384
LC 32000Hz 5.0 (C L R Ls Rs) -- 128,144,160,192,224,256,288,320,384,448
LC 32000Hz 5.1 (C L R Ls Rs LFE) -- 128,144,160,192,224,256,288,320,384,448
LC 32000Hz 6.0 (C L R Ls Rs Cs) -- 144,160,192,224,256,288,320,384,448,512,576
LC 32000Hz 6.1 (C L R Ls Rs Cs LFE) -- 144,160,192,224,256,288,320,384,448,512,576
LC 32000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 192,224,256,288,320,384,448,512,576,640
LC 32000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 192,224,256,288,320,384,448,512,576,640
LC 32000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 192,224,256,288,320,384,448,512,576,640,768
LC 44100Hz Mono -- 32,40,48,56,64,72,80,96,112,128,144,160,192,224,256
LC 44100Hz Stereo -- 64,72,80,96,112,128,144,160,192,224,256,288,320
LC 44100Hz 3.0 (C L R) -- 96,112,128,144,160,192,224,256,288,320,384,448
LC 44100Hz 4.0 (L R Ls Rs) -- 128,144,160,192,224,256,288,320,384,448,512,576,640
LC 44100Hz 4.0 (C L R Cs) -- 128,144,160,192,224,256,288,320,384,448,512,576,640
LC 44100Hz 5.0 (C L R Ls Rs) -- 160,192,224,256,288,320,384,448,512,576,640,768
LC 44100Hz 5.1 (C L R Ls Rs LFE) -- 160,192,224,256,288,320,384,448,512,576,640,768
LC 44100Hz 6.0 (C L R Ls Rs Cs) -- 192,224,256,288,320,384,448,512,576,640,768,960
LC 44100Hz 6.1 (C L R Ls Rs Cs LFE) -- 192,224,256,288,320,384,448,512,576,640,768,960
LC 44100Hz 7.0 (C L R Ls Rs Rls Rrs) -- 224,256,288,320,384,448,512,576,640,768,960
LC 44100Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 224,256,288,320,384,448,512,576,640,768,960
LC 44100Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 256,288,320,384,448,512,576,640,768,960,1280
LC 48000Hz Mono -- 32,40,48,56,64,72,80,96,112,128,144,160,192,224,256
LC 48000Hz Stereo -- 64,72,80,96,112,128,144,160,192,224,256,288,320
LC 48000Hz 3.0 (C L R) -- 96,112,128,144,160,192,224,256,288,320,384,448
LC 48000Hz 4.0 (L R Ls Rs) -- 128,144,160,192,224,256,288,320,384,448,512,576,640
LC 48000Hz 4.0 (C L R Cs) -- 128,144,160,192,224,256,288,320,384,448,512,576,640
LC 48000Hz 5.0 (C L R Ls Rs) -- 160,192,224,256,288,320,384,448,512,576,640,768
LC 48000Hz 5.1 (C L R Ls Rs LFE) -- 160,192,224,256,288,320,384,448,512,576,640,768
LC 48000Hz 6.0 (C L R Ls Rs Cs) -- 192,224,256,288,320,384,448,512,576,640,768,960
LC 48000Hz 6.1 (C L R Ls Rs Cs LFE) -- 192,224,256,288,320,384,448,512,576,640,768,960
LC 48000Hz 7.0 (C L R Ls Rs Rls Rrs) -- 224,256,288,320,384,448,512,576,640,768,960
LC 48000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 224,256,288,320,384,448,512,576,640,768,960
LC 48000Hz 8.0 (C L R Ls Rs Rls Rrs Cs) -- 256,288,320,384,448,512,576,640,768,960,1280
HE 32000Hz Mono -- 12,16,24,32,40
HE 32000Hz Stereo -- 24,32,40,48,56,64,80
HE 32000Hz 4.0 (L R Ls Rs) -- 48,56,64,80,96,112,128,160
HE 32000Hz 5.1 (C L R Ls Rs LFE) -- 64,80,96,112,128,160,192
HE 32000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 96,112,128,160,192,224,256
HE 44100Hz Mono -- 16,24,32,40
HE 44100Hz Stereo -- 32,40,48,56,64,80
HE 44100Hz 4.0 (L R Ls Rs) -- 64,80,96,112,128,160
HE 44100Hz 5.1 (C L R Ls Rs LFE) -- 80,96,112,128,160,192
HE 44100Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 112,128,160,192,224,256
HE 48000Hz Mono -- 16,24,32,40
HE 48000Hz Stereo -- 32,40,48,56,64,80
HE 48000Hz 4.0 (L R Ls Rs) -- 64,80,96,112,128,160
HE 48000Hz 5.1 (C L R Ls Rs LFE) -- 80,96,112,128,160,192
HE 48000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 112,128,160,192,224,256
HE 88200Hz Mono -- 32,40,48,56,64,80,96,112,128,160
HE 88200Hz Stereo -- 64,80,96,112,128,160,192,224,256,320
HE 88200Hz 4.0 (L R Ls Rs) -- 128,160,192,224,256,320,448,640
HE 88200Hz 5.1 (C L R Ls Rs LFE) -- 160,192,224,256,320,448,640
HE 88200Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 224,256,320,448,640,1120
HE 96000Hz Mono -- 32,40,48,56,64,80,96,112,128,160
HE 96000Hz Stereo -- 64,80,96,112,128,160,192,224,256,320
HE 96000Hz 4.0 (L R Ls Rs) -- 128,160,192,224,256,320,448,640
HE 96000Hz 5.1 (C L R Ls Rs LFE) -- 160,192,224,256,320,448,640
HE 96000Hz 7.1 (C Lc Rc L R Ls Rs LFE) -- 224,256,320,448,640,1120

=> will look into reproducing the issue

Cu Selur

Ps.: send you a pm with a link to a dev version which should fix the issue,...

[Dclose]

btw. to see the restrictions regarding bit rates from the qaac encoder you can call 'qaac --formats':
You overestimate my ability to know how to "call" formats. :D But that's fine. It's great that Hybrid can use QAAC, and multiple tracks, btw. Lots of other encoders don't do that.

Reason isn't DPL/DPL2, but simply that Hybrid seems to ignore the downmix when you use drag&drop.
Though Hybrid did keep the settings when drag&dropping to Stereo downmix.

It looks like the new fix fixed it. Thank you. I hadn't even considered it could be linked to drag&dropping.

[Selur]

Hybrid rev 2017.05.03.1:
*fixed*

Vapoursynth: a bunch of small changes
filtering: handling of rotate when avisynth/vapoursynth is used
audio: audio channel count not properly taken into account on source change when downMix was involved
audio: qaac datarate adjustement

*changed*

general: high bit depth input handling
general: Vapoursynth automatically installed on 64bit systems (Windows only); not enabled by default.

*added*

Vapoursynth: nnedi3-resize, DegrainMedian
VCEEnc: support for some additional options

-> downloads: http://www.selur.de/downloads

Cu Selur

[Havokdan]

Hybrid rev 2017.05.03.1:
(...)
general: Vapoursynth automatically installed on 64bit systems (Windows only); not enabled by default.
(...)


Why disabled?

[Selur]

Why disabled?
Because it's Avisynth or Vapoursynth. (can be easily enabled through the misc.ini)
+ No time and motivation to adjust the NSIS Installer script. (aside from minimal changes I haven't adjusted the script for ages and it would need a complete overhaul)

[Havokdan]

Because it's Avisynth or Vapoursynth. (can be easily enabled through the misc.ini)
+ No time and motivation to adjust the NSIS Installer script. (aside from minimal changes I haven't adjusted the script for ages and it would need a complete overhaul)

Thanks :thanks:

I did a clean install here, and Vapousynth is enabled instead of Avisynth. Windows 10 x64

[Selur]

Yeah, I know, I accidentally copied a misc.ini into the installer. -> Adjusting (or deleting) the misc.ini will fix that.

[Selur]

Hybrid rev 2017.03.06.1:
*fixed*

Vapoursynth: Interframe
audio: vbr stack position on Linux/Mac
general: removed misc.ini from installer (Windows only)
vceenc: typo 'vbv-buffsize' -> 'vbv-bufsize' (Windows only)


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: to Windows users: You can switch between vapoursynth and avisynth by adjusting the misc.ini file.

[Selur]

Hybrid rev 2017.03.19.1:
*fixed*

Hybrid complaining about missing 'bitrate' and 'targetSizemode' depending on the encoder saved as default 'Video Handling'
x265: rdrefine was always enabled with rdo >=5
x265: glitch in 'minimize call'
general: some glitches with Avisynth/Vapoursynth preview
general: reset cut settings on disabling
Vapoursynth: cut + Vapoursynth script
Vapoursynth: FilterView make sure variable&dualVar have the same color space

*changed*

Vapoursynth: using vfrtocfr
Avisynth: TDeint/TFM with QTGMC support
FilterView/Split: Adjustment to options

*added*

Vapoursynth: DGDecIM support
Vapoursynth: sRestore
Output: .ts as output container
misc: allow to enforce language and country

-> downloads: http://www.selur.de/downloads

Cu Selur

[Havokdan]

I cannot remux any file to mkv with this latest version, worked in previous versions.

[Selur]

Argh,.. I know why it happens,.. will fix => fixed locally, can send you a dev version for testing if you tell me what OS you are using :)

[Havokdan]

Windows 10 x64

[Selur]

-> check your pm

[Havokdan]

-> check your pm

It worked here, thank you.

[Selur]

Hybrid rev 2017.03.26.1:
*fixed*

Vapoursynth: Interframe
muxing: replace "- with '-symbol in tags
avisynth: non yv12 handling
muxing: remuxing with video passthrough
avisynth: missing GRunT.dll

*changed*

x265: adjusted to new slower/very slow presets

*added*

Vapoursynth: set color resizer
misc.ini: skip nvidia checks with: nonnvidia=true

-> downloads: http://www.selur.de/downloads

Cu Selur

[InTheLight]

2017.03.26.1

I noticed that under the Audio Tab/Audio Encoding Options/Filtering, whenever I set "Downmix" to custom and regardless of what I place for the remix call, when I add the project to the queue I receive the error message "Unsupported downmix call 2, custom!" twice.

Is this a bug or am I butchering my custom sox call?

Let me know if you would like a log.

[EDIT]
2017.04.09 - 11:09:24_Windows 7-32bit_2017.03.26.1 - level 8: creating sox call,...
2017.04.09 - 11:09:24_Windows 7-32bit_2017.03.26.1 - level 8: soxRemixCall input count 2, mode: custom
2017.04.09 - 11:09:24_Windows 7-32bit_2017.03.26.1 - level 8: Unsupported downmix call 2, custom!

[Selur]

It's a bug. :)

[Selur]

Hybrid rev 2017.04.10.1:
*fixed*

audio: fix sox custom remix
x264: cbr handling
avisynth: using QTGMC with TIVTC/TDeint
Vapoursynth: DVD input handling
tools: some problems loading tools

*changed*

x265: added support for --limit-sao
vpx: added support for --row-mt
Vapoursynth: set frame properties (color matrix/range) for input to make sure they are set
ui: enable tool tip side bar by default
ui: only show available deinterlacers


-> downloads: http://www.selur.de/downloads

Cu Selur

[digitall.h]

Dear selur,
Thank you very much for keeping this program updated.
I'm using latest Hybrid 2017.04.10.1 Linux QT 5.21 version, under Debian.
Pop-up text when pointing buttons appear as a black pop-up with no readable text inside.
I did not change system colour scheme lately, and this did not happen with previous versions.
Of course I'm a long time Hybrid user and I know the purpose of those buttons, it's just a question of fixing the GUI.

Best

[Selur]

I didn't change anything on my build system which is a Kubunutu 64bit 14.04 with Qt 5.2.1, so I doubt the it's the gui at fault.

My guess is Debian changed something in their Qt libraries (they did something similar in the past) or their color schema.
As a workaround you could:
a. disable the tool tips
b. enable the tooltip sidebar
(Config->View->Design)

Can't say more since the info that you use a Debian system in itself doesn't allow me to reproduce your environment inside a VM.

Cu Selur

[digitall.h]

Sorry if the information was not enough.
Please tell me which further information you would need.
I don't want to make you lose your time since this seems just to be my problem.

I think I explained it bad. I perfectly see tooltips, so no need to disable tooltip and enable tooltip side bar.
What is shown wrong are the texts (i called them pop-up, probably wrong) when you point on buttons and not on text or checkboxes.

[Selur]

a. What Debian system do you use exactly? What version? What window manager? If you know where I can download an install image post a link to the site. I need enough information to install such a system inside a VM.
b. The text that appear once you point at something are the tool-tips. So now I'm even more confused now. :)
You can try whether it helps to deinstall and reinstall Hybrid.

[digitall.h]

My Debian system:
Kernel: 3.16.0-4-amd64 x86_64 (64 bit)
Desktop: KDE 4.14.2
Distro: SolydXK 8 solydxk

So it is based in Debian 8.

Related to tool-tips confusion I cannot explain it better: when you point with the mouse to text or checkboxes or slidedown menues there appears a tool-tip, in my system a yellow square with black text in it. When you point to a button appears a dark square that used to be white with black text in it, but as it is now black there cannot be read anything.

I will try to get a snapshot and send you to explain it better.

[Selur]

Color selection isn't done by my code at all. :)
Sounds like some theme setting in your distro. :)

[Selur]

Got your email an here's my reply:
Ah, now I see what you meant.
To reproduce this I downloaded latest debian iso, installed it inside a VM selected KDE as desktop manager installed Hybrid and I can reproduce the problem in the VM.

Seems like this is a problem with their kde color theme/configuration.

Starting 'Colors' I selected the 'Obisidian Coast' theme, applied it, started Hybrid and tool tips were readable.


=> not a bug in Hybrid, but a strange default color theme of debian.
Seeing that tooltips work fine with 'Obisidian Coast' this can be fixed by properly configuring the colors on your system.
Since I don't know you color preferences and Hybrid doesn't enforce any color theme, but uses the one of the desktop manger there is nothing for me to do.

Cu Selur

Ps.: posted this here so other which might have the same problem can see the solution. (adjusting the color theme of the desktop manager)

[digitall.h]

Thank you for your answer.
I will try to adjust system colour scheme, and see if it makes look bad other applications.

Best

[Dclose]

I've always had QAAC problems when installing Hybrid. I just installed it to Windows 7 64-bit, after installing Quicktime, and it happened again.

The rest of Hybrid is fine, but QAAC will cause a job to crash when it tries to encode the audio. What seems to be happening is Hybrid doesn't want to use the 64-bit QAAC.exe file. Or doesn't want to use the qaac.exe file that comes with Hybrid, if there is one. I'm not sure where Hybrid is getting its qaac.exe file from.

Downloading QAAC from the main QAAC page, the 64-bit exe is called qaac64.exe. Renaming it to qaac.exe and putting it in the Hybrid folder, the job in hybrid still crashes when starting the audio, so the problem isn't just a filename problem.

The fix I've always had to do is copy (what I assume is the) 32-bit version of qaac's exe file called qaac.exe to the hybrid folder. Then it all works fine. If Hybrid installs its own qaac.exe file, then whether it's 32-bit or 64-bit, the job always crashes at the start of audio encoding on 64-bit Windows.

It's done it to me for a long time, and just did it again on a desktop where the Hybrid install was fine with Windows 7 32-bit but now I added Windows 7 64-bit for dual-booting, and with a fresh Hybrid install the qaac audio crashed as usual until I downloaded a qaac package and used the 32-bit qaac.exe.

I don't recall seeing anyone else mention that problem, but it's been consistent for me on multiple computers. If anyone gets that problem, that's the fix.

[Selur]

Wild guess: you only installed the 32bit Version of the 'Apple Application Support'-package and not the64bit version thus, 64bit version of qaac will crash and 32bit version won't.
Tested here. For me both versions work if the appropriate 'Apple Application Support' is installed.

[Dclose]

Wild guess: you only installed the 32bit Version of the 'Apple Application Support'-package and not the64bit version thus, 64bit version of qaac will crash and 32bit version won't.
Tested here. For me both versions work if the appropriate 'Apple Application Support' is installed.
Probably, since I didn't specifically try to install either of those. I don't recall at the moment if Hybrid makes a specific note of that during install, (other than mentioning Quicktime), but if it does I guess I don't read it as it's intended. Even on forums when people ask about getting QAAC to work, the main thing I recall is, "Install Quicktime. It has the needed files." But there's more to it than that.

[Selur]

Probably, since I didn't specifically try to install either of those.
64bit 'Apple Application Support' should come with 64bit iTunes/Quicktime not the 32bit version.
-> best bet nowadays is probable downloading a 64bit version of iTunes, open the installer with 7zip, extract the AppleApplicationSuppot.msi and install that. :)
(Sadly there is no separate download for the application support,..)
App & Features under Windows 10 shows whether the 64bit and or the 32bit version is installed, I guess the same should be true for Win7.

[Selur]

Hybrid rev 2017.04.23.1:
*fixed*

jobs: Don't generate unneeded Vapoursynth script during job generation
ffmpegnvenc: glitch when selecting constant quantizer encoding not showing the field to set the quantizer
Vapoursynth: Interframe 'overwrite Alg.' selection
Vapoursynth: inital vsCustomStack position
Vapoursynth: don't show resize for preview in Vapoursynth script view
Vapoursynth: deinterlacer adjustment selecting DGDecNV even when not present
Vapoursynth: loading of content with d2vsource (now using rff=False for PAL source)
QSVEnc: some minor fixes
Avisynth: Flash3kyuu deband defaults

*changed*

Avisynth: Avisynth preview now shows 'distributor()' again

*added*

x265: support for dynamic HDR-10
Avisynth: ResampleMT support
Avisynth: support amDCT
Vapoursynth: MVToolsFPS frame interpolation
Vapoursynth: using d2vSource and d2vWitch also for H.264 raw/transportstreams


-> downloads: http://www.selur.de/downloads

Cu Selur

[Dclose]

Hybrid rev 2017.04.23.1

Based on the timing of this release, I assume it also includes:

2. Lambda tables for 8, 10, and 12-bit encoding revised, resulting in significant enhancement to subjective visual quality.
Some people in the x265 thread are very excited over that.

[Selur]

Yes, the x265 version is like always up-to-date and since it's not anything that needs to be en-/disabled it's not mentioned in the changelog.

Cu Selur

[Selur]

Hybrid rev 2017.05.06.1:
*fixed*

input: workaround for latest MakeMKV changes
general: minimizing folder path could result in an empty folder
general: resolution restriction for blu-ray
Vapoursynth: reset defaults
Avisynth: reset defaults
Avisynth: defaults for fft3dgpu, Flash3k

*changed*

general: forum link updated to forum.selur.net
disabled console output
changing max mpls playlist size from 50GB to 100GB
log: removed details from log and moved them to the report since I'm annoyed of people posting the log and sending me parts of the log by email. (next escalation step will be to remove the log)

*added*

Vapoursynth: VsSSIQ support


-> downloads: http://www.selur.de/downloads

Cu Selur

[stax76]

input: workaround for latest MakeMKV changes

What did MakeMKV change?

[Selur]

What did MakeMKV change?
They started to include the cover, which got Hybrids internal ffmpeg<>mplayer<>mediaInfo id mapping confused since mplayer&mediainfo reported 1 Video stream and ffmpeg reported two. :)

Cu Selur

[Dclose]

I don't recall at the moment if Hybrid makes a specific note of that during install, (other than mentioning Quicktime), but if it does I guess I don't read it as it's intended. Even on forums when people ask about getting QAAC to work, the main thing I recall is, "Install Quicktime. It has the needed files." But there's more to it than that.
To follow up on that, installing Quicktime player apparently doesn't auto-install the Apple Support Package file needed for QAAC to work with 64-bit Windows, but during the last update-install of Hybrid I installed iTunes and saw it auto-installing the support package during the install, and then as expected QAAC worked for 64-bit Windows 7/8.
Hybrid rev 2017.05.06.1:

log: removed details from log and moved them to the report since I'm annoyed of people posting the log and sending me parts of the log by email. (next escalation step will be to remove the log)
I was wondering what happened with the now sparse log. Sometimes I'd forget what CFR or other settings I used on an encode, and I could go and see it in the log. Now the only useful thing in there for most people is the time the encoding took. I don't know if there's another hopefully easy way to see what settings were used on a prior encode.

[Selur]

I don't know if there's another hopefully easy way to see what settings were used on a prior encode.
In case the job is still inside the job-queue you could look at the encoding call by disabling the 'minimize job command lines' option.

[den78]

Hi, Selur.
Just got one question. Why doesn't mediainfo show x265 codec information? Everything is fine with x264.

I've sent you a debug just in case.

[sneaker_ger]

Mkv with mkvtoolnix version 11? There was a bug that got fixed in 12.

[Selur]

yup, updating mkvtoolnix should fix it.

[den78]

Yep, that fixed it. Thanks, guys.

[cocobanana69]

Hi Salur,

First I must say nice work about Hybrid video converter.

I’m still favoring the XviD-style ripping (my favorite ripping app was GordianKnot).

Features I’m missing in Hybrid, please add them:

-auto de-delay for delayed audios
-auto converting DVD/Blu-Ray subtitles (if I choose) to text (for ex. to .srt or .sub)
-GTK3 interface version (for Linux)
-Opening image files (iso,nrg,daa etc)
-Different field alignment (1st Main, 2nd Crop/Resize, 3rd Filtering, 4th Video codec options, 5th Audio encoding options, 6th Chapter, 7th Config, 8th Jobs, 9th Logs)
-Multilanguage translations (I need Serbian Cyrillic and Russian)
-MKV tagging panel, for writing MKV tags like studio, year, director, actors, soundtrack info…
-Compressibility check option before encoding with XviD

https://s11.postimg.org/lw1kvx6ub/01_compressibility_check.jpg

-Preview window like in GordianKnot

https://s23.postimg.org/5mi98dw7v/02_preview_panel.jpg

-Bits/pixel with compressibility test panel (like in GordianKnot) (for XviD encoding)

https://s24.postimg.org/pcutm0arp/03_Bits-_Pixels_Panel.jpg

-Resize & aspect ratio panel (like in GordianKnot) with auto near original aspet-ratio resolutions suggestions

https://s21.postimg.org/vm21yl8sn/04_Crop_Resolution.jpg

-Credits encoding options allowing multiple setups (for XviD encoding, for example I want to encode 1st setting of credits starting from frame 136377 to 140200 at quant: 20, then from frame 140201 to 140800 regularly as the rip is, then from frame 140801 to 150744 also at quant: 20 and in greyscale)

https://s21.postimg.org/qnsz6u2h3/05_Credits.jpg

Features I’m missing in ffmpeg & mencoder (and to be ported to Hybrid):
-lanczos3 & lanczos4
-all avisynth filters missing (should be ported)

Audio feautures I’m missing (I also miss thes for music encoding on Linux):
-fraunhofer-like vbr mp3 encoder or lame without psychoacoustic frequencies cut-off and with high-shelf frequencies cutter (customly set) at 16KHz (I already messaged the fre:ac team for such encoder).

And please post the source code.
Thanks,

[Selur]

auto de-delay for delayed audios
Already possible, inside the Audio-tab you can set the audio delay you want for the current stream before adding it to the audio queue,.. (Audio->Main->Audio Source->A/V delay (ms).

auto converting DVD/Blu-Ray subtitles (if I choose) to text (for ex. to .srt or .sub)
Don't know of any decent/stable/reliable and portable way that this could be done cross plattform.
-> won't probably come.
GTK3 interface version (for Linux)
unsure about what is missing there, normally Hybrid should work fine under Gnome,..

Opening image files (iso,nrg,daa etc)
not going to happen

Different field alignment (1st Main, 2nd Crop/Resize, 3rd Filtering, 4th Video codec options, 5th Audio encoding options, 6th Chapter, 7th Config, 8th Jobs, 9th Logs)
won't happen

Multilanguage translations (I need Serbian Cyrillic and Russian)
I don't speak either an I lack the time to translate all the texts.
-> probably won't happen unless I win the lottery

MKV tagging panel, for writing MKV tags like studio, year, director, actors, soundtrack info…
Already possible: Config->Output->Mkv Tags

Compressibility check option before encoding with XviD
I don't believe in compression tests, so that won't happen. :)

Preview window like in GordianKnot
Both the Avisynth and the Vapoursynth Preview windows should be similar.

Resize & aspect ratio panel (like in GordianKnot) with auto near original aspet-ratio resolutions suggestions
Don't like the panel and Hybrid already tries to minimize the any aspect error. -> won't change.

Credits encoding options allowing multiple setups (for XviD encoding, for example I want to encode 1st setting of credits starting from frame 136377 to 140200 at quant: 20, then from frame 140201 to 140800 regularly as the rip is, then from frame 140801 to 150744 also at quant: 20 and in greyscale)
Aside from gray scale, the normal Xvid->Zones control should already cover that.

all avisynth filters missing (should be ported)
Hybrid already supports tons of Avisynth (Filtering->Avisynth) and Vapoursynth (Filtering->Vapoursynth) filters depending of the choice one makes. (tested on Windows and Linux, not on Mac)

Audio feautures I’m missing (I also miss thes for music encoding on Linux):
-fraunhofer-like vbr mp3 encoder or lame without psychoacoustic frequencies cut-off and with high-shelf frequencies cutter (customly set) at 16KHz (I already messaged the fre:ac team for such encoder).
Hybrid atm. only supports ffmpeg and lame for mp3 encoding, since I'm not aware of any other free mp3 encoders on Linux. (about the cutoff, if you can tell me the ffmpeg/lame command line I could add support for it)

Cu Selur

[von Suppé]

In config tab, I always had "Prefer FFmpeg as decoder" checked.

It lately draws my attention that during a recode "ffmpeg_32.exe *32" is taking a lot of CPU and memory resources, having significant effect on the transcoding time. Did something change?

BTW what decoder would Hybrid use if I would uncheck it?

[Selur]

what decoder would Hybrid use if I would uncheck it?
Hybrid uses either: ffmpeg or mencoder or avs2yuv or VSPipe depending on the input and what it deems a good choice. :)

Did something change?
Not really. My guess is you now just use some filter which requires a lot of resources.
(Also I updated qtgmc iirc. so that might influence the ram and cpu usage)

[HeiLi]

Hi, i have been using it for 5-6 weeks hybrid but last 4-5 encode always gives random grey screen like this

http://i.imgur.com/P8UozAd.jpg

How can i fix that?

(Add later debug report)

[Havokdan]

Hi, i have been using it for 5-6 weeks hybrid but last 4-5 encode always gives random grey screen like this

http://i.imgur.com/P8UozAd.jpg

How can i fix that?

(Add later debug report)

This happens to me when I try to use the encoders that use the GPU and the original file is 10 bits.

[HeiLi]

This happens to me when I try to use the encoders that use the GPU and the original file is 10 bits.

I dont use gpu or 10bits encode x264 but last 3-4 encode like this and sometimes encode bluray to x264 7-8 hours sometimes 48 hours i dont understand

[Selur]

Looks like a decoding problem of the source without any details the only thing I can say is that using another decoder might help.
In case the source if vc-1 using a gpu based decoder like FRIM (Avisynth only), DGDecNV (Avisynth&Vapoursynth), DGDecIM (Vapoursynth), might help.
If the source isn't vc-1 any other decoder might help.

[HeiLi]

Looks like a decoding problem of the source without any details the only thing I can say is that using another decoder might help.
In case the source if vc-1 using a gpu based decoder like FRIM (Avisynth only), DGDecNV (Avisynth&Vapoursynth), DGDecIM (Vapoursynth), might help.
If the source isn't vc-1 any other decoder might help.

i dont use gpu decoder always use QTGMC, add debug txt but hybrid so slow encode now sometimes faster

[Selur]

Might also be some ram problem or some problem due to overclocking or heat issues.
I don't need a DebugOutput of the processing. I need a debug output of the file analysis and job creation to know what you are doing.

[HeiLi]

Might also be some ram problem or some problem due to overclocking or heat issues.
I don't need a DebugOutput of the processing. I need a debug output of the file analysis and job creation to know what you are doing.

I have 32 gb ram and xeon e3-1231v3 cpu ?



debug output (http://www8.zippyshare.com/v/HxAjIJEg/file.html)

[Selur]

Encoding call:
"C:\PROGRA~1\Hybrid\ffmpeg_32.exe" -y -loglevel fatal -threads 8 -i "C:\Users\Administrator\AppData\Local\Temp\2\encodingTempSynthSkript_19_58_47_5210.avs" -an -sn -vsync 0 -pix_fmt yuv420p -f rawvideo - | "C:\PROGRA~1\Hybrid\x264.exe" --preset superfast --pass 1 --bitrate 2048 --profile high --level 4.1 --direct auto --qcomp 0.5 --slices 4 --aq-mode 0 --sar 40:33 --non-deterministic --range tv --stats "C:\Users\ADMINI~1\AppData\Local\Temp\2\Lethal_Weapon_2_t01_new_19_58_47_5210_06.stats" --demuxer raw --input-depth 8 --input-res 1280x646 --input-csp i420 --fps 24000/1001 --output NUL -
looks fine, Avisynth script:
SetMemoryMax(1536)
SetMTMode(5,8) # change MT mode
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\LoadDll.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\LSMASHSource.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\dfttest.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\RgTools.dll")
LoadDll("C:\PROGRA~1\Hybrid\AVISYN~1\fftw3.dll")
# loading source: D:\bd\Lethal_Weapon_2_t01.mkv
# input luminance scale tv
LWLibavVideoSource("D:\bd\Lethal_Weapon_2_t01.mkv",cache=false,stacked=true,format="YUV420P8",repeat=true)
# current resolution: 1920x1080
SetMTMode(2) # change MT mode
# cropping to 1920x800
Crop(0,140,0,-140)
# current resolution: 1920x800
# denoising using dfttest
dfttest()
# grain handling
Removegrain()
# scaling
Spline36Resize(1280,646)
distributor()
return
looks fine, no QTGMC used (source is progressive so no deinterlacing needed).
Videosource is VC-1/WVC1. Disable 'Avisynth->Filtering->Misc->LibavVideoSource instead of FFmpegSource', enabled 'Avisynth->Filtering->Misc->Support FRIM' and check the Avisynth script in Hybrid to see if Hybrid uses FRIM and whether that helps with the decoding problem.
Also if you can reproduce the problem with a small sample, it would help if you can share that sample with me so that I can try whether this also happens here.
You can also try, whether enabling 'Avisynth->Filtering->Misc->Force DirectShowSource' helps. Remember to check the Avisynth Preview, it the preview doesn't work or shows those Problems the encoding will too, so no reason to encode then.
Sadly since you don't have any card in your system that could be used for video decoding, that option is useless.
-> If neither FRIM or DirectShowSource helps the problem might be that no available decoder can handle the source. (sadly there is nothing I can do about that)

Cu Selur

[Music Fan]

Hi Selur,
I'm gonna be a little bit out of topic but as I see that HeiLi has 32 GB ram, I wonder when it could be useful.
I read several times that more than 8 GB ram was rarely useful.
I still have only 4 GB and its usage is always under 100 %, would I earn speed anyway with 8 GB ? I mean for video encoding and to open 3 or 4 applications at the same time, I don't play games.

[HeiLi]

Encoding call:
"C:\PROGRA~1\Hybrid\ffmpeg_32.exe" -y -loglevel fatal -threads 8 -i "C:\Users\Administrator\AppData\Local\Temp\2\encodingTempSynthSkript_19_58_47_5210.avs" -an -sn -vsync 0 -pix_fmt yuv420p -f rawvideo - | "C:\PROGRA~1\Hybrid\x264.exe" --preset superfast --pass 1 --bitrate 2048 --profile high --level 4.1 --direct auto --qcomp 0.5 --slices 4 --aq-mode 0 --sar 40:33 --non-deterministic --range tv --stats "C:\Users\ADMINI~1\AppData\Local\Temp\2\Lethal_Weapon_2_t01_new_19_58_47_5210_06.stats" --demuxer raw --input-depth 8 --input-res 1280x646 --input-csp i420 --fps 24000/1001 --output NUL -
looks fine, Avisynth script:
SetMemoryMax(1536)
SetMTMode(5,8) # change MT mode
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\LoadDll.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\LSMASHSource.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\dfttest.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\AVISYN~1\RgTools.dll")
LoadDll("C:\PROGRA~1\Hybrid\AVISYN~1\fftw3.dll")
# loading source: D:\bd\Lethal_Weapon_2_t01.mkv
# input luminance scale tv
LWLibavVideoSource("D:\bd\Lethal_Weapon_2_t01.mkv",cache=false,stacked=true,format="YUV420P8",repeat=true)
# current resolution: 1920x1080
SetMTMode(2) # change MT mode
# cropping to 1920x800
Crop(0,140,0,-140)
# current resolution: 1920x800
# denoising using dfttest
dfttest()
# grain handling
Removegrain()
# scaling
Spline36Resize(1280,646)
distributor()
return
looks fine, no QTGMC used (source is progressive so no deinterlacing needed).
Videosource is VC-1/WVC1. Disable 'Avisynth->Filtering->Misc->LibavVideoSource instead of FFmpegSource', enabled 'Avisynth->Filtering->Misc->Support FRIM' and check the Avisynth script in Hybrid to see if Hybrid uses FRIM and whether that helps with the decoding problem.
Also if you can reproduce the problem with a small sample, it would help if you can share that sample with me so that I can try whether this also happens here.
You can also try, whether enabling 'Avisynth->Filtering->Misc->Force DirectShowSource' helps. Remember to check the Avisynth Preview, it the preview doesn't work or shows those Problems the encoding will too, so no reason to encode then.
Sadly since you don't have any card in your system that could be used for video decoding, that option is useless.
-> If neither FRIM or DirectShowSource helps the problem might be that no available decoder can handle the source. (sadly there is nothing I can do about that)

Cu Selur

Check with avisynth preview there is no problem but hybrid sometimes so slow encoding speed (01-03 fps cpu use %99) but when reboot my pc first encode faster (20-30 fps cpu use %99 ) Why?

[yensteel]

Getting a hang of the interface. I encountered problems though.

I don't know what I'm doing wrong, but every time I tried, no matter qaac, fdk or faac the resulting file always contains two audio, the original and the encoded one. The first files of some of my batches sometimes doesn't even have the encoded version just the original. This occurs in different videos, even though they are from different sources.

I chose auto add (all) (auto add first also same issue)

Audio Encoding options checked
merge audio streams checked (because of 6 channels)

[Selur]

@HeiLi:
Check with avisynth preview there is no problem but hybrid sometimes so slow encoding speed (01-03 fps cpu use %99) but when reboot my pc first encode faster (20-30 fps cpu use %99 ) Why?
Wild guess is that one of the dlls used either directly has some problem with your system or another tool is interfering.

@yensteel: Far too few details to say anything. Create a debug output which shows the problem and I can look into it.
@MusicFan: When you don't use the available RAM adding more will not speed up encoding. (using a RAM drive might help a bit)

[Music Fan]

using a RAM drive might help a bit
I didn't know this ; I see that it simulates a hard drive on a part of the RAM whose available size is therefore reduced, thus I wonder how it may help to gain speed.:confused:

[Selur]

Depending on what you do it helps with remuxing, indexing and general file access. Encoding will probably only gain from it is otherwise your hdd is a bottleneck.
In Avisynth using a higher or lower MemoryMax can also influence the speed of script, but also in general not much.

[Selur]

Hybrid rev 2017.10.03.1:
*fixed*

Avisynth on Linux handling
fixed progress for ffmpeg encoding
Vapoursynth: bunch of typos
general: problem with 'ignore all input time codes'
audio: dts-hd handling
bunch of small bugs (I forgot about)

*changed*

updated dependencies
Switching between Avisynth and Vapoursynth should be possible now (Filtering->Support)
renamed Main- to Base-tab
lot of code refactoring
loosend dgdecnv restrictions
vsedit directly pipes to encoder if possible
x265: updates to newer version
forceCBR tsMuxeR muxing when cbr is used

*added*

Avisynth: FrameRateConverter


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: starting Friday I will be on a China tour and probably not online a lot (if at all) so don't be surprised in case I don't react. :)

[Gravitator]

Здравствуйте!
http://i89.fastpic.ru/big/2017/1013/31/a46c1182113fe7ceab08d921572d1131.jpg
I open any x264 and x265 files through AVS preview (did not try on others) :
Tick the info >
1. Click the next frame (or twice) > click the reverse action / previous frame (or twice);
2. Next > previous > next.
- I get a gluck of the discrepancy of the frame number in the program window and the text inside.

[Gravitator]

Is there a plan for adding a frame comparison function?

[alter4]

How can I encode vertical (portrait) video, when I open my video.mp4 from a smartphone in hybrid, it is opened in landscape view and I don't know what to do. Guys could you please help.

[jd17]

Does anyone else face an issue with --hdr-opt in the latest version of Hybrid?

I use Hybrid rev 2017.10.05.1 on a Windows x64 machine.
x265 is stable 2.5+5-0b3ba15 from here: http://www.msystem.waw.pl/x265/
(VS 2017 AVX2)

I added --hdr-opt under "Misc" as custom command line addition, like I used to.
Hybrid also confirms this in the bottom summary of commands.
The result however shows "no-hdr-opt" in MediaInfo", which is also clearly reflected in a higher bitrate.

Can you help me out?

[jd17]

I repeated the test with the x265 that comes with Hybrid rev 2017.10.05.1 (= 2.5+20).
Unfortunately, the issue is still present -> no-hdr-opt. :(

So, I went back to rev 2017.05.06.1 and replaced the integrated x265 with the stable 2.5+5-0b3ba15 again.
Again, I got the same issue! :(

Then I looked at my previous HDR testencodes and saw that I had always used x265 2.5+1-adbcc90bdef3 before.
Unfortunately I deleted that from my HDD, so I downloaded the next best thing, the stable 2.4+13.
Even with that I got no-hdr-opt! :(

As a last resort I used a text compare tool to see what might trigger this and I finally found the root cause.
Somehow, --transfer was set to 2 instead of 16, although I always had the right box checked (smpte-st-2084).
I then entered --transfer 16 manually as custom command line and voila! hdr-opt works again! :)


Is it possible that x265 no longer accepts the string "smpte-st-2084", but only "smpte2084" or the integer 16?

[den78]

Hi, Selur.

Vspipe.exe is crashing here.
https://i.imgur.com/hPCZKOg.png

I've sent you the debug file.

[Selur]

@den78: Won't probably get around to look into it before next week. :( But I got it marked for investigation.
@jd17: Yes, x265 changes the naming of the matrices. Hybrid should allow to select them manually adding a custom addition shouldn't be needed.

[den78]

^ It seems that it can't handle TS (DVB-T2) files properly. It doesn't crash when it's a remuxed MKV, but with it i have A/V sync issues.

Edit:
I remuxed it to mp4 & it can handle it with no issues. But, yeah, nevertheless, please look into it when you have the time. Thanks.
& thank you for vapoursynth. Results are way better with it.

[Selur]

Problem is probably that the transportstream headers are just too broken for the used source filter to handle it properly. (or broken time codes)
-> nothing can do about it

[bob88]

can anybody tell me how to check avisynth version and how to update to latest version? thanks.

[Music Fan]

Simply type
version()
in an avs file & open it in Vdub.
For an update, AVISynth+ is maybe preferable.

edit : sorry, I didn't saw we were in the Hybrid topic, there's maybe another solution to verify the version.

[Selur]

Hybrid does come with the latest Avisynth 32bit MT which is the only version is it tested with so installing another version might cause problems,...

[Selur]

Small question to the Windows users: "Is anyone still using a 32bit Windows installation with Hybrid?"
(Background: Qt isn't providing 32bit Windows builds out of the box through their update routines for MSVC 2017, so I'm thinking about dropping 32bit Windows support.)

Cu Selur

[bob88]

Selur,
Would it be possible to integrate 64bit Avisynth in Hybrid?
I want to use 64bit Mdegrain.

[Selur]

Nope. No plan to support Avisynth 64bit in any way.

[bob88]

why???
its faster. faster is better.

[Selur]

Avisynth 64bit is way unstable doesn't support all the filters Hybrid uses. Avisynth+ 64bit might be an alternative, but I'm more into supporting more Vapoursynth stuff than Avisynth+.
(+ Would be a lot of work, writing code, testing stuff,... )

Cu Selur

[bob88]

Are the currently integrated vapoursynth filters 64bit?

[Selur]

Yes, Hybrid only supports 64bit Vapoursynth.

[bob88]

I just did a few tests with Vapoursynth's degrainers. Compared to Avisynth's 32bit Mdegrain I got the same speeds and the results qualitywise were less satisfying.

[Selur]

So your best bet is either not use Hybrid or find an alternative to MDegrain which runs under Vapoursynth that I could support in Hybrid, since I don't plan to support Avisynth 64bit. (or stick with the 32bit Avisynth version)

Cu Selur

[bob88]

how about this SMDegrain?
https://forum.doom9.org/showthread.php?t=166582

[Selur]

Already comes with Hybrid, but isn't used other than in QTGMC atm. .
Looking just at the parameter list:
SMDegrain(input, tr=2, thSAD=300, thSADC=None, RefineMotion=False, contrasharp=None, CClip=None, interlaced=False, tff=None, plane=4, Globals=0, pel=None, subpixel=2, prefilter=-1, mfilter=None, blksize=None, overlap=None, search=4, truemotion=None, MVglobal=None, dct=0, limit=255, limitc=None, thSCD1=None, thSCD2=130, chroma=True, hpad=None, vpad=None, Str=1., Amp=0.0625)
Adding support for it will take me a while. (lot of tooltips&documentation)
-> will look into it, hopefully I get it done during the next week

[bob88]

ok Im super noob so I dont know what QTGMC is but Ill be waiting with anticipation for your update.
Thank you!

[alter4]

How can I encode vertical (portrait) video, when I open my video.mp4 from a smartphone in hybrid, it is opened in landscape view and I don't know what to do. Guys could you please help.

Sorry to repeat, Could anybody suggest something?

[Selur]

@alter4: You could use 'Crop/Resize->Misc->Misc->Rotate output' to rotate the video. :)

Cu Selur

[Vesdaris]

Hey Selur

I've been experimenting with nvenc hevc encoding and I always keep getting crashes

Could you check the log, please?

https://pastebin.com/ZKBSpsxL

I suppose Im doing something wrong here..I dunno

Thank you

[Selur]

Vapoursynth script looks fine to me:
# Imports
import os
import sys
import ctypes
Dllref = ctypes.windll.LoadLibrary("F:/Hybrid/vsfilters/Support/libfftw3f-3.dll")
# Loading Support Files
Dllref = ctypes.windll.LoadLibrary("F:/Hybrid/vsfilters/Support/OpenCL.dll")
import vapoursynth as vs
core = vs.get_core()
# Import scripts folder
scriptPath = 'F:/Hybrid/vsscripts'
sys.path.append(os.path.abspath(scriptPath))
# Loading Plugins
core.std.LoadPlugin(path="F:/Hybrid/vsfilters/GrainFilter/AddGrain/AddGrain.dll")
core.std.LoadPlugin(path="F:/Hybrid/vsfilters/DenoiseFilter/FFT3DFilter/vsfft3dfilter.dll")
core.std.LoadPlugin(path="F:/Hybrid/vsfilters/DenoiseFilter/DFTTest/DFTTest.dll")
core.std.LoadPlugin(path="F:/Hybrid/vsfilters/DenoiseFilter/KNLMeansCL/KNLMeansCL.dll")
core.std.LoadPlugin(path="F:/Hybrid/vsfilters/ResizeFilter/NNEDI3/libnnedi3.dll")
core.std.LoadPlugin(path="F:/Hybrid/vsfilters/Support/libmvtools.dll")
core.std.LoadPlugin(path="F:/Hybrid/vsfilters/Support/temporalsoften.dll")
core.std.LoadPlugin(path="F:/Hybrid/vsfilters/Support/scenechange.dll")
core.std.LoadPlugin(path="F:/Hybrid/vsfilters/Support/fmtconv.dll")
core.std.LoadPlugin(path="F:/Hybrid/vsfilters/SourceFilter/LSmashSource/vslsmashsource.dll")
# Import scripts
import havsfunc as havsfunc
# Loading F:\TEMP\Howards.End.S01E03.1080p.HDTV.H264-MTB[rarbg]\test.mkv using LWLibavSource
clip = core.lsmas.LWLibavSource(source="F:/TEMP/Howards.End.S01E03.1080p.HDTV.H264-MTB[rarbg]/test.mkv", format="YUV420P8", cache=0)
# making sure input color matrix is set as 709
clip = core.resize.Point(clip, matrix_in_s="709")
# Making sure input color range is set to TV (limited) range.
clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)
# setting field order to what QTGMC should assume
clip = core.std.SetFrameProp(clip=clip, prop="_FieldBased", intval=2)
# Deinterlacing using QTGMC
clip = havsfunc.QTGMC(Input=clip, Preset="Fast", TFF=True)
clip = clip[::2]
# adjusting output color from: YUV420P8 to YUV420P10
clip = core.resize.Bicubic(clip=clip, format=vs.YUV420P10)
# Output
clip.set_output()
Encoding call seems fine too:
"F:\Hybrid\Vapoursynth\vspipe.exe" "F:\CORBEAU\encodingTempSynthSkript_15_34_40_3010.vpy" - --y4m | "F:\Hybrid\ffmpeg.exe" -y -threads 8 -f yuv4mpegpipe -i - -an -sn -vsync 0 -strict -1 -pix_fmt yuv420p10le -vcodec hevc_nvenc -preset 5 -profile:v 1 -level 0 -rc constqp -cq 18 -2pass 0 -rc-lookahead -1 -surfaces 32 -nonref_p 1 -strict_gop 1 "F:\CORBEAU\test_15_34_40_3010_02.mkv"
(assuming your graphic card support 10bit hevc encoding; mine doesn't)

Cu Selur
What happens if you call the encoding call inside a Windows Command Prompt?

[Vesdaris]

OK.. I changed hevc to nvenc (not nvenc ffmpeg) and it works now.

I was wondering what settings would you recommend using for maximum quality? I mean these ones;

- adaptive quantization (enable it?) Strength can't be set though

- number of ref frames. 3 is default. Would 5 be a better choice quality\compression wise?

- It seems I can't enable B-frames..Is it supposed to be like this on nvenc-hevc?

-gop size max? default 0

- lookahead? default 16

Thank you!

[Selur]

adaptive quantization (enable it?) Strength can't be set though
Strength for adaptive quantizer is only available for H264. Yes, I would enable it.
Would 5 be a better choice quality\compression wise?
yes
It seems I can't enable B-frames..Is it supposed to be like this on nvenc-hevc?
Yes. The VPU doesn't support B-frames which is one of the main drawbacks.

gop size max? default 0
Quality/Compression ratio wise 0 is the best, for playback ease I would recommend something like 10 or 20 times the fps.

lookahead? default 16
Higher lookahead should provide better compression/quality, but will also require more gpu ram.

Cu Selur

[Gravitator]

С наступающим :)
Hybrid v2017.10.05.1, windows x64.
Сhoose the preset slow (the cu lossless panel is active and responds) > off cutree > on cutree - panel cu lossles attenuated / blocked (x265).

[Selur]

Started Hybrid
Set 'Best->Processing->Video' to 'x265'
Selected 'x265->Base->Preset and Tune->Preset' to 'slow' and applied it. Resulting command line was:
x265 --preset slow --input - --output-depth 10 --y4m --profile main10 --high-tier --level-idc 4.1 --qpstep 1 --crf 18.00 --qpfile GENERATED_QP_FILE --vbv-maxrate 50000 --vbv-bufsize 50000 --range limited --colormatrix bt709 --output OUTPUTFILE
enabled 'x265->Quantization->QuadTree->cu lossless' command line changed to:
x265 --preset slow --input - --output-depth 10 --y4m --profile main10 --high-tier --level-idc 4.1 --qpstep 1 --crf 18.00 --qpfile GENERATED_QP_FILE --cu-lossless --vbv-maxrate 50000 --vbv-bufsize 50000 --range limited --colormatrix bt709 --output OUTPUTFILE
disabled 'x265->Quantization->QuadTree->cu lossless' command line changed to:
x265 --preset slow --input - --output-depth 10 --y4m --profile main10 --high-tier --level-idc 4.1 --qpstep 1 --crf 18.00 --qpfile GENERATED_QP_FILE --vbv-maxrate 50000 --vbv-bufsize 50000 --range limited --colormatrix bt709 --output OUTPUTFILE

-> Can't reproduce the issue like I understood your description of the problem.

Cu Selur

[Gravitator]

Video :)
https://files.videohelp.com/u/227452/Bug.mkv

[Selur]

Ah okay, got it. Was a small bug, fixed locally.
cu-lossless should be disabled while cu-tree is disabled and not the other way around, will send you a link for a dev version which fixes this.

Cu Selur

[Gravitator]

Please add information about the frame type (I-P-B) for Vapoursynth :rolleyes:

[Selur]

@Graviator:Will think about it.

[digitall.h]

Hi selur,
I'm using Hybrid v2017.10.05.1, Linux 64bit qt521
When I load a BluRay and select a mpls play list, and Hybrid is analysing streams, Hybrid crashes. I reproduce the problem with several BD.
It does not happen in this same BD with Hybrid v2017.05,06.1

I send you a debug file.

:thanks:

[Selur]

No debug output here so far,... in case you didn't use djcjs repository make sure that your ffmpeg and mencoder/mplayer builds are build with libbluray.

[Selur]

Okay, got an email now,..
When you call:
"/usr/bin/mediainfo" --Language=en --LogFile="/home/MKV/mediaInfoLog.txt" "--Full" "/home/Data/GET_OUT/BDMV/PLAYLIST/00243.mpls"
Whats the content of '/home/MKV/mediaInfoLog.txt' ?

Cu Selur

[digitall.h]

You were faster answering than I was sending the debug file
:D

I built ffmpeg with libbluray. Here is my version:
ffmpeg version 2.6.git Copyright (c) 2000-2015 the FFmpeg developers
built with gcc 4.9.2 (Debian 4.9.2-10)
configuration: --prefix=/home/ximo/ffmpeg_build --pkg-config-flags=--static --extra-cflags=-I/home/ximo/ffmpeg_build/include --extra-ldflags=-L/home/ximo/ffmpeg_build/lib --bindir=/home/ximo/bin --enable-gpl --enable-libass --enable-libfreetype --enable-libmp3lame --enable-libbluray --enable-libvpx --enable-libx264 --enable-nonfree --enable-x11grab
And it worked OK with previous Hybrid, did you change anything related to ffmpeg?

Is djcjs repository a Debian repo?

Thank you

[Selur]

Is djcjs repository a Debian repo?
Yes.
did you change anything related to ffmpeg?
No. Haven't touched anything related to the analysis and Blu-ray for quite some time.

According to the debug output you send the analysis abort since it doesn't fine enough data from mediaInfo,... (see my previous post)

[digitall.h]

The content of '/home/MKV/mediaInfoLog.txt' is long.
I send you the file by mail.

[Selur]

Seems like mediainfo 00243.mpls has 100 video tracks,.. wtf? -> best try to remux that stuff with something like mkvmerge or similar before feeding it to Hybrid or directly open the right mpls file.

Cu Selur

[Selur]

Also just stumbled over a bug in current libbluray/ffmpeg which seems to cause ffmpeg to throw errors.
-> best remux with mkvmerge/mkvmake to mkv before feeding blu-rays to Hybrid.

[digitall.h]

You're right, I'm sure.
But,
keeping all the rest of elements (bluray and ffmpeg) it works with previous Hybrid version and not with last version. I never needed to remux bluray streams to make it work with Hybrid.
And even there are several mpls playlists, I'm able to choose the correct one.
The problem comes later and it seems to happen when the streams are being analysed.

And just with this last version of Hybrid. With previous one there's no error.

[Selur]

Like I wrote sems to be a bug in libbluray/ffmpeg, assuming it's the same bug I stumbled here upon.
For me ffmpeg freezes on analysis, see: https://ffmpeg.zeranoe.com/forum/viewtopic.php?f=7&t=5368

[digitall.h]

My libbluray/ffmpeg version is working OK with previous Hybrid, and crashes with last Hybrid.
So the bug in libbluray/ffmpeg is just crashing last Hybrid and not older Hybrid.
With no change in libbluray/ffmpeg part of the code between Hybrid versions.

You think this is the reason?

[Selur]

Nope, the libbluray problem I have is unrelated to Hybrid and would also crash on old versions.
No clue whats causing your crash. Would need a debug output level 9 of the analysis that crashes to look into it.

Cu Selur

Ps.: For some Blu-Rays it also helps to enable 'Config->Input->Input Analysis->Use MediaInfo for mpls'.

[digitall.h]

Thank you selur, I will try to also use mediainfo for analysis and post if there's a difference. But as I said, I'm using the same configuration and analysis method in old and new Hybrid, and new crashes where old does not. So my guess is that it is also not related to analysis method.

I will generate a level 9 debug and send you.

I'm sorry for bugging you with this that may be not of general interest...
:o

[Selur]

But as I said, I'm using the same configuration and analysis method in old and new Hybrid
What you call the old version which I basically stopped working the day it was released (which is basically 1/2 year ago). ;)
Basically all the tools I use with Hybrid changed since then and Hybrid itself also gained ~1.5k lines of code and also changed internally thanks to refactoring. ;)
-> References to old versions do not help me at all unless that old version is rather 'new'. :)

I'm sorry for bugging you with this that may be not of general interest...
No clue whether this a bug in Hybrid or your setup since I don't really know whats happening.

In case the only problem is the "MediaInfo found no 'length' info!" popup, for 00243.mpls, it probably can be ignored.
(General problem with Blu-Ray structures is that they often contain dummy tracks which can cause problems for lots of tools. ;))

[digitall.h]

selur,
I just made a reference to the old Hybrid version to explain that my setup is working OK since it works with an older Hybrid version without crashing Hybrid.
And because you said that nothing changed in Hybrid in relation to how other tools are being used.
Tools may be a bit outdated, but are working OK.

I explained myself bad. The mpls selector window pops-up well, and I'm able to select a suitable mpls. Then Hybrid starts stream analysis and it is during this phase that Hybrid crashes. Of course analysis is done with some external tool, but the way now Hybrid send commands to the tools or the way Hybrid interacts with those tools, make it Hybrid crash where oooolder Hybrid version do not crash.
You may send me commands that Hybrid internally send to that tools (libbluray/ffmpeg?) so I can hunt down where the error emerges.

[Selur]

When prefiltering through ffmpeg is enable, Hybrid uses:
ffmpeg -analyzeduration 200M -probesize 200M -i bluray:"PATH TO BLURAY"
to get the usable playlists.
It later analyses the mpls files with mplayer, ffmpeg and mediainfo.
In the debug output you send me, Hybrid aborted with 'Couldn't find any video data with mediainfo!' when analysing the mediaInfo output. (I might have a sanity restriction somewhere, since - like mentioned before - the dummy playlist lists 100 video streams,..)

[digitall.h]

selur,
I tried preselecting mpls with ffmpeg as you suggested.
As expected I got a shorter list of mpls to select from in the mpls selection window, but once a mpls was selected and analysis started Hybrid crashed again. The problem is not in the number of mpls showed in the selection window.

This is the output of
ffmpeg -analyzeduration 200M -probesize 200M -i bluray:"PATH TO BLURAY"
ffmpeg version 2.6.git Copyright (c) 2000-2015 the FFmpeg developers
built with gcc 4.9.2 (Debian 4.9.2-10)
configuration: --prefix=/home/fmpeg_build --pkg-config-flags=--static --extra-cflags=-I/home/ffmpeg_build/include --extra-ldflags=-L/home/ffmpeg_build/lib --bindir=/home/ximo/bin --enable-gpl --enable-libass --enable-libfreetype --enable-libmp3lame --enable-libbluray --enable-libvpx --enable-libx264 --enable-nonfree --enable-x11grab
libavutil 54. 23.101 / 54. 23.101
libavcodec 56. 37.100 / 56. 37.100
libavformat 56. 31.102 / 56. 31.102
libavdevice 56. 4.100 / 56. 4.100
libavfilter 5. 16.101 / 5. 16.101
libswscale 3. 1.101 / 3. 1.101
libswresample 1. 1.100 / 1. 1.100
libpostproc 53. 3.100 / 53. 3.100
bdj.c:253: libbluray-j2se-0.6.2.jar not found.
bdj.c:355: BD-J check: Failed to load libbluray.jar
[bluray @ 0x2c83c80] 6 usable playlists:
[bluray @ 0x2c83c80] playlist 00276.mpls (0:05:26)
[bluray @ 0x2c83c80] playlist 01215.mpls (1:44:05)
[bluray @ 0x2c83c80] playlist 01264.mpls (0:03:39)
[bluray @ 0x2c83c80] playlist 01214.mpls (0:05:28)
[bluray @ 0x2c83c80] playlist 01213.mpls (0:08:50)
[bluray @ 0x2c83c80] playlist 01265.mpls (0:23:03)
[bluray @ 0x2c83c80] selected 01215.mpls
[mpegts @ 0x2c84460] PES packet size mismatch
Last message repeated 1 times
[mpegts @ 0x2c84460] probed stream 30 failed
[mpegts @ 0x2c84460] DTS discontinuity in stream 29: packet 4 with DTS 1707185, packet 5 with DTS 1821786
[mpegts @ 0x2c84460] DTS discontinuity in stream 26: packet 20 with DTS 2101526, packet 21 with DTS 2594862
[mpegts @ 0x2c84460] DTS discontinuity in stream 27: packet 20 with DTS 2101520, packet 21 with DTS 2594869
[mpegts @ 0x2c84460] DTS discontinuity in stream 28: packet 28 with DTS 2596876, packet 29 with DTS 2839001
[mpegts @ 0x2c84460] DTS discontinuity in stream 25: packet 28 with DTS 2596866, packet 29 with DTS 2839010
[mpegts @ 0x2c84460] DTS discontinuity in stream 29: packet 28 with DTS 2596860, packet 29 with DTS 2839017
[mpegts @ 0x2c84460] DTS discontinuity in stream 26: packet 44 with DTS 2979854, packet 45 with DTS 3390708
[mpegts @ 0x2c84460] DTS discontinuity in stream 25: packet 52 with DTS 3392757, packet 53 with DTS 3818644
[mpegts @ 0x2c84460] DTS discontinuity in stream 29: packet 52 with DTS 3392728, packet 53 with DTS 3818673
[mpegts @ 0x2c84460] DTS discontinuity in stream 28: packet 52 with DTS 3392716, packet 53 with DTS 3818685
[mpegts @ 0x2c84460] DTS discontinuity in stream 27: packet 52 with DTS 3392668, packet 53 with DTS 3818732
[mpegts @ 0x2c84460] DTS discontinuity in stream 26: packet 60 with DTS 3820695, packet 61 with DTS 4246562
[mpegts @ 0x2c84460] DTS discontinuity in stream 29: packet 60 with DTS 3820692, packet 61 with DTS 4246565
[mpegts @ 0x2c84460] DTS discontinuity in stream 27: packet 60 with DTS 3820668, packet 61 with DTS 4246588
[NULL @ 0x2cb9060] start time for stream 8 is not set in estimate_timings_from_pts
[NULL @ 0x2cb9ba0] start time for stream 9 is not set in estimate_timings_from_pts
[NULL @ 0x2cba6e0] start time for stream 10 is not set in estimate_timings_from_pts
[NULL @ 0x2cbb220] start time for stream 11 is not set in estimate_timings_from_pts
[NULL @ 0x2cbbd60] start time for stream 12 is not set in estimate_timings_from_pts
[NULL @ 0x2cbc8a0] start time for stream 13 is not set in estimate_timings_from_pts
[NULL @ 0x2cbd3e0] start time for stream 14 is not set in estimate_timings_from_pts
[NULL @ 0x2cd8040] start time for stream 15 is not set in estimate_timings_from_pts
[NULL @ 0x2cd8a60] start time for stream 16 is not set in estimate_timings_from_pts
[NULL @ 0x2cd9480] start time for stream 17 is not set in estimate_timings_from_pts
[NULL @ 0x2cd9f60] start time for stream 18 is not set in estimate_timings_from_pts
[NULL @ 0x2cdb6e0] start time for stream 20 is not set in estimate_timings_from_pts
[NULL @ 0x2cdc220] start time for stream 21 is not set in estimate_timings_from_pts
[NULL @ 0x2cdcd60] start time for stream 22 is not set in estimate_timings_from_pts
[NULL @ 0x2cdd8a0] start time for stream 23 is not set in estimate_timings_from_pts
[NULL @ 0x2cde3e0] start time for stream 24 is not set in estimate_timings_from_pts
[mpegts @ 0x2c84460] Could not find codec parameters for stream 8 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 9 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 10 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 11 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 12 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 13 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 14 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 15 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 16 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 17 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 18 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 20 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 21 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 22 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 23 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 24 (Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
[mpegts @ 0x2c84460] Could not find codec parameters for stream 30 (Unknown: none): unknown codec
Consider increasing the value for the 'analyzeduration' and 'probesize' options
Input #0, mpegts, from 'bluray:/home/Data/GET_OUT/':
Duration: 01:44:04.95, start: 11.623256, bitrate: 39316 kb/s
Program 1
Stream #0:0[0x1011]: Video: h264 (High) (HDMV / 0x564D4448), yuv420p, 1920x1080 [SAR 1:1 DAR 16:9], 23.98 fps, 23.98 tbr, 90k tbn, 47.95 tbc
Stream #0:1[0x1100]: Audio: dts (DTS-HD MA) ([134][0][0][0] / 0x0086), 48000 Hz, 5.1(side), fltp
Stream #0:2[0x1101]: Audio: dts (DTS) ([130][0][0][0] / 0x0082), 48000 Hz, 5.1(side), fltp, 768 kb/s
Stream #0:3[0x1102]: Audio: dts (DTS) ([130][0][0][0] / 0x0082), 48000 Hz, 5.1(side), fltp, 768 kb/s
Stream #0:4[0x1103]: Audio: dts (DTS) ([130][0][0][0] / 0x0082), 48000 Hz, 5.1(side), fltp, 768 kb/s
Stream #0:5[0x1104]: Audio: dts (DTS) ([130][0][0][0] / 0x0082), 48000 Hz, 5.1(side), fltp, 768 kb/s
Stream #0:6[0x1105]: Audio: ac3 (AC-3 / 0x332D4341), 48000 Hz, stereo, fltp, 192 kb/s
Stream #0:7[0x1106]: Audio: ac3 (AC-3 / 0x332D4341), 48000 Hz, stereo, fltp, 192 kb/s
Stream #0:8[0x1200]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:9[0x1201]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:10[0x1202]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:11[0x1203]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:12[0x1204]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:13[0x1205]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:14[0x1206]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:15[0x1207]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:16[0x1208]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:17[0x1209]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:18[0x120a]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:19[0x120b]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090), 1920x1080
Stream #0:20[0x120c]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:21[0x120d]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:22[0x120e]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:23[0x120f]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:24[0x1210]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090)
Stream #0:25[0x1211]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090), 1920x1080
Stream #0:26[0x1212]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090), 1920x1080
Stream #0:27[0x1213]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090), 1920x1080
Stream #0:28[0x1214]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090), 1920x1080
Stream #0:29[0x1215]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090), 1920x1080
No Program
Stream #0:30[0x1fff]: Unknown: none
At least one output file must be specified

[Selur]

okay,...
bdj.c:253: libbluray-j2se-0.6.2.jar not found.
bdj.c:355: BD-J check: Failed to load libbluray.jar
a. looks like there is something amiss with the libbluray build.
b. can't say anything about the crash without a debug output, but you should now that by now

[digitall.h]

Ok.
Re to b., it was just a ffmpeg call. There was no Hybrid there.
Or do you refer to the crash after changing the gui behaviour with 'preselecting mpls'? This did not change anything: I selected the mpls as I was doing before and Hybrid crashed during analysis phase.

Re to a., yes, libbluray.jar missing. This warning has always been there. Something related to playing BD java menus, not with analysing streams.
And I have always been able to use Hybrid with this libbluray/ffmpeg version showing this error.

So, in your opinion, this may be the error that makes last Linux Hybrid version crash.

[digitall.h]

Hi,
I'm looking at djcj repository.
I tried to ask in selur forum but I'm afraid I'm not a member anymore, most probably due to inactivity (sorry for that).

I'm not sure djcj is a Debian repository, but Ubuntu ppa.
Please correct me if wrong and please tell me how to add an Ubuntu ppa to Debian.
It is usually risky to add a ppa to Debian because differences in versions and basically the rest of the system.
I did not understand what hybrid-tools package will install afterwards, if it will install ffmpeg, and if it will conflict with avconv that is what Debian selected instead of ffmpeg.

If this is not the right place for these questions, please make me know and point me in the right direction.

[Selur]

a. I usually only delete 0-posters,... the forum moved a while back from http://forum.selur.de (read only now) to https://forum.selur.net/ so you might simply never have created a user at selur.net.
b. you are right it's a Ubuntu repository, not sure whether they work on debian (might be worth a try). About adding a ppa to debian,.. here's the first hit when I google that: http://www.webupd8.org/2014/10/how-to-add-launchpad-ppas-in-debian-via.html
c. problem isn't that Hybrid doesn't work with avconv, problem is that avconv and all vanilla build from the ubuntu/debian main repositories are usually not compiled with libass, libbluray, etc. -> unless you compile a lot of dependencies manually (or have someone like djcj doing it for you) you simply lack proper tools for Hybrid to work with
d. that djcj decided to download the tools with a small wrapper tool was his decision.
-> best ast djcj about this,... if I remember correctly he also has a user account in this board.

Cu Selur

[digitall.h]

b. I'm sorry I explained it bad.
I know how to add a ppa to Debian. I did it for mkvmerge for instance.
When it comes to single applications it is not problematic, they may not work properly if they depend in some different version of required tools, but they do not interfere in general system.
But a package that adds ffmpeg to the system I'm afraid that may conflict with avconv. I will do as you suggest and ask djcj.

And yes, I compiled ffmpeg myself because vanilla avconv was not compiled with bluray.
:thanks:

[Selur]

-> you might also want to compile mplayer&mencoder :)

[digitall.h]

They have always been working OK for me.
I never felt the need.
:rolleyes:

[Selur]

Hybrid rev 2017.12.23.1:
*fixed*

Vapoursynht: Waifu2x call
tools: accept both 'yes' and '1' during feature check for NVEncC and QSVEncC
small problems with filter preview of Avisynth and Vapoursynth
x265: fixed cuTree <> cuLossless connection
preview: ccd avisynth filter preview
Avisynth: DeNoise/-Block 'before' fixes

*changed*

analysis: ffmpeg probesize now 200M
Vapoursynth: in preview use ClipInfo stead of FrameProperties
mkv: timecodes start differently with mkvmerge 18.0+
Read MasterDisplay and Max CLL/FALL with MediaInfo copy to NVEncC and x265.
changed: FFV1 use bitDepth of input

*added*

Vapoursynth: option to change filter category order
Vapoursynth: added experimental 'HDR10ToSDR' option (https://forum.doom9.org/showthread.php?t=174415)
Vapoursynth: added BlindDeHalo3
Vapoursynth: openCl for QTGMC, DAA, Santiang
Vapoursynth: added SMDegrain
Avisynth: added BalanceBorders
Avisynth: added NNEDI3CL to Yadifmod
Avisynth: added GPU support for NNEDI3 resizing using NNEDI3CL instead of NNEDI3
Avisynth: added SMDegrain
Avisynth: Mpeg2Source postprocessing option


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: This is the last release which supports 32bit Windows. (Folks, it's really time to ditch 32bit OS&machines)

[Selur]

Hybrid rev 2017.12.24.1:
*fixed*

Stupid bug which broke mplayer/mencoder calls handling m2ts&co files.

*added*

Vapoursynth: FixTelecinedFades
Avisynth: Preview got 'scrolling' option, which zooms to 100% and adds scrollbars.


-> downloads: http://www.selur.de/downloads (http not https)

Cu Selur

Ps.: This is the last release which supports 32bit Windows. (Folks, it's really time to ditch 32bit OS&machines)

[Havokdan]

Hybrid rev 2017.12.24.1:
*fixed*

Stupid bug which broke mplayer/mencoder calls handling m2ts&co files.

*added*

Vapoursynth: FixTelecinedFades
Avisynth: Preview got 'scrolling' option, which zooms to 100% and adds scrollbars.


-> downloads: http://www.selur.de/downloads (http not https)

Cu Selur

Ps.: This is the last release which supports 32bit Windows. (Folks, it's really time to ditch 32bit OS&machines)

Problem with me:

[Selur]

working on it,.. I guess, try download in 20min again :)

[Selur]

updated download, hopefully that fixes the issue,.. if not please 'ctr+c' the content of any popup and post it since it might take a few days for an attachment to get approved here.

[Havokdan]

https://i.imgur.com/ImK7GuP.png

[Selur]

Redownload should work now. :)

[Havokdan]

Redownload should work now. :)

Works now, but:

https://i.imgur.com/biZoOHZ.png

[Selur]

Not sure what you are referring to,.. for me the it looks like this:
https://s14.postimg.org/6ohmlwzy7/main.png
(on Windows 10 64bit; 5k display)

Cu Selur

Ps.: I witched to the latest Qt version (Qt 5.10) but didn't change much else,..

[Havokdan]

I was referring to the Windows 98 look.

[Selur]

Seems like the Qt folks decided to change the default styling on at least Win7.
It didn't change on Win10, which is why I didn't notice it.
As a workaround I will add an option in upcoming releases to switch the Window style of Hybrid.
You yourself can also test it by trying to call Hybrid with:
1.
--style windowsvista
2.
--style Windows
3.
--style Fusion

'Windows Vista' seems to be the default style I normally see under Win10.
Side note: On Linux and Mac are other styles available.

->
a. What OS are you using (details please)? (so I know where I can test)
b. Does it help to switch the window styles and which styles do change the layout for you?

Cu Selur

Ps.: Seems like tool-tips of image-buttons are black on some styles. -> looking into that atm.

[Havokdan]

Windows 10 x64 1709 build 16299.125, intel hd 4000 and nvidia gtx 670mx (notebook)

[Selur]

Okay, since you send my via PM that 'windowsvista' isn't available on your system: Did you change the default design in Windows? On my Win10 64bit the view didn't change.

[Havokdan]

Okay, since you send my via PM that 'windowsvista' isn't available on your system: Did you change the default design in Windows? On my Win10 64bit the view didn't change.

Default, only change dpi to 100%

[Selur]

Strange no clue why the style isn't available then. -> looking into it

[Selur]

Think I found the problem. :) They moved "windowsvista" into a separate dll, which on my dev system is loaded by default,.. -> working on it

[locotus]

Strange no clue why the style isn't available then. -> looking into it

Also last update (12-24)crash on start complaiming about
QT plugin "windowsin". No problem with 12-23 update.
Running W10, with creator's update.

Thanks.

[Selur]

Just downloaded the version from my homepage and installed it on my Win10 notebook.
No crash here on startup. -> please redownload, try again and report back. Thanks

Cu Selur

[locotus]

Just downloaded the version from my homepage and installed it on my Win10 notebook.
No crash here on startup. -> please redownload, try again and report back. Thanks

Cu Selur

Thanks, that did the trick.

[Selur]

Hybrid rev 2017.12.25.1:
*fixed*

gui: Windows style problems due to missing dll
Avisynth: SMDeGrain version mixup
tooltip: black tooltips on window style change
SystemTrayIcon not hidden, when not minimized
Avisynth: SMDegrain, loading LSMod for contrastsharp != 0
x265: quantizer group minimum 8


-> downloads: http://www.selur.de/downloads (http not https)

Cu Selur

[jonesjrgr]

Hi to everyone!!!
i have i huge problem with the latest version of Hybrid (2017.12.25.1)..It crashes everytime i try to open a HDR file..It doesn't happen with other files (x264,x265 etc..).Just only with HDR (x265) files.I'm frustrated because i wanted to try this experimental HDR2SDR feature..I can't get an error/debug file,because it crashes in the beginning.Oh,i checked with a previous version of Hybrid (2017.10.5.1),and it works flawlessly (without THAT feature)...I un/installed both version,cleanup registry,but nothing....The only thing i noticed,is the number of files between the 2 versions (Older version had so many more)..Anyway,thanx in advance for any help.

[Selur]

Will probably release a new version tonight which addresses this, problem is that I didn't know that mediainfo also reported 'Display P3' as colorimetric for HDR.

[Selur]

Hybrid rev 2017.12.27.1:
*fixed*

crash when loading HDR source with 'Display P3' colormetric, assuming D3D65 and not theater for the time being.


-> downloads: http://www.selur.de/downloads (http not https)

Cu Selur

[Alex_A]

Hybrid rev 2017.12.27.1:
*fixed*

crash when loading HDR source with 'Display P3' colormetric, assuming D3D65 and not theater for the time being.


-> downloads: http://www.selur.de/downloads (http not https)

Cu Selur

Instead of version 2017.12.27.1, the old version of 2017.12.25.1 was installed

[Selur]

Argh,... will uppdate the download in ~20min, should be fine then.
(updated now, thanks for reporting)

Cu Selur

[jones1913]

Hey, since ver 2017.12.25.1 Hybrid looks odd here on Manjaro-Linux:
https://abload.de/thumb/hybrid_def0psge.png (http://abload.de/image.php?img=hybrid_def0psge.png)



When I downgrade to 2017.12.23.1 all is normal:
https://abload.de/thumb/hybrid_ok7rsm3.png (http://abload.de/image.php?img=hybrid_ok7rsm3.png)

I have tried with --style parameter but that has no effect. Other qt5 applications look totally normal here.

[Selur]

Hybrid now takes the first available style, you can change the style under Config->View->Style.
What styles are listed there for you and which does look 'normal'? :)

Cu Selur

[jones1913]

Ok, works now. The new "style" dropdown was way out of my screen borders, but I found it now.
First style is "bb10dark" which is somehow defect. When I set it system wide all looks like crap.

Other styles are: bb10light, cleanlooks, gtk2, cde, motif, plastique, qt5ct-style, Windows, Fusion

Wouldn't it be the best to use default system style instead first in the list (like it was in the past)?

[Selur]

Wouldn't it be the best to use default system style instead first in the list (like it was in the past)?
May be, added it for testing and since it didn't cause any trouble on my systems, I left it in. :)

[jones1913]

:thanks: for making this cross platform application, btw.

[Selur]

Looking for a few volunteers which use Windows 64bit (Win7/8/Vista/10) to test whether deinstalling the current Hybrid version they use and installing a current dev version, in which I'm testing a new installer (InnoSetup), produces a working Hybrid installation.
I'm especially interested whether this works on Windows installations which use unicode&co (Russian, Arabic, Chinese) systems.
For those those that are interested please try: --deleted---rl] (will delete the file in a few days)

-> if you test the build please report any problems together with what OS and language combination triggered the problem. Thanks!

Cu Selur

[Magik Mark]

Selur,

"Multipass opt analysis" has been renamed to:

--analysis-save <filename> Dump analysis info into the specified file. Default Disabled
--analysis-load <filename> Load analysis buffers from the file specified. Default Disabled

Can we have a work around?

Thanks

[Selur]

Can we have a work around?
If you can help me understand this. :)

Looking at the documentation:
http://x265.readthedocs.io/en/latest/cli.html?highlight=analysis-save#cmdoption-multi-pass-opt-analysis
http://x265.readthedocs.io/en/latest/cli.html?highlight=analysis-save#cmdoption-multi-pass-opt-distortion
http://x265.readthedocs.io/en/latest/cli.html?highlight=analysis-save#cmdoption-analysis-save
http://x265.readthedocs.io/en/latest/cli.html?highlight=analysis-save#cmdoption-analysis-load
http://x265.readthedocs.io/en/latest/cli.html?highlight=analysis-save#cmdoption-analysis-reuse-file
and your statement that
"Multipass opt analysis" has been renamed
I'm confused :)

How I understood it so far was:
a. you enable '--multi-pass-opt-analysis' and/or '--multi-pass-opt-distortion' and x265 will use file named 'x265_analysis.dat' and save/load it into/from the current working directory.
b. your can use '--analysis-reuse-file' to specify specify a custom file name and location for the analysis file to use.
So Hybrid set for example:
--multi-pass-opt-analysis --multi-pass-opt-distortion --analysis-reuse-file "theData.analysis"
during both paths of a 2pass encoding.

Now, alternatively to using '--multi-pass-opt-analysis' and/or '--multi-pass-opt-distortion' you can now use:
'--analysis-save <filename>' and '--analysis-load <filename>' together with '--analysis-reuse-level <1..10>' to determine which data is save/loaded.
So Hybrid would need to set for example:
'--analysis-save "theData.analysis" --analysis-reuse-level 5
during the first pass and
'--analysis-load "theData.analysis" --analysis-reuse-level 5
during the 2nd pass, if it would support this during 2pass encoding.

So how I understood it, was that:

Analysis save/load was ment as a speed-up option in case you wan to reuse analysis data. Which only seems like a good idea if you reencode the same source twice or use 2pass without 'Fast 1st pass'.
The other two options are ment to optimize the analysis and distortion calculation.

Since I thought that the analysis load/save thing wasn't something that was really useful I haven't supported it so far.
(May be I'll add support for it in case 2pass without 'Fast 1st pass' is used. Don't think I'll add support for it during 1pass encoding.)

-> How did you understand the documentation? For me it doesn't look like a renaming,.. So I'm unclear how this should be properly used. :)

Cu Selur

Ps.: I suspect that 'analysis-reuse-mode' which is referred to in the documentation means 'analysis-save'/'analysis-load'.

[Magik Mark]

The “renamed” word came from x25 forum. I think one of the developers said that. Maybe we could do some experiments ourselves and see what happens


Sent from my iPhone using Tapatalk

[von Suppé]

Hi Selur

Would it be possible to place the temp-files for each job in a separate folder (in the default temp path) which is named after the input file?

Secondly, can it be that the latest version of Hybrid on the Videohelp.com download page is the wrong one? Carefully taking my first steps with UHD x265 encoding, "Hybrid stopped working" when importing HDR video. The version of your own downloadlink works though. Running Win7 64bit.

Much obliged

[DJ-1]

Hi, I've been using hybrid to extract subtitles from bluray source..
Result is a .Sup
But not everything likes one subtitle format..
Is it possible to 'extract + convert' with Hybrid & I just specify a format .Pgs for instance
?
Thanks.
(yeah these are external subtitles)

Sent from my HTC_M10h using Tapatalk

[von Suppé]

PGS is the format used on BD. Demuxing them will give you a SUP file, which is image-based.
If you want to convert them into text-based subtitles you'll have to use OCR. I don't believe Hybrid does that on the fly.
I can't think of a tool that can do this all automatically. Human error-correction is always necessary due to "reading faults".
If you want to give this a try, take a look at SubtitleEdit. It has a OCR function but as said, you gotta be there during process.

[x265]

Hi. When i open .d2v files in hybrid it gets stuck and i get the message filtering input. There was a field order transition that was fixed by dgindex. How do i open .d2v files?

https://i.imgur.com/eG4arCv.png

[von Suppé]

Hi selur,

This weekend I'm gonna try out a hardware accelerated/supported x265 encode on a friend's pc. He has a nVidia GTX 1080Ti videocard.
Since this is my first try, is there any extra drivers or software to install before I go encoding and need I pay attention to other things?

cheers
von Suppé

[Selur]

@Suppé: x265 doesn't support any gpu based or assisted encoding atm.
@x265: Hybrid doesn't support .d2v as Input.

[Selur]

Hybrid rev 22018.01.28.1:
*fixed*

x265: only allow strictCBR when vbv/profile&level is set
x264: rgb input handling
Vapoursynth: Hysteria call
Subtitle: add font
View: NVEncC progress indication

*added*

Avisynth: support for SuperToon
Avisynth: high bit depth handling
Vapoursynth: using vsnedi3 instead of nnedi3 now
Vapoursynth: support .vpy input
Vapoursynth: tonemap filter
Input: support image sequences as input
Encoder: experimental support for videotoolbox - gpu encoding on mac
filtering: added experimental support for ffmpegs toneMap, note that Mplayer preview will not work with this, since Mplayer doesn't support toneMap. Better use toneMap through Vapoursynth.

*changed*

container: make xps_inband optional for mp4, for Quicktime playback compatibility
264: assume unified build starting with r289
Windows: file locations and installer -> Windows users need to deinstall old versions completely


-> downloads: http://www.selur.de/downloads

Cu Selur

[von Suppé]

Thanks for the update Selur.

Does it mean that, from now on, every time a new version comes out, Windows-users always have to deinstall the old version first?
Or is it just a one time thing for now?

[Selur]

Hybrid rev 22018.01.28.2:
*fixed*

x265: typo strctCbr -> strictCbr

-> downloads: http://www.selur.de/downloads

--------------------
Or is it just a one time thing for now?
Just this time since the location of the files changed, I'm using a new installer and the to account for that some internals needed to change.
-> That the settings need to be deinstalled doesn't happen normally unless some radical changes inside the code had to be made.

Cu Selur

[Havokdan]

Hybrid rev 22018.01.28.2:
*fixed*

x265: typo strctCbr -> strictCbr

-> downloads: http://www.selur.de/downloads

--------------------

Just this time since the location of the files changed, I'm using a new installer and the to account for that some internals needed to change.
-> That the settings need to be deinstalled doesn't happen normally unless some radical changes inside the code had to be made.

Cu Selur

I did a clean install of the previous version, will I have to do this?

[x265]

hybrid is crashing for me with this script.

copypaste from vapoursynth script view


# Imports
import os
import sys
import ctypes
Dllref = ctypes.windll.LoadLibrary("C:/Program Files/Hybrid/64bit/vsfilters/Support/libfftw3f-3.dll")
# Loading Support Files
Dllref = ctypes.windll.LoadLibrary("C:/Program Files/Hybrid/64bit/vsfilters/Support/OpenCL.dll")
import vapoursynth as vs
core = vs.get_core()
# Import scripts folder
scriptPath = 'C:/Program Files/Hybrid/64bit/vsscripts'
sys.path.append(os.path.abspath(scriptPath))
# Loading Plugins
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/GrainFilter/AddGrain/AddGrain.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/DenoiseFilter/FFT3DFilter/vsfft3dfilter.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/DenoiseFilter/DFTTest/DFTTest.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/DenoiseFilter/KNLMeansCL/KNLMeansCL.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/EEDI3.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/ResizeFilter/nnedi3/vsznedi3.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/libmvtools.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/temporalsoften.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/scenechange.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/fmtconv.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/SourceFilter/LSmashSource/vslsmashsource.dll")
# Import scripts
import havsfunc
# Loading C:\Users\Kannan\Desktop\title00.mkv using LWLibavSource
clip = core.lsmas.LWLibavSource(source="C:/Users/Kannan/Desktop/title00.mkv", format="YUV420P8", cache=0)
# making sure input color matrix is set as 470bg
clip = core.resize.Point(clip, matrix_in_s="470bg")
# Making sure input color range is set to TV (limited) range.
clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)
# with preset 'Very Slow' and 'Placebo' QTGMC is limited to 8bit YUV/GRAY only due to FF3dFilter
# setting field order to what QTGMC should assume
clip = core.std.SetFrameProp(clip=clip, prop="_FieldBased", intval=2)
# Deinterlacing using QTGMC
clip = havsfunc.QTGMC(Input=clip, Preset="Placebo", TFF=True, opencl=False)
# cropping the video to 690x410
clip = core.std.CropRel(clip=clip, left=16, right=14, top=82, bottom=84)
# Output
clip.set_output()




log

Creating job for current source
-> Creating jobs for source 1,...
-> Generating calls for: C:\Users\Kannan\Desktop\movie.mkv
adding x264 calls for source: 1
createJobs for C:\Users\Kannan\Desktop\movie.mkv
creating attachment extraction subjobs
creating video subjobs
creating muxing subjobs
optimizing the subJobs
-> finished creating subjobs for current source
added new job with id 19_19_22_5510
-> should be there after processing: C:\Users\Kannan\AppData\Local\Temp\19_19_22_5510_03.264
-> 19_19_22_5510_03_video crashed: set absolute paths crashed,...
Aborting '19_19_22_5510_03_video' due to:
Job 19_19_22_5510 finished!

[Selur]

Hybrid rev 2018.02.03.1:
*fixed*

Vapoursynth: missing '"'-around fontDir when handling vapoursynth subtitles
Avisynth: fixed some small stuff
Video: small bug in fps<>fraction conversion

*added*

audio: option to remove negative delays through ffmpeg (cutting)


-> downloads: http://www.selur.de/downloads

--------------------
@x265: video crashed: set absolute paths crashed,...
Seems like a tool is missing. Wild guess: You didn't deinstall your previous Hybrid install properly.


Cu Selur

[x265]

hi selur. Can you add ASTDR derainbow script to next release of hybrid?

[Selur]

This one http://avisynth.nl/index.php/ASTDR ? Have you tried it? Is it any good?
Next release is planned for this weekend,..
Not much documentation. -> definitely not in the next release
Can you provide the min&max values for strength, tempsoftth, tempsoftrad, tempsoftsc, blstr, tht, FluxStv, dcm, thSAD? (otherwise this will have to wait till I find the time to look all these up in all the different filters it uses etc. -> no clue when/if that will happen)

[Selur]

Hybrid rev 2018.02.18.1:
*fixed*

Vapoursynth: resize + crop preview
Input: crash when first loading a file and then enabling DVD input
Avisynth: Preview not working for content with color matrix 'fcc'
x264: removed unnedded empty cleanUpCall

*added*

Vapoursynth: FastLineDarken, Toon, ResampleHQ, GradFun3, Flash3kDeband

*changed*

vpx: adjusted to vpx 1.7
Vapoursynth: using AssumeFPS to set frame rate on input to handle 'prefer original frame rate'
Input: FFmpeg GPU decoding make thread limitation optional



-> downloads: http://www.selur.de/downloads

Cu Selur

[Magik Mark]

Selur,

Just wondering, does the app offer:

1. Auto Crop feature for letter boxed videos?
2. awarpsharp2

Thanks

[Selur]

regarding 1.: Letterboxing is done by the Letterbox option, not the auto cropping
regarding 2.: yes (Filtering->Vapoursynth->Sharpen->Sharpener ; Filtering->Avisynth->Sharpening->Sharpening)

[Magik Mark]

Thanks. Are all the filters x64 but already?


Sent from my iPhone using Tapatalk

[Selur]

That depends whether Avisynth or Vapoursynth is used:

Avisynth (MT): 32bit
Vapoursynth: 64bit


Cu Selur

[x265]

This one http://avisynth.nl/index.php/ASTDR ? Have you tried it? Is it any good?
Next release is planned for this weekend,..
Not much documentation. -> definitely not in the next release
Can you provide the min&max values for strength, tempsoftth, tempsoftrad, tempsoftsc, blstr, tht, FluxStv, dcm, thSAD? (otherwise this will have to wait till I find the time to look all these up in all the different filters it uses etc. -> no clue when/if that will happen)

Yeah that's the one. I will try to find the values and send them to you when i can.


How do i update hybrid from within the app itself? when i click "check for release" under tools section it just says "new version available". Is it possible to update hybrid automatically?

[Selur]

No there isn't a way to update Hybrid, mainly because managing the whole thing is too much of a hassle. (maintaining server, compiling and updating tools,...)
The 'check for release' button simple checks the current installed version against the version.txt on the server. :)

Cu Selur

[phazer11]

Selur. First off thanks for being so helpful. Could I bother you to add Anime IVTC to your filter lists please?
http://avisynth.nl/index.php/AnimeIVTC

Also perhaps you could add an advanced mode where we could change some of the values. For example, SMDegrain has a built in prefilter mode of 4 which is a built in KNLMeans denoiser, but as it is we can't change the script values.

[Selur]

Could I bother you to add Anime IVTC to your filter lists please?
Looking at the parameter count:
AnimeIVTC (clip i, int "mode", int "aa", int "precision", int "killcomb", int "cache", bool "ifade", float "sfthr"*, bool "sfshow"* , bool "chrfix" , bool "blend" , bool "normconv" , int "pattern" , int "pass", bool "rendering" , int "bbob", int "cbob", string "edimode", int "degrain", int "omode" , int "cthresh"* , int "blockx"* , int "blocky"* , int "MI"* , int "tfmm"* , int "pp"* , int "metric"* , int "micmatching"* , int "i1", int "i2", int "e1", int "e2", int "e3", int "p1", int "p2" , bool "dchr"* , bool "palf"* , bool "tcfv1"* , bool "nvfr"* , bool "real30p"* , bool "autoAssuf"* , int "ediandnn"* , bool "o3025cfr"* , int "overlap", int "pel", int "search", bool "nnedi3pel"** , string "credconv" , bool "mode22"* , clip "ediext"* , string "bob4p"* , clip "extbob"* , string "extbobf"* , bool "repwithtdeint"* , float "dark", int "thin", int "sharp", int "smooth", bool "stabilize", int "tradius", int "aapel", int "aaov", int "aablk", string "aatype")
I'm not planning to add that. :) Might think about it in case there ever is a Vapoursynth version for it.

Also perhaps you could add an advanced mode where we could change some of the values.
When Hybrid isn't offering an option it is normally because of on of thre reasons:
a. I didn't know about the option
b. I couldn't find an explanation for it
c. I didn't come up with a way how to add it into the gui or I didn't think anyone would use the option. :)
-> if you think Hybrid is lacking an option, let me know and the chances are good that I'll add support for it.
For example, SMDegrain has a built in prefilter mode of 4 which is a built in KNLMeans denoiser, but as it is we can't change the script values.
Didn't add mode 4 because it wasn't mentioned over at http://avisynth.nl/index.php/SMDegrain -> will add support for it :)

Cu Selur

[phazer11]

Yeah it was a sneaky add-in for SMDegrain they only really mention it in the change log on that page in the section for 3.1d and 3.1.1d

[mytbyte]

I need help with Hybrid encoding (via x264) to a constant bitrate (of say 25 mbps). Any of the options available, no matter what bitrate or filesize or factor I specify produces higher than a 3 mbps video. It's a highly compressable flat color animaton with little delta between frames, but I need at least 25 mbps overall to be accepted for projection. Any ideas? Thx.

[Selur]

Any of the options available, no matter what bitrate or filesize or factore I specify produces higher than a 3 mbps video.
Assuming that statement is correct (which I highly doubt) and even lossless encoding or correctly setting vbv values and bitrate/crf doesn't help bitrate stuffing might be an option.
So you could try cbr output and m2ts as container, Hybrid then would use tsMuxeRs cbr option which should add stuffing to the output.

[sneaker_ger]

Or create noise. :devil:

[Selur]

Yup altering the source itself by adding noise resizing etc. should also change the bitrate requirement.

[mytbyte]

Thx. Noise is not an option i.e. re-rendering of any kind ...I did check cbr-hd option but it's not inserted into the command line...also help says something about it requires bitrate control. is there something else I need to click or set?

haven't tried lossless though but counting on 4:4:4 playback is most likely not an option...

P.S. I can't get higher bitrate than 2-3 mbps in MPE as well...

[Selur]

requires bitrate contril.
what? never heard of it. not sure at the requirements of it atm. you can try adding it as a custom addition '--nal-hrd cbr',...

[sneaker_ger]

Try
--vbv-bufsize 25000 --filler
or
--vbv-bufsize 25000 --nal-hrd cbr


Just --filler or --nal-hrd cbr without --vbv-bufsize isn't enough.

[Selur]

Did a quick check,there's a bug in Hybrid which is why this option can't be used other than through a custom addition atm. (will fix for the next release)

[mytbyte]

@selur: thx, never needed to alter the command line so can't quicky locate how to do it :D

Ctrl-F10 for ctr-hd says "Requires bitrate mode ratecontrol" but I guess it just means it can't be used with CRF encoding mode...

[Selur]

not sure if that is still right, should probably work with crf too, test it and the whether it helps.
'x264->Misc->Import/Additions->Addition' is where you can add custom additions to the command line

[x265]

Hybrid goes beyond 100% with the current source (dvd) that i'm reencoding.

script:

SetMemoryMax(768)
SetMTMode(5,4) # changing MT mode
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\LoadDll.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\AddGrainC.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\dfttest.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\EEDI2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\eedi3.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\FFT3DFilter.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\masktools2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\mvtools2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\nnedi.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\nnedi2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\SSE2Tools.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\TDeint.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\VerticalCleanerSSE2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\PlanarTools.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\nnedi3.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\RgTools.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\TIVTC.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\Average.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\GRunT.dll")
LoadCPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\yadif.dll")
LoadDll("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\fftw3.dll")
LoadDll("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\LIBFFT~1.DLL")
Import("C:\Program Files\Hybrid\32bit\avisynthPlugins\QTGMC.avsi")
Import("C:\Program Files\Hybrid\32bit\avisynthPlugins\SMDegrain.avsi")
Import("C:\Program Files\Hybrid\32bit\avisynthPlugins\AnimeIVTC.avsi")
Import("C:\Program Files\Hybrid\32bit\avisynthPlugins\Srestore.avsi")
LoadCPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\ffms2.dll")
# loading source: C:\Users\kyy\Desktop\brs\cycl\title00.mkv
# input luminance scale tv
FFVideoSource("C:\Users\Kannan\Desktop\MYENCO~1\cyclist\title00.mkv")
# deinterlacing
SetMTMode(2) # changing MT mode
AssumeTFF()
QTGMC(Preset="Very Slow", ediThreads=2)
# removing ghosting
srestore(frate=25.000,omode=6)
# cropping to 694x476
Crop(4,2,-22,-2)
# current resolution: 694x476
# filtering
return last


https://i.imgur.com/35JzEZ3.png


Another thing can you add an option to change the denoiser used with qtgmc?. It used fft3dfilter by default. I would like to use knlmeanscl and sometimes Dfttest.

[Selur]

Hybrid goes beyond 100% with the current source (dvd) that i'm reencoding.[/quote]
your source is an mkv not a dvd,... From the looks of it my guess is that Hybrid got the frame count somewhere wrong. Can't really tell from the avisynth script.

[code]Another thing can you add an option to change the denoiser used with qtgmc?
Depending on the preset used QTGMC uses fft3dgpu or dfttest:
# Noise presets Slower Slow Medium Fast Faster
Denoiser = default( Denoiser, Select( npNum, "dfttest", "dfttest", "dfttest", "fft3df", "fft3df" ) )
I could add the 'Denoiser' option which would allow to switch between "dfttest", "fft3df", and "KNLMeansCL" when using the 'Custom' preset selection.

Cu Selur

[xueyue]

16302
This is my BUG feedback
The display inside is normal，But your software cannot be used.https://wx3.sinaimg.cn/mw690/0071U8zSgy1fpzr9ndiulj314o0i8mz3.jpg
However, the vapoursynth file I generated with your software is not usable.
https://wx1.sinaimg.cn/mw690/0071U8zSgy1fpzrbbfmn3j30my0dcwem.jpg
Please give me some help，My English is very poor, the above words are translated by Google.

[xueyue]

I tried to use Hybrid_2018.02.18.1,2018.02.03.1,2018.01.28.2.
But it still not worked and prompted a problem：Didn't find a 'frame rate' info node!
I don't konw how to solve this error,Because my scripts are correct.
Please give me some help if you can see.:thanks:

[Selur]

From the looks of it you are feeding Hybrid with a vapoursynth script which it can't process properly, but I will have to wait foe a moderator to approve the attachment before I can look into it.
What output do you get when you call:
"Path to vspipe" --info "Path to your script" -
(with the necessary replacements)

The Python exception probably happens since you load libawarpsharp2 inside your script and also enabled it in Hybrid,..

[xueyue]

My vapoursynth script is:
import vapoursynth as vsimport sys
import havsfunc as haf
import mvsfunc as mvf
core=vs.get_core()
c=core.lsmas.LWLibavSource(source=r"E:\111_output.mp4")c=haf.MCTemporalDenoise(c,settings="high")
c.set_output()
When i run this scrip:
The hybrid return:Didn't find a Â’frame rateÂ’ info node！
The HybridDebugOutput say:
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: VPYAnalyser calling:"D:\Hybrid\64bit\Vapoursynth\VSPipe.exe" --info C:\Users\19960824\Desktop\test.vpy
-2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: VPYAnalyser - process finished()
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: -> Analyzing collected data...
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: Analyzing output:
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: analyzing video data,...
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: -> found video_bitrate: 0.000
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: -> found video_length: nan
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: -> found video_aspectRatio: 1
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: -> found video_format: raw
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: VPYAnalyser finished, mode 0.
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: -> Found no infos for Color.
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: -> Setting Format to 'raw'
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: -> Setting Bit rate to '0.000'
2018.04.03 - 20:25:40_Windows 10-64bit_2018.02.18.1 - level 4: -> Found no infos for Frame rate.
My havsfunc’s version is R29,vapoursynth’s version is R43,L-SMASH–works is R929.and my system is Windows10 home 1709.All of the above are 64-bit.
When i use other scripts which are always get the same error,such as svpflow.
Please help me and thank you very much.

[Selur]

So what happens if you call:
"D:\Hybrid\64bit\Vapoursynth\VSPipe.exe" --info C:\Users\19960824\Desktop\test.vpy
inside a Windows command prompt?

Either Hybrid parses that output wrong, or vspipe thinks something is wrong with your script.
Oh, I think I see the problem. It looks like you are feeding a script that seems to use a globally installed Vapoursynth script rather than the portable one that comes with Hybrid.
So your script relies on the filters from your installed Vapoursynth which Hybrids Vapoursynth has no access to.
-> Adjust your script to to not rely on the auto loading and you should probably be fine.

Cu Selur

[Selur]

Hybrid rev 2018.04.08.1:
*fixed*

x264: cbr signaling
x265: only use radl with fixed gop size
Avisynth: fixed TemporalDegrain HQ != 0
Avisynth: QTGMC 'Very Slow' and 'Placebo'
Subtitle: srt to ass
Subtitle: srt overwrite styling
Exraction: DGIndex extract audio from DVD
Audio: channel count on custom downmix
Vapoursynth: fixed d2v indexing call
Vapoursynth: fixed Preview aspect ratio bug
Vapoursynth: fixed f3kdb.Deband parameter names
Filter: tonemap call when using ffmpeg (no Vapoursynth or Avisynth)
Decode: use ffmpeg32 when converting to UT and filtering with Avisynth
Jobs: crash when output name contains tool name

*added*

Vapoursynth: support for MCTemporalDenoise
Vapoursynth: support for PSharpen
Vapoursynth: support for MCDegrainSharp
Vapoursynth: support for EdgeCleaner
Vapoursynth: support for TTempSmooth
Vapoursynth: support for LSFMod
Avisynth: support for DeScratch, DeSpot, DeFlicker
Resize: Allow to crop uneven. Note that this will enforce a conversion to RGB or 4:4:4 which might cause a slow down depending on the rest of the filter chain.
x265: support for 'extend-gop' and 'gop-lookahead' option
Vapoursynth: Mac now supports the 'VapourSynth macOS Installer' (https://forum.doom9.org/showthread.php?t=173453). Special thanks to l33tmeatwad for his help! see: https://forum.selur.net/showthread.php?tid=359

*changed*

Avisynth/Vapoursynth: added 'Denoiser' option to QTGMC
Vapoursynth: Santiag added more parameters for NNEDI3 and EEDI3
Vapoursynth: more vsDaa NNedi3-Parameter
Vapoursynth: disabling cropView when enabling FilterView and vice versa
Vapoursynth: support DGDecNV deinterlacing and use native Vapoursynth filter mode
Resize: Increase resolution max to 32768x32768
NVEncC: not usins nvhsp anymore for HDR patching since NVEncC allows to signal maxcll/maxfall and master display information.
Changed: 'ColorMatrix' now support Rec.20202
Avisynth: SMDeGrain add support for prefilter mode 4 which uses KNLMeansCL
Input: handling of mixed input, now 'mixed'-content will be treated as normal interlaced
Automation: disabled automatic telecine/interlaced check by default
Crop: use 'Avisynth' crop style as default
Jobs: abort when time codes are used and inputFPS != outputFPS


-> downloads: http://www.selur.de/downloads

Cu Selur

[MrScientist]

Hi, I'm a new Hybrid user. Let me begin by thanking you for a nice tool. I use QSVenc (i7-7700k). When I try to change the bitrate based on target mode (1 BD 25, 2 BD 25, etc.), the target sizes and bitrates don't change. And if I choose Target mode: custom, the target size is greyed out. This is only for QSVenc. If I switch to x264, x265, etc., the calculation works correctly. Thank you again.

[Selur]

Sounds like a bug, I'm really busy today and tomorrow and will probably not have time to look into this until the weekend.
Once I've looked into it I'll report back to you.

Cu Selur

[Magik Mark]

Selur,

Does the app now supports rearranging of jobs in the queue? How about editing it and sending it back to the queue?

[Selur]

Rearranging jobs (not subjobs) is possible the whole time.
Changing jobs is not possible and probably won't happen, since it would require some major rewrite of the whole tool.
(A job doesn't include all the analysis results of the source or the state of the gui when the job was created, so restoring that state isn't possible.)
-> if this is a major let down for you, you best look for another tool.

Cu Selur

[hin12]

I get this on startup (Ubuntu 18.04):

Detected the following cards which seem to lack NVIDIA PureVideo support: Corporation GK104 [GeForce GTX 660 Ti]
No card found that's supported by nvenc, the CUDA encoding library. -> Removing CUDA based encoding support. Card name was compared against: https://developer.nvidia.com/cuda-gpus

Hardware h264 encoding works fine with ffmpeg, so the 660 Ti should be added to the list.

Also, the default option for opus audio should be to keep the input sample rate, not to resample it to 48000Hz, see: https://hydrogenaud.io/index.php/topic,115672.0.html

[Selur]

Detected the following cards which seem to lack NVIDIA PureVideo support: Corporation GK104 [GeForce GTX 660 Ti]
The card is in the compatible list, Hybrid just couldn't handle the naming scheme that is used, fixed that here locally on the 21th of April so the next release shouldn't show that message.

Hardware h264 encoding works fine with ffmpeg,
What settings do you use for hardware encoding in ffmpeg, because as as I know nvenc doesn't support such an old card.

Also, the default option for opus audio should be to keep the input sample rate, not to resample it to 48000Hz, see: https://hydrogenaud.io/index.php/topic,115672.0.html
The default resampling option is not dependent on the audio format it's something I chose to use in Hybrid.


Cu Selur

[hin12]

What settings do you use for hardware encoding in ffmpeg, because as as I know nvenc doesn't support such an old card.

https://developer.nvidia.com/nvidia-video-codec-sdk

It's a Kepler card, so it only supports H264 with YUV 4:2:0 and 8-bit color.

I used this command to test it:

ffmpeg -i input.mp4 -c:v h264_nvenc -pixel_format yuv420p -preset default output.mp4

[xueyue]

I want to add a logo to the video and I use the "Add Logo" in filtering of vapoursynth.This is the result,please check it.https://wx3.sinaimg.cn/mw690/0071U8zSgy1frh3u5jse9j30nh0k9jt6.jpg
https://wx3.sinaimg.cn/mw690/0071U8zSgy1frh3u69a5pj30ye0k1jt9.jpg
And my script can be used normally.
https://wx1.sinaimg.cn/mw690/0071U8zSgy1frh3u538r9j30nf0isjs5.jpg

[Selur]

argh,.. damn typo will fix. :)

[Selur]

Hybrid rev 2018.05.21.1:
*fixed*

Avisynth: minor bugs I don't remember in detail :)
Avisynth: LSFMod parameters not working due to intererence with Vapoursynth
x264: pc scale handling
ffmpeg: yadif call bug
x265: defaults/presets minimization, adjust to source code for x265 presets not the stuff listed in the documentation
QSVEnc: targetSizeMode
Vapoursynth: D2v Witch indexing call on Mac/Linux
Audio: AudioAnalyzer tsmuxer path
Subtitle: crash when font name start with whitespace

*added*

filter: support custom letterboxing
Avisynth: HQDN3D, MSharpen
Vapoursynth: DeBlockPP7, FFMS2K
muxing: tsMuxeR not adding audio language to streams

*changed*

tools: Handle 'Corporation ***** [GeForce *]'-names
filter: ffmpeg set zscale rangein and range
audio: No longer using NamedPipe, raw audio decoding is always done through ffmpeg
nvenc: added support of NVEncCs 'preset' selection
x265: added support for '--single-sei', '--max-ausize-factor', '--refine-inter', '--refine-intra', '--dynamic-refine', '--refine-mv'
subtitle: using ffmpeg for ogm subtitle extraction
input: fixed bug when handling vfr input time codes
Avisynth: support changing the filter order (Filtering->Avisynth->Misc->Custom Filter Order)
Avisynth: specifying multiple custom script additions like in Vapoursynth (Filtering->Avisynth->Custom)
Avisynth: telling Hybrid not to Hybrid defaults (Filtering->Avisynth->Misc->Don't hide defaults)
Avisynth: Avisynth script as input and FilterView
Avisynth: wine missing in avsViewer call on non-Windows systems
Vapoursynth: added more options to addLogo to control whether the alpha should be used and inverted or not.
Vapoursynth: DAA, added OpenCL device selection
Vapoursynth: Santiag, added device selection
Vapoursynth: allow to set openCL device for QTGMC
Tools: vsedit is now named vsViewer to avoid confusion with the normal Vapoursynth Editor


-> downloads: http://www.selur.de/downloads

Cu Selur

[Havokdan]

Thanks for the new version, I love the app, admirable your work, which continues quite successfully.

[Selur]

*fixed*

Mac: forgot to rename vsedit to vsViewer</li>
subitle: Hybrid freezes on ogm subtitle extraction when output already existed

*added*

Vapoursynth: "don't hide defaults"-option

*changed*

audio: use FFmpeg for ogm audio extraction
input: use FFmpeg for ogm subtitle extraction


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

Hybrid rev 2018.05.25.1:
*fixed*

x264: Minimize command line with ' --color-matrix undef' in it.</li>
Vapoursynth: a bunch of small bugs related to default values
Vapoursynth: FFT3DFilter Sigma2-4 not working properly
Avisynth: a bunch of small bugs related to default values
Tools: Tools not properly kept on 'Reset all', required to start Hybrid two times

-> downloads: http://www.selur.de/downloads

Cu Selur

[Dclose]

I was using last month's build fine for x265. 2018.04.08? Installed the newer 05.22.1. x265 encodes crash when the video conversion starts. I put audio on ignore to rule that out. Files load fine, the job starts fine, extracts fine, but then the log says Hybrid can't create the video file, and the job crashes.

x264 encoding works fine.

I installed 2018.04.08 again, and x265 works fine again.

That was a couple days ago. Figured I'd wait a couple days to see if a new version gets put out or what.

Now 2018.05.25.1 is released. It does the same thing. x265 encoding job crashes after extraction when it starts to encode the video. x264 works fine.

Also, I had some jobs Waiting in queue this time when I installed the new 25.1 over last month's .04.08. I didn't let the encode finish, but it went past the extraction part and started to encode video fine in x265.

If I'm the only one having this problem, no big deal, I'll just go back to .04.08 and try again with a future release.

Windows 10 64-bit.

[Selur]

In case you used Hybrid back in 2017, did you completely deinstall it (and all it's settings) when the new installer was introduced and the file locations changed?
If you didn't that is the cause of the crash now, since you got a x265 in you Hybrid folder that still belongs to an old install.
-> deinstalling Hybrid completely making sure that the Hybrid folder and all the settings are deleted and installing Hybrid anew should fix the issue.
(newer builds will all crash)

If in case I'm wrong with my speculation: read http://www.selur.de/support create a debug output and share it with me.

Cu Selur

[Dclose]

Yes, I used Hybrid in 2017, and in 2018. I probably did a full, new install in November or December of 2017 since that's when I switched to Windows 10. I was using the April 2018 build fine, and x264 worked with the new build, so I assumed the full deinstall mentioned a couple pages back wasn't the problem.

But that was the problem. Or something related to it. It works now after a full deinstall/reinstall. Thank you.

If I recall, Cut Support works well, except I think the only problems I've had with it is the encode never stops when using 2160p and/or 2160p HDR content. The encode goes past 100% completion and will go to 400% and higher until I stop it. The result is still viewable in its unmuxed ".265" format file. So, even with having to manually stop it, it's still good enough to make test samples of the video.

Question: Are QAAC aac audio updates included in new Hybrid builds? Or should the user do a manual update of QAAC files if wanting QAAC updated? The tooltip says to have an updated Quicktime, so is that the way to update it?

[Selur]

Question: Are QAAC aac audio updates included in new Hybrid builds? Or should the user do a manual update of QAAC files if wanting QAAC updated? The tooltip says to have an updated Quicktime, so is that the way to update it?
Hybrid does not include the latest qaac from the 22th since I wasn't aware of it. Since the last few updates didn't change anything major, that's okay, but you can update it manually.
I probably should change that tool-tip, what you need is an up-to-date 'Apple Application Support'-package which nowadays comes with iTunes. (Sadly Apple doesn't provide a separate download for it.)

About the cut support: Would need a debug output level 9 of the job creation and processing to see what's happening there and to be able to reproduce and fix it.

Cu Selur

[Dclose]

Resize isn't changing the resolution. 2018.05.25.1 Windows 10. Ryzen 1700.
Full uninstall and uninstall settings. Also doesn't work with the prior 2018.05.22 build.

Works fine with 2018.04.08.01.

Drag and drop file. Tried multiple files.
The Crop/Resize tab will show the correct "Output resolution:" of what I set it for. The job in Jobs will show the resolution I set it at in the command line. The Preview/CropView sample test resizes correctly and makes the preview video smaller.

But the file encodes in the original resolution.

Also, a different problem I noticed when trying to resize with x264 is the job progress goes over 100%, I think even over 1000%, and the output video is a green screen with a scrambled image, even when watching in VLC, which should play most anything. I basically only care about x265 at this point, so just mentioning it. Maybe it's just something with my computer, since if that green scrambled video output were a common problem, I assume someone else would have mentioned it already.

The debug file is for the x265 resize.

Oh, and for QAAC in the prior post, I didn't know there was a QAAC update on the 22nd either. I was just wondering if QAAC ever got updated by Hybrid, at all, ever. Looking at the files, I couldn't tell. But apparently it does get updated at least once in a while, so that's good.

[Selur]

Can't access the attached file since it waits for moderator approval -> either share it through another way with me or remind me to look at it once it's a approved.
-> regarding the resize seems to be a problem with the call, which I need the debug output for this

Updated qaac here locally, a few days ago, but really busy atm. to didn't get around to make a new release, may be on the weekend. As a side note: don't seem any changes which are relevant to Hybrid.

[Dclose]

Installed the dev version you sent. It made me smile seeing a resized video encoding at 100 fps. So, resize problem is fixed.

Also, x264 no longer shows a progress of over 100% during an encode, and the output was no longer a scrambled green video. So, that's fixed too, whatever the problem was.

:thanks:

If I wasn't clear about qaac, I don't care about always having the newest updated version of it in Hybrid. It's just that I couldn't tell from the Hybrid files if qaac was ever updated, ever. So not being able to tell made me wonder if I was still using a version of qaac from years ago. But you said it gets updated into Hybrid once in a while, so great.

[Selur]

Yes, I usually update it:
a. whenever someone especially asks for an updated version
or
b. I see something in the changelog which is used in Hybrid.

Cu Selur

[Vesdaris]

Selur

If it's not much of a trouble, could you, please, post recommended settings for qsvenc h264 encoding? I've got 3770k CPU, so I presume some settings are not supported. I want to get out of this old "horse" as much "quality" as possible ;)

Thank in advance.

[Selur]

Good news: All settings Hybrid allows you to use should be supported. :)
(Hybrid uses the output of 'QSVEncC --check-features' to determine what is supported an what not. :))

[Selur]

Hybrid rev 2018.06.03.1:
*fixed*

Avisynth preview: RGB content handling
x265: medium preset minimize
DivX265 width&height need to be mod8.

*changed*

Filtering: only support toneMap and zscale when FFmpeg is compiled with it
Qt: switched to Qt 5.11 on Windows


-> downloads: http://www.selur.de/downloads

Cu Selur

[rbauer]

Hybrid rev 2018.06.03.1:

Does exist a Hybrid's portable (no registry/sys folder entries) version?

Many thanks :thanks:

[Selur]

you can make one if you want to:
-> [INFO] *hidden* Hybrid options,... -> make Hybrid portable (Windows only) (https://forum.selur.net/showthread.php?tid=10)

[rbauer]

you can make one if you want to:
-> [INFO] *hidden* Hybrid options,... -> make Hybrid portable (Windows only) (https://forum.selur.net/showthread.php?tid=10)
Thanks, many useful.

Edit: done! Extracted setup installer. I'm on 64 bit win platform, so can I delete "32bit" folder, right?

[Selur]

I'm on 64 bit win platform, so can I delete "32bit" folder, right?
No!

---------------

Hybrid rev 2018.06.04.1:
*fixed*

Decoder: FFmpeg call


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: atm. only Windows and Linux are updated will update Mac version later today.

[rbauer]

No!

I tried and it works fine without "32bit" folder :confused:

thanks

[Selur]

Avisynth won't work.

[Selur]

updated to Hybrid rev 2018.06.04.2

[Magik Mark]

Hi Selur!

Is there a way to activate ffmpeg for audio encoder? Together with its options? Such as dynaudnorm feature?

[Gravitator]

Халоу!
Hybrid 2018.06.04.1
x265> preset:uf/vf/fr/fast > unchecking the SAO flag - causes the SAO window to be disabled.

x265 info:Maximum motion vector search range:
Sets the radius in which neighboring pixels could influence the motion vector for the current pixel.

(command line match: --merange <integer>; default: 60)
Is it not 57?

Hybrid 2018.06.04.2
Filtering > Vapoursynth > Misc > Use DGDecNV/IM when avalible - the checkbox is disabled when you try to tick.

[Selur]

@Magik Mark:
Is there a way to activate ffmpeg for audio encoder? Together with its options? Such as dynaudnorm feature?
No, Hybrid always uses sox for audio filtering.

@Graviator:
regarding x265: yup default is 57 according to the source code: https://bitbucket.org/multicoreware/x265/src/07defe235cde1949c55464952ee1dbdc10baf5e4/source/common/param.cpp?at=default&fileviewer=file-view-default
Filtering > Vapoursynth > Misc > Use DGDecNV/IM when avalible - the checkbox is disabled when you try to tick.
they are both only enabled in case Hybrid found them

[Magik Mark]

Selur,

Does x265 supports the ff switches:?

--refine-intra 4 --dynamic-refine --analysis-reuse-level 10

Dynamic refine has prerequisites by the way

[Selur]

Does x265 supports the ff switches:?
What?
--refine-intra 4 --dynamic-refine --analysis-reuse-level 10
all three can be set in Hybrid unter x265->Base->2pass Analysis

[Magik Mark]

Selur,

it's greyed out. Can't access it. Any prerequisites for that?

[Magik Mark]

Selur,

I did a clean install of my system. I can't remember on how to double the size of hybrid window. Can you share it again?

Thanks

[Selur]

it's greyed out. Can't access it. Any prerequisites for that?
'x265->Base->2pass Analysis' is about 2pass options, so 'x265->Base->General Settings->Encoding settings' need to be set to 'specific filesize/bitrate (2-pass)'.
Additionally to use the options you mentioned, 'Save RPS' and 'Save analysis' need to be enabled.

I can't remember on how to double the size of hybrid window. Can you share it again?
[INFO] *hidden* Hybrid options,... (https://forum.selur.net/showthread.php?tid=10)->misc.ini options->'change GUI fontsize' is probably what you want.

Cu Selur

[Magik Mark]

Selur,

Activating --dynamic-refine . You need the following setting as well according to x265 docs:

--refine-intra 4 --analysis-reuse-level 10

[Selur]

What doc are you reading? (Really interested, would a link would be nice.)
What I know of is:
--dynamic-refine, --no-dynamic-refine
Dynamically switches --refine-inter levels 0-3 based on the content and the encoder settings. It is recommended to use --refine-intra 4 with dynamic refinement. Default disabled.
source: https://x265.readthedocs.io/en/latest/cli.html?highlight=dynamic-refine#cmdoption-dynamic-refine
And that only states that it is recommended to use --refine-intra 4, not needed like you stated.
Also there is no mentioning of '--analysis-reuse-level 10'.

Only thing I see where it advised to use '--analysis-reuse-level 10' is when using '--scale-factor', see: https://x265.readthedocs.io/en/latest/cli.html?highlight=analysis-reuse-level%20#cmdoption-scale-factor
-> I'll be happy to fix any requirements in Hybrid if you can show me some code or official statement which shows that '--analysis-reuse-level 10' is needed.

Cu Selur

[Selur]

Hybrid rev 2018.06.18.1:
*fixed*

Vapoursynth: adjusted AddLogo to handle Logos with uneven resolutions
x265: --analysis-reuse-level
Avisynth: SMDEGrain defaults
qtgmc sharpness
Avisynth: CustomName fixed for AutoAdjust

*changed*

Filtering: remove logo path in case logo doesn't exist
Vapoursynth: better handle uneven resolutions and RGB input

*added*

Vapoursynth: DeCross
Vapoursynth: Histogram for preview


-> downloads: http://www.selur.de/downloads

Cu Selur

[Revan654]

Great program, one quick Question and on issue I have with your tool.

1. Is there a 64Bit version of your avs viewer? It's a great little tool to use to see the changes. I can't use it with some of my filters due to the filters are 64Bit. It would really help specially when trying to de-noise some videos.

2. One thing I've noticed when using it, Maybe it's the filter not 100% sure. It's when you use Rec.2020 it dishes out errors and refuses to load any preview. When using Rec.601 and Rec.709 it's perfectly fine. I'm not trying to do some massive color changes like 601 to 2020.

[Selur]

about the errors: would need more details to reproduce the issue. (I usually don't use avsViewer as a standalone tool :))
about a 64bit version: Hybrid only uses 32bit filters and thus only comes with a 32bit avsViewer.
btw. source code of avsViewer is available over at https://github.com/Selur/avsViewer/
(compiled a 64bit version and added it to my Google Drive (https://drive.google.com/drive/folders/0B_WxUS1XGCPASUZibG5XZkRfeTg?usp=sharing).)

Cu Selur

[Revan654]

about the errors: would need more details to reproduce the issue. (I usually don't use avsViewer as a standalone tool :))
about a 64bit version: Hybrid only uses 32bit filters and thus only comes with a 32bit avsViewer.
btw. source code of avsViewer is available over at https://github.com/Selur/avsViewer/
(compiled a 64bit version and added it to my Google Drive (https://drive.google.com/drive/folders/0B_WxUS1XGCPASUZibG5XZkRfeTg?usp=sharing).)

Cu Selur

Thanks, I couldn't recall the error or the name of the error. I will try to create the issue when I get home.

[Selur]

Hybrid rev 2018.06.23.1:
*fixed*

Synth: bunch of fixes for Avisynth and Vapoursynth filtering
Avisynth: custom filter order and additions
Vapoursynth: always using DeCross

*changed*

Vapoursynth: some additional color space limitations added to some filters
general: code for loading profiles
ffmpeg: changed 'limited' to 'tv' and 'full' to 'pc'


-> downloads: http://www.selur.de/downloads

Cu Selur

[Havokdan]

Hybrid rev 2018.06.23.1:
*fixed*

Synth: bunch of fixes for Avisynth and Vapoursynth filtering
Avisynth: custom filter order and additions
Vapoursynth: always using DeCross

*changed*

Vapoursynth: some additional color space limitations added to some filters
general: code for loading profiles
ffmpeg: changed 'limited' to 'tv' and 'full' to 'pc'


-> downloads: http://www.selur.de/downloads

Cu Selur

This version does not open here. Windows 10 x64. Error message:

https://i.imgur.com/aDM6Cfc.png

[Selur]

Thanks for reporting!
damn, looks like I missed to update one of the windows.dll plugins,..
-> pleas download again, should be fixed now

Cu Selur

[Selur]

Hybrid rev 2018.08.05.1:
*fixed*

General: properly detect gpus with 'Mobile' in it's name
Analysis: Image/Video reporting in MediaInfo for an Image sequence
Audio: aften ac3 encoding
Avisynth: Image Sequence + Denoiser + only filter dark
Vapoursynth: Range conversion by setting the '_ColorRange' beforehand according to 'X' in 'X->Y'
Avisynth: pc scale handling
Vapoursynth: set range on resize, full range -> x264/x265
Audio: fixed dts encoding
Avisynth: QTGMC Sharpness value
Avisynth: FrameRateConverter + don't hide defaults
Vapoursynth: vsCustomAdditions
Avisynth: custom 'before End', 'defore DeInterlace'
Vapoursynth: vsDeGrainMedianModeY limit to 5

*changed*

Avisynth: using Avisynth+ now
Tools: adjusted to latest QSVEncC
Avisynth: removed avsCutSupport button, since avsViewer doesn't support it atm.
Input: restrict accept button in ImageSequenceDialo

*added*

Avisynth: TUnsharp, SmoothCurve, SmoothTweak(16), SmoothLevels16, SmoothLevels, DeblockPP7, Deblock, ColorBalance, BucketMedian, MedianBlur2
Avisynth: ColorYUV support which allows LumaRange change
Vapoursynth: MiniDeen, HQDN3D
Video: support some of NVEncCs VPP filters
Video: NVEncC support for 'use encoder only'-option
Input: ffmpeg hwaccel choice

-> downloads: http://www.selur.de/downloads

Cu Selur

[mparade]

Hello there,

When setting both ConvertBits(16) and Crop "function" on a 8-bit source within the same script file before feeding it into x264-10bit encoder, which feature should be set as first?

Thanks for the help in advance.

[Selur]

Shouldn't matter since Crop isn't bit depth dependent.
Also: If you are using Hybrid, there should be no need to add anything lik a 'ConvertBits(16)' manually. Hybrid will adjust the Avisynth/Vapoursynth script depending to the output bit depth of the encoder.
-> ARGH, it doesn't do that anymore, must have lost that when I switched to Avisynth+ (will add again; bit depth conversion atm. is done through ffmpeg)

Cu Selur

[mparade]

Thank you very much!

[Gravitator]

Салам!
You can add for HEVC the ability to select a simplified %25,50,75,100 CPU load (in the base tab)? It is convenient for ordinary users.
For x264 do not see the setting of the threading?

[Selur]

I have no clue how to control either x264 or any hevc encoder that way ( + personally I see no usage for such a feature.)
-> if either x264 or x265 supports that I can add an option for it
If you want such a feature you can try to contact the x264 and x265 developers. (same for all other encoders where you want that option)

[ABDO]

HI @Selur
thank you for this amazing program
is it possible to add MSU color_enhancement Filter
http://www.compression.ru/video/color_enhancement/index_en.html
it is "VirtualDub Plugin" that we can use as "HDRAGC" in avisynth, the key benefit on use it, that it can enhance shadows as hdragc+ compress highlights also.
it would be nice if you can add it to your great program
thank you

[Selur]

Hybrid already has 'HDRAGC' through: https://forum.doom9.org/showthread.php?t=93571
That aside I'm not so keen on adding a VirtualDub filter, since:
a. I have no clue how the VirtualDub filter would need to be called through Avisynth. (main problem)
b. Looking at the license I would need get permission from MSU Video Group to redistribute the filter.
In case you can give me some info on how to access this filter through Avisynth I can look into the permission form MSU Video Group.

-> Judging by https://forum.doom9.org/showthread.php?p=162158 it doesn't look like it's possible to use 'MSU Color Enhancement'-filter in Avisynth, thus I can't add it to Hybrid.

Cu Selur

[ABDO]

Hybrid already has 'HDRAGC' through: https://forum.doom9.org/showthread.php?t=93571
yes, 'HDRAGC' one of the most important filter i have ever used, it always take any video used with it to another fantastic world throw Hybrid

-> Judging by https://forum.doom9.org/showthread.php?p=162158 it doesn't look like it's possible to use 'MSU Color Enhancement'-filter in Avisynth, thus I can't add it to Hybrid.
Cu Selur
thank you very much to taking some time to look at it, i wish some day we can have hightlight compress filter to work with 'HDRAGC'.
please keep you great work in your Unique program :thanks:

[Revan654]

2. One thing I've noticed when using it, Maybe it's the filter not 100% sure. It's when you use Rec.2020 it dishes out errors and refuses to load any preview. When using Rec.601 and Rec.709 it's perfectly fine. I'm not trying to do some massive color changes like 601 to 2020.

I had a chance to use the app again and test rec 2020.

It seem the issue is due to the fact that 2020cl is used. If you change it from 2020cl to 2020ncl there is no issues. I think cl requires additional colorspace functions to proper work inside VS (I think Transfer has to be used as well). I don't see a option for transfer so it has to be entered manually.

It might be worth looking into, Could also be a bug within VS as well. I know others have ran into the issue as well when trying to use 2020cl.

------

One additional question, Is there reason why you never considered updating AVS filters to 64 Bit? I'm guessing it's due to the fact that many filters only have 32 Bit versions.

[Selur]

Is there reason why you never considered updating AVS filters to 64 Bit? I'm guessing it's due to the fact that many filters only have 32 Bit versions.
Multiple reasons:

yes, a bunch of filters in Hybrid are 32bit only
64bit used to be really buggy
I decided to go for 64bit Vapoursynth since it: a. performed better for me b. it's available on Mac and Linux too (I hate to use Wine (https://www.winehq.org) since it's always buggy for me and I always have to build the latest version and then even have to try to patch that version and still be stuck with bugs)

My main reasons nowadays not to look into Avisynth 64bit is that:
a. it's Windows only
b. I'm to lazy to change tons of lines of code to add switches depending on whether 64bit or 32bit Avisynth is used. (I recently changed in Hybrid from Avisynth MT to Avisynth+ MT which was annoying enough. ;))

Cu Selur

[Forteen88]

Selur, you wrote in Hybrid that:
"Use gpu for decoding":
When enabled Hybrid will add '-hwaccel auto' to ffmpeg decoding calls.

This option has no effect if the selected hwaccel is not available or not supported by the chosen decoder (only mpeg-2, H.264, VC-1 and WMV3 content are supported atm.).
Note that most acceleration methods are intended for playback and will not be faster than software decoding on modern CPUs. Additionally, ffmpeg will usually need to copy the decoded frames from the GPU memory into the system memory, resulting in further performance loss. So this option might not provide any additional speed and thus is mainly useful for testing.Does that mean that the program DGDecNV (which uses Nvidia's CUDA to decode video-frames) is also pointless and is only mainly useful for testing? I haven't done benchmarks on any of those GPU-decoders yet.

EDIT: OK, thanks for the answer Selur.

[Selur]

Does that mean that the program DGDecNV (which uses Nvidia's CUDA to decode video-frames) is also pointless and is only mainly useful for testing?
No. The 'use gpu for decoding' option only does anything when the decoding is done through ffmpeg on it's own. It has nothing to do with any Avisynth or Vapoursynth filters.
-> So if you use Avisynth or Vapoursynth this option does nothing. :)

[Revan654]

The VSEditor does not load and seems to crash.

2018.09.17 - 23:34:03_Windows 10-64bit_2018.08.05.1 - level 9: using script:
# Imports
import vapoursynth as vs
core = vs.get_core()
# Loading Plugins
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/ResizeFilter/Waifu2x/Waifu2x-w2xc.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/fmtconv.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/SourceFilter/LSmashSource/vslsmashsource.dll")
clip = core.lsmas.LWLibavSource(source="to.Source.File.Path.Hidden.mkv", format="YUV420P8", cache=0)
# making sure input color matrix is set as unspec
clip = core.resize.Point(clip, matrix_in_s="unspec",range_s="limited")
# making sure frame rate is set to 25/1
clip = core.std.AssumeFPS(clip, fpsnum=25, fpsden=1)
# Setting color range to TV (limited) range.
clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)
original = clip
# resizing clip to 960x540
# adjusting bit depth to 32bit
clip = core.fmtc.bitdepth(clip=clip,bits=32)
# resizing using Waifu2x
clip = core.w2xc.Waifu2x(clip=clip, scale=4, noise=0)
# adjusting bit depth back to 8bit
clip = core.fmtc.bitdepth(clip=clip,bits=8)
# adjusting resizing
clip = core.fmtc.resample(clip=clip, w=960, h=540, kernel="spline64", interlaced=False, interlacedd=False)
# resizing original to 960x540
original = core.fmtc.resample(clip=original, kernel="spline16", w=960, h=540, interlaced=False, interlacedd=False)
# adjusting output color from: YUV420P16 to YUV420P8 for x264Model (i420)
clip = core.resize.Spline36(clip=clip, format=vs.YUV420P8, range_s="limited")
# adjusting for FilterView
if original.format.id != clip.format.id:
if (original.format.color_family == vs.RGB and clip.format.color_family != vs.RGB):
original = core.resize.Spline36(original, format=clip.format.id, matrix_s="470bg", range_s="limited")
elif (original.format.color_family == clip.format.color_family):
original = core.resize.Spline36(original, format=clip.format.id, range_s="limited")
else:
original = core.resize.Spline36(original, format=clip.format.id, matrix_in_s="470bg", range_s="limited")
stacked = core.std.StackHorizontal([original,clip])
# Output
stacked.set_output()
2018.09.17 - 23:34:03_Windows 10-64bit_2018.08.05.1 - level 9: indexing finished,..
2018.09.17 - 23:34:03_Windows 10-64bit_2018.08.05.1 - level 9: startPreview call: "C:\Program Files\Hybrid\64bit\Vapoursynth\vsViewer.exe" --preview-only "C:\Users\Revan\AppData\Local\Temp\tempPreviewVapoursynthFile23_34_03_200.vpy" --aspect 1 --matrix Rec601 --listen 7404_VPY_ --listen-to HYBRID --crop off#0#0#0#0
2018.09.17 - 23:34:03_Windows 10-64bit_2018.08.05.1 - level 9: adjustProcessEnvironmentOfProcess,...
2018.09.17 - 23:34:03_Windows 10-64bit_2018.08.05.1 - level 9: adjusted ProcessEnvironmentOfProcess
2018.09.17 - 23:34:03_Windows 10-64bit_2018.08.05.1 - level 9: changingProcessPriority to: -19
2018.09.17 - 23:34:03_Windows 10-64bit_2018.08.05.1 - level 9: Preview::indexingIsFinished 'VAPOURSYNTHPREVIEW' ignored,..
2018.09.17 - 23:34:03_Windows 10-64bit_2018.08.05.1 - level 9: Got unhandled IPC message: loading C:\Users\Revan\AppData\Local\Temp\tempPreviewVapoursynthFile23_34_03_200.vpy
2018.09.17 - 23:34:06_Windows 10-64bit_2018.08.05.1 - level 9: player finished,...
2018.09.17 - 23:34:06_Windows 10-64bit_2018.08.05.1 - level 9: ExitCode: -1073740791, status: 1

The second issue, I can't post any logs since it's tied to start-up. Everytime I try to start Hybrid the splash screens loads but then the program just crashes. The second time always loads flawlessly. I think it's something to do with looking for GPU's or with QT. This PC has a very old GPU and is AMD, not sure if that has any affect on the start-up.

[Selur]

About the Vapoursynth script:
What error does the Cropview show when you load that script?
Does the normal Vapoursynth preview work? If not what error does the Vapoursynth Preview show?
(Also loading can take a while since Waifu2x is really slow, too slow to use in my option,...)

Without knowing anything about the source other than that that it is YUV 4:2:0 8bit I got no clue.
-> Not enough data to reproduce.
Everytime I try to start Hybrid the splash screens loads but then the program just crashes. The second time always loads flawlessly.
If you set a default output folder and enable the debug output at level 9, save those settings and restart you should get a debug output of the startup.
Wild guess would be that QSVEncC, VCEEncC or NVEncC causes the crash. Adding those to the ignorelist could workaround this, but need to know more details to know for sure.

Cu Selur

[Revan654]

About the Vapoursynth script:
What error does the Cropview show when you load that script?
Does the normal Vapoursynth preview work? If not what error does the Vapoursynth Preview show?
(Also loading can take a while since Waifu2x is really slow, too slow to use in my option,...)

Without knowing anything about the source other than that that it is YUV 4:2:0 8bit I got no clue.
-> Not enough data to reproduce.

If you set a default output folder and enable the debug output at level 9, save those settings and restart you should get a debug output of the startup.
Wild guess would be that QSVEncC, VCEEncC or NVEncC causes the crash. Adding those to the ignorelist could workaround this, but need to know more details to know for sure.

Cu Selur

Without Waifu2x it loads fine.

It's unlikely worth the effort to deal with Waifu2x. With separate tests outside of Hybrid, It loads but it's ungodly slow. Not worth using without being used by a High End GPU.

I'll try use the ignorelist to see what happens.

--------

btw, Since it seems you use ipcServer for sending requests to VSEditor. Would you happen to have any docs or sites detailing on usage of it? I can't seem to find any get docs for the process and how to link everything up using VS2017.

[Selur]

Hybrid uses a modified old version of VSEditor, normal VSEditor doesn't support any ipc calls.
see: https://github.com/Selur/vapoursynth-editor

Cu Selur

[Revan654]

Hybrid uses a modified old version of VSEditor, normal VSEditor doesn't support any ipc calls.
see: https://github.com/Selur/vapoursynth-editor

Cu Selur

I get that, I was wonder how ipc works, I can't find any info on it when I do searches. Atlease in the area of programming.

[Selur]

Hybrid rev 2018.10.07.1:
*fixed*

Filtering: FFmpeg deinterlacer
Output: File name generation using chapter names
Subtitle: convertion from ass -> txt
Vapoursynth: Histogram in Preview
Windows: progress indication after minimize and restore

*changed*

Adjusted to QSVEncC 3.09 (Windows only)
Analysis: subtitle analysis through MediaInfo, due to changes in MediaInfo
Analysis: ignore audio languages which are not ISO 3166-1 alpha-2/3 conform
Analysis: handling apcm_adx audio in mpeg container, since MediaInfo reports this wrong atm.
Avisynth: set output depth explicitly
General: Close Hybrid on Windows in case it's running with administrator rights.
General: monitoring and file input.
Misc: misc.ini got an 'runAsAdmin' option, if you use it don't complain in case Hybrid messes up your system
Tags: tags clearing on file change.
Video: Removed NVEnc MB check, since NVEncC doesn't read max MB and doesn't report it anymore (Windows only)

*added*

Filtering: FFmpeg vidstab support
Vapoursynth: mClean
Video: NVEnc, key-frames on chapter start option (Windows only)
Video: QSVEnc HDR signaling support (Windows only)
Video: AV-1 support through aomenc and rav1e

*removed*

Filtering: 'mcdeint' support since it was removed from mencoder/mplayer

-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: Not releasing any Mac versions for the time being. (I don't have access to a mac anymore.)

[ABDO]

great work Selur, thank you

[Gravitator]

Салам!
Add more fine-tuning hundredths of --qcomp for x265 (0.65, 0.75).

[Selur]

@Gravitor: no problem, will add it for the next release

[Gravitator]

aomenc
.mp4 - aomenc video output is not allowed for .mp4 output!
.mkv - AV1 is only allowed as raw or inside a webm container!

ravie
.mp4 - ravie video output is not allowed for .mp4 output!
.mkv - AV1 is only allowed as raw or inside a webm container!

[Selur]

Please either post proper sentences stating what you want or refrain from posting.
Having to wildly guess what you might want isn't something I'm inclined to do.
So not sure what you want, but since the specification for av1 inside mkv isn't finalized it's not supported by mkvmerge atm. so Hybrid doesn't allow AV1 in mkv. Afaik there is no specification for muxing av1 inside an mp4 container, which is why Hybrid doesn't support muxing AV1 into an mkv container.

Cu Selur

[sneaker_ger]

AV1 in mkv spec is finalized. AV1 in mp4 is also possible. Youtube uses mp4.

[Selur]

AV1 in mkv spec is finalized.
Ah nice, missed that (https://forum.doom9.org/showthread.php?p=1854553#post1854553)
-> So after the next release of mvktoolnix and Hybrid, I can remove the mvk muxing restriction. :)
AV1 in mp4 is also possible.
Nice, a lot of changes in the last week. :)

-> did some test and from the looks of it I can add mkv and mp4 muxing support for av1 (and mp4 muxing support for vp9).

Cu Selur

[tola5]

Hi first sorry I not post so much info are in bed little sick but can try too get somevlog later too day if it are.
It just I update to the new version and I get a lot lower fps when I encoding now it are X265 crf 18 and medium All orther standard . Before the update I got 13-14 fps now it like 6-7 got a amd 2700X there use 7 core only in a vm under unraid but it was the same before the update
All the best
Tobias

[Selur]

Don't think I changed anything which should slow down the processing, wild guesses:
a. you are doing some filtering which slows down the processing
b. your sources have different resolutions
...
-> don't think this is Hybrids fault, log won't help

[tola5]

Don't think I changed anything which should slow down the processing, wild guesses:
a. you are doing some filtering which slows down the processing
b. your sources have different resolutions
...
-> don't think this is Hybrids fault, log won't help

wird just install the old version of Hybrid and the new one once more
test same file and if you have not change what filtering there are as standard it the same only change to preset to medium CRT are 18 both time

[Selur]

sorry, no clue what that is intended to mean :)

[tola5]

sorry, no clue what that is intended to mean :)

it ok sorry my whiting are not the best.
I try to install the old Hybrid_2018.08.05 to test one more time .
I test the same file in Hybrid_2018.08.05 and Hybrid_2018.10.07
both are set to X265 and CRT 18 and medium
you talk abut it cold be a filer setting but nothing is change other in to medium so if it not change some standard I don't know

[Selur]

whiting <- that's a fish where I live,..
cold <- the opposite of warm ;)

Okay, I get what you are doing, but make sure to:
a. deinstall Hybrid completely between the changes
b. create a separate debug output level 9 of each encoding

Cu Selur

[Dclose]

It just I update to the new version and I get a lot lower fps when I encoding now it are X265 crf 18 and medium All orther standard . Before the update I got 13-14 fps now it like 6-7 got a amd 2700X
I updated to the newest version last night. Drag-and-dropped a file, used my usual settings, (checked them multiple times), but the encoding speed was 3x faster than usual, and the estimated file size output was 4x bigger than expected.

I was curious what the output video would look like, but the encoding crashed towards the end. I closed Hybrid, re-opened Hybrid, drag-and-dropped the file again, and the speed returned to normal (slower) and the output has turned out as expected, on multiple files.

This is on Windows 10 and AMD 1700, and after a reboot right before updating. I didn't think anything of it until reading tola5's post and seeing he's also using AMD Ryzen. Though his speed went slower, not faster. I don't use any presets.

I don't know why closing and re-opening Hybrid fixed it, but it did.

[Zetti]

MKVToolNix v28.0.0 is released.

[Selur]

I know, as was Vapoursynth R45.
(can't get portable R45 running with vsedit/vsViewer atm.)

[Selur]

Hybrid rev 2018.10.21.1:
*fixed*

Analysis: mediainfo dts-hd detection
Input: Input par not properly reset when 'Input PAR' was manually changed for previous source
Audio: output created in temp folder in some cases with audio only output
Video: splitting an mkv with split chapter, deleting of temp file
Jobs: minimize job description for webm muxing
Filtering: crash when using ToneMap (No XSynth)
Video: crash when changing values in 'ffmpeg (NVEnc)'

*changed*

Video: AV1, adjusted aomenc bit-depth and color sampling restrictions
Cosmetics: x265, finer control over qcomp
Input: disabling crop when loading a new single source and autocrop isn't enabled
Input: disabling 'convert output par to' when loading a new single source
Output: support mkv(av1), mp4(av1), mp4(vp9) (requires latest mkvmerge and mp4box)
Vapoursynth: updated to R45

*added*

Video: NVEncC, trim used with cut (Windows only)
Video: NVEncC, only add VUI parameters when they are enabled (Windows only)
GUI: Added warning when drivers seem to be too old to work with NVEncC (Windows only)
Vapoursynth: HDRToSDR by DG support (Windows only)

-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

Hybrid rev 2018.11.01.1:
*fixed*

output: mp4 output naming during batch processing
analysis: mkv default&forced track detection

*added*

video: support rav1e '--keyint'

*added*

Qt Linux: min. version is now 5.5.1

-> downloads: http://www.selur.de/downloads

Cu Selur

[Dclose]

Windows 10.
Hybrid 2018.8.05.1 had been working great. :thanks:

Updated to 2018.10.21 and got the problem mentioned on your forum about drag and dropping multiple files and then output or something renames them in a way which overwrites previous output files. You made a dev fix version and then released 2018.11.01.

Updated to 2018.11.01 today. Hybrid crashes and closes while loading a drag and dropped file. Reading the top Hybrid header as it loads the file, it looks like it goes through "extracting" fine but then hangs on something about "MP4audio." The input files are mkv.

I spent a couple hours messing with it. Different versions, updating, uninstalling completely, rebooting, installing. I've had previous troubles with 2018.10.21, such as it crashes, Windows gives a popup window of "x265.exe" still being open, and Hybrid wouldn't open again until after rebooting the computer. Something like that. Sorry for lacking more specifics. I figured it was a one-time thing with my computer or with that particular Hybrid version.

After a full install of 2018.10.21, then update to 2018.11.01, 2018.11.01 wouldn't open and gave errors that it can't find a few avs something .dll files. Something about AVCSEE and GotTools.

Reinstalling 2018.8.05.1 worked again, and then updating it to 2018.11.01 allowed 2018.11.01 to open again. Though it did the same "MP4audio" (?) crash and close when drag and dropping a file again. So, the error was repeatable, and with different files. Some have AC3, others have AAC. I also downloaded 11.01 a second time in case the first setup file was corrupted somehow during download.

Completely uninstalled Hybrid again, tried 2018.10.07.1, and that one is working completely fine so far as I type this.

I don't know what's going on with 2018.11.01, (and 2018.10.21.1 to a lesser extent). Maybe it's just me. But I saw a thread about someone having problems when loading a file, and the problem was being diagnosed as something with the audio. So, figured I'd mention this and the "MP4audio" crash to help diagnose it in case it's not just me.

[Selur]

Something about AVCSEE and GotTools.
No clue,...

Cu Selur

[Selur]

Hybrid rev 2018.11.11.1:
*fixed*

models: error messages about models which weren't created due to hardware/os limitations
jobs: Avisynth, job id wasn't replaced

-> downloads: http://www.selur.de/downloads

Cu Selur

[blackhd]

I would like to know how to choose NVENCC as encoder?
I bought a 2070 recently, but I cannot find it in the drop down list of video and "Config->Tools".

How can I activate it?

:thanks:

[Selur]

You can't since Hybrid doesn't support those atm. :)
What is the exact naming Hybrid reports your card? (see Log)
(I don't own one and https://developer.nvidia.com/cuda-gpus doesn't list those currently., same goes for all 20XX cards.)

[blackhd]

You can't since Hybrid doesn't support those atm. :)
What is the exact naming Hybrid reports your card? (see Log)
(I don't own one and https://developer.nvidia.com/cuda-gpus doesn't list those currently., same goes for all 20XX cards.)

Detected the following cards which seem to lack NVIDIA PureVideo support: Intel(R) HD Graphics 530 GeForce RTX 2070

Is this correct?
Since I found some of the changelog starting with "Video: NVEncC",
I think I can use NVEncC in hybrid:(

[Selur]

Is this correct?
Yes, that is the info I needed.
I think I can use NVEncC in hybrid
You might be right, but only if the card is in Hybrids internal 'supported card list'.
-> will create a modified dev version and send you a link via PM (gimme ~1/2 hour) :)

Cu Selur

[blackhd]

Yes, that is the info I needed.

You might be right, but only if the card is in Hybrids internal 'supported card list'.
-> will create a modified dev version and send you a link via PM (gimme ~1/2 hour) :)

Cu Selur

Thank you very much.:thanks:

[Gravitator]

Веселых выходных :)
Hybrid v2018.11.11.1
There is no x265 response to on/off "weighted b-frames" when using presets: slower, veryslow, placebo.

[Selur]

Seems to be a bug in the 'Minimize command line' functionality, so disabling 'x265->Misc->Random Stuff->Minimize command line' should help.
Will try to fix it over the weekend, but I'm rewriting some other stuff atm. so not totally use whether I find the time for it.

Cu Selur

[Gravitator]

Thank!
---
What do you think about XVC? Very similar to x265 ... The only difference is in smoothing (x265 - wavy, XVC - classical).

[Selur]

Last time I tried it was really slow, it could not be decoded through libav/ffmpeg which made it a Deadline end format for me. All in all it had potential, but was not really usable. Did any of that change?

Cu Selur

[Gravitator]

Hybrid v2018.11.25.1

DivX265
There is no explanation/help for the encoding mode "constant quantizer", the "average bitrate" mode in the help is named as "variable bitrate".

VPX
There is no explanation/help for the encoding mode "1st & 2nd pass", no explanation "lossless".

AOM
The encoding mode "2pass" is not available for coding, but is present in the help; no explanation "lossless".
Tune: SSIM, SDEF, DAALA - are supported? Ends without coding and errors.

Filtering
Vapoursynth>DeNoise>Denoise: DFTTest, MCTemporalDenoise, mClean - сannot start in the preview.

[Selur]

Filtering
Vapoursynth>DeNoise>Denoise: DFTTest, MCTemporalDenoise, mClean - сannot start in the preview.
Only thing that seems to be broken is mClean, the rest works fine here. (seems like a dependency for tmedian is missing)
About the rest: since you didn't care to post details about the errors the preview reported don't expect me to investigate anymore
-> will look mClean it for the next public release,

VPX
There is no explanation/help for the encoding mode "1st & 2nd pass", no explanation "lossless".
I know lots of tool tips are empty since I couldn't motivate myself to write something.

DivX265
There is no explanation/help for the encoding mode "constant quantizer"
see above

, the "average bitrate" mode in the help is named as "variable bitrate".
this is intended

AOM
The encoding mode "2pass" is not available for coding, but is present in the help;
will probably remove the help; still not sure when I will add 2pass encoding (1pass is to slow to really use, so I see no point in 2pass encoding atm.)

no explanation "lossless".
not explanation needed,...

Tune: SSIM, SDEF, DAALA - are supported? Ends without coding and errors.
should be:
--tune=<arg> Distortion metric tuned with
psnr, ssim, cdef-dist, daala-dist

Cu Selur

[Gravitator]

Report problems with filters.

[Gravitator]

Quote:
no explanation "lossless".
not explanation needed,...

Warning about maximum bitrate, encoding duration, time? For Dummies.
----

VPX
For the "lossless" encoding mode, a bitrate regulator is present (for AOM it is turned off).

[Selur]

Report problems with filters.
remind me again in case the attachment is approved before the next public release of Hybrid.
Warning about maximum bitrate, encoding duration, time?
max bitrate: less then uncompressed
encoding duration/time: no clue, longer or higher resolution file -> longer encoding duration
For Dummies.
from the homepage:
Hybrid is intended for advanced users.
It doesn't aim to be a tool used by everybody.
If you don't know the basics about contains, video formats, etc. Hybrid is not meant for you.

VPX
For the "lossless" encoding mode, a bitrate regulator is present (for AOM it is turned off).
That's a bug in the gui, will remove it in VPX too. (lossless should only be available when VP9 is used)

[Selur]

Hybrid rev 2018.12.09.1:
*fixed*

jobs: deleting *.sub files
Vapoursynth: mClean, missing libtemporalmedian dependency
Subs: use more digits when calculation stretch factor
Avisynth: selectEvery + Avisynth Filter View
Avisynth: Preview filter reload
Jobs: vpx mkv checks
x265: minimize command line + weightb and preset slower+
models: error messages about models which weren't created due to hardware/os limitations
Avisynth: job id wasn't replaced when using dgindex

*added*

jobs: parallel subjob-processing, process subjobs out of order and in parallel if possible
muxing: use FFmpeg, enforces the use of ffmpeg as muxer, might be useful for bitrate stuffing

*changed*

Qt Windows: using Qt 5.12 now
code: some rearraning, which might speedup the startup when Avisynth is disabled.


-> downloads: http://www.selur.de/downloads

Cu Selur

[Gravitator]

x265 gives coding failure.

[Selur]

can't access the file,...

[Selur]

Hybrid rev 2018.12.23.1:
*fixed*

subs: extract dvb as vobsub subtitles when they are inside an mkv container
QSVEnc: remember HEVC 10bit support
Jobs: image sequence job handling
Jobs: mp4fpsmod input detection
Jobs: NVEncC input detection
Vapoursynth: handling 'cfr output'

*changed*

decode: enforce cfrDecode on scaleFPS


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

Hybrid rev 2019.01.06.1:
*added*

subtitle: subtilte only extraction for mkv input
subtitle: support extraction of dvb_subs as vobsub subtitles when they are inside an mkv container

*fixed*

video: Xvid 2pass bitrate adjustment
automation: handling of aborted subtitle extraction

*changed*

internal: changed the logging
decoding: enforce cfr decode on fps scaling


-> downloads: http://www.selur.de/downloads

Cu Selur

[Gravitator]

Hybrid dev v2019.01.18.1

RAV1E
Add: --primaries, --matrix, --transfer.

AOM
Mode "Adaptive quantization" is not available by default.

[Selur]

AOM
Mode "Adaptive quantization" is not available by default.
yup, that was because of a typo. :) -> will be fixed in the next release

RAV1E
Add: --primaries, --matrix, --transfer.
will think about it.

[Gravitator]

AOM, RAV1E, VPX
It is possible to expand the information in containers: color space, sub. saturation, matrix coefficient, bit stream, coding library, program settings?

[Selur]

(most of the options you mentioned are not characteristics of a container but a video stream, so I assume you want more data on the currently selected video stream)
In theory:
Hybrid could show way more of the data collected about a specific video stream.
In praxis:
Since:
a. you could simply use MediaInfo (Hybrid uses it was it main source for video header data; it also uses a mencoder, ffmpeg, tsmuxer,...)
b. I have no clue where to present that data in a meaningful way.
this is probably not going to happen.

Here's the data Hybrid usually keeps internally for a video stream:
ID: mp4_95d5d2d8dfbe3d9336dc68c1ba286704_6334
Typ: Video
Input: F:\TestClips&Co\files\HDR\HDR10\4K sun HDR test.mp4
Stretching:
General Data:
<input_data value="0" name="VideoID">
<input_data value="0" name="ffmpegID"/>
<input_data value="HEVC" name="Format"/>
<input_data value="Base Media / Version 2" name="Container Profile"/>
<input_data value="125" name="Frame count"/>
<input_data value="25" name="Frame rate"/>
<input_data value="10" name="Bit depth"/>
<input_data value="Main 10@L5.1@Main" name="Profile"/>
<input_data value="wpp / ctu=64 / min-cu-size=8 / max-tu-size=32 / tu-intra-depth=1 / tu-inter-depth=1 / me=3 / subme=7 / merange=57 / rect / no-amp / max-merge=3 / temporal-mvp / no-early-skip / rdpenalty=0 / no-tskip / no-tskip-fast / strong-intra-smoothing / no-lossless / no-cu-lossless / no-constrained-intra / no-fast-intra / no-open-gop / no-temporal-layers / interlace=0 / keyint=24 / min-keyint=1 / scenecut=40 / rc-lookahead=25 / lookahead-slices=4 / bframes=12 / bframe-bias=0 / b-adapt=2 / ref=4 / limit-refs=3 / limit-modes / weightp / no-weightb / aq-mode=1 / qg-size=32 / aq-strength=1.00 / cbqpoffs=0 / crqpoffs=0 / rd=4 / psy-rd=2.00 / rdoq-level=2 / psy-rdoq=1.00 / no-rd-refine / signhide / deblock=0:0 / sao / no-sao-non-deblock / b-pyramid / cutree / no-intra-refresh / rc=crf / crf=13.0 / qcomp=0.60 / qpmin=0 / qpmax=51 / qpstep=4 / vbv-maxrate=40000 / vbv-bufsize=40000 / crf-max=0.0 / ipratio=1.40 / pbratio=1.30" name="Encoding settings"/>
<input_data value="CFR" name="Frame rate mode"/>
<input_data value="3840" name="Original width"/>
<input_data value="2160" name="Original height"/>
<input_data value="3840" name="width"/>
<input_data value="2160" name="height"/>
<input_data value="progressive" name="Interlacing"/>
<input_data value="4:2:0" name="Color"/>
<input_data value="BT.2020 non-constant" name="Color Matrix"/>
<input_data value="smpte2084" name="Color Transfer"/>
<input_data value="BT.2020" name="Color Primes"/>
<input_data value="G(13250,34499)B(7499,2999)R(34000,15999)WP(15634,16450)L(10000000,1)" name="Master Display"/>
<input_data value="hvc1" name="Codec"/>
<input_data value="tv" name="Luminance scale"/>
<input_data value="1" name="Aspect ratio"/>
<input_data value="1" name="mp4ID"/>
<input_data value="false" name="Raw video"/>
<input_data value="5.000" name="Length"/>
<input_data value="0" name="Chapter count"/>
<input_data value="0-0" name="Chapter selection"/>
<input_data value="false" name="Needs YUY2"/>
<input_data value="0.000" name="Rotation"/>
<input_data value="1" name="Angles"/>
<input_data value="false" name="vmf"/>
<input_data value="0" name="Title"/>
<input_data value="5" name="Selection length"/>
<input_data value="25" name="Frame rate"/>
<input_data value="1" name="Video stream count"/>
<input_data value="0" name="Start pos"/>
<input_data value="false" name="DVDInput"/>
</input_data>
sub. saturation
don't even know that that is meant to be,..

Cu Selur

[Gravitator]

Subsampling saturation :)
Like "SEI info" for x265 (to see more information in MediaInfo). Need to ask the developers themselves?

[DJ-1]

Hi, I`m looking at test a few clips at ripping old animated DVDs,
I updated to the latest Hybrid, did a clean install of Hybrid.

Video = x264 (no preset selected)
Audio = ignore
no crop / resize
filtering = Avisynth Mode - Deinterlace - QTGMC (Slower) preset + bobbed output + Avisynth / Frame - sRestore to 23.976

LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\LoadDll.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\DGDecodeNV.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\VFRtoCFR.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\AddGrainC.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\dfttest.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\EEDI2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\eedi3.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\FFT3DFilter.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\masktools2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\mvtools2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\nnedi.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\nnedi2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\SSE2Tools.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\TDeint.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\VerticalCleanerSSE2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\PlanarTools.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\MedianBlur2.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\nnedi3.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\RgTools.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\TIVTC.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\Average.dll")
LoadPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\GRunT.dll")
LoadCPlugin("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\yadif.dll")
LoadDLL("C:\PROGRA~1\Hybrid\32bit\AVISYN~1\libfftw3f-3.dll")
Import("C:\Program Files\Hybrid\32bit\avisynthPlugins\QTGMC.avsi")
Import("C:\Program Files\Hybrid\32bit\avisynthPlugins\SMDegrain.avsi")
Import("C:\Program Files\Hybrid\32bit\avisynthPlugins\AnimeIVTC.avsi")
Import("C:\Program Files\Hybrid\32bit\avisynthPlugins\Srestore.avsi")
SetFilterMTMode("DEFAULT_MT_MODE", MT_MULTI_INSTANCE)
# loading source: K:\Family.Guy.S01E04.Mind.Over.Murder.480p.MakeMKV.AC3.MPEG-BHD.mkv
# input color sampling YV12
# input luminance scale tv
DGSource(dgi="K:\ENCODES\Hybrid\Temp\mkv_fe6fc7556eb489c4c62547ab17a1741d_41.dgi",fieldop=2)
vfrtocfr(timecodes="K:\ENCODES\Hybrid\Temp\synthTempTimeCodes__mkv_fe6fc7556eb489c4c62547ab17a1741d_41.tc", fpsnum=24000, fpsden=1001)
# current resolution: 720x480
# deinterlacing
QTGMC(Preset="Slower", ediThreads=2)
# removing ghosting
srestore(frate=23.976,omode=6)
# filtering
# adjusting bit depth current 8 to target 8
PreFetch(16)
return last

i do get a crash - debug https://www.dropbox.com/s/6weagz9mogcaezo/HybridDebugOutput.txt?dl=0

i do get several popups from Hybrid before the crash (which the contents maybe in the debug, but i took screens of them anyhow) https://www.dropbox.com/sh/d637wfto2mqi813/AAAwv7rbpBcYtultiB2xw1gOa?dl=0

hope that helps.

Thanks.

[Gravitator]

Hybrid dev v2019.01.19.1

AOM
It's time to update the settings ... --aq-mode 4 is excluded.
--help
--aq-mode = <arg> Adaptive quantization mode (0: off (default), 1: variance 2: complexity, 3: cyclic refresh)

An empty box of color sampling at - Profile:Main > In-Output bit-depth:10bit (Changing the profile leads to the inability to choose --420 for 10bit).

[Selur]

@Graviator:
Subsampling saturation
Sorry that combination of words does not hold any information for me.
I know luma subsampling or color spaces like 4:2:0, I know of subsampling in context of signaling or rate control, but no clue what 'subsampling saturation' refers to.

AOM
It's time to update the settings ... --aq-mode 4 is excluded.
--help
--aq-mode = <arg> Adaptive quantization mode (0: off (default), 1: variance 2: complexity, 3: cyclic refresh)
I will remove it.

An empty box of color sampling at - Profile:Main > In-Output bit-depth:10bit (Changing the profile leads to the inability to choose --420 for 10bit).
will fix

@DJ-1: I think I fixed those. Will send you a link to my current dev version in a few minutes.

Avisynth Mode - Deinterlace - QTGMC (Slower) preset + bobbed output + Avisynth / Frame - sRestore to 23.976
Hope you checked that T/VIVTC isn't a better solution.
Also ignoring using Config->Internals->Handling->Ignore all input timecodes with be a good idea.

Cu Selur

[Gravitator]

@Graviator:

Sorry that combination of words does not hold any information for me.
I know luma subsampling or color spaces like 4:2:0, I know of subsampling in context of signaling or rate control, but no clue what 'subsampling saturation' refers to.

This is exactly what is read in the Russian version of MediaInfo.
In English now I see "Сhromа subsampling".
Chromа = Saturation :o
-------
Launch Hybrid (In the line we see the resolution of the frame --input-res 48x48) > I apply the frame crop function (↑132/↓132) > Add the any video file to the program - The cropping did not apply (--input-res 48x48 still on the command line).

[Selur]

Launch Hybrid, ...
doesn't happen here so I probably fixed it already

This is exactly what is read in the Russian version of MediaInfo.
In English now I see "Сhromа subsampling".
Chromа = Saturation
Ah, lost in translation. :)
Still probably not going to add any additional source info view.

If you just want to know the chroma subsampling, load a source and either look into the Avisynth or Vapoursynth script preview.
The Avisynth script contains: '# input color sampling YV12' and the Vapoursynth script also should have info about the input chroma sampling, like 'YUV420P8'.

Cu Selur

[Gravitator]

Hybrid dev v2019.01.19.1

X265
It's time to update the settings ... --rd 0 is excluded.
--help
--rd <1..6> Level of RDO in mode decision 1:least....6:full RDO. Default 3

[Selur]

https://x265.readthedocs.io/en/default/cli.html?highlight=--rdo#mode-decision-analysis
is strange, it lists rdo 0, but doesn't put it in the allowed range,... -> will adjust Hybrid

[DJ-1]

ok, i got the dev version, cheers.
same job, with the `ignore input timecodes` setting enabled
job ran successfully without popup errors/msgs etc, although slowly (only using 18% CPU)....but without crash was the main part.
debug : https://www.dropbox.com/s/smmrmemx91ykx1e/successfull%20HybridDebugOutput.txt?dl=0

i did try it after without enabling `ignore input timecodes` out of curiosity...
crashed
debug: https://www.dropbox.com/s/0a3bhrf0w2m34cu/HybridDebugOutput.txt?dl=0

..yeah i did look on doom9 forum to see what other peoples experiences were with the DVD sources i have (US Family Guy) & the outcome seemed there was a choice of either
tdeint (at double rate (mode1 i think)) + sRestore
or
QTGMC Default preset + bobbed output + sRestore

.....this was just the start of the testing...

cheers.
shall i leave the `ignore input timecodes` off all the time then?, its not set by default.

[Selur]

shall i leave the `ignore input timecodes` off all the time then?, its not set by default.
Normally Hybrid automatically ignores the time codes when input from MakeMKV is detected, not totally sure why that didn't happen. :)
-> enable it when you feed Hybrid with input from MakeMkv

Cu Selur

[DJ-1]

Normally Hybrid automatically ignores the time codes when input from MakeMKV is detected, not totally sure why that didn't happen. :)
-> enable it when you feed Hybrid with input from MakeMkv

Cu Selur

ok, I use makemkv its seems goog at what it does, i assume it makes a mess of timecodes then?

as far as the low CPU usage goes, its fine until sRestore is added, only using arou 18% sRestore included.

...1 thing have noticed is that if i remove a job before its ended in Hybrid, it completely close Hybrid, no prompt like usual shutdown...no error.

thanks.

[Selur]

It does add vfr time codes for cfr content so yes,...
About sRestore&Co and CPU usage, using Vapoursynth instead of Avisynth might help a bit.

Cu Selur

[DJ-1]

got the dev version, thanks, assuming you were referring to the removing of jobs in progress, seems ok now.

thanks.

[Selur]

assuming you were referring to the removing of jobs in progress, seems ok now.
Yes, happy that was fixed.

Cu Selur

[DJ-1]

Yes, happy that was fixed.

Cu Selur

correction, i switched to Vaporsynth, started to run a job, see that CPU usage was fine, cleared the job (`remove all` button) & Hybrid closed again, no errors/msgs

thanks

[Selur]

Will look into it. :)

Cu Selur

[Gravitator]

Hybrid dev v2019.01.20.1

AOM
There is no reaction (on the command line) to changes in any parameters from the Signal VUI.

[Selur]

Lol, forgot to add actually trigger the code I added. :)
Will send you another link to a dev build in a bit.

[Gravitator]

Will there be any new options for the AOM encoder in the near future?

[Selur]

You would have to ask the aomenc authors for that. No clue whether there is anything new planned in the near future.

[Gravitator]

Then how can I add parameters (Into Hybrid command line) to the AOM encoder?
In --help interesting news...

--enable-interinter-wedge
--enable-interintra-wedge
--enable-smooth-interintra
--enable-smooth-intra

[Selur]

Then how can I add parameters to the AOM encoder?
You use aomenc without Hybrid.
Looking at the aomenc options, there are a bunch of options Hybrid doesn't support atm. :
--enable-cdef=<arg> Enable the constrained directional enhancement filter (0: false, 1: true (default))
--enable-restoration=<arg> Enable the loop restoration filter (0: false, 1: true (default))
--enable-rect-partitions=<a Enable rectangular partitions (0: false, 1: true (default))
--enable-dual-filter=<arg> Enable dual filter (0: false, 1: true (default))
--enable-intra-edge-filter= Enable intra edge filtering (0: false, 1: true (default))
--enable-order-hint=<arg> Enable order hint (0: false, 1: true (default))
--enable-tx64=<arg> Enable 64-pt transform (0: false, 1: true (default))
--enable-dist-wtd-comp=<arg Enable distance-weighted compound (0: false, 1: true (default))
--enable-masked-comp=<arg> Enable masked (wedge/diff-wtd) compound (0: false, 1: true (default))
--enable-interintra-comp=<a Enable interintra compound (0: false, 1: true (default))
--enable-smooth-interintra= Enable smooth interintra mode (0: false, 1: true (default))
--enable-diff-wtd-comp=<arg Enable difference-weighted compound (0: false, 1: true (default))
--enable-interinter-wedge=< Enable interinter wedge compound (0: false, 1: true (default))
--enable-interintra-wedge=< Enable interintra wedge compound (0: false, 1: true (default))
--enable-global-motion=<arg Enable global motion (0: false, 1: true (default))
--enable-warped-motion=<arg Enable local warped motion (0: false, 1: true (default))
--enable-filter-intra=<arg> Enable filter intra prediction mode (0: false, 1: true (default))
--enable-smooth-intra=<arg> Enable smooth intra prediction modes (0: false, 1: true (default))
--enable-paeth-intra=<arg> Enable Paeth intra prediction mode (0: false, 1: true (default))
--enable-cfl-intra=<arg> Enable chroma from luma intra prediction mode (0: false, 1: true (default))
--enable-obmc=<arg> Enable OBMC (0: false, 1: true (default))
--enable-palette=<arg> Enable palette prediction mode (0: false, 1: true (default))
--enable-intrabc=<arg> Enable intra block copy prediction mode (0: false, 1: true (default))
--enable-angle-delta=<arg> Enable intra angle delta (0: false, 1: true (default)) source: aoemenc --help
https://jmvalin.ca/papers/AV1_tools.pdf offers some insight into what some of them are meant to do and from the looks of it, disabling them seems like a mistake.

Cu Selur

[Gravitator]

AOM
Add:
--film-grain-test=<arg> Film grain test vectors (0: none (default), 1: test-1 2: test-2, ... 16: test-16)
--film-grain-table=<arg> Path to file containing film grain parameters
--max-intra-rate=<arg> Max I-frame bitrate (pct)
--max-inter-rate=<arg> Max P-frame bitrate (pct)

[Selur]

--film-grain-test=<arg> Film grain test vectors (0: none (default), 1: test-1 2: test-2, ... 16: test-16)
--film-grain-table=<arg> Path to file containing film grain parameters
No clue what they do, especially ' --film-grain-table'.
-> not going to happen unless you can provide an explanation to the options.

--max-intra-rate=<arg> Max I-frame bitrate (pct)
--max-inter-rate=<arg> Max P-frame bitrate (pct)
Would need to know the default values.

Cu Selur

[Gravitator]

Hybrid dev v2019.01.22.1

AOM
2-pass mode ( bitrate mode is enabled) > change to any other mode > go back to 2-pass - bitrate mode is disabled.

[Selur]

'bitrate mode is enabled' -> will fix, should only be enabled for 'average bitrate (1-pass)'

[Selur]

Hybrid rev 2019.02.23.1:
*added*

Vapoursynth: MLDegrain, MDering, aaf, DaaMod, Tweak, SmoothGrad
Avisynth: MPEG2DecPlus

*fixed*

Analysis: ImageSequence try 'Image' and 'Video'
muxing: no explixit PAR (during muxing) signaling when XPS is enabled
muxing: hevc brand signaling
Avisynth/Vapoursynth: ffmsindex call for mkv(vp9) content
Avisynth: use ConvertBits with full=true for PC scale
Vapoursynth: d2vsource call, accidentally contained 'rff=False' for NTSC sources.
Jobs: to wav job creation
Preview: MPlayer not properly closing

*changed*

video: adjusted autoAltRef for vp9 and av1
x265: rearranged x265->Singaling tab
Vapoursynth: added OpenCL parameter to SMDeGrain
lots of internal code changes
Avisynth: lower threads in Prefetch

*removed*

Vapoursynth: DGDecIM support since it's not developed any more and doesn't work with latest drivers


-> downloads: http://www.selur.de/downloads

Cu Selur

[DJ-1]

Hi, is there a plan to add aac he v2
And/or the xhe-aac ( https://www.iis.fraunhofer.de/en/ff/amm/prod/digirundfunk/digirundf/xheaac.html ) ?

Cheers.

Sent from my Pixel using Tapatalk

[Selur]

Hybrid (for years) supports fhgaacenc, neroaac and fdkaac which all support he aac v2.
Note that you can only use he aac v2 with stereo content and it's only recommend on low bit rates,....

Cu Selur

[DJ-1]

Hybrid (for years) supports fhgaacenc, neroaac and fdkaac which all support he aac v2.
Note that you can only use he aac v2 with stereo content and it's only recommend on low bit rates,....

Cu SelurAh ok, my bad, hadn't noticed.. had always used either passthru or aac LC

With the fdk box ticked (using he V2)
Is the xHe aac playback (decoder only?) & V2 being latest encoder side?

Also am testing with he aac V2 & if I choose mode_ vbr, I only get max of 3 settings & apparently there are 5.

Cheers.

Sent from my Pixel using Tapatalk

[Selur]

yes, fdkaac doesn't support xHE-AAC for encoding. (afaik)
Don't know of any available encoder that does, also iirc. only real addition xhe-aac delivered was support for DRM, so even if there was an encoder available I would probably not include it into Hybrid.

Also am testing with he aac V2 & if I choose mode_ vbr, I only get max of 3 settings & apparently there are 5.
vbr 4&5 are only available for AAC LC

As a side note there are also MPEG-4 AAC LD, MPEG-4 AAC ELD, but I never bothered supporting them in Hybrid since they are low delay modes which isn't really interesting since Hybrid doesn't do live encoding where low delay would be needed.

Cu Selur

[DJ-1]

Ok thanks, was just just testing & comparing Vs Opus..but would also need to factor in compatibility etc.,
Cheers.

Sent from my Pixel using Tapatalk

[den78]

Hello Selur.
I get error message when applying x265 preset in the new version, saying:
x265Model doSet->found no field: singleSei

[Selur]

got it, removed the value and forgot to remove the call when defaults are applied (simply ignoring it should be fine)
-> will fix

[Gravitator]

Hybrid v2019.02.23.1

x265
Preset: veryfast, faster, fast > off/on Loop Filter or SAO - then it is blocked by SAO.
Preset: medium, slow, slower, veryslow, placebo > Adaptive b-frame decicsion: none > off/on Loop Filter or SAO - SAO goes off.

[Selur]

will look into it, not sure when, busy atm.

[Selur]

For sao to be available loop filter and adaptive b-frames need to be enabled.
Problem is that changing the 'Adaptive b-frame decicsion' didn't properly tricker the check whether sao is available or not.

[Gravitator]

Hybrid v2019.02.23.1

x265
Preset: veryfast, faster, fast > off/on Cu lossless - there is no reaction in the command line.

RAV1E
Developers have added new coding options.

[Selur]

x265
Preset: veryfast, faster, fast > off/on Cu lossless - there is no reaction in the command line.
Another visual bug:
cu lossless requires:
1. cuTree enabled
2. Adaptive Quantization
3. Adaptive Quantization strength > 0
4. Rate Distortion Optimization > 2
Developers have added new coding options.
okay,...

[Selur]

Hybrid rev 2019.03.01.1:
*fixed*

QSVEnc: typo when using HDR signaling
avi: container tagging
Avisynth: ConvertBits, don't use dither (stupid idea, since it's already handled when needed)
Video: timecode files not created on all video formats
x265: cu lossless not disabled when it has no effect
Vaporusynth: mClean call
QSVEnc: HDR signaling (typo)

*changed*

video: timcode handling with delays
video: support rav1e bitrate&2pass bitrate encoding
Vapoursynth: nnedi3cl support to nnedi3aa


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

Hybrid rev 2019.03.04.1:
*fixed*

x264: 2pass call
muxing: input detection in mencoder muxing call


-> downloads: http://www.selur.de/downloads

Cu Selur

[mzso]

Hello!
Does this support directshow/vfw codecs? I struggle to find a tool that does.

[Selur]

When using Avisynth instead of Vapoursynth (lower right corner in the Filtering tab), you can:
a. enable 'Filtering->Avisynth->Misc->Prefer AviSource for avi input' which will use AviSource for .avi input which uses vfw decoders assuming the needed 32bit vfw decoder is installed in the system
b. enable 'Filtering->Avisynth->Misc->Force DirectShowSource' which will use DirectShowSource as decoder for the input.eerrrrrrrrrrrrrrrrrrrr

Hybrid does not support encoder using a vfw encoder.

Cu Selur

Ps.: You probably have to use Virtual Dub 2 (https://sourceforge.net/projects/vdfiltermod/) if you want to use vfw encoders. (Don't know of any tool which allows to use DirectShow encoder filters.)

[Selur]

Hybrid rev 2019.03.14.1:
*fixed*

Avisynth: loading FFms2k
muxing: input detection in mencoder muxing call

*changed*

x264: 2pass call
input: MakeMKV input handling
Avisynth: updated extension version,..
jobs: enforce negative audio delay fixing with ffmpeg for mov output
jobs: enforce cfr decoding for mov output

*added*

Avisynth: ChannelMixer


-> downloads: http://www.selur.de/downloads

Cu Selur

[Sharc]

When I select NVEnc as video encoder the framerate is always set to --fps 29.970, disregarding the discovered framerate of the source.

I didn't find a way to set or change the output framerate, or did I miss something?

[Selur]

Can't reproduce this here.
Would need a detailed description on what you do and a debug output level 9 of it.
Looking at the NVEnc command line the default starts with '--fps 25', when I load a source that is 29.97 it changes to '--fps 29.970', when I IVTC telecined content it changes to '23.976' fps.
Wild guess: some old defaults that you didn't reset when you should have or some old profile is used,... (or you enabled some filter which causes Hybrid to output 29.97 fps)

Cu Selur

[Sharc]

Can't reproduce this here.
Would need a detailed description on what you do and a debug output level 9 of it.
Looking at the NVEnc command line the default starts with '--fps 25', when I load a source that is 29.97 it changes to '--fps 29.970', when I IVTC telecined content it changes to '23.976' fps.
Wild guess: some old defaults that you didn't reset when you should have or some old profile is used,... (or you enabled some filter which causes Hybrid to output 29.97 fps)

Cu Selur
Oooops! Sorry, I had the speed change still ticked. Stupid me :o

[Sharc]

Everything went fine along the job queue, until step "11_muxing" when tsmuxer crashed for whatever reason. Despite tsmuxer crashing the next step "12_cleanup" was executed, deleting the .264 encoded file, so all was lost.

I had "Parallel subjob processing" enabled. Would disabling it have prevented the file deletion?

[Selur]

I had "Parallel subjob processing" enabled. Would disabling it have prevented the file deletion?
It should. (Alternatively tell Hybrid to keep the intermediate files.) What happens is that the subjob processing routines see that there is no running subjob which needs the .264 files and thus the deletion is okay. I could probably write a workaround for this behavior, but that will take some time,..

Everything went fine along the job queue, until step "11_muxing" when tsmuxer crashed for whatever reason.
And without a details that allow me to reproduce this, that is the way it is.

Cu Selur

[Sharc]

It should. (Alternatively tell Hybrid to keep the intermediate files.) What happens is that the subjob processing routines see that there is no running subjob which needs the .264 files and thus the deletion is okay. I could probably write a workaround for this behavior, but that will take some time,..
No rush, thanks. I'll keep the intermediate files for now.

And without a details that allow me to reproduce this, that is the way it is.

Sure. I'll try to collect more info if it should happen again. I am currently experimenting with blu-ray compliant GPU encodes with NVEnc and NVEnc(ffmpeg).

[Sharc]

I made a few tests with the experimental NVEnc(ffmpeg). It seems to work here with -rc-lookahead=0 (or omitted). It fails for non-zero values. So this setting may perhaps be greyed out unless it's just an issue of the current ffmpeg or a problem with my system. According to the ffmpeg doc this parameter is valid, so I don't know. The default value is 0 anyway, so as long as I don't change it it's no problem.

[Selur]

Seems like rc-lookahead only works with H.264 and not H.265, since
ffmpeg -y -threads 8 -i "F:\TestClips&Co\files\Test-AC3-5.1.avi" -map 0:0 -an -sn -vf zscale=rangein=tv:range=tv -pix_fmt yuv420p -vsync 0 -vcodec h264_nvenc -preset 5 -profile:v 2 -level 0 -rc vbr_2pass -2pass 1 -qmin 0 -qmax 51 -b:v 1500k -spatial-aq 0 -temporal-aq 0 -rc-lookahead 32 -surfaces 32 -no-scenecut 0 -refs 3 -nonref_p 1 -strict_gop 1 -bsf:v h264_mp4toannexb "E:\Temp\Test-AC3-5.1_00_21_02_2110_02.264"
works fine here.
This is exactly why I keep the 'experimental': Couldn't find any documentation about the dependencies of the options and I'm too lazy to test everything. ;)
-> I'll make sure rc-lookahead is only 'usable' (and added to the command line) when H.264 is used in the next release.

Cu Selur

Ps.: let me know if you stumble over other restrictions. ;)

[Groucho2004]

Seems like rc-lookahead only works with H.264 and not H.265
I'm not sure if it works but have you tried -x265-params (in bold below)?
libx265 AVOptions:
-crf <float> E..V..... set the x265 crf (from -1 to FLT_MAX) (default -1)
-forced-idr <boolean> E..V..... if forcing keyframes, force them as IDR frames (default false)
-preset <string> E..V..... set the x265 preset
-tune <string> E..V..... set the x265 tune parameter
-profile <string> E..V..... set the x265 profile
-x265-params <string> E..V..... set the x265 configuration using a :-separated list of key=value parameters



Couldn't find any documentation about the dependencies of the options
"ffmpeg -h full"

Also: https://trac.ffmpeg.org/wiki/Encode/H.265

[Selur]

@Groucho2004: we are talking '-vcodec hevc_nvenc' and '-vcodec h264_nvenc' here,... not general H.264 and H.265 :D

[Sharc]

Seems like rc-lookahead only works with H.264 and not H.265, …..

It doesn't work here with H.264 either. I tried your script and it failed. Only difference is that my input is a .m2ts file


P.S.

You have added -bsf:v h264_mp4toannexb to your script. I think this should be added per default when selecting .m2ts as output. Currently it is missing.

[Selur]

The call works fine here. :/
You did adjust all the paths (input, output, ffmpeg) and you are using the ffmpeg version which came with Hybrid, correct?
What does ffmpeg report when you call your modified call?

[Sharc]

Another observation with .m2ts as target:
The files play fine on all players which I have tried. The exception are VEGAS products. Neither VEGAS Movie Studio nor VEGAS DVD architect handle these files properly (freezing or similar).
I found a workaround by remuxing the .m2ts file again with tsmuxer, this time with insertSEI, contSPS, in the .meta.
Now I wonder whether insertSEI, contSPS, should be added per default - unless it has other unwanted implications which I don't know.

[Selur]

Don't see any harm in doing so. Will send you a link to a dev version in a few minutes for testing.

Cu Selur

[Sharc]

The call works fine here. :/
You did adjust all the paths (input, output, ffmpeg) and you are using the ffmpeg version which came with Hybrid, correct?
What does ffmpeg report when you call your modified call?

Hmmm… I have ffmpeg 4.1.1 64 bit in the system path. Maybe this makes a difference. FFmpeg reports that -rc-lookahead does not apply (discarding it), and I am also getting a warning that -rc vbr_2pass is obsolete and should be replaced by -rc vbr_hq.

I need to double check with the ffmpeg version which comes with Hybrid.

[Selur]

The deprecated warning is no problem.
Last ffmpeg I shipped with Hybrid on Windows was:
ffmpeg version N-93331-gc2a221c5ae Copyright (c) 2000-2019 the FFmpeg developers
built with gcc 8.3.0 (Rev2, Built by MSYS2 project)
configuration: --disable-autodetect --enable-amf --enable-bzlib --enable-cuda --enable-cuvid --enable-d3d11va --enable-dxva2 --enable-iconv --enable-lzma --enable-nvenc --enable-zlib --enable-sdl2 --disable-debug --enable-ffnvcodec --enable-nvdec --enable-libmp3lame --enable-libopus --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libx265 --enable-libdav1d --enable-fontconfig --enable-libass --enable-libbluray --enable-libfreetype --enable-libmfx --enable-libmysofa --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libsnappy --enable-libsoxr --enable-libspeex --enable-libtheora --enable-libtwolame --enable-libvidstab --enable-libvo-amrwbenc --enable-libwavpack --enable-libwebp --enable-libxml2 --enable-libzimg --enable-libshine --enable-gpl --enable-avisynth --enable-libxvid --enable-libaom --enable-libopenmpt --enable-version3 --enable-libkvazaar --enable-libxavs --enable-libxavs2 --enable-mbedtls --extra-cflags=-DLIBTWOLAME_STATIC --extra-libs=-lstdc++ --extra-cflags=-DLIBXML_STATIC --extra-libs=-liconv --extra-cflags=-DKVZ_STATIC_LIB
libavutil 56. 26.100 / 56. 26.100
libavcodec 58. 47.103 / 58. 47.103
libavformat 58. 26.101 / 58. 26.101
libavdevice 58. 6.101 / 58. 6.101
libavfilter 7. 48.100 / 7. 48.100
libswscale 5. 4.100 / 5. 4.100
libswresample 3. 4.100 / 3. 4.100
libpostproc 55. 4.100 / 55. 4.100

[Sharc]

Your version is newer. I stuck with the last stable release 4.1.1. from Zeranoe.
I'll upgrade.....

[Selur]

I usually build ffmpeg&co on the day I release Hybrid, so the versions of the tools should be up-to-date usually.

[Selur]

Hybrid rev 2019.05.05.1:
*fixed*

jobs: unwarrented warning about 'You are creating a job without audio, while the input has audio!' during batch processing
ffmpeg: tonemap sd range missing
Avisynth: amDCT Prefetch missing
Muxing: Mp4Box + chapters + Parallel subjob processing
NVEnc: 'Parallel subjob processing' + keyfile caused crash
Vapoursynth: LSFMod options matching code (missed version)
Avisynth: NNEDI3Resize missing prescreen mode 0
x26X: parallel subjob processing .mbtree-system cleanup triggered too early
Muxing: MP4Box 'flattening'-call handling
jobs: Subjobs that took longer to delete
ffnvenc: bsf written
Vapoursynth: rotation handling
VP9: color space grayed initially out
NVEnc: file size calculation adjustment
Subtitle: subtitle order mixup
filtering: 'Force 4:4:4 downsize' with 'no XSynth'
Avisynth: LSMASHVideoSource + hight bit depth source
Vaporoursynth: MCDegrainSharp, libmvtools dependency
jobs: show fps when mencoder is used
muxing: cosmetic glitch caused ffmpeg for ts muxing to be disabled
jobs: mencoder getting stuck on mencoder job
jobs: mencoder progress indication during Xvid encoding
jobs: Abort subjob processing if another subjob is aborted/postponed/stopped/...

*changed*

Vapoursynth: default filter order (rotation before resize)
Avisynth: use Prefetch() with "cpu count / 2" instead of "cpu count" when MT-Threads is set to 0
mencoder: use fractions when specifiying ofps
muxing: added 'insertSEI, contSPS' to tsMuxeR muxing call

*added*

filtering: 'rotate' 'auto' option
Avisynth: MvToolsFPSmod
Vapoursynth: added TimeCube to apply LUTs for Color Correction
Avisynth: MCDegrainSharp


-> downloads: http://www.selur.de/downloads

Cu Selur

[Gravitator]

Hybrid v2019.05.05.1

x264
Add any video file > Hardware: (DXVA > Blu-ray > DXVA) - results in blocking the "Resize" checkbox.

[Selur]

That's probably right, since 'Blu-ray' might enforce 'Convert output to PAR' which enforces 'Resize' and switching back to DXVA doesn't disable it.
-> works like intended

[Selur]

Hybrid rev 2019.06.09.2:
*fixed*

NVEnc: init bug


Hybrid rev 2019.06.09.1:
*fixed*

Vapoursynth: embedding idx/sub and sup subtitles
Filtering: crop only bug
Xvid: decimal/fraction in mencoder
Files: Do not allow semicolon or other separator used by Hybrid inside input and output file names.
Mp4Box: mp4box vfr handling
Avisynth: HDRAGC min grain option
Avisynth: Flash3K Deband defaults

*changed*

NVEnc: allow aq-strength for H.265
NVEnc: rearranges layout
Decoder: prefer ffmpeg as decoder for > 8Bit output
FFNvEnc: tier-support added missing level 51
QSVEnc: fixed --weightp support
x265: adjusted to 'committed 4583000' "Changed the params max-merge to 3, b-intra enabled, limit-ref and early-skip enabled for improved performance for high res"
Vapoursynth: replaced 'knwvsfunc.py' with 'G41Fun.py'

*added*

NVEnc: support --dhdr10-info, --nonrefp, --no-i-adapt, --no-b-adapt, --direct, --adapt-transform, --weightp, --strict-gop
Tags: option to set copyright info


-> downloads: http://www.selur.de/downloads

Cu Selur

[StainlessS]

Selur,
Your screenshot link in 1st post is 404. [JFYI]

[Selur]

Hybrid rev 2019.08.25.1:
*fixed*

jobs: removed unneeded copy call during subtitle speed-change
Avisynth: FrameRateConverter
Analysis: adjusted to MediaInfo reporting 'TrueHD / AC-3' streams as just 'AC-3'
Analysis: chapter info parsing from mediainfo
video: some ffmpeg nvenc options
subs: automatic extraction of PGS subtitles from transport streams
Vapoursynth: waifux dependency loading
jobs: mkvmerge call input detection checks
muxing: tsMuxer call
analysis: raw audio handling
jobs: Avisynth/Vapoursynth: DGIndexNV log cleanup
analysis: AutoCrop for Blu-ray/mpls input
muxing: force 1:1 signaling in case input and output PAR differ
Avisynth: Preview refresh problem
video: NvEncLogicService errors/crash
video: support adaptive quantization nvenc / ffnvenc

*changed*

Vapoursynth: adjusted to lates DFTTest changes
jobs: process priority only on Windows systems
analysis: ffmpeg direct timecode extraction
Vapoursynth: used sub.TextFile for .srt subtitles (VsFilter is Windows only :/)
Vapoursynth: adjusted to HDR10ToSDRDG 1.12
Analysis: if time code count and indicated frame count differ more than 1000 frames use time code count as frame count
Global: don't check for avsviewer when no Avisynth is detected, don't check for VSPIPE or VSViewer, when Vapoursynth isn't detected
video: adjusted to x265 3.1, rewrote x265-"minimize command line"
Preview: report Avisynth and Vapoursynth Errors not through popups, but through an entry into the log

*added*

gui: option to 'ask for confirmation' in multiple places
gui: optional warnings when deleting/resetting/... stuff
jobs: auto save after adding jobs option
video: support for const quality in ffmpeg (nvenc)
video: support for NVEncCs 'constant quality'-mode
video: x265, support for adaptive quantization mode 4
video: x265, support hierarchical motion estimation (--hme/--hme-search)
video: nvenc, support --dhdr10-info, allow aq-strength for H.265, --nonrefp, i/b adapt
jobs: 'ignore crashed'-option
Vapoursynth: vs subtitle blend option
Vapoursynth: 'custom font path'-option

*removed*

audio: support for libvo (since ffmpeg dropped support for it back in 2016)


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: working on providing a Vapoursynth package or build script for Debian based Linux systesm, but stuck atm. :/

[Selur]

Hybrid rev 2019.08.26.1: (Windows only release)
*fixed*

switched back to Vapoursynth R46, since R47 breaks vsViewer and vsedit.


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

Hybrid rev 2019.09.29.1:
*fixed*

Jobs: d2v Indexer output detection
x265: removed '--no-intra-refresh'
Input: ffmpeg direct time code extraction
Jobs: x265, don't create .cutree-deletion job when '--no-cutree' is used.
Jobs: 'MPEG-4 ASP + mkv stretch'-handling on passthrough
FilterPreview: with interlaced content, that doesn't get deinterlaced while 'No Avisynth during Preview' is disabled

*changed*

Vapoursynth: make vsFilterMod optional (Windows only)
Vapoursynth: handle chapter selection
x265: support selective SAO
Preview: mplayer preview + crop force refresh
Removed old compatibility code for old platforms (XP/Vista/SnowLeopard/...)
Input: small stuff, disable stretch detection (since it only works for video atm.)

*added*

Vapoursynth: Cnr2, TemporalDegrain filter
Synth: support 'prefer HW decoding' for LWLibAvSource


-> downloads: http://www.selur.de/downloads

Cu Selur

[Jamaika]

Hi Selur

Have you tried to add codec jctvc or jvetvvc to ffmpeg? Does 'media-autobuild_suite' enable this?

Cu Jamaika

[Selur]

Nope, I've never tried jctvc or jvetvvc and MBAS doesn't support them.

[Jamaika]

Well done with hybrid.
I ask out of curiosity because I was interested in your question about the new svt hevc codec?

[Selur]

Thought about adding support for the SVT encoders to Hybrid, but that turns out to be dead end for now, since the seem to be more broken then working atm. ;)

[Jamaika]

Maybe, have you added SVT to x265 in the new Hybrid?

[Selur]

No I haven't added support for SVT since compiling x265 with SVT support failed in MBAS and I prefer to use the standalone application since it offers access to more of the switches.
Atm. STVs HEVC encoder (same for x265 with SVT) isn't really useful for me since you can't specify VUI flags for luma scale (pc/tv) or color matrix, thus ending up with wrong colors on playback.
(Also with piping not working properly in the current stand alone StvHevcEncApp, I wouldn't recommend using this encoder for anything but general technology tests,..)

Cu Selur

[Rumbah]

I just noticed that with Hybrid 2019.09.30.1 the x264 --b-adapt setting is one too low:

From the documentation:
--b-adapt <integer> Adaptive B-frame decision method [1]
Higher values may lower threading efficiency.
- 0: Disabled
- 1: Fast
- 2: Optimal (slow with high --bframes)
In the advanced x264 Hybrid setting it says and sets optimal=1 and fast=0. The presets Hybrid sets get it wrong, too.

[aj89]

HI

How ican remove gui encode frome media info
i dont like show hybrid name
https://imgur.com/t9kYymb

[Selur]

Hybrid rev 2019.11.03.1:
*fixed*

VideoModel: loading a preset overwrote the resolution&c
Vapoursynth: SMDeGrain Ui glitch
Vapoursynth: KNLMeansCL, color adjustment for cmode
Vapoursynth: LWLibavSource: don't add prefer_hw for AVC with high bit depth
Vapoursynth: CTMF dependency missing for TemporalDegrain depending of Input and/or settings

*changed*

Analyse: Always using '--File_TestContinuousFileNames=0' with MediaInfo.
Avisynth: adjust interlaced crop handling
NVEncC: adjusted progress indication
Subtitle: Teletext handling
muxing: explicilty set empty 'name'-tags when using MP4box to avoid a bug in MediaInfo
Video: NVEncC support for --multiref-l0/--multiref-l1
Video: adjusting To Kvazaar updates

*added*

Vapoursynth: FilterView 'difference'-mode
Video: support SVT-HEVC
Avisynth: added support for TemporalDegrain2


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

Arghh,.. email notification doesn't seem to work in the forum atm.

@Rumbah: Opening Hybrid, 'x264->Frames->B-Frame Settings->Adaptive b-frame decision' odes list 'fast', 'optimal' and 'disabled', but the command line seems wrong.
-> will look into it => disabled and optimal are switched :/
@aj89: If you don't like Hybrid, don't use it. (I don't see any valid reason for not adding that info during the muxing.)

[von Suppé]

Hi Selur,

I get an error when trying to recode a video (pls from a BD) with x264 (with "--bluray compat" in the commandline). It says: "Current output framerate 29.97 is not allowed.
29.97 fps is allowed for bluray when video is interlaced. The source is a BD.

MediaInfo on the source-playlist:

Scan type: MBAFF
Scan type, store method: Interleaved fields
Scan order: Top Field First

I recall having these types of video transcoded earlier succesfully.
Am I doing something wrong? Using your latest version.

Cheers
von Suppé

[Selur]

Need more details, for example what framer rate and resolution does you video have?

Hybrid should check that you use one of the following resolution and frame rate combinations:

1920/1440x1080 23.976p/24p/25i/29.97i
1280x720 23.97p. 24p, 50p, 59,940p
720x576 25i
720x480 29.97i

and only complain if blu-ray comatibility is enabled and none of these combinations is used.
If you aim for interlaced output, make sure to enable either 'x264->Frames->Advanced Frame Settings->Interlaced encoding' or 'x264->Frames->Advanced Frame Settings->Force bff' otherwise x264 and Hybrid would assume that the output should be progressive.

Cu Selur

Ps.: I assume 'x264->Base->Restriction Settings->Hardware' is enabled and set to 'Bluray/AVCHD' and you didn't add '--bluray-compat' through a custom command line addition.

[von Suppé]

MediaInfo states:

Width: 1920 pixels
Height: 1080 pixels
Frame rate: 29.970 (30000/1001) FPS

Checking "Interlaced encoding" both with and without "Force bff"checked gave me same error "Crashed with exit status 0"

I also remuxed playlist into mkv and tried that as input, same errors.

Edit: Got it, sorry. My bad, my guess. Audio was set to "ignore". Changed to "passthrough all", now it's running.

[Selur]

Don't realls get what the audio has to with it all, but if you found the problem happy that worked out. :)

Cu Selur

[Selur]

Hybrid rev 2019.11.23.1:
*fixed*

muxing: ffmpeg indicating vfr even when video is cfr
video: x265 HDR-10+ signaling
video: maxCLL typo related to STV-HEVC encoding (can't be used there anyways so more a cosmetic fix)
video: NVEncC multi level typo bug
Avisynth: crop not working with custom resizer
muxing: mp4 language passthrough
gui: VIDEOHANDLING problem on Linux

*changed*

profiles: Do not allow to load Global Profiles while a source is loaded.
Vapoursynth: Make sure that after QTGMC the clip is preceived as frame based.
video: NVEncC allow L0 and L1 references at the same time
video: x265 added support for 'scenecut-aware-qp' with 'scenecut-window' and 'max-qp-delta'


-> downloads: http://www.selur.de/downloads

Cu Selur

[ReinerSchweinlin]

I did my first test runs with hybrid and have to say the amount of capabilities in hybrid is impressive, thank you for this nice peace of software :)
I wanted to resize some stuff with waifu2x which works good. Since my testfiles are all yu12 (XVID, old Cartoons, etc..) - would it be possible to change the intermediate Colourspace to YU12 or RGB24 instead of RGB48 - would speed benefit from it? I don´t need 1080p to 8k resizes, but want to restore some rare old footage, some captured from VHS.

Any advice on a potent grafix card for waifu2x resize up to around 300 $/€? It seems NVIDIA is supportet better but some ATI Cards have mor bang for the buck (maybe use vulkan with ati)?

[ReinerSchweinlin]

I just noticed one thing: after resizing with waifu2x, the resulting file seems to have a different color range than the original one - the script seems to convert to TV levels, black now is dark grey and bright areas are a little dimmer (maybe PC->TV Conversion forced?) - any way I can change that in the settings? Looked around, but can´t find it.

[Sharc]

With avisynth enabled (or similar with vapoursynth):
Tab Filtering -> Avisynth -> Tab Color -> Tickbox ColorYUV -> Levels …..

[ReinerSchweinlin]

Thank you. I have nothing selcted and "ange conversion" says: "do not"... But it does... I tried setting levels manualy, didn´t help. The script says:
# Setting color range to TV (limited) range.
clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)
maybe this is causing the color change?

[ReinerSchweinlin]

I did another check - if waifu2x is disabled, colours are passed through unchanged. With waifu2x enabled, range is altered automaticaly.

[ReinerSchweinlin]

I just stumpled across this:
https://forum.doom9.org/showthread.php?t=176961
Could this be implemented in Hybrid? I know, I´d be needing a bigger GPU, but I have some personal rare footage I´d like to throw at it..

[Selur]

Hybrid rev 2019.12.01.1:
*fixed*

video: x265 make sure 'fixed quantizer' zones use integer

*changed*

muxing: Hybrid doesn't send 'und' to MP4Box but empty language strings

*added*

muxing: option to overwrite not only container but additionally stream aspect ratio


-> downloads: http://www.selur.de/downloads

----
Last I checked the main problem with lots of these neural network solutions is finding the tools and the dependencies that need to be available to incorporate them into Hybrid.
Often the dependencies are huge (in file size) and some times they are by their license not allowed to be distributed in parts or in a whole through others.
-> so if you can list a tool with its dependencies and those dependencies can be shipped with Hybrid and are not humongous I could add that tool/filter to Hybrid, atm. there are no plans for this.

"With waifu2x enabled, range is altered automaticaly. "
Hybrid only sets the range detected by the analysis.


Cu Selur

[Havokdan]

I some time without converting files, but experiencing this version, I realized that does not carry the QSVEnc profiles, I create one and try to match a aac audio, but always carries the codec x264.

[Selur]

I don't really understand what you are speaking of. :)
In case QSVEnc isn't snown in case:
a. no Intel CPU is detected
b. 'QSVEncC --check-features' reports 'QSV unavailable.' or nor H.264 and H.265 features.
Usually this is due to conflict of QSVEncC with the CPU drivers.
-> What does 'QSVEncC --check-features' report on your system ?

Cu Selur

[Havokdan]

I don't really understand what you are speaking of. :)
In case QSVEnc isn't snown in case:
a. no Intel CPU is detected
b. 'QSVEncC --check-features' reports 'QSV unavailable.' or nor H.264 and H.265 features.
Usually this is due to conflict of QSVEncC with the CPU drivers.
-> What does 'QSVEncC --check-features' report on your system ?

Cu Selur

I'm sorry, I'm using an online translator, I wanted to say this, I create a profile for QSVEnc, so I try to create an "AV-Combo" profile with him, however, when trying to load the profile, video codec that is loaded it is x264.

[ReinerSchweinlin]

Thanx for the updates!!


----
Last I checked the main problem with lots of these neural network solutions is finding the tools and the dependencies that need to be available to incorporate them into Hybrid.
Often the dependencies are huge (in file size) and some times they are by their license not allowed to be distributed in parts or in a whole through others.
-> so if you can list a tool with its dependencies and those dependencies can be shipped with Hybrid and are not humongous I could add that tool/filter to Hybrid, atm. there are no plans for this.

Unfortunately, my programming skills are too basic to really get a grasp of everything needed to implement this filter, but as far as I understand it, this
https://github.com/imPRAGMA/VSGAN
simply seems to be a vapourynth port of ESRGAN.
Reading through the install-files, it looks straight forward to me - all tthat´s needed to add is the model which calculations are done with.
Some are found here:
https://upscale.wiki/wiki/Model_Database
which seem to work, according to some forum members...
I wouldn´t mind downloading these manualy :)

"With waifu2x enabled, range is altered automaticaly. "
Hybrid only sets the range detected by the analysis.

thanx, I will investigate my source file further to see if there is weird colour setting in it that causes the range change...
Does Hybrid do a different analsys if wafu2x is enabled? If I leave eveyrthing unchanged and simply turn waifu2x on or off, I get different colour ranges, so my assumption is, that it has something to do with waifu2x, not hybrids interpretation of the source file analysis.

[Selur]

I wanted to say this, I create a profile for QSVEnc, so I try to create an "AV-Combo" profile with him, however, when trying to load the profile, video codec that is loaded it is x264.
Ah okay, will try to reproduce this tomorrow. :)
(Haven't used QSVEnc for ages since my encoding machine uses an AMD CPU and my other machines only have older Intel CPUs where GPU encoding wasn't that appealing,..)
-> will fix, happens not only with QSVEnc, so it's easier to reproduce and fix. :)

Does Hybrid do a different analsys if wafu2x is enabled? If I leave eveyrthing unchanged and simply turn waifu2x on or off, I get different colour ranges, so my assumption is, that it has something to do with waifu2x, not hybrids interpretation of the source file analysis.
No the analysis is right but unless a rgb <> yuv conversion is needed the color range is only used for flagging in the output.
In case a rgb <> yuv conversion is used the assumed color matrix will be used during that conversion.
Here's an example:
# Imports
import vapoursynth as vs
core = vs.get_core()
import ctypes
# Loading Support Files
Dllref = ctypes.windll.LoadLibrary("I:/Hybrid/64bit/vsfilters/ResizeFilter/Waifu2x/w2xc.dll")
# Loading Plugins
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/ResizeFilter/Waifu2x/Waifu2x-w2xc.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/Support/fmtconv.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/SourceFilter/FFMS2/ffms2k.dll")
# Loading source using FFMS2K
clip = core.ffms2.Source(source="F:/TESTCL~1/files/test.avi",cachefile="E:/Temp/avi_078c37f69bb356e7b5fa040c71584c40_853323747.ffindex",format=vs.YUV420P8,alpha=False)
# making sure input color matrix is set as 470bg
clip = core.resize.Point(clip, matrix_in_s="470bg",range_s="limited")
# making sure frame rate is set to 25
clip = core.std.AssumeFPS(clip, fpsnum=25, fpsden=1)
# Setting color range to TV (limited) range.
clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)
# resizing clip to 1280x704
# adjusting color space from YUV420P8 to RGB48 for Waifu2x-w2xc
clip = core.resize.Bicubic(clip=clip, format=vs.RGB48, matrix_in_s="470bg", range_s="limited")
# current color sampling: RGB48
# adjusting bit depth to 32bit
clip = core.fmtc.bitdepth(clip=clip,bits=32)
# resizing using Waifu2x
clip = core.w2xc.Waifu2x(clip=clip, scale=2)
# current color sampling: RGB48
# adjusting output color from: RGB48 to YUV420P8 for x264Model (i420)
clip = core.resize.Bicubic(clip=clip, format=vs.YUV420P8, matrix_s="470bg", range_s="limited")
# Output
clip.set_output()
Since you didn't share any details I have no way to reproduce your problem and can't really help.

Cu Selur

[ReinerSchweinlin]

Hi Selur,

thank you for your answers.
I tried some stuff and found it: After setting range conversion to "limited->full" in Filtering-vs-colour, the colors stay the same.

The script looks like this now:

# Imports
import vapoursynth as vs
core = vs.get_core()
import ctypes
# Loading Support Files
Dllref = ctypes.windll.LoadLibrary("C:/Program Files/Hybrid/64bit/vsfilters/ResizeFilter/Waifu2x/w2xc.dll")
# Loading Plugins
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/ResizeFilter/Waifu2x/Waifu2x-w2xc.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/Support/fmtconv.dll")
core.std.LoadPlugin(path="C:/Program Files/Hybrid/64bit/vsfilters/SourceFilter/LSmashSource/vslsmashsource.dll")
# Loading G:\video\test.mkv using LWLibavSource
clip = core.lsmas.LWLibavSource(source="G:/video/test.mkv", format="YUV420P8", cache=0, prefer_hw=3)
# making sure input color matrix is set as 470bg
clip = core.resize.Point(clip, matrix_in_s="470bg",range_s="limited")
# making sure frame rate is set to 24
clip = core.std.AssumeFPS(clip, fpsnum=24, fpsden=1)
# Setting color range to TV (limited) range.
clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)
clip = core.resize.Point(clip, range_in_s="limited", range_s="full")
# Setting color range to PC (full) range.
clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=0)
# resizing clip to 1280x960
# adjusting color space from YUV420P8 to RGB48 for Waifu2x-w2xc
clip = core.resize.Bicubic(clip=clip, format=vs.RGB48, matrix_in_s="470bg", range_s="full")
# current color sampling: RGB48
# adjusting bit depth to 32bit
clip = core.fmtc.bitdepth(clip=clip,bits=32)
# resizing using Waifu2x
clip = core.w2xc.Waifu2x(clip=clip, scale=2, noise=0, block=1024)
# current color sampling: RGB48
# adjusting resizing
clip = core.fmtc.resample(clip=clip, w=1280, h=960, kernel="lanczos", interlaced=False, interlacedd=False)
# adjusting output color from: RGB48 to YUV420P8 for FFvHuffModel (i420)
clip = core.resize.Bicubic(clip=clip, format=vs.YUV420P8, matrix_s="470bg", range_s="full")
# Output
clip.set_output()

While testing some of the other Waifu2x Tools, I noticed quite a bit of speed increase when using RGB24Bit intermediate colour space without noticable quality change with old anime encoded in XVID (which are Yu12). Do you think this could speed up conversion in Hybrid, too? After turning off the noise reduction in Waifu2x and encoding to Huffyuf, upscaling a 640x480 file 2fold in size, I get 1fps on an gtx960... (i5 Kabylake).

when using wafu2x in Foto mode while keeping the resize factor to 1, it´s sometimes possible to clear up files with real life footage from noise, blocking, artefacts... In Hybrid, only upscaling is possible, as soon as the resizer is set to the input size, it leaves out waifu2x in the qeue. Could you implement "keep the resolution but still run it through waifu2x" please?

Ah, on the download page I still get the previous version, although its named different.

Another thing I noticed: While trying to distribute work onto cpu/gpus, I wanted to switch to qsv decoding. But no matter what I select in the config-input-decoding section, it´s always the nvidia card decoding.

Thanx for your help!

[Selur]

Ah, on the download page I still get the previous version, although its named different.
nope. just forgot to update the revision info properly. :)
While trying to distribute work onto cpu/gpus, I wanted to switch to qsv decoding. But no matter what I select in the config-input-decoding section, it´s always the nvidia card decoding.
I don't think Hybrid atm. does support using qsv for decoding with Vapoursynth.
It might be possible if I change the HW decoding support Hybrid uses with LibavSMASHSource, atm. Hybrid uses either "prefer_hw=0", or "prefer_hw=3", but I could extend this to :
- 0 : Use default software decoder.
- 1 : Use NVIDIA CUVID acceleration for supported codec, otherwise use default software decoder.
- 2 : Use Intel Quick Sync Video acceleration for supported codec, otherwise use default software decoder.
- 3 : Try hardware decoder in the order of CUVID->QSV. If none is available then use default software decoder.
-> will look into it.

[Selur]

Looked into https://github.com/imPRAGMA/VSGAN from what I gather Hybrid would need to install:
* PyTorch
* Numpy
* mvsfunc
* CUDA (NVIDIA GPU Support, optional but without this everything will be extremely slow)
* vsgan
-> unless there is a portable package I could ship with Hybrid or provide as a separate download I won't write support for this.
Will send you a PM in regard to LibavSMASHSource 'prefer_hw' and allowing waifu2x while no real resizing is done,...

While testing some of the other Waifu2x Tools, I noticed quite a bit of speed increase when using RGB24Bit intermediate colour space without noticable quality change with old anime encoded in XVID (which are Yu12). Do you think this could speed up conversion in Hybrid, too?
Yes, using RGB24 instead of RGB48 which Vapoursynth uses by default might speed-up stuff a bit. How much are we talking about here. (Not really sure it's worth the effort to support this.)

Cu Selur

[ReinerSchweinlin]

Looked into https://github.com/imPRAGMA/VSGAN ... unless there is a portable package I could ship with Hybrid or provide as a separate download I won't write support for this.

ok, thanx for looking into it. Maybe the vapoursynthport will advance over time and I hope there will be a portable, simple solution available in the future. Postponed for the moment..

Will send you a PM in regard to LibavSMASHSource 'prefer_hw' and allowing waifu2x while no real resizing is done,...

Cool, thank you! Will try


Yes, using RGB24 instead of RGB48 which Vapoursynth uses by default might speed-up stuff a bit. How much are we talking about here. (Not really sure it's worth the effort to support this.)

My comparison is based video2x, which converts a videofile into intermediate PNG Files, throws all of them into waifu2x and then joines the image sequence back together into a videofile. Of course, the overhead here is much bigger than with hybrid which pipes all the steps into skript, so the speed difference might vary.

in Video2x, The speed difference on my system is between 50% and 80% speedup when using RGB24 instead of RGB48 (temp on SATA SSD)..

I could try to test this myself - my amateur skills give me the idea to tap into the skript of hybrid, play around with the settings and run it manualy to measure the speed difference. I´d have to do some digging, did some avisynth back in the days... Maybe I get it done - do you think this could be an approach to test it (so I can contribute a little work to the topic) ?

[Selur]

Maybe I get it done - do you think this could be an approach to test it (so I can contribute a little work to the topic) ?
You can see the script (Filtering->Vapoursynth->Vapoursynth Script View), copy the script into a new file and call vspip:
VSPipe.exe -p <PATH TO SCRIPT> NUL
and ist should output something like:
Frame: 115/5000 (2.62 fps)
(on a small test using RGB24 instead of RGB48 didn't give any speed gain; but that was only one test)

Cu Selur

[ReinerSchweinlin]

Thanx!

Report:
using RGB24 instead of RGB48 has almost no effect with my examples - maybe upscaling HD->4K would benefit more, but my GFX Card is too small and slow - with SD Sources I see no speed difference

Using Waifu2x with scale 1 ("enhance, no upscale") uses

# resizing using Waifu2x
clip = core.w2xc.Waifu2x(clip=clip, scale=2, block=1024)

In the VS Script. So it´s still upscaled and then resized to the original resolution. So it takes the same time as upscaling 2x.
Would it be possible to set

# resizing using Waifu2x
clip = core.w2xc.Waifu2x(clip=clip, scale=1, block=1024)
?

Thanx to your hint above on how to call vspipe manualy, I did a speedtest:

# resizing using Waifu2x
0,95fps <- clip = core.w2xc.Waifu2x(clip=clip, scale=2)
4,33fps <- clip = core.w2xc.Waifu2x(clip=clip, scale=1, block=1024)


Regarding HW-Decode with QSV: So far no luck, have to investigate more, seems decoding is still done in SW or Nvidia Card.

[Selur]

Looking into it.
---
HW decoding requires LWLibAVSource (or DGDecNV) to be used and for QSV prefer_hw should be set to 2, script wise Hybrid seems to properly allow to set the value, so if QSV isn't used working it might be a driver problem or a bug in LWLibAVSource.

[ReinerSchweinlin]

Hey. Waifu works fine now, many thanks for looking into this! The colour issue is gone and scale=1 works, too :)

Assigning all the GPU Power still is a little bumpy - when doing NVENC (RTX 2060) Encodes, I assign NLMEANS CL to the same card (CUDA) and decoding to QSV - which won`t work. In Handbrake, it works right away and with others, too, so drivers etc.. seem to be ok..
I have to do some more testing on this to give a clearer report - for the meantime, this is no problem, since the speed difference is small.

Christmas probably kicking in @all, so time will be spent elsewhere :)

[Selur]

Hybrid rev 2020.01.11.1:
*fixed*

analyse: HDR parameter import rounding
Synth: ignore timecodes for output if FrameInterpolation or Restore is
cosmectice: AudioTab, AudioInfo on track change
Synth: ImageSource handling
video: qsv encoding mode
tools: manual set tools could cause crashes
Vapoursynt: Waifu2x limited range compression by using RGBS
video: vsPipe Problem gefixed
video: NVEnc only singal VUI in case VUI Signaling is enabled
audio: handling of multiple trueHD streams with ffmpeg
jobs: mkvmerge muxing timecodes parallel processing
video: x265 support histogram scence change detection

*changed*

audio: dts-es muxing
Synth: allow to set LibWAV HW decoding preference
Vapoursynth: allow Waifu2X even when resize resolution is the same as the crop resolution.

*added*

Vapoursynth: added support for EdgeFixer
Synth: 'SelectRangeEvery' for Avisynth and Vapoursynth (video only!)


-> downloads: http://www.selur.de/downloads

Cu Selur

[Havokdan]

I already made the request to the StaxRip, do the same for the Hybrid now automatically import the Hybrid is possible tags/fields nfo files (https://kodi.wiki/view/NFO_files) when the support such target format information?

[Selur]

Not totally sure I get that. :)
Do you want Hybrid to:
a. create kodi-NFO-files from the tags of a file, or
b. import tag information from an NFO file, or
c. do something different I totally missed?

Cu Selur

[Havokdan]

B

I want to import the nfo file information to the video file in a format that supports such information, the information in the nfo file that usually use is the standard kodi.

[Selur]

Hmm, looking at what tags Hybrid supports (for the main containers):

Container tags:

Name
Genre
Comments
Copyright
Cover

Stream tags:

Title
Language


and the tags Kodi allows (https://kodi.wiki/view/NFO_files/Movies#nfo_Tags)
I don't see anything for Copyright in the movie info and the Cover info would probably best match to the 'thumb' info, but to be useful Hybrid would need to download stuff (not likely to happen).

I see the following mapping:
Container->Name <=> movie->title
Container->Genre <=> movie->genre
Container->Comments <=> movie->plot
and
for audio streams:
Stream->Title <=> movie->fileinfo->streamdetails->audio->language
There is no language for video streams.

Since normally at least the audio language tags are usually properly flagged in the files itself, the only real gain from this would be to import the container tags.
(Importing the audio language could also be done, but probably not worth the trouble.)

-> don't really think that it importing the container tags is worth the effort, but if you find a few other users which think this is a nice feature I can probably write some code for it.

Cu Selur

Ps.: for mkv output adding a 'import for nfo' under 'Config->Output->MKV Tags' might be more useful, since there would be a bigger overlap of the kodi-NFO data and the mkv tags.

[Havokdan]

Hmm, looking at what tags Hybrid supports (for the main containers):

Container tags:

Name
Genre
Comments
Copyright
Cover

Stream tags:

Title
Language


and the tags Kodi allows (https://kodi.wiki/view/NFO_files/Movies#nfo_Tags)
I don't see anything for Copyright in the movie info and the Cover info would probably best match to the 'thumb' info, but to be useful Hybrid would need to download stuff (not likely to happen).

I see the following mapping:
Container->Name <=> movie->title
Container->Genre <=> movie->genre
Container->Comments <=> movie->plot
and
for audio streams:
Stream->Title <=> movie->fileinfo->streamdetails->audio->language
There is no language for video streams.

Since normally at least the audio language tags are usually properly flagged in the files itself, the only real gain from this would be to import the container tags.
(Importing the audio language could also be done, but probably not worth the trouble.)

-> don't really think that it importing the container tags is worth the effort, but if you find a few other users which think this is a nice feature I can probably write some code for it.

Cu Selur

Ps.: for mkv output adding a 'import for nfo' under 'Config->Output->MKV Tags' might be more useful, since there would be a bigger overlap of the kodi-NFO data and the mkv tags.

Ok, thanks :)

[Selur]

Hybrid rev 2020.01.12.1:
*fixed*

decode: crash when ffmpeg was not compiled with 'zscale'

-> downloads: http://www.selur.de/downloads

Cu Selur

[Gravitator]

Хай!
Hybrid v2020.01.12.1

I use a two-second sample ProRes for encoding.
The culprit of the memory consumption of Deinterlacer: QTGMC (ffmpeg not eating).

[Selur]

No surprise, QTGMC will use quite a bit of memory and cpu power depending on the resolution and settings,...
Not sure why you mention ffmpeg, you are using VSPipe and x265, ffmpeg isn't used.

Cu Selur

[Gravitator]

No surprise, QTGMC will use quite a bit of memory and cpu power depending on the resolution and settings,...Deinterlacer: QTGMC - eats 4GB at any encoding settings (Overcoming the memory limit leads to the reset of information on the virtual memory HDD - fatal system throttling / only reboot).
Not sure why you mention ffmpeg, you are using VSPipe and x265, ffmpeg isn't used.Deinterlacer: FFMpeg - eats 80MB.
This only happens when using ProRes (Drop sample?).

[Gravitator]

Checking the new release leads to the collapse of Hybrid.

[Gravitator]

AOM
10bit mod & Profile: Main, High, Professional - оn any settings, disabling the "64-pt transform" leads to an instant encoding finish (Empty *.mkv or nothing at the output);
8bit mod & Profile: Main - ok, High - no, Professional - no.

[Gravitator]

RAV1E
Add new features: --threads, --first/second-pass (Instead of --pass 1/2), --rdo-lookahead-frames.

[Selur]

Deinterlacer: QTGMC - eats 4GB at any encoding settings
You can:
a. limit the thread count
b. limit the cache size
in Hybrid which will allow you to reduce the memory footprint.

Deinterlacer: FFMpeg - eats 80MB.
So ? Using a simple Yadif (or most others simpler deinterlacing methods) instead of QTGMC will use less memory.

Checking the new release leads to the collapse of Hybrid.
fixed locally.

AOM
10bit mod & Profile: Main, High, Professional - оn any settings, disabling the "64-pt transform" leads to an instant encoding finish (Empty *.mkv or nothing at the output);
8bit mod & Profile: Main - ok, High - no, Professional - no.
may be the command line options changed

RAV1E
Add new features: --threads, --first/second-pass (Instead of --pass 1/2), --rdo-lookahead-frames.
will think about it

[Gravitator]

It’s not normal that for 2 second (60 frames) 1080p Prores video, it consumes 4GB or more (There is no control of memory consumption).
-----
This only happens with ProRes in the *.mkv container! Then comes the call to VSPipe (*.mov moderate memory consumption and without VSPipe).

[Selur]

It’s not normal that for 2 second (60 frames) 1080p Prores video, it consumes 4GB or more (There is no control of memory consumption).
You are right QTGMC should not be used at all for 1080p content unless you explicitly tell Hybrid to do so.

Using:
# Imports
import os
import sys
import ctypes
# Loading Support Files
Dllref = ctypes.windll.LoadLibrary("I:/Hybrid/64bit/vsfilters/Support/libfftw3f-3.dll")
import vapoursynth as vs
core = vs.get_core()
# Import scripts folder
scriptPath = 'I:/Hybrid/64bit/vsscripts'
sys.path.append(os.path.abspath(scriptPath))
# Loading Plugins
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/GrainFilter/AddGrain/AddGrain.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/DenoiseFilter/FFT3DFilter/fft3dfilter.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/DenoiseFilter/DFTTest/DFTTest.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/Support/EEDI3.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/ResizeFilter/nnedi3/vsznedi3.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/Support/libmvtools.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/Support/temporalsoften.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/Support/scenechange.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/Support/fmtconv.dll")
core.std.LoadPlugin(path="I:/Hybrid/64bit/vsfilters/SourceFilter/LSmashSource/vslsmashsource.dll")
# Import scripts
import havsfunc
# Loading F:\TestClips&Co\files\interlaceAndTelecineSamples\interlaced\avc - interlaced.m2ts using LWLibavSource
clip = core.lsmas.LWLibavSource(source="F:/TESTCL~1/files/INTERL~1/INTERL~1/AVC-IN~1.M2T", format="YUV420P8", cache=0, prefer_hw=0)
# making sure input color matrix is set as 709
clip = core.resize.Point(clip, matrix_in_s="709",range_s="limited")
# making sure frame rate is set to 29.970
clip = core.std.AssumeFPS(clip, fpsnum=30000, fpsden=1001)
# Setting color range to TV (limited) range.
clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)
# setting field order to what QTGMC should assume
clip = core.std.SetFrameProp(clip=clip, prop="_FieldBased", intval=2)
# Deinterlacing using QTGMC
clip = havsfunc.QTGMC(Input=clip, Preset="Fast", TFF=True)
# make sure content is preceived as frame based
clip = core.std.SetFieldBased(clip, 0)
clip = clip[::2]
# Output
clip.set_output()
Vapoursynth uses ~4.2GB, using preset 'placebo' it uses an additional ~300MB here.
Settin "Filtering->Vapoursynth->Misc->Limit Cache MB" to '1024', less than 1500MB get used.

Checking the new release leads to the collapse of Hybrid.
the option is removed now; too much trouble fixing it properly

Cu Selur

[Selur]

AOM
10bit mod & Profile: Main, High, Professional - оn any settings, disabling the "64-pt transform" leads to an instant encoding finish (Empty *.mkv or nothing at the output);
8bit mod & Profile: Main - ok, High - no, Professional - no.
calling it in a command line I got a few messages:
TX64 can only be disabled on I420 images.
8bit + profile 1: Profile 1 requires 4:4:4 color format
8bit + profile 2: Profile 2 bit-depth < 10 requires 4:2:2 color format

The main surprise here is the color space restriction since this is more restrictive than https://en.wikipedia.org/wiki/AV1#Profiles which doesn't allow High profile with 4:2:0,...
-> created an issue entry over at: https://github.com/mozilla/aom/issues/10

[Selur]

Hybrid rev 2020.02.16.1:
*fixed*

Video: QSVEnc encoding mode mapping.
Vapoursynth: custom script mismapping
Vapoursynth: SelectRangeEvery and Tweak
Vapoursynth: Fixed HDR to SDR DG call
Vapoursynth: HDRToSDR
Vapoursynth: LWLivavSource use track instead of streamindex
Avisynth: removed 8bit enforcement for Tweak and Levels
Video: aomenc - tx64 handling
Profile: error message regarding 'input' when loading a GlobalDataModel
Muxing: mp4 video&audio tagging
Video: NVEnc, arib-std-b67 vs. arib-srd-b67
Muxing: h264 avi muxing
Analysis: mediainfo video bitrate indication
Avisynth: handling high bitdepth content
Decoder: FFmpeg decoder call

*changed*

Avisynth: force LibavSource for AV1 content
Avisynth: make sure BalanceBorders isn't used when Top/Bottom/Left/Right isn't mod 2

*added*

Vapoursynth: option to save/load custom script content
Video: rav1e added "--threads X", "--switch-frame-interval X", "--reservoir-frame-delay X" and "--rdo-lookahead-frames X"
Video: NVEncC added support for '--psnr', '--ssim'

*removed*

check for update
Synth: removed FFms2k (it's changes were incorporated in ffms2)


-> downloads: http://www.selur.de/downloads

Cu Selur

[Gravitator]

Hybrid v2020.02.16.1

RAV1E
Unable to make the first pass, as in the TEMP folder is not created file *.ivf

[Selur]

I see the problem, will fix it when I find the time.
-> fixed locally, will be fine during next release

[Gravitator]

1st and 2nd pass works, but there is a problem with complex coding 2pass (The output is an empty container).

[Selur]

no clue what's causing this,... also I can't access the files attached there unless a moderator approves them.
-> got it

[Selur]

Hybrid rev 2020.02.17.1:
*fixed*

video: rav1e 2pass calls

*changed*

video: x265, adjusted to 3.3 branch


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

Hybrid rev 2020.02.19.1:
*fixed*

Video: x265, messed up during 3.3. branch adjustment

*changed*

Internals: removed some abandoned code which causes unneeded slow downs after files with lot of meta data were analysed


-> downloads: http://www.selur.de/downloads

Cu Selur

[Gravitator]

Hybrid v2020.02.19.1

AOM
Noise reduction and AQ are not available for: abr, cq, cqr encoding mode?
2-pass > abr, cq, cqr, lossless > 2-pass - noise reduction is blocked.

[Selur]

Please, try to make whole sentences it's getting harder and harder to guess what you want to say.
No clue what cqr is.
Combining cq (= constantquantizer) with aq (adaptive quantizer) makes no sense.
Combining lossless with noise reduction doesn't make sense either.
Last I checked adpative quantization in aomenc requires 2pass encoding, so anything that isn't 2pass encoding can't use it.
Might be that some disabling of features is wrong in the gui, but without details I'm not looking into it.

Cu Selur

[Gravitator]

Might be that some disabling of features is wrong in the gui, but without details I'm not looking into it.
Cu Selur

2-pass mod - noise reduction box is unlocked > change the mode to: abr (or) q, cq, lossless > return back to 2-pass mod - noise reduction box is blocked (The reason is to switch --auto-alt-ref=1 to --auto-alt-ref=0).

[Gravitator]

Combining cq (= constantquantizer) with aq (adaptive quantizer) makes no sense.
I have observed improvement (the reflection of the water and the edge of the trees) > Example (https://files.videohelp.com/u/227452/AOM%20%28q_aq0_aq1%29.zip)
aomenc -w 1920 -h 1080 --threads=0 --passes=1 --pass=1 --cpu-used=8 --i420 --input-bit-depth=10 --bit-depth=10 --end-usage=q --cq-level=54 --auto-alt-ref=1 --aq-mode=0 --webm -o k:\aq0.webm
aomenc -w 1920 -h 1080 --threads=0 --passes=1 --pass=1 --cpu-used=8 --i420 --input-bit-depth=10 --bit-depth=10 --end-usage=q --cq-level=55 --auto-alt-ref=1 --aq-mode=1 --webm -o k:\aq1.webm
-------
AOM
New! --tune=psnr, ssim, vmaf_with_preprocessing, vmaf_without_preprocessing, vmaf (Removed CDEF and DAALA)
New! --disable-trellis-quant=<arg> Disable trellis optimization of quantized coefficients (0: false 1: true 2: true for rd search 3: true for estimate yrd serch (default))

[Selur]

Okay, if they now support 1pass with adaptive quantizer I could add support for it in Hybrid. (Like I wrote, that didn't work last time I checked.)

[Gravitator]

x265
Encoding problem with any settings when "Recursion skip" is disabled.

[Selur]

seems like the syntax changed and I missed it.

[Selur]

Hybrid rev 2020.05.01.1:
*fixed*

Monitoring: some small bugs with the monitoring feature
Muxing: removed '-r' as input option for ffmbc, since it causes 'Option framerate not found.', works fine with ffmpeg
Jobs: FFmpeg job optimizer, properly handle extern audio
Vapoursynth: glitch with Vapoursynth Filter Preview
Avisynth: Deen mode restictions
Avisynth: dfttest ui mapping issue
Muxing: ui crash when video calls didn't generate a video output
Video: NVEnc: UI init problem regarding HDR selection
Vapoursynth: fix odd/even selection when not using bob
Vapoursynth: when using Stab include mvtools
Synth: output bit depth adjustment on VideoModel bitDepth change for some VideoModels
Vapoursynth: support for Bwdif deinterlacer
Muxint: muxing: title&language passthrough during mp4 muxing
GUI: font size adjustments (misc.ini)


*added*

Video: SVT-AV1 support
Audio: 'ignore encode and use passthrough if format doesn't change' option
Synth: added FineDehalo support through Avisynth and Vapoursynth
Audio: eac3 encoding support
Filtering: added bwdif for FFmpeg


*changed*

Vapoursynth: QTGMC always set TFF according to 'order' even for progressive input
Video: Support new NVEncC rotate (and transpose) and adjust it to FilterModels rotate option
Video: NVEncC: aq handling
Video: NVEncC: supporting '--input-option' for Blu-ray support
Video: NVencC: support for HDR to SDR
Avisynth: handling of interlaced filtering
Avisynth: mftoons now use masktools2 instead of masktools
Avisynth: add trim according to cut&chapter selection
Video: x265 using --rskip-edge-threshold and --rskip X
Vapoursynth: allow Waifu2x scale =1
Video: support more aomenc options
Vapoursynth: support for 'Waifu2x NCNN Vulkan' (Windows only)
Synth: Updated svpflow dlls and adjusted Avisynth Interframe2.avsi
Vapoursynth: removed support for Waifu2x on Linux since I can't get it to compile


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: For Debian/Ubuntu Linux Users: I tried writing a build-vapoursynth and build-plugins script, see: https://github.com/Selur/hybrid-vapoursynth-addon Hope someone get's inspired and makes this more usable. :)

[ReinerSchweinlin]

Thanx a lot for the update :) Nice to see the Vulkan implementation of waifu2x together with the 1x upscale option - its very usefull to clean up anime stuff!!

One thing I noticed: Doing just this Waifu2x cleanup, I can´t output to fv1 anymore, which worked before. So I just Prores which is fine for the moment.

Hybrid really become my favorite tool for all kinds of filtering and conversion!

Thanx!

[MadMonkey57]

Hi there,

Thx Selur for this wonderful piece of software.

I'm reencoding an HEVC / HDR video with NVEnc (FFmpeg).
Encoding video stream works fine, but patching for HDR signaling fails.

My system: ubuntu 20.04, python 3.8.2

# ./nvhsp -full_range tv -colorprim bt2020 -colormatrix bt2020nc -transfer smpte-st-2084 -maxcll 1000 -maxfall 300 -chromaloc 2 -masterdisplay "G(13250,34500)B(7500,3000)R(34000,16000)WP(15635,16450)L(10000000,1)" "/tmp/sample_new.265" "/tmp/sample_new_hdr.265"

Traceback (most recent call last):
File "./nvhsp", line 802, in __new__
return _cache[auto]
KeyError: '-full_range,tv,-colorprim,bt2020,-colormatrix,bt2020nc,-transfer,smpte-st-2084,-maxcll,1000,-maxfall,300,-chromaloc,2,-masterdisplay,G(13250,34500)B(7500,3000)R(34000,16000)WP(15635,16450)L(10000000,1),/tmp/sample_new.265,/tmp/sample_new_hdr.265'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "./nvhsp", line 637, in tokenparser
length = int(length)
ValueError: invalid literal for int() with base 10: '-full_range'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "./nvhsp", line 806, in __new__
_, tokens = tokenparser(auto)
File "./nvhsp", line 647, in tokenparser
raise ValueError("Don't understand length '{0}' of token.".format(length))
ValueError: Don't understand length '-full_range' of token.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "./nvhsp", line 4301, in <module>
b = Bits(','.join(sys.argv[1:]))
File "./nvhsp", line 808, in __new__
raise CreationError(*e.args)
__main__.CreationError: Don't understand length '-full_range' of token.

Any idea?

UPDATE: BTW it works fine in Windows

[Selur]

No clue. Seems like nvshp aborts since it can't properly parse'-full_range tv', but according to the code (https://github.com/SK-Hardwired/nv_hevc_hdr_patcher/blob/master/nvhsp.py) it should work.
in line 29 range_list is defined as:
range_list = ['tv','full']
then in line 38 we have:
full_range = range_list.index(args.full_range)
so full_range should be zero.
My guess is that this is due to some change in Python.
You best contact the nvhsp author since I can't really say anything more,... (https://github.com/SK-Hardwired/nv_hevc_hdr_patcher)

Cu Selur

[Selur]

Hybrid rev 2020.06.21.1:
*fixed*

Video: rav1e 2pass calls
Video: x265, minimize command line problem with constant quantizer
Vapoursynth: interframe + sRestore
Avisynth: pad to make sure dfttest gets mod16 input
Vapoursynth: forgot to add Bwdif.dll to th the install (Windows only)
Muxing: passhtrough huffuyv
Video: do not add maxCll 0,0
Analyse: Blu-ray input analysis
Avisynth: imagesource bug
Avisynth: ConvertToYUV where colorMatrix was accidentally added
Video: Xvid 2pass stats file bug
Jobs: ffindex deleted to early
Jobs: trying to delete x264 mbtree file evenwhen it should not be there
Audio: dialog norm make sure that '-dialognorm 0' instead of '-dialognorm -0' is used
Input: ffmpeg time code extraction
Analyse: dts-xx detection
Handling: aid problem avi preview und tokenizing
Video: x265: separate 2pass passes
Jobs: progress indication when using qsvencc
Video: FFMpeg NVEncC, multi-ref was accidentally disabled
Chapters: switchting within sources with multi input and chapter import
Jobs: ffmpeg job minimization

*changed*

Avisynth: removed SSE(2)Tools
Avisynth: interplaced preview handling
Avisynth: moved histrogram control into Hybrid itself
Avisynth: removed AnimeIVTC since it's not required for QTGMC any more
Avisynth: switched to Avisynth+ 3.6
Avisynth: removed PlanarTools
Avisynth: restrict DeRainbow to 8bit
Avisynth: during ConvertToXY add ChromaInPlacement when converting from YV12 and ChromaOutPlacement when converting to YV12 in case MPEG-1 or DV Video input is detected.
Avisynth: allow to enforce ChromaInPlacement/ChromaOutPlacement during ConvertToXY when converting to or from YV12
Cut: allow to set cut by frame number
Timecods: the way Hybrid tries to fix time codes
Preview: force color space conversion on mplayer preview
Input: added image sequence option to set base fps
Audio: restrict eac3 bit rates
Avisynth: extended color support for some filters
Tools: handling version info with newer tsMuxeR
XSynth: using trim with cut support

*added*

Vapoursynth: Anime4KCPP resizer, JincResize, DeDot, CAS and adjusted LSFMod to use CAS
Vapoursynth: added option to lower resolution before resize
Avisynth: MToon, DFMDeRainbow
Audio: handling of 8bit pcm
Video: dynamic HDR with NVEncC
Audio: ffmpeg ac3 & eac3: dialog normalization

*removed*

Video: FFmpeg NVEnc (Windows only)


-> downloads: http://www.selur.de/downloads

Cu Selur

[ReinerSchweinlin]

Thanx a lot !!!

[Selur]

Hybrid rev 2020.07.11.1:
*fixed*

Vapoursynth: Anime4KCPP when using GPU
Vapoursynth: limit Santiag Pcsnr when using GPU
Avisynth: limit SplineResize100/144 to YV12, YUY2, RGB24, RGB32
Avisynth: FrameRateConverter custom fps grayed out
Avisynth: interlaced handling
General: aborting close didn't work
Communication: send ssl mail
General: some more cut handling
Jobs: MPEG-4 ASP workaround
Jobs: tsMuxeR output file checks
Tools: BDSup2Sub++ detection
Video: x264 apply to preset
Audio: ac3 reencoding
Video: 2pass vpx/aomenc encoding
Analysis: mpeg audio layer 1/2/3 detection
General: gui error/freeze when extration of subtitle wasn't possible
Avisynth: 'show YUV values' and 'histogram' for YUY2 input

*changed*

Jobs: using long IDs
General: sometimes crashing on start up
Avisynth: smdegrain limit/limitc double values
General: always use DebugOutput Level 9
Container: added matroska extensions: m3kd
Vapoursynth: replaceed fft3dfilter with neo_fft3d (adjusted QTGMC and MCTemporalDenoise in havsfunc accordingly)

*added*

audio: add option to clear audio queue on processing change
Vapoursynth: support for SRMD


-> downloads: http://www.selur.de/downloads

Cu Selur

[Gravitator]

Hybrid v2020.07.11.1

VP9
The Adaptive quantization window is blocked for 2-pass mode.

AOM
Max frames to lag: 25 expanded to 70.
Pyramid height (max): 4 to 5.
There seems to be no window for Disable trellis quant: 0, 1, 2.
CPU utilization modifier: 8 to 9.

[Gravitator]

x265
Refine CTU: on + Coding unit tree adaptive quantization: off - endless coding is obtained for the second pass.

[Selur]

Tested it here.
For the first pass I used:
"I:\Hybrid\64bit\ffmpeg.exe" -y -loglevel fatal -noautorotate -nostdin -threads 8 -i "F:\TestClips&Co\files\test.avi" -map 0:0 -an -sn -vf zscale=rangein=tv:range=tv -pix_fmt yuv420p10le -strict -1 -vsync 0 -f yuv4mpegpipe - | "I:\Hybrid\64bit\x265.exe" --input - --output-depth 10 --y4m --profile main10 --limit-modes --no-early-skip --no-open-gop --opt-ref-list-length-pps --pass 1 --slow-firstpass --bitrate 1500 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --limit-refs 0 --ssim-rd --psy-rd 2.50 --rdoq-level 2 --psy-rdoq 10.00 --aq-mode 0 --deblock=-1:-1 --limit-sao --no-repeat-headers --range limited --colormatrix bt470bg --stats "E:\Temp\test_generated.stats" --analysis-save "E:\Temp\test_generated.analysis" --analysis-reuse-level 6 --no-multi-pass-opt-analysis --output "E:\Temp\test.265"
which finished fine with:
y4m [info]: 640x352 fps 25/1 i420p10 sar 1:1 unknown frame count
raw [info]: output file: E:\Temp\test.265
x265 [info]: HEVC encoder version 3.4+7-g38774073d
x265 [info]: build info [Windows][GCC 10.1.0][64 bit] 10bit
x265 [info]: using cpu capabilities: MMX2 SSE2Fast LZCNT SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
x265 [warning]: analysis-load-reuse-level can be set only when analysis-load is enabled. Resetting analysis-load-reuse-level to 0.
x265 [info]: Main 10 profile, Level-2.1 (Main tier)
x265 [info]: Thread pool created using 16 threads
x265 [info]: Slices : 1
x265 [info]: frame threads / pool features : 4 / wpp(6 rows)
x265 [warning]: Source height < 720p; disabling lookahead-slices
x265 [info]: Coding QT: max CU size, min CU size : 64 / 8
x265 [info]: Residual QT: max TU size, max depth : 32 / 1 inter / 1 intra
x265 [info]: ME / range / subpel / merge : hex / 57 / 2 / 3
x265 [info]: Keyframe min / max / scenecut / bias : 25 / 250 / 40 / 5.00
x265 [info]: Cb/Cr QP Offset : -2 / -2
x265 [info]: Lookahead / bframes / badapt : 20 / 4 / 2
x265 [info]: b-pyramid / weightp / weightb : 1 / 1 / 0
x265 [info]: References / ref-limit cu / depth : 3 / off / off
x265 [info]: AQ: mode / str / qg-size / cu-tree : 1 / 0.0 / 32 / 1
x265 [info]: Rate Control / qCompress : ABR-1500 kbps / 0.60
x265 [info]: tools: limit-modes rd=3 ssim-rd psy-rd=2.50 rdoq=2 psy-rdoq=10.00
x265 [info]: tools: rskip mode=1 signhide tmvp b-intra strong-intra-smoothing
x265 [info]: tools: deblock(tC=-1:B=-1) sao stats-write
x265 [info]: frame I: 3, Avg QP:16.35 kb/s: 3518.73
x265 [info]: frame P: 120, Avg QP:13.21 kb/s: 3109.91
x265 [info]: frame B: 306, Avg QP:17.58 kb/s: 770.49
x265 [info]: Weighted P-Frames: Y:0.8% UV:0.8%
x265 [info]: consecutive B-frames: 3.3% 0.0% 58.5% 21.1% 17.1%
encoded 429 frames in 15.97s (26.86 fps), 1444.09 kb/s, Avg QP:16.35

then I started a second pass with:
"I:\Hybrid\64bit\ffmpeg.exe" -y -loglevel fatal -noautorotate -nostdin -threads 8 -i "F:\TestClips&Co\files\test.avi" -map 0:0 -an -sn -vf zscale=rangein=tv:range=tv -pix_fmt yuv420p10le -strict -1 -vsync 0 -f yuv4mpegpipe - | "I:\Hybrid\64bit\x265.exe" --input - --output-depth 10 --y4m --profile main10 --limit-modes --no-early-skip --no-open-gop --opt-ref-list-length-pps --pass 2 --bitrate 1500 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --limit-refs 0 --ssim-rd --psy-rd 2.50 --rdoq-level 2 --psy-rdoq 10.00 --aq-mode 0 --deblock=-1:-1 --limit-sao --no-repeat-headers --range limited --colormatrix bt470bg --stats "E:\Temp\test_generated.stats" --analysis-load "E:\Temp\test_generated.analysis" --analysis-reuse-level 6 --no-multi-pass-opt-analysis --no-dynamic-refine --refine-ctu-distortion 1 --output "E:\Temp\2020-07-12@16_54_18_4110_03.265"
which starts fine:
y4m [info]: 640x352 fps 25/1 i420p10 sar 1:1 unknown frame count
raw [info]: output file: E:\Temp\2020-07-12@16_54_18_4110_03.265
x265 [info]: HEVC encoder version 3.4+7-g38774073d
x265 [info]: build info [Windows][GCC 10.1.0][64 bit] 10bit
x265 [info]: using cpu capabilities: MMX2 SSE2Fast LZCNT SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
x265 [warning]: analysis-save-reuse-level can be set only when analysis-save is enabled. Resetting analysis-save-reuse-level to 0.
x265 [info]: Main 10 profile, Level-2.1 (Main tier)
x265 [info]: Thread pool created using 16 threads
x265 [info]: Slices : 1
x265 [info]: frame threads / pool features : 4 / wpp(6 rows)
x265 [warning]: Source height < 720p; disabling lookahead-slices
x265 [info]: Coding QT: max CU size, min CU size : 64 / 8
x265 [info]: Residual QT: max TU size, max depth : 32 / 1 inter / 1 intra
x265 [info]: ME / range / subpel / merge : hex / 57 / 2 / 3
x265 [info]: Keyframe min / max / scenecut / bias : 25 / 250 / 40 / 5.00
x265 [info]: Cb/Cr QP Offset : -2 / -2
x265 [info]: Lookahead / bframes / badapt : 20 / 4 / 2
x265 [info]: b-pyramid / weightp / weightb : 1 / 1 / 0
x265 [info]: References / ref-limit cu / depth : 3 / off / off
x265 [info]: AQ: mode / str / qg-size / cu-tree : 1 / 0.0 / 32 / 1
x265 [info]: Rate Control / qCompress : ABR-1500 kbps / 0.60
x265 [info]: tools: limit-modes rd=3 ssim-rd psy-rd=2.50 rdoq=2 psy-rdoq=10.00
x265 [info]: tools: rskip mode=1 signhide tmvp b-intra strong-intra-smoothing
x265 [info]: tools: deblock(tC=-1:B=-1) sao stats-read

but then nothing happens.
x265 is running, but seemingly not doing anything.

Also tried with HEVC "encoder version 3.4+12-geff904199" same effect.

My guess is that:
analysis-save-reuse-level can be set only when analysis-save is enabled. Resetting analysis-save-reuse-level to 0.
is the problem, but atm. no clue how the x265 folks think that this feature should be used.

Cu Selur

[Gravitator]

x265
Crash on any settings with Recursive skip: 3.

[Selur]

Recursive skip: 3.
'--rskip' should not have an additional parameter,..
No clue what you are writing about and like always you do not really share details,... :/
-> forget that I was confused since I seemingly left an abandoned checkbox in the gui ;)

[Selur]

Got it:
Invalid Recursion skip mode. Valid modes 0,1,2
seems like rskip 3 was removed,.. see: https://bitbucket.org/multicoreware/x265/commits/df0886d58b867e7bd219ec3f8197bab22417795c

Cu Selur

[Gravitator]

AOM
Encoding crash when enabling Signal VUI/Chroma loc.: colocated (Any mode and settings).

[Selur]

regarding the x265 options, see: https://forum.doom9.org/showthread.php?p=1918240#post1918240
-> I'll remove the whole 'Save analysis'-option and the option related to it, since they don't make sense for Hybrid.

regarding aomenc: I'll look into it.

[Selur]

Hybrid rev 2020.07.11.1:
*fixed*

Avisynth: limit LSFMod to YV12
Startup/Shutdown crashes

*added*

Vapoursynth: LGhost


-> downloads: http://www.selur.de/downloads

Cu Selur

[Richard1485]

The default parameters for NVenc in Hybrid include:


-spatial-aq 0 -temporal_aq 0


But if I copy the command into the terminal and run it, ffmpeg outputs a warning that the temporal_aq option is unused -- unless I change it to:


-spatial-aq 0 -temporal-aq 0


Is this just a typo in the GUI or would it affect the output?

I was using the parameters set by Hybrid as the starting point for some tests with ffmpeg on Ubuntu 20.04 when I noticed the message.

[Sharc]

NVEnc(ffmpeg) has been abandoned in Hybrid. Only NVEncC is supported in newer releases.
Edit: .... abandoned for Windows only; sorry.

[Selur]

@richard1485: typo -> fixed for the next release, should not change the output. Since it's the default.

Cu Selur

[Sharc]

According to recent update of ffmpeg 4.3.1, ffmpeg -h encoder=h264_nvenc indicates that both syntax are accepted, i.e.
'-spatial-aq' or '-spatial_aq'
'-temporal-aq' or '-temporal_aq'
I didn't test it though.

[Selur]

Hybrid rev 2020.07.11.1:
*fixed*

Vapoursynth: saving profiles
Vapoursynth QTGMC bob
General: restting Defaults caused two instances to open
Jobs: tsMuxeR temp and input file handling

*changed*

Filtering: Removed the "force deinterlace" option, instead 'Overwrite input scan type to' should be used.
Tools: adjusted to latest NVEncC 'check-features' changes
Audio: remove unneeded '-ac' in ffmpeg call

*added*

Avisynth: 64bit Support, support 32bit and 64bit Avisynth jobs at the same time in the jobqueue (Windows only)


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

Since the https://github.com/Selur/hybrid-vapoursynth-addon is clearly broken (building Vapoursynth seems to work, building plugins seems to work, but vscript and vspipe fails) and I lack the time/knowledge/motivation to spend tons of time to look into fixing this I'm thinking about dropping Linux support as a whole and turn Hybrid into a Windows only tool.

Spending most of my free time trying to get shell scripts working is way to frustrating to keep this going.

Cu Selur

[god_md5]

why i have this bug?

[Selur]

image is still waiting for approval nothing I can do,...

[ReinerSchweinlin]

....dropping Linux support as a whole and turn Hybrid into a Windows only tool.

Spending most of my free time trying to get shell scripts working is way to frustrating to keep this going.

Totally understandable. Let me take the opportunity to thank you for the wonderfull programm you are building here all this time. This has some eral nice gems - I often use the AI-Upscaling Capabilities - which makes hybrid really unique in the way it is incorporating them into a simple GUI. Thanx !

[Selur]

Hybrid rev 2020.10.10.1:
*fixed*

Avisynth: handling QTGMC field order
Video: nvencc version check
Vapoursynth: Cnr2, typo sceneChroma -> scenechroma
Video: NVEncC bluray input with 'encoder only' not starting encoding
Avisynth+: no preRo for Text subittle
Vapoursynth: sRestore output frame count calculation fixex: Preview not working with crop
Vapoursynth: handle TimeCude filename changes
Synth: some colorMatrix fixes
Avisynth: DFMDeRainbow load mvtools for motion compensation
Vapoursynth: Fix Combed to IVTC
Analysis: aspect ratio detection
Avisynth: 64bit preview
Vapoursynth: restrict VFM Mi to blockx*blocky
Video: problem with aoemenc and svthevc
Vapoursynth: fixed Santiang typevh typo to typev, "AntiAliasing" - Nedi3aa vs "AnitAliasing - Nedi3aa"
Vapoursynth: deleting profiles
Vapoursynth: MAA missing dependecies
Analysis: prefer stream aspect ratio
Video: scanorder handling especially for DNxHD
Tools: tsMuxeR version check

*changed*

Cosmetics: Bunch of cosmetic changes
Video: Using xvid_encraw instead of mencoder for XviD encoding.
Vapoursynth: using D2VSource instead of MPEG2DecPlus
Tools: use 64bit of vStripFree, Telxcc and MP4FPSmod (Windows only)
Synth: added QTGMC custom addition
Synth: splitting QTGMC into "QTGMC - Deinterlacer" (InputType = 0) and "QTGMC - Filter" (InputType > 0)
Vapoursynth: Anime4KCPP support for HDNLevel Option
Vapoursynth/Avisynth: in custom section you can now overwrite width, height, scantype and colorformat using: Note that these values are only respected inside the Vapoursynth/ Avisynth script.
Frame rate and frame count need to be globally overwrite using "Filtering->Misc->Overwrite output". All other values also need to be overwritten globally if you set them.
Vapoursynth: Levels handling
Avisynth: added QTGMC MatchPreset, MatchPreset2, MatchEnhance
Crop: lowered min Crop-Width-Mult from 0.75 to 0.7
Avisynth: TDecimate added additional parameters
Vapoursynth: VIVTC added Deinterlacer option
Avisynth: Flash3KDeband adjust output bit depth
Avisynth: TemporalDegrain requires 8bit input
Vapoursynth: KNLMeans adjust RGBX
Vapoursynth: TimeCube RGB handling
Vapoursynth: restrict filters based on Hqdn3d to 8bit
Tools: look also for '3D controller" in lspci output for video card
Analysis: added option to prefer stream luma range
Video: changed ProRes models to proxy, lt, standard, hq, 4444, 4444x
Avisynth: added RemoveDirt
Output: added 'Config->Output->MKV Contains->Matroska advanced video signaling->Import HDR data from source'

*added*

Video: Cineform HD support
Filtering: option to flip video horizontally

*removed*

Tools: nvhsp not used on Windows, hardly working on Linux&Mac, sadly ffmpeg does still not support signaling max cll/fall etc
Tools: not using ffmbc any more
Avisynth support on MacOS



-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: Added Mac OS back to the release OSs.

[Sharc]

:thanks:

Can't try it.
It starts to initialize then closes ....

[Selur]

Hmm,.. works fine here. :/
Can you try if it works if you move the content of your settings folder?

[Sharc]

Yes, this helped. I just lost my configurations, however.

Edit:
Copying the former settings back piece by piece and see where it crashes .........

[Selur]

If you figure out what it is let me know and I can probably fix it. :)

Cu Selur

[Sharc]

I could so far recover most of it, but disableIcons=true in the misc.ini has no effect .... (I prefer the test buttons).

[Selur]

Yeah, that option is removed due to complains that it breaks the layout, not sure whether it will return.

Cu Selur

[Selur]

Hybrid rev 2020.10.11.1:
*fixed*

Vapoursynth: TDeintMod in VIVTC and FixCombed
Muxing: always used ffmpeg for mkv muxing

*changed*

Vapoursynth: add 'Use RGB' for Limiter and Levels and use it as default.
Vapoursynth: extend compatible TimeCube color spaces by RGBS

*added*

Filtering: "fieldmatch + yadif + deint + decimate2" as IVTC through FFmpeg method.

*removed*

Vapoursynth: support for mClean due to missing TemporalMedian library. (MacOS only)


-> downloads: http://www.selur.de/downloads

Cu Selur

[Gravitator]

Hybrid v2020.10.11.1

x265
Enabled hevc-aq and disabled psy-rd results no-hevc-aq (In the MInfo output file).

[Selur]

might be a bug in 'minimize command line', not totally sure, will look at it when the debug output is approved,...

[Selur]

defaults:
x265 --input - --output-depth 10 --y4m --profile main10 --limit-modes --no-early-skip --no-open-gop --opt-ref-list-length-pps --crf 18.00 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --qpfile GENERATED_QP_FILE --limit-refs 0 --ssim-rd --psy-rd 2.50 --rdoq-level 2 --psy-rdoq 10.00 --aq-mode 0 --deblock=-1:-1 --limit-sao --no-repeat-headers --range limited --colormatrix unknown --output OUTPUTFILE

activated hevc-ac:
x265 --input - --output-depth 10 --y4m --profile main10 --limit-modes --no-early-skip --no-open-gop --opt-ref-list-length-pps --crf 18.00 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --qpfile GENERATED_QP_FILE --limit-refs 0 --ssim-rd --psy-rd 2.50 --rdoq-level 2 --psy-rdoq 10.00 --aq-mode 0 --hevc-aq --no-cutree --deblock=-1:-1 --limit-sao --no-repeat-headers --range limited --colormatrix unknown --output OUTPUTFILE

set psy-rd to 0:
x265 --tune psnr --input - --output-depth 10 --y4m --profile main10 --limit-modes --no-early-skip --no-open-gop --opt-ref-list-length-pps --crf 18.00 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --qpfile GENERATED_QP_FILE --limit-refs 0 --ssim-rd --rdoq-level 2 --psy-rdoq 10.00 --hevc-aq --no-cutree --deblock=-1:-1 --limit-sao --no-repeat-headers --range limited --colormatrix unknown --output OUTPUTFILE

-> used encoding call:
ffmpeg -y -loglevel fatal -noautorotate -nostdin -threads 8 -i "F:\TestClips&Co\files\test.avi" -map 0:0 -an -sn -vf zscale=rangein=tv:range=tv -pix_fmt yuv420p10le -strict -1 -vsync 0 -f yuv4mpegpipe - | x265 --tune psnr --input - --output-depth 10 --y4m --profile main10 --limit-modes --no-early-skip --no-open-gop --opt-ref-list-length-pps --crf 18.00 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --limit-refs 0 --ssim-rd --rdoq-level 2 --psy-rdoq 10.00 --hevc-aq --no-cutree --deblock=-1:-1 --limit-sao --no-repeat-headers --range limited --colormatrix bt470bg --output "E:\Temp\2020-10-15@17_37_08_9710_01.265"
results in:

x265 [info]: HEVC encoder version 3.4+26-ga82c6c7a7
x265 [info]: build info [Windows][GCC 10.2.0][64 bit] 10bit
x265 [info]: using cpu capabilities: MMX2 SSE2Fast LZCNT SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
x265 [info]: Main 10 profile, Level-2.1 (Main tier)
x265 [info]: Thread pool created using 16 threads
x265 [info]: Slices : 1
x265 [info]: frame threads / pool features : 4 / wpp(6 rows)
x265 [warning]: Source height < 720p; disabling lookahead-slices
x265 [info]: Coding QT: max CU size, min CU size : 64 / 8
x265 [info]: Residual QT: max TU size, max depth : 32 / 1 inter / 1 intra
x265 [info]: ME / range / subpel / merge : hex / 57 / 2 / 3
x265 [info]: Keyframe min / max / scenecut / bias : 25 / 250 / 40 / 5.00
x265 [info]: Cb/Cr QP Offset : -2 / -2
x265 [info]: Lookahead / bframes / badapt : 20 / 4 / 2
x265 [info]: b-pyramid / weightp / weightb : 1 / 1 / 0
x265 [info]: References / ref-limit cu / depth : 3 / off / off
x265 [info]: Rate Control / qCompress : CRF-18.0 / 0.60
x265 [info]: tools: limit-modes rd=3 ssim-rd rdoq=2 psy-rdoq=10.00 rskip mode=1
x265 [info]: tools: signhide tmvp b-intra strong-intra-smoothing
x265 [info]: tools: deblock(tC=-1:B=-1) sao

there is no info that hevc-aq would be ignored,...

MediaInfo reports:
Writing library : x265 3.4+26-ga82c6c7a7:[Windows][GCC 10.2.0][64 bit] 10bit
Encoding settings : cpuid=1111039 / frame-threads=4 / numa-pools=16 / wpp / no-pmode / no-pme / no-psnr / no-ssim / log-level=2 / input-csp=1 / input-res=640x352 / interlace=0 / total-frames=0 / level-idc=0 / high-tier=1 / uhd-bd=0 / ref=3 / no-allow-non-conformance / no-repeat-headers / annexb / no-aud / no-hrd / info / hash=0 / no-temporal-layers / no-open-gop / min-keyint=25 / keyint=250 / gop-lookahead=0 / bframes=4 / b-adapt=2 / b-pyramid / bframe-bias=0 / rc-lookahead=20 / lookahead-slices=0 / scenecut=40 / hist-scenecut=0 / radl=0 / no-splice / no-intra-refresh / ctu=64 / min-cu-size=8 / no-rect / no-amp / max-tu-size=32 / tu-inter-depth=1 / tu-intra-depth=1 / limit-tu=0 / rdoq-level=2 / dynamic-rd=0.00 / ssim-rd / signhide / no-tskip / nr-intra=0 / nr-inter=0 / no-constrained-intra / strong-intra-smoothing / max-merge=3 / limit-refs=0 / limit-modes / me=1 / subme=2 / merange=57 / temporal-mvp / no-frame-dup / no-hme / weightp / no-weightb / no-analyze-src-pics / deblock=-1:-1 / sao / no-sao-non-deblock / rd=3 / selective-sao=4 / no-early-skip / rskip / no-fast-intra / no-tskip-fast / no-cu-lossless / b-intra / no-splitrd-skip / rdpenalty=0 / psy-rd=0.00 / psy-rdoq=10.00 / no-rd-refine / no-lossless / cbqpoffs=-2 / crqpoffs=-2 / rc=crf / crf=18.0 / qcomp=0.60 / qpstep=4 / stats-write=0 / stats-read=0 / ipratio=1.40 / pbratio=1.30 / aq-mode=0 / aq-strength=0.00 / no-cutree / zone-count=0 / no-strict-cbr / qg-size=64 / no-rc-grain / qpmax=69 / qpmin=0 / no-const-vbv / sar=1 / overscan=0 / videoformat=5 / range=0 / colorprim=2 / transfer=2 / colormatrix=5 / chromaloc=0 / display-window=0 / cll=0,0 / min-luma=0 / max-luma=1023 / log2-max-poc-lsb=8 / vui-timing-info / vui-hrd-info / slices=1 / opt-qp-pps / opt-ref-list-length-pps / no-multi-pass-opt-rps / scenecut-bias=0.05 / hist-threshold=0.03 / no-opt-cu-delta-qp / no-aq-motion / no-hdr10 / no-hdr10-opt / no-dhdr10-opt / no-idr-recovery-sei / analysis-reuse-level=0 / analysis-save-reuse-level=0 / analysis-load-reuse-level=0 / scale-factor=0 / refine-intra=0 / refine-inter=0 / refine-mv=1 / refine-ctu-distortion=0 / limit-sao / ctu-info=0 / no-lowpass-dct / refine-analysis-type=0 / copy-pic=1 / max-ausize-factor=1.0 / no-dynamic-refine / no-single-sei / no-hevc-aq / no-svt / no-field / qp-adaptation-range=1.00 / no-scenecut-aware-qpconformance-window-offsets / right=0 / bottom=0 / decoder-max-rate=0 / no-vbv-live-multi-pass
-> 'no-hevc-aq'

hevc-aq documentaion:
--hevc-aq

Enable adaptive quantization It scales the quantization step size according to the spatial activity of one coding unit relative to frame average spatial activity. This AQ method utilizes the minimum variance of sub-unit in each coding unit to represent the spatial complexity of the coding unit.
source: https://x265.readthedocs.io/en/latest/cli.html?highlight=hevc-aq#cmdoption-hevc-aq

disabling 'minimize command line' results in:
x265 --frame-threads 0 --wpp --no-pmode --no-pme --log-level 2 --input - --output-depth 10 --y4m --profile main10 --high-tier --min-cu-size 8 --ctu 64 --qg-size 32 --no-opt-cu-delta-qp --tu-intra-depth 1 --tu-inter-depth 1 --limit-tu 0 --max-tu-size 32 --no-hme --me hex --subme 2 --merange 57 --no-analyze-src-pics --limit-modes --no-rect --no-amp --max-merge 3 --no-early-skip --rskip 1 --temporal-mvp --rdpenalty 0 --no-tskip --strong-intra-smoothing --no-constrained-intra --no-open-gop --gop-lookahead 0 --scenecut 40 --scenecut-bias 5 --no-scenecut-aware-qp --no-hist-scenecut --opt-ref-list-length-pps --keyint 250 --min-keyint 0 --max-ausize-factor 1 --radl 0 --bframes 4 --no-weightb --no-b-intra --bframe-bias 0 --b-adapt 2 --b-pyramid --ref 3 --weightp --rc-lookahead 20 --slices 1 --lookahead-threads 0 --qpstep 4 --qpmin 0 --qpmax 69 --qblur 0.50 --cplxblur 20.0 --crf 18.00 --opt-qp-pps --no-rc-grain --cbqpoffs -2 --crqpoffs -2 --ipratio 1.4 --pbratio 1.3 --nr-intra 0 --nr-inter 0 --qpfile GENERATED_QP_FILE --limit-refs 0 --rd 3 --no-fast-intra --ssim-rd --no-rd-refine --no-psy-rd --rdoq-level 2 --psy-rdoq 10.00 --signhide --no-splitrd-skip --qcomp 0.60 --qp-adaptation-range 1 --no-aq-motion --aq-mode 0 --hevc-aq --no-cutree --no-const-vbv --vbv-maxrate 0 --vbv-bufsize 0 --vbv-init 0.9 --vbv-end 0.000 --no-hrd --no-aud --info --no-idr-recovery-sei --deblock=-1:-1 --sao --selective-sao 4 --no-sao-non-deblock --limit-sao --no-repeat-headers --no-temporal-layers --log2-max-poc-lsb 8 --no-fades --no-psnr --no-ssim --no-interlace --range limited --colormatrix bt470bg --no-single-sei --no-hdr10 --output "E:\Temp\test.265"
and
ffmpeg -y -loglevel fatal -noautorotate -nostdin -threads 8 -i "F:\TestClips&Co\files\test.avi" -map 0:0 -an -sn -vf zscale=rangein=tv:range=tv -pix_fmt yuv420p10le -strict -1 -vsync 0 -f yuv4mpegpipe - | x265 --frame-threads 0 --wpp --no-pmode --no-pme --log-level 2 --input - --output-depth 10 --y4m --profile main10 --high-tier --min-cu-size 8 --ctu 64 --qg-size 32 --no-opt-cu-delta-qp --tu-intra-depth 1 --tu-inter-depth 1 --limit-tu 0 --max-tu-size 32 --no-hme --me hex --subme 2 --merange 57 --no-analyze-src-pics --limit-modes --no-rect --no-amp --max-merge 3 --no-early-skip --rskip 1 --temporal-mvp --rdpenalty 0 --no-tskip --strong-intra-smoothing --no-constrained-intra --no-open-gop --gop-lookahead 0 --scenecut 40 --scenecut-bias 5 --no-scenecut-aware-qp --no-hist-scenecut --opt-ref-list-length-pps --keyint 250 --min-keyint 0 --max-ausize-factor 1 --radl 0 --bframes 4 --no-weightb --no-b-intra --bframe-bias 0 --b-adapt 2 --b-pyramid --ref 3 --weightp --rc-lookahead 20 --slices 1 --lookahead-threads 0 --qpstep 4 --qpmin 0 --qpmax 69 --qblur 0.50 --cplxblur 20.0 --crf 18.00 --opt-qp-pps --no-rc-grain --cbqpoffs -2 --crqpoffs -2 --ipratio 1.4 --pbratio 1.3 --nr-intra 0 --nr-inter 0 --limit-refs 0 --rd 3 --no-fast-intra --ssim-rd --no-rd-refine --no-psy-rd --rdoq-level 2 --psy-rdoq 10.00 --signhide --no-splitrd-skip --qcomp 0.60 --qp-adaptation-range 1 --no-aq-motion --aq-mode 0 --hevc-aq --no-cutree --no-const-vbv --vbv-maxrate 0 --vbv-bufsize 0 --vbv-init 0.9 --vbv-end 0.000 --no-hrd --no-aud --info --no-idr-recovery-sei --deblock=-1:-1 --sao --selective-sao 4 --no-sao-non-deblock --limit-sao --no-repeat-headers --no-temporal-layers --log2-max-poc-lsb 8 --no-fades --no-psnr --no-ssim --no-interlace --range limited --colormatrix bt470bg --no-single-sei --no-hdr10 --output "E:\Temp\2020-10-15@17_44_14_4510_01.265"
results in:
x265 [info]: HEVC encoder version 3.4+26-ga82c6c7a7
x265 [info]: build info [Windows][GCC 10.2.0][64 bit] 10bit
x265 [info]: using cpu capabilities: MMX2 SSE2Fast LZCNT SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
x265 [info]: Main 10 profile, Level-2.1 (Main tier)
x265 [info]: Thread pool created using 16 threads
x265 [info]: Slices : 1
x265 [info]: frame threads / pool features : 4 / wpp(6 rows)
x265 [warning]: Source height < 720p; disabling lookahead-slices
x265 [info]: Coding QT: max CU size, min CU size : 64 / 8
x265 [info]: Residual QT: max TU size, max depth : 32 / 1 inter / 1 intra
x265 [info]: ME / range / subpel / merge : hex / 57 / 2 / 3
x265 [info]: Keyframe min / max / scenecut / bias : 25 / 250 / 40 / 5.00
x265 [info]: Cb/Cr QP Offset : -2 / -2
x265 [info]: Lookahead / bframes / badapt : 20 / 4 / 2
x265 [info]: b-pyramid / weightp / weightb : 1 / 1 / 0
x265 [info]: References / ref-limit cu / depth : 3 / off / off
x265 [info]: Rate Control / qCompress : CRF-18.0 / 0.60
x265 [info]: tools: limit-modes rd=3 ssim-rd rdoq=2 psy-rdoq=10.00 rskip mode=1
x265 [info]: tools: signhide tmvp strong-intra-smoothing deblock(tC=-1:B=-1)
x265 [info]: tools: sao

MediaInfo report:
cpuid=1111039 / frame-threads=4 / numa-pools=16 / wpp / no-pmode / no-pme / no-psnr / no-ssim / log-level=2 / input-csp=1 / input-res=640x352 / interlace=0 / total-frames=0 / level-idc=0 / high-tier=1 / uhd-bd=0 / ref=3 / no-allow-non-conformance / no-repeat-headers / annexb / no-aud / no-hrd / info / hash=0 / no-temporal-layers / no-open-gop / min-keyint=25 / keyint=250 / gop-lookahead=0 / bframes=4 / b-adapt=2 / b-pyramid / bframe-bias=0 / rc-lookahead=20 / lookahead-slices=0 / scenecut=40 / hist-scenecut=0 / radl=0 / no-splice / no-intra-refresh / ctu=64 / min-cu-size=8 / no-rect / no-amp / max-tu-size=32 / tu-inter-depth=1 / tu-intra-depth=1 / limit-tu=0 / rdoq-level=2 / dynamic-rd=0.00 / ssim-rd / signhide / no-tskip / nr-intra=0 / nr-inter=0 / no-constrained-intra / strong-intra-smoothing / max-merge=3 / limit-refs=0 / limit-modes / me=1 / subme=2 / merange=57 / temporal-mvp / no-frame-dup / no-hme / weightp / no-weightb / no-analyze-src-pics / deblock=-1:-1 / sao / no-sao-non-deblock / rd=3 / selective-sao=4 / no-early-skip / rskip / no-fast-intra / no-tskip-fast / no-cu-lossless / no-b-intra / no-splitrd-skip / rdpenalty=0 / psy-rd=0.00 / psy-rdoq=10.00 / no-rd-refine / no-lossless / cbqpoffs=-2 / crqpoffs=-2 / rc=crf / crf=18.0 / qcomp=0.60 / qpstep=4 / stats-write=0 / stats-read=0 / ipratio=1.40 / pbratio=1.30 / aq-mode=0 / aq-strength=0.00 / no-cutree / zone-count=0 / no-strict-cbr / qg-size=64 / no-rc-grain / qpmax=69 / qpmin=0 / no-const-vbv / sar=1 / overscan=0 / videoformat=5 / range=0 / colorprim=2 / transfer=2 / colormatrix=5 / chromaloc=0 / display-window=0 / cll=0,0 / min-luma=0 / max-luma=1023 / log2-max-poc-lsb=8 / vui-timing-info / vui-hrd-info / slices=1 / opt-qp-pps / opt-ref-list-length-pps / no-multi-pass-opt-rps / scenecut-bias=0.05 / hist-threshold=0.03 / no-opt-cu-delta-qp / no-aq-motion / no-hdr10 / no-hdr10-opt / no-dhdr10-opt / no-idr-recovery-sei / analysis-reuse-level=0 / analysis-save-reuse-level=0 / analysis-load-reuse-level=0 / scale-factor=0 / refine-intra=0 / refine-inter=0 / refine-mv=1 / refine-ctu-distortion=0 / limit-sao / ctu-info=0 / no-lowpass-dct / refine-analysis-type=0 / copy-pic=1 / max-ausize-factor=1.0 / no-dynamic-refine / no-single-sei / hevc-aq / no-svt / no-field / qp-adaptation-range=1.00 / no-scenecut-aware-qpconformance-window-offsets / right=0 / bottom=0 / decoder-max-rate=0 / no-vbv-live-multi-pass
-> 'hevc-aq'

so it's an issue with 'minimize command line'.

so looking at command line before setting psy-rd to zero:
x265 --input - --output-depth 10 --y4m --profile main10 --limit-modes --no-early-skip --no-open-gop --opt-ref-list-length-pps --crf 18.00 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --qpfile GENERATED_QP_FILE --limit-refs 0 --ssim-rd --psy-rd 2.50 --rdoq-level 2 --psy-rdoq 10.00 --aq-mode 0 --hevc-aq --no-cutree --deblock=-1:-1 --limit-sao --no-repeat-headers --range limited --colormatrix unknown --output OUTPUTFILE
and with psy-rd to 0:
x265 --tune psnr --input - --output-depth 10 --y4m --profile main10 --limit-modes --no-early-skip --no-open-gop --opt-ref-list-length-pps --crf 18.00 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --qpfile GENERATED_QP_FILE --limit-refs 0 --ssim-rd --rdoq-level 2 --psy-rdoq 10.00 --hevc-aq --no-cutree --deblock=-1:-1 --limit-sao --no-repeat-headers --range limited --colormatrix unknown --output OUTPUTFILE


before: --psy-rd 2.50 --aq-mode 0 --hevc-aq
after: --tune psnr -hevc-aq

checking tune psnr:

if (!strcmp(tune, "psnr"))
{
param->rc.aqStrength = 0.0;
param->psyRd = 0.0;
param->psyRdoq = 0.0;
}

see: https://bitbucket.org/multicoreware/x265_git/src/master/source/common/param.cpp (line 566-571)
so: "--aq-strength 0 --psy-rd 0 --psy-rdoq 0"
since "--aq-strength 0" should be the same as "--aq-mode 0".

now looking again at before and after:
before: --psy-rd 2.50 --aq-mode 0 --hevc-aq
after: --tune psnr --hevc-aq = --aq-mode 0 --psy-rd 0 --psy-rdoq 0 --hevc-aq
since after also has '--psy-rdoq 10.00'
after = --aq-mode 0 --psy-rd 0 --psy-rdoq 10.00 --hevc-aq

-> don't see where Hybrid is making a mistake. :( those command lines should be identical unless '--tune psnr' has other side effect that I don't know about.

Cu Selur

[Selur]

removing command line minimization for 'tune', Hybrid would use:
x265 --input - --output-depth 10 --y4m --profile main10 --limit-modes --no-early-skip --no-open-gop --opt-ref-list-length-pps --crf 18.00 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --qpfile GENERATED_QP_FILE --limit-refs 0 --ssim-rd --no-psy-rd --rdoq-level 2 --psy-rdoq 10.00 --aq-mode 0 --hevc-aq --no-cutree --deblock=-1:-1 --limit-sao --no-repeat-headers --range limited --colormatrix bt470bg --output OUTPUTFILE
and
ffmpeg -y -loglevel fatal -noautorotate -nostdin -threads 8 -i "F:\TestClips&Co\files\test.avi" -map 0:0 -an -sn -vf zscale=rangein=tv:range=tv -pix_fmt yuv420p10le -strict -1 -vsync 0 -f yuv4mpegpipe - | x265 --input - --output-depth 10 --y4m --profile main10 --limit-modes --no-early-skip --no-open-gop --opt-ref-list-length-pps --crf 18.00 --opt-qp-pps --cbqpoffs -2 --crqpoffs -2 --limit-refs 0 --ssim-rd --no-psy-rd --rdoq-level 2 --psy-rdoq 10.00 --aq-mode 0 --hevc-aq --no-cutree --deblock=-1:-1 --limit-sao --no-repeat-headers --range limited --colormatrix bt470bg --output "E:\Temp\2020-10-15@18_07_32_7810_01.265"
resulting in:
x265 [info]: HEVC encoder version 3.4+26-ga82c6c7a7
x265 [info]: build info [Windows][GCC 10.2.0][64 bit] 10bit
x265 [info]: using cpu capabilities: MMX2 SSE2Fast LZCNT SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
x265 [info]: Main 10 profile, Level-2.1 (Main tier)
x265 [info]: Thread pool created using 16 threads
x265 [info]: Slices : 1
x265 [info]: frame threads / pool features : 4 / wpp(6 rows)
x265 [warning]: Source height < 720p; disabling lookahead-slices
x265 [info]: Coding QT: max CU size, min CU size : 64 / 8
x265 [info]: Residual QT: max TU size, max depth : 32 / 1 inter / 1 intra
x265 [info]: ME / range / subpel / merge : hex / 57 / 2 / 3
x265 [info]: Keyframe min / max / scenecut / bias : 25 / 250 / 40 / 5.00
x265 [info]: Cb/Cr QP Offset : -2 / -2
x265 [info]: Lookahead / bframes / badapt : 20 / 4 / 2
x265 [info]: b-pyramid / weightp / weightb : 1 / 1 / 0
x265 [info]: References / ref-limit cu / depth : 3 / off / off
x265 [info]: Rate Control / qCompress : CRF-18.0 / 0.60
x265 [info]: tools: limit-modes rd=3 ssim-rd rdoq=2 psy-rdoq=10.00 rskip mode=1
x265 [info]: tools: signhide tmvp b-intra strong-intra-smoothing
x265 [info]: tools: deblock(tC=-1:B=-1) sao
and MediaInfo:
x265 3.4+26-ga82c6c7a7:[Windows][GCC 10.2.0][64 bit] 10bit
Encoding settings : cpuid=1111039 / frame-threads=4 / numa-pools=16 / wpp / no-pmode / no-pme / no-psnr / no-ssim / log-level=2 / input-csp=1 / input-res=640x352 / interlace=0 / total-frames=0 / level-idc=0 / high-tier=1 / uhd-bd=0 / ref=3 / no-allow-non-conformance / no-repeat-headers / annexb / no-aud / no-hrd / info / hash=0 / no-temporal-layers / no-open-gop / min-keyint=25 / keyint=250 / gop-lookahead=0 / bframes=4 / b-adapt=2 / b-pyramid / bframe-bias=0 / rc-lookahead=20 / lookahead-slices=0 / scenecut=40 / hist-scenecut=0 / radl=0 / no-splice / no-intra-refresh / ctu=64 / min-cu-size=8 / no-rect / no-amp / max-tu-size=32 / tu-inter-depth=1 / tu-intra-depth=1 / limit-tu=0 / rdoq-level=2 / dynamic-rd=0.00 / ssim-rd / signhide / no-tskip / nr-intra=0 / nr-inter=0 / no-constrained-intra / strong-intra-smoothing / max-merge=3 / limit-refs=0 / limit-modes / me=1 / subme=2 / merange=57 / temporal-mvp / no-frame-dup / no-hme / weightp / no-weightb / no-analyze-src-pics / deblock=-1:-1 / sao / no-sao-non-deblock / rd=3 / selective-sao=4 / no-early-skip / rskip / no-fast-intra / no-tskip-fast / no-cu-lossless / b-intra / no-splitrd-skip / rdpenalty=0 / psy-rd=0.00 / psy-rdoq=10.00 / no-rd-refine / no-lossless / cbqpoffs=-2 / crqpoffs=-2 / rc=crf / crf=18.0 / qcomp=0.60 / qpstep=4 / stats-write=0 / stats-read=0 / ipratio=1.40 / pbratio=1.30 / aq-mode=0 / aq-strength=0.00 / no-cutree / zone-count=0 / no-strict-cbr / qg-size=64 / no-rc-grain / qpmax=69 / qpmin=0 / no-const-vbv / sar=1 / overscan=0 / videoformat=5 / range=0 / colorprim=2 / transfer=2 / colormatrix=5 / chromaloc=0 / display-window=0 / cll=0,0 / min-luma=0 / max-luma=1023 / log2-max-poc-lsb=8 / vui-timing-info / vui-hrd-info / slices=1 / opt-qp-pps / opt-ref-list-length-pps / no-multi-pass-opt-rps / scenecut-bias=0.05 / hist-threshold=0.03 / no-opt-cu-delta-qp / no-aq-motion / no-hdr10 / no-hdr10-opt / no-dhdr10-opt / no-idr-recovery-sei / analysis-reuse-level=0 / analysis-save-reuse-level=0 / analysis-load-reuse-level=0 / scale-factor=0 / refine-intra=0 / refine-inter=0 / refine-mv=1 / refine-ctu-distortion=0 / limit-sao / ctu-info=0 / no-lowpass-dct / refine-analysis-type=0 / copy-pic=1 / max-ausize-factor=1.0 / no-dynamic-refine / no-single-sei / hevc-aq / no-svt / no-field / qp-adaptation-range=1.00 / no-scenecut-aware-qpconformance-window-offsets / right=0 / bottom=0 / decoder-max-rate=0 / no-vbv-live-multi-pass

-> so seems like '--tune psnr' does have additional side effects that were not documented (or where I do not know where it is documented).

Cu Selur

Ps.: created and entry in the x265 issue tracker: https://bitbucket.org/multicoreware/x265_git/issues/575/can-someone-clear-up-why-tune-psnr and a post in the x265 thread here https://forum.doom9.org/showthread.php?p=1925879.

[Gravitator]

AOM
Profile: high, 8bit&10bit, color sampling: 420 - crash encoding;
Profile: pro, 8bit&10bit, color sampling: 420 - crash encoding.

[Selur]

Yeah, seems like the 'high'-profile is now restricted to 4:4:4 as the error it gives is: "Profile 1 requires 4:4:4 color format" and for the 'pro'-profile I get 'Profile 2 bit-depth <= 10 requires 4:2:2 color format"

I assumed:
Main Profile (0) is 8/10-bit 4:2:0 or Monochrome (4:0:0)
High Profile (1) is 8/10-bit 4:2:0, 4:4:4, Monochrome
Professional Profile (2) is 8/10/12-bit, 4:2:0, 4:2:2, 4:4:4, Monochrome
but seems like they limited it to:
Main Profile (0) is 8/10-bit 4:2:0
High Profile (1) is 8/10-bit 4:4:4
Professional Profile (2) is 8/10/12-bit, 4:2:2, 4:4:4

[sneaker_ger]

AFAIK:
Main Profile (0) is 8/10-bit 4:2:0
High Profile (1) is 8/10-bit 4:4:4
Professional Profile (2) is 12-bit with any sub-sampling; plus 8, 10 bit with 4:2:2
So there is (and always has been) 0 choice about the profile when encoding with a given bitdepth and sub-sampling.

[Selur]

ah okay,...

[Gravitator]

AOM
Golden frame dist max: 1 - crash with any settings.

[Selur]

I can reproduce the crash, it even gives and error:
Failed to control codec: Invalid parameter
max_gf_interval out of range [AOMMAX(2, extra_cfg->min_gf_interval)..(MAX_LAG_BUFFERS - 1)]
-> so it needs to be at least 2 and below 'max frames to lag' -1

[Selur]

Hybrid rev 2020.12.12.1:
*fixed*

Synth: Avisynth, color matrix seleciton (missing '.')
x264: min/max quant maximum (missing + 18)
Synth: Avisynth, 64bit/32bit switching filter path initialization problem
Synth: Avisynth, small fixes related: RemoveDirt, TFM
Synth: Avisynth, QTGMC filter call
Synth: Vapoursynth, fixed SSIQ
Synth: Vapoursynth, ImageReader always outputs RGB24
Synth: Vapoursynth, HQDering default, LSFMod gui control
Synth: Vapoursynth, ff3dFilter extended supported color spaces
Synth: Vapoursynth, LUTDeCrawl and LUTDeRainbow crashes when 'mask' was used
Synth: Avisynth, import Zs_RF_Shared for edgecleaner
Synth: Avisynth, a few bugs related to interlaced handling in Avisynth
Synth: Avisynth, ColorYUV 'Off' control U/V mixup
Synth: Avisynth, FastLineDarkenMOD call
refresh resize&co on manual input PAR change
Synth: Vapoursynth, typos 'sceneChroma' -> 'scenechroma','internations'->'interrations',...
Synth: respect filter order when calculating output frame count and frame rate.
Synth: handle RGB input
Synth: sRestore support mode -1,-3,-4
Synth: Avisynth, ColorMatrix change preview
Synth: av1 aomenc color space restriction
Synth: error about "overWriteOutputFrameRateValue" when Vapoursynth wasn't present
Muxing: always used ffmpeg for mkv muxing

*changed*

x265: using --scenecut-aware-qp <0..3> and --masking-strength <string> instead of --[no-]scenecut-aware-qp, --scenecut-window <0..1000>, --qp- delta-ref <0..10> and --qp-delta-nonref <0..10
x265: support for --vbv-live-multi-pass, --max-vbv-fullness, --min-vbv- fullness
x265: support for --min-vbv-fullness & --max-vbv-fullness
Synth: Avisynth, removed 'enableAvisynthForDVD' it's always assumed to be enabled now
Synth: Vapoursynth, added overwrite Smooth/Super/Vectors to Interframe
UI: removed simple view for Xvid and x264
Synth: Reset Avisynth and Vapoursynth filter order when filter count changed.
Synth: Avisynth´, force use of D2VSource instead of DGDecode when 64bit Avisynth is used (there does not seem to be any working 64bit DGDecode.dll)
UI: Rearranged the Vapoursynth filters into groups
Tools: FFMpeg, using '-video_size' instead of '-s'
Tools: FFMpeg use ignore_editlist for mov/mp4/3gb input
Synth: Avisynth, limit QTGMC imports
Synth: enforce Crop < Rotate < Resize < Letterbox order in custom filter orders
Synth: Vapoursynth, now also allows interlaced output of scripts
Video: FFvHuFF set output color space
Muxing: use '-of avi' instead of '-of lavf' when muxing to .avi using mencoder
Synth: Vapoursynth, support RGBS in Limiter and Levels
x265: disabled 'replace tune' in 'minimize command line', since atleast '--tune psnr' has some unknown side effects, see: https://bitbucket.org/multicoreware/x265_git/issues/575/can-someone-clear-up-why-tune-psnr

*added*

Output: PNG output, use together with 'raw' as 'container' *experimental*
Synth: Avisynth: old SpotRemover script by Didée (http://forum.doom9.net/showthread.php?p=1402690#post1402690)
Synth: VsAutoWhite script


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: Note that Vapoursynth now allows to change the filter order of each filter individually, but this requires to reset the default filter order.

[Selur]

Hybrid rev 2020.12.13.1:
*fixed*

Decode: deinterlacing through FFmpeg wasn't used when Avisynth/ Vapoursynth was used.
Avisynth: custom before 'End'-section was missing
(e)ac3 dialog normalization ui glitch


-> downloads: http://www.selur.de/downloads

Cu Selur

[stax76]

@Selur

Did you try Anime4KCPP under AviSynth? It gives an access violation here.

[Selur]

@stax76: tried it, didn't work for me either. Did not spend time on it since it works on Vapoursynth.

Cu Selur

[stax76]

OK, thanks.

[Selur]

Hybrid rev 2021.01.31.1:
*fixed*

Vapoursynth: libneo-fft3d.so -> libneofft3d.so
Vapoursynth: DaaMod fix 'no field vsDaaExpMod'
Sytnh: DVD handling Avisynth/Vapoursynth
Synth: Remove Trim from FilterOrder, always trim directly after source filter.
Avisynth: 64bit use MPEG2Dec_64
Decode: deinterlacing through FFmpeg wasn't used when Avisynth/ Vapoursynth was used.
Avisynth: custom before 'End'-section was missing
UI: (e)ac3 dialog normalization ui glitch
Vapoursynth: IVTCCustomDeinterlacerGPU ui glitch
Subtitle: subtitle extraction form MP4 files
Preview: title left/right mixup
x265: strictCBR handling
Vapousynth: MClean
Vapoursynth: Denois filters on their own were ignored
Audio: qaac language passthrough
Audio: ffmpeg aac reencode + ffmpeg cut causing audio glitch
Vapoursynth: VFM color space restriction
Synth: SelectEvery and SelectRageEvery
Vaporsyunth: AddGrain
NVEnc: L1/0 reference min value and gui glitch
Vapoursynth: make sure filters that change resolution or frame rate are applied to input&output in FilterView.

*changed*

Framework: switchted to Qt6 (Mac&Windows only)
Jobs: mplayer try to skip unnecessary muxing for video only encodings.
AutoCrop crash
Preview: add '-vf scale' during MPlayer Preview
UI: bunch of cosmetic changes
Synth: adjustements adjust to latest DGDecNV DGSource
Vapoursynth: Anime4k supported color formats
MPlayer: added 'ni' to mplayer dvd analysis call

*added*

NVEnc: support for --vpp-smooth
Vapoursynth: GLSL Resizer support
Vapoursynth: "only apply" from some filters.
Vapoursynth: support for GrainFactory3
Video: custom addition support for aomenc and vpx
Video: constrained quality 2pass VPX/AV1


-> downloads: http://www.selur.de/downloads

Cu Selur

[DzeSezMee]

any tutorial for installing on ubuntu 18.04?

[Selur]

not that i know of.
It's basically download Hybrids package, extract it and start and start the hybrid/Hybrid binary.
For adding Vapoursynth support see: https://github.com/Selur/hybrid-vapoursynth-addon

Cu Selur

[DzeSezMee]

not that i know of.
It's basically download Hybrids package, extract it and start and start the hybrid/Hybrid binary.
For adding Vapoursynth support see: https://github.com/Selur/hybrid-vapoursynth-addon

Cu Selur

How to fix this error?
https://i.ibb.co/7nN7KXT/Jepretan-Layar-2021-03-26-pukul-22-26-06.png (https://ibb.co/pfrCdxs)

[Selur]

What error do you mean?
-> Ah, now I see a screeshot.
That happens when using vnc iirc. (see: https://forum.selur.net/Thread-Install-Hybrid-on-Ubuntu-18-04-VNC + https://forum.selur.net/Thread-Hybrid-fatal-IO-error-11-on-x-server-1-0)
May be forcing a 'platform' like mentioned in the thread helps.

[Selur]

Hybrid rev 2021.04.05.1:
*fixed*

Synth: remove fulldepth parameter from DGSource
Output: mkv HDR tagging
Tools: mp4box version check
NVEncC: crop&co
Avisynth: FrameRateConverter: don't add Num/DeNum when double is used
Output: generate output name: add cut start&end frame info
Vapoursynth: removed vsAnime4KLevel
x264 'aud' GUI option not connected to model
Vapoursynth FilterPreview crash with Letterboxing
aomenc: fixed do not use adaptive quantization with lossless
Filtering: yadif - custom: field order ignored
fixed double calculation deinterlace/ivtc
x264: gui falsely indicated 'me=uhm' for preset 'slow' even when 'me=hex' was used.
Vapoursynth: try shortening file names for VSFilterMod (Windows only)
UI: dependencies adjustment when max value changed
x264: initial qt_max value.
aomenc: gui glitch
rav1e: 2pass crash
NVEnc: progress indication
Audio: problem with audio cut
Vapoursynth: script profile saving missing some filters.
Input: fixed image sequence pattern detection (bug cause by switch to qt6)

*changed*

UI: removed deinterlacer 'none', since it was just decorational
Vapoursynth: label Vapoursynth preview in 'Filter-' and 'Filter-Split'- view
Avisynth: CCD added ccdMT option
Avisynth: AssRenderer set matrix
Avisynth: Blur selection fixed cosmetics
Vapoursynth: mClean: use vcm instead of vcmod
Analysis: interpret 0.0000 as PAR as 1

*added*

Vapoursynth: "Always use Vapoursynth" to tell Hybrid to use Vapoursynth even when it's not needed
Vapoursynth: Filter-Queue (see: [INFO] About Vapoursynth Filter Order/Queue (https://forum.selur.net/Thread-INFO-About-Vapoursynth-Filter-Order-Queue))
Vapoursynth: DetailSharpen
Vapoursynth: 'Apply Only to' + 'Every'
Avisynth: CAS, vsMSharpen
Decode: ffmpeg vsync option
x265: support --(no-)eos/b
NVEncC: support warpsharp


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

Hybrid rev 2021.04.06.1:
*fixed*

Jobs: batch job creation


-> downloads: http://www.selur.de/downloads

Cu Selur

[ReinerSchweinlin]

Hybrid rev 2021.04.06.1:
*fixed*

Jobs: batch job creation


-> downloads: http://www.selur.de/downloads

Cu Selur
Nice :) Thanx!

[Selur]

Hybrid rev 2021.04.12.1:
*fixed*


Jobs: vpx version check
Avisynth: IVTC - TFM add additional options
Vapoursynth: fixed QTGMC bob


-> downloads: http://www.selur.de/downloads

Cu Selur

[Gravitator]

Салам!
How can I avoid resetting the encoding thread to the processor when changing the profile in x265?

[Gravitator]

I mean the preset :o

[Gravitator]

Bloated font.
https://images4.imagebam.com/f2/29/9f/ME1224W_o.jpg

[Selur]

How can I avoid resetting the encoding thread to the processor when changing the profile in x265?
You can't presets are rooted on the default values, thus Hybrid resets first to the default values and then applies the changes associated especially with the preset.

Bloated font.
will look at it

[Selur]

Hybrid rev 2021.07.17.1:
*fixed*

Decode: ffmpeg rgb -> yuv conversion with pc scale
Subtitle: do not convert subtitles to ttxt when muxing srt to mp4 with ffmpeg
Vapoursynth: embed sup subtitle
Avisynth: AutoAdjust only using general parameters when autoBalance was a enabled
Vapoursynth: ResampleHQ call
Vapoursynth: crash when fmtconv was used multiple times with 'spline16'
Avisynth: 64bit ImageSequence handling
Decode: add swapuv when using eq2 on 422 sources in FFmpeg
Xvid: encoding progress when using xvid_encraw
Vapoursynth: meta overwrite evaluation
Vapoursynth: Santiag h/v
Vapoursynth: Stab 'mirror'-parameter
Frontend: fixed font size in some tool-tips
Analyser: crash on some calculation decimal to fraction combinations
Vapoursynth: TIVTC + gpu
Audio: eac3 passthrough to mp4
Muxing: ffmpeg chapter generation
Muxing: ffmpeg + avc + parallel subjobs
Avisynth: don't use lsb with Avisynth+
Vapoursynth: GLSFilter calls when used with FilterQueue
Vapoursynth: Filter Queue handling
x265: adjust --uhd-bd do not remove options
x265: ultrafast profile
Audio: fixed 'auto (add all)' accidentally adding an audio track 2 times
Vapoursynth: Histogram depending on type restrict color space
Vapoursynth: filter preview crash for SRMD resizer
Vapoursynth: crash when ToneMap was used on rgb content
Avisynth: inserting custom code before QTGMC failed.
Audio: '-ignore_editlist' accidentally used even when handling pipeinput from sox
Crop: crop value not always properly reset on source change
Jobs: don't use mkvmerge with flac audio since mkvmerge doesn't support cutting flac fixed ass style overwrite
Vapoursynth: AddLogo (reenabled EdgeFixer)
Vapoursynth: set Fades didn't enable (on Linux)
Decode: ffmpeg -> x264 (use rawvideo)
Decode: x264 use raw video instead of y4m pipe when not using a script
Synth: SelectRangeEvery frame count/rate calculation of output
Filtering: fix rotation flag
Avisynth: tv/pc luma mapping on import
Autocrop: fix QSVCEnc detect fix

*changed*

Vapoursynth: changed sangnom.SangNom to sangnom.SangNomMod calls in scripts
Files: sort file names explicitly when collecting them from the OS
Frontend: AVStats: do not adjust based on selection/cut
Muxing: adjust chapter&qp file to cut
Vapoursynth: use 'accurate=1' when using VsFilterMod and high bit depth content. (slower but more accurate)
Vapoursynth: stepped resize support
Vapoursynth: FrameRateConverter
Synth: Hybrid now only auto-refreshes the Vapoursynth/Avisynth preview if "Filtering->Misc->Auto-refresh" is enabled, otherwise refresh needs to be triggered manually
Synth: F5 refreshes the Avisynth/Vapoursynth preview when 'auto-refresh' is disabled
Vapoursynth: allow multiple 'Framerate Interpolation'-methods when using 'Filter Queue'
Tools: remove 32bit qaac support
Internal: fraction handling
Synth: by default remove FFmpeg/Mencoder deinterlacers when using AviSynth
Vapoursynth: adjusted to Waifu2x cnn to r4
Vapoursynth: dgdecnv use default activated
Vapoursynth: added option to force a specific chroma location when converting from and to YUV420
Avisynth: changed script names
Avisynth: lifing some 8bit restrictions
adjusting SVT AV1 to removed options
Video: NVEncC: check that "NVEncC->Only encoder" is used only on supported formats
Video: x265: hme-range support
Video: svt-av1 ui glitch when switching encoding typ
DGDecNV: remove license.txt check of DGDecNV support, since it's free now
Video: x265 removed signal end of bitstream/videostream
Analysis: if scantype is interlaced but unknown after interlaced checked, assume tff if inpuHeight > 576
x264/x265: extend some options from single to double didget fl

*added*

Vapoursynth: debanding through vs-placebo
Vapoursynth: FrameRateConverter support
Vapoursynth: Detail-Preserving Image Downscaler (https://github.com/ WolframRhodium/VapourSynth-dpid/)
Vapoursynth: added support for RIFE (https://github.com/ HomeOfVapourSynthEvolution/VapourSynth-RIFE-ncnn-Vulkan) (Windows only atm.)
Vapoursynth: support for VSGAN through an 'addon' (Windows only), download only by request since it's huge
Vapoursynth: SpotLess from G41Fun
Vapoursynth: support for SmoothLevels from havsfunc
Vapoursynth: TIVTC from https://github.com/dubhater/vapoursynth-tivtc
Vapoursynth: added 'Flip vertical' to Santiag


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: BIG THANKS to DG for allowing me to bundle DGDecNV with Hybrid!

[Selur]

Hybrid rev 2021.07.18.1:
*fixed*

Vapoursynth: missing libraries and old scripts in MacOS package
Users might have to reset FilteringModel defaults and Vaporusynth filter order.

*changed*

Vapoursynth: changed sangnom.SangNomMod to sangnom.SangNom calls


-> downloads: http://www.selur.de/downloads

Cu Selur

[RedDwarf1]

I have been using this to quickly encode to HEVC with my Turing nVidia graphics card. The relatively, compared with other software, advanced options is a big reason why I use it.

However I have found that the encodes only come out with CAVLC and not CABAC entropy encoding. Can anyone tell me why this is so? Is it all that nVidia NVEnc can manage or is a setting that I am using incorrect? Is there a way to enable that to improve the compression/quality?

Thanks in advance for all input and help with this issue.

[Gravitator]

The sample was taken from this post >https://forum.doom9.org/showpost.php?p=1948242&postcount=8
Video
ID : 1
ID in the original source medi : 4113 (0x1011)
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4.1
Format settings : CABAC / 4 Ref Frames
Format settings, CABAC : Yes
Format settings, Reference fra : 4 frames
Codec ID : V_MPEG4/ISO/AVC
Duration : 1 min 0 s
Bit rate mode : Variable
Bit rate : 17.9 Mb/s
Maximum bit rate : 38.0 Mb/s
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Frame rate : 2 176.592 FPS
Original frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.004
Stream size : 11.4 GiB
Language : English
Default : Yes
Forced : No
Original source medium : Blu-ray
Is it better for a Hybrid to use information from "Frame rate" or "Original frame rate"? The first has sky-high rates and this leads to sky-high rates of coding time calculation in the program :scared:

[Sharc]

The sample was taken from this post >https://forum.doom9.org/showpost.php?p=1948242&postcount=8
[/SIZE]
Is it better for a Hybrid to use information from "Frame rate" or "Original frame rate"? The first has sky-high rates and this leads to sky-high rates of coding time calculation in the program :scared:
Framerate Mode "Variable" is also suspicious :eek:

[Selur]

Is it better for a Hybrid to use information from "Frame rate" or "Original frame rate"?
In this case definitly 'Original frame rate', also if the source really is a Blu-ray vfr is wrong, so ignoring time codes when importing with Hybrid is probably a good idea too.
(wondering if 'email notification' will ever work again for subscriped threads,...)

Cu Selur

[Gravitator]

Hydrid v2021.07.18.1

x265
Any change in the "Hierarchical M.E. range:" parameter will automatically save to the configuration profile.

[Selur]

x265
Any change in the "Hierarchical M.E. range:" parameter will automatically save to the configuration profile.

Can't reproduce.
What I did:

start Hybrid
enable "x265->Motion->Motion estimation->Hierarchical M.E."
set "x265->Motion->Motion estimation->Hierarchical M.E." to "full/full/full"
closed Hybrid
opened Hybrid

-> settings are not automatically saved
Yes, if you save a 'Configuration' those settings are naturally saved, same if you save a global profile or similar.

Cu Selur

[Gravitator]

I apologize for the inaccuracy.
Apply the currently selected configuration "Ультра" - the parameter Hierarchical range ME range: 16/32/48 will be loaded > change to 16/16/16 > re-apply the currently selected configuration "Ультра" - the original 16/32/48 did not load.
It seems that this happens if the name is in Russian.

[Selur]

Still can't reproduce this here.

What I did:

Start Hybrid
save a x265 configuration under "Ультра"
enable "x265->Motion->Motion estimation->Hierarchical M.E."
set "x265->Motion->Motion estimation->Hierarchical M.E. range" to "full/full/full"
change "Encoding mode" to "average bitrate (1-pass)"
apply the "Ультра" x265 configuration

-> "Encoding mode", "Hierarchical M.E." and the "Hierarchical M.E. range" values have all been properly reset.

It seems that this happens if the name is in Russian.
Could it be that the profile isn't loaded at all?
May be some file system encoding vs Windows language coding issue?
-> no clue atm.

Cu Selur

[Gravitator]

Could it be that the profile isn't loaded at all?

The profile is loaded, but it is not possible to restart this setting again.
May be some file system encoding vs Windows language coding issue?

It seems that this problem also affects the font (it becomes very bold).
>Demo (https://files.videohelp.com/u/227452/2021.08.08%20%28Hybrid_v2021.07.18.1_prof_font%29.mkv)

[Gravitator]

Activates/Aktivates (In the info field when selecting encoders).
Video handling:
Allows you to select the video encoder that Hybrid will use:
available options:
- x264 (MPEG-4 AVC aka. H.264) activates the x264-Tab
- XviD (MPEG-4 ASP) activates the Xvid-Tab
- VPX activates the VPX-Tab
- CUDA H.264 (MPEG-4 AVC aka. H.264) activates the CUDA H.264-Tab (Windows only)
- x265 (MPEG-4 HEVC aka. H.265) aktivates x265-Tab
- DivX265 (MPEG-4 HEVC aka. H.265) aktivates DivX265-Tab
- Kvazaar (MPEG-4 HEVC aka. H.265) aktivates Kvazaar-Tab
- passthrough = copy input clip to output without any changes (no reencoding)
- ignore (no video processing, needed for audio only output)

[Selur]

The too large font for 'Target mode&Co' is a bug with some tooltips which happend when switching to Qt6 -> will fix for the shown instances.
About the other 'bug', looking at the video:
- you apply "rskip"and the range ist set to 16,32,32
- you manually change the rante to 0,0,0
- you apply "rskip"and the range ist set to 16,32,32
- "Ультра" and the range stays 16,32,32
- you manually change the range to 0,0,0
and then the cinfiguration is not applied (no clue, why)
- you apply "rskip"and the range ist set to 16,32,32

-> tried the sam here, couldn't reproduce.

I got an idea,..

Ccould it be that you "Ультра" profile is an old profile which was created with an old Hybrid version and the profile simply doesn't contain any:
<HybridData value="true" name="hierarchicalME"/>
<HybridData value="umh" name="hme1"/>
<HybridData value="umh" name="hme4"/>
<HybridData value="hex" name="hme6"/>
<HybridData value="48" name="hmer1"/>
<HybridData value="32" name="hmer4"/>
<HybridData value="16" name="hmer6"/>

values?

If you open a profile with

Try:
1. applying you "Ультра" profile
2. saving it again
and see whether it now properly resets the setitng since the profile now contains the hme options,...

Cu Selur

[Gravitator]

Re-saving helped!

VPX
VP8->apply 'Conf:vp8 Hybrid Default' - the encoder was changed to VP9.

[Selur]

the encoder was changed to VP9.
Yes, it's the default in current versions.
Hybrid from time to time changes the Hybrid-defaults, but it does not change the program-defaults saved by the user.

Cu Selur

[Gravitator]

x265
Enc mode:ABR > minimize command line: off > start encoding - crash.

[Gravitator]

Input: *.yuv > Error (sorry, only supporting, YUV4MPEG2 raw atm.) > OK > the GUI freezes.

[Selur]

regarding x265: command line is complaing about 'no-script-cbr' being an unkown option.
Funny thing is that the command line help:
--[no-]strict-cbr Enable stricter conditions and tolerance for bitrate deviations in CBR mode. Default disabled
and the latest documentation https://x265.readthedocs.io/en/latest/cli.html?highlight=no-strict-cbr#cmdoption-strict-cbr mentions it.
-> created a bug report over at https://bitbucket.org/multicoreware/x265_git/issues/602/unkown-option-no-strict-cbr

Input: *.yuv > Error (sorry, only supporting, YUV4MPEG2 raw atm.) > OK > the GUI freezes.
I can reproduce the issue, I will look into it.

Cu Selur

Ps.: will send you a fix regarding the yuv-opening freeze. Will wait for feedback from the x265 devs regarding '--no-strict-cbr'.

[Selur]

Hybrid rev 2021.12.05.1:
*fixed*

Video: some models didn't load the saved defaults
Video: aoemenc:,2pass fails when custom addition was used
Vapoursynth: 12bit scaling bug
Vapoursynth: reset filters + removed use Deinterlace checkbox
Vapoursynth: HQDN3D parameter fix
Vapoursynth: handling Vapoursynth input scripts in case they are not further processed
Vapoursynth: TemporalGrain2 support
Avisynth: QTGMC doesn't support YUV411
Analysis/Muxing: handling mkv supporting that a single attachment can have multiple ids.
Jobs: handling jobs with 99+ subjobs
Audio: stereo to mono downmix
Input: Input dvd title minimum length was active even then the option was disabled.
Extraction: use more quotes when using mplayer for extraction on MacOS and Linux
Jobs: option to clean temp folder
Filtering: wrong letterbox values on reload of source
Vapoursynth: letterbox + telecine handling
Analysis: filter audio by language before initialization
Video: ProRes 4444 handling + activate Controls
Input: freeze on unsupported raw input
Vapoursynth: DGDecNV deinterlace call for tff
Audio: handling opus raw audio as Audio Input
Vapoursynth: sRestore was missing ChangeFPS when 'frate' was set while 'omode' wasn't 6.
GUI: some small glitches
Cut: cut end frame rounding
Video: NVEnc deinterlace
Video: ffnvenc accidentally removed on Linux systems.
Model: reset input when reloading defaults on non-empty input
Audio: pcm input handling
Avisynth: LSFMod now hdb compatible
Analysis: TimeCodeGrabber ffmpeg
Jobs: extrating mov to raw
Audio: ffmpeg->sox->ffmpeg encoding
Avisynth: RemoveDirt split typo
Avisynth: FrameRateConverter, MvToolsFPS
Vaoursynth: trim + Vapoursynth FilterPreview
Audio: ac3 dialognorm, audioNormalization
Muxing: Cineform muxing
Avisynth: 'Filtering->Avisynth->Denoise' en-/disable not saved in profiles.
Vapoursynth: forgot adjuster for SRMD
Vapoursynht: drop alpha channel when using LWLibavSource
Muxing: remuxing vfr mov to m2ts by ignoring the time codes
Decoding: vidstab when ffmpeg encoders are used
Vapoursynth: fixed ResampleHQ call when stepped resize is used
Audio: audio only audio job creation
Vapoursynth: vsHDRToSDRDGLight init value

*changed*

Video: x265: support old 16bit x265 from 2015 (misc.ini, supportX26516bit=true; Windows only)
Decode: mplayer/mencode use '-demuxer lavf' not '-demuxer mkv' for webm content.
Video: encoding type "specific filesize/bitrate (1st pass)" -> "specific filesize/bitrate (2pass - 1st pass)" and "specific filesize/bitrate (2nd pass)" ->"specific filesize/bitrate (2pass - 2nd pass)" to limit confusion
Video: NVEnc - added '--chroma-qp-offset X'
Avisynth: TemporalDegrain2 supports 8-16bit YUV
NVEnc: color range values
Muxing: mkv mpeg-2 interlaced handling
Jobs: create sub-folders in temp per job (experimental)
Vapoursynth: tcanny sigma to 0.5
Synth: allow to force bit depth when using flash3kdb
Vapoursynth: FilterOrder cosmetics: a. used filters are bold b. filters can be moved through mouse drag&drop
Decode: not uisng '-demuxer mkv' but for mkv/webm compatible input (this might cause problems, but we will see)
Analysis: changed MakeMKV workarounds,...
Input: add 'iso' to the file extensions to ignore
Vapoursynth: sRestore min threshold to 1
Jobs: use json options file for mkvtoolnix
Jobs: explicitly set LANG and LC_ALL in qprocess environment
Muxing: allow passthrough ProRes to mkv
Vapoursynth: allow GLS Resizer&Filter for Mac
Vapoursynth: updated ReadMPLS filter on Windows, fixed code
Vapoursynth: "lower resolution before resize" can now be restricted to width/height
Avisynth: FrFun7 now available for 64bit Avisynth
Video: support FFV1 set output color space.
Avisynth: disable HDRAGC for 64bit
Avisnyth: more Histogram options separateFieldInterlacedHandling
Vapoursynth: switch to API4
Analysis: adjust to MediaInfo for HDR transfer detection
Video: Xvid encraw doesn't have '-i ..' anymore
Vapoursynth: removed openCL option from vsMCDeGrain since there's no TCannyCL anymore.
Analysis: removed unnecessary code Analysis if container frame rate more than 20 times the stream frame rate, use stream frame rate. x264 version: convert 0.164.x to version 5000 vp9: restrict cpuUsed to
Filtering: ffmpeg tonemap added npl-parameter

*added*

Video: ediaaCuda (Windows only)
Vapoursynth: muvsfunc.SSMI_downscale = SSIMD
Vapoursynth: DPIR, vsRealESRGAN, BasicVSR++, TorchRIFE, vs-swinir, vs-hinet support (in pytorch addon)
Avisynth: Spotless
Vapoursynth: only apply range option
Vapoursynth: chunked encoding support for 1pass encoding methods (*experimental*)


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: Note that Hybrid switched to Vapoursynth API4 so Vapoursynth, filters and tools usind Vapoursynth need to be updated. Also deinstall old Hybrid before installing this release.

[Gravitator]

Avisynth preview: Info - does not work.

[Selur]

Will look at it during the week. (Since I'm not really using Avisynth much nowadys I hadn't noticed.)
Looks like it never worked with LWLibavVideoSource, but only with FFMpegSource2 and I deleted the FFMS2.avsi (thought it wasn't needed) which contains FFInfo.
So if you extract FFMS2.avsi (https://github.com/FFMS/ffms2/releases) into your avisynthPlugins folders at least when using FFMpegSource2 it should work.
-> it's a bug in avsViewer -> send you a link to fixed avsViewer builds via PM.

Cu Selur

[Selur]

@Gravitator: Seems like this is related to Avisnyth++ itself. If you disable multithreading through enabling 'Filtering->Avisynth->Misc->Script->Miscellaneous->Disable MT' frame detection works fine, but it will slow down all other filters.

[Gravitator]

@Gravitator: Seems like this is related to Avisnyth++ itself. If you disable multithreading through enabling 'Filtering->Avisynth->Misc->Script->Miscellaneous->Disable MT' frame detection works fine, but it will slow down all other filters.
This helped with the correct display of information about I/P/B frames:o

[Selur]

Foudn another way to fix this without having to disable MT, send you a link to modified avsViewer builds.

Cu Selur

[Gravitator]

1. Replacement C:\Program Files\Hybrid\32bit\avsViewer.exe leads to activation of Info for LibavVideoSource.
2. Adding files avsViewer.exe and FFMS2.avsi in C:\Program Files\Hybrid\32bit\avisynthPlugins leads to activation of Info for FFmpegSource.

The new fix does not work without disabling MT. How to switch to 64bit mode?

[Selur]

a. avsViewer should be in 32bit folder not in 32bit/avisynthPlugins folder.
b. for me this happens:
when I use Avisynth 32bit and load a source where LWLibavVideoSource is used triggering the Info-button properly adds Info() to the preview script.
when I use Avisynth 32bit and load a source where FFVideoSourceis used triggering the Info-button properly adds FFInfo() to the preview script, but at the wrong position. (not above Prefetch)
when I use Avisynth 64bit and load a source where LWLibavVideoSource is used triggering the Info-button properly adds Info() to the preview script.
when I use Avisynth 32bit and load a source where FFVideoSourceis used triggering the Info-button properly adds FFInfo() properly before Prefetch().
-> will look into it.
How to switch to 64bit mode?
"Config->Internals->Avisynth->Avisynth type"

[Selur]

Hybrid rev 2021.12.12.1:
*fixed*

Avisynth: RGBAdjust controls
Avisynth: FRIM handling
Vapoursynth: some popups when HINet isn't present


*changed*

add chapters to non-mp4 files when muxing with ffmpeg


-> downloads: http://www.selur.de/downloads

Cu Selur

[Gravitator]

v1.12.12.2021

avsViewer32
Prefer FFmpegSorce: does not show I/P/B without disconnecting MT;
Support DGdecNV: "Avisynth preview error: -> DGSource: Could not open DGIndexNV.ini. (C:\Users\GRAVITATOR\AppData\Local\Temp\tempPreviewAvisynthFile19_16_30_812.avs, line 6)" - the window does not start.

avsViewer64
Support DGdecNV: "Avisynth preview error: -> DGSource: Could not open DGIndexNV.ini. (C:\Users\GRAVITATOR\AppData\Local\Temp\tempPreviewAvisynthFile19_33_15_828.avs, line 6)" - the window does not start;
LibavVideoSorce: "Avisynth preview error: -> Import: unable to locate "" (try specifying a path), error=0x57 (C:\Users\GRAVITATOR\AppData\Local\Temp\tempPreviewAvisynthFile19_23_57_531_tmp.avs, line 12)" - error when enabling info and the window is closed when the button for the next frame is pressed;
Prefer FFmpegSorce: Info - press the button for the next frame > I dont know what FFSAR_NUM means ([Scriipt Clip] line 1).

x254/x265
Changing the encoding mode (ABR, 2-pass, b-frames) at a bitrate of 1000kbps results in the message "Current settings would result in a file size below 1 MB. -> changing Filesize to 1MB".

[Selur]

I dont know what FFSAR_NUM means
-> not chaning since it would require a rewrite of FFMS2.avsi

x254/x265
Changing the encoding mode (ABR, 2-pass, b-frames) at a bitrate of 1000kbps results in the message "Current settings would result in a file size below 1 MB. -> changing Filesize to 1MB".
Can't reproduce, only should happen if the warning is correct.

about DGDecNV: Try starting DGIndexNV.exe, it should recreate the file. Alternatively you can create empty DGDecNV.ini files.
about Info: probably forgot to update avsViewer

-> seems like I should think about dropping 32bit Avisynth from Hybrid

Cu Selur

[Gravitator]

Should be changed to DGIndexNV.ini
> C:\Program Files\Hybrid\64bit\Avisynth\avisynthPlugins\DGIndexNV.exe
> C:\Program Files\Hybrid\32bit\avisynthPlugins\DGIndexNV.exe

avsViewer64
Warning "FFSAR_NUM ([Scriipt Clip] строка 1)" is disabled via MT off.

x264/x265
Changing Filesize to 1MB > demo (https://files.videohelp.com/u/227452/x264_x265%201Mb%20spamming.mkv)

[Selur]

x264/x265
Changing Filesize to 1MB > demo

This behaviour is correct.
Hybrid does not support encoding to file sizes below 1 MB through bit rate based methods.


Should be changed to DGIndexNV.ini
> C:\Program Files\Hybrid\64bit\Avisynth\avisynthPlugins\DGIndexNV.exe
> C:\Program Files\Hybrid\32bit\avisynthPlugins\DGIndexNV.exe

No clue what you intended to say wih that.

avsViewer64
Warning "FFSAR_NUM ([Scriipt Clip] строка 1)" is disabled via MT off.
Yes, but I see no script change in Hybrid that could prevent it aside from dropping multithreading which is not happing.
It's this way for years.

Cu Selur

[Gravitator]

This behaviour is correct.
Hybrid does not support encoding to file sizes below 1 MB through bit rate based methods.
Why is it triggered on b-frames? How to disable it?

DGDecNV: Try starting DGIndexNV.exe, it should recreate the file. Alternatively you can create empty DGDecNV.ini files.
An empty file helped.

[Gravitator]

I propose to expand the viewer :rolleyes:
If you hold click and down on the original photo, the filtered photo will be displayed (as on photo comparison sites).

[Selur]

Why is it triggered on b-frames? How to disable it?
It has nothing to do with b-frames. It's triggered if you tell Hybrid to use a specific bit rate for encoding and that bit rate would result in a file size below 1 MB.

I propose to expand the viewer
If you hold click and down on the original photo, the filtered photo will be displayed (as on photo comparison sites).
Not happening, but I welcome patches for avsViewer if someone wants to implement it.
Personally I'm fine with interleaving filtered and original using 'Filtering->Avisynth->Misc->Preview->SplitView Placement->interleaved'.

Cu Selur

[Gravitator]

It has nothing to do with b-frames.

At the end of the video(previously gave a link to a demo), there is a display of the message output when changing the b-frame at the encoders.
https://i.ibb.co/WcddFZw/1mb.jpg (https://ibb.co/zSLLbpY)

[Gravitator]

AOM
Incorrect encoding progress in %.

[Selur]

B-frames influences the container overhead and as such influences the bit rate calculation.
Since Hybrids calculation assumes a minimum file size of 1 MB (this will not change), either:
a. do not use such short content
b. do not use such low bit rates when using such short samples
c. do not use encoding modes which use bit rates when using such short samples
d. do not use Hybrid

AOM
Incorrect encoding progress in %.
I know, it's this way for quite some time, haven't had time to look into it. (think I fixed some of it and atm. it's mainly missing a *100)

[Gravitator]

Why can't I use short videos with a transfer rate of 1 Mbit/s? Discrimination!

[Selur]

Just for clarification: Hybrid it's a restiction of the file size to 1MB not a transfer rate restriction.

[Selur]

Hybrid rev 2022.03.19.1:
*fixed*

Analysis: FFmpeg color matrix detection
Input: make sure fps&frame count in encoders get updated on loading a source
aomenc: noise settings for 2pass constrained quality
Vapoursynth: split view 'diff' for RGB content
Preview: do not use MPlayer for Preview when Avisynth 64bit is used
Avisynth: profile: all Misc settings should now be saved in the Avisynth-profile
Avisynth: YUY2 needs to converted to YV16 for output.
Vapoursynth: mClean handling
Vapoursynth: fallback downsizer selection
Vapoursynth: missing PlaceboDebandThreshold
Vapoursynth: missing AviSource library loading
Vapoursynth: vsSMDegrain NNedi3 subpixel selection
Vapoursynth: stepped resizing
Vapoursynth: fixed vsRIFE missing miscfilters import when sceneChangeDetection was used.
Vapoursynth: adds mod4 for vsMCTemporalDenoise&avsTTempSmooth
Vapoursynth: clean up model combis
Vapoursynth: allow to set ColorMatrix flag when using timeCube
NVEncC: Level<>Bitrate mixup
FFMpeg: added -bsf:v h264_mp4toannexb for H.264 .avi muxing
UTVideo: use yuv44p for ULRG


*added*

Vapoursynth: support filter masking (max and edge mask)
Vapoursynth: FillDrops, CCD, parameterized GLSResizer
Vapoursynth: 12bit color handling
Vapoursynth: parameterized GLSL: adaptive-sharpen, LumaSharpen, ThinLine, A4K Darken
Vapoursynth: option to limit Libav threads in VapoursynthTab
Avisynth: support Histogram "RGBLevels", "CMYLevels", "RGBParade"
Avisynth: use avsInfo/avsInfo64 selection depending on whether Avisynth 32bit or 64bit selection
NVEncC: -vpp-colorspace lut3d=".." support


*changed*

Avisynth: Updated TemporalDegrain2 to 2.4.2
Vapoursynth: Adjusted to VSGAN 1.6.4
Avisynth Script input: force FFmpeg autoCrop detection
Vapoursynth: moved aaf and TemporalDegrain to 'rescued.py'
Vapoursynth: MCTemporalDenoise nCPU
Vapoursynth: save misc in profiles
FFMpeg now using matrix instead of out_color_matrix
GUI: fullscreen, font size adjustment,...
FFMpeg: explicitly call YUV->RGB conversion with matrix on PNG creation and PNG encoder call
Vapoursynth: KillerSpots and RemoveDirt disabled since RemoveDirt is broken for Vapoursynth
Vapoursynth: '# require colorformat XXX' for custom sections
Vapoursynth: changed default behaviour regarding loading libraries and added vsLoadLibrariesOnNonWindows to misc.ini on non-Windows OSs
Vapoursynth: use KNLMeansCL-Wrapper from havsfunc for speedUp on YUV content when all channels are filtered
Output: allow Hevc for mov container

*removed*

misc.ini: fontSize and toolTipFontSize options


-> downloads: http://www.selur.de/downloads

Cu Selur

[locotus]

Latest update seems to had broken something in NVenc X265, conversion stops after audio's and finishes the job.

[Selur]

Works fine here.
NVEncC version was updated, so you might have to update your NVIDIA drivers.
If updating the drivers doesn't help provide a debug output of the issue, maybe I can see something in the debug output.

Cu Selur

[locotus]

NVidia runing drivers are V511.79, updated on 02/14/2022. Uninstalling and going back to the previous Hybrid version, works normally.

[Selur]

Just noticed that I broke Vapoursynth script creation during job creation. So if you used Vapoursynth tha tis probably it.
-> will release a fixed version tomorrow.

Cu Selur

[locotus]

Just noticed that I broke Vapoursynth script creation during job creation. So if you used Vapoursynth tha tis probably it.
-> will release a fixed version tomorrow.

Cu Selur

Thanks a lot.

[Selur]

Hybrid rev 2022.03.20.1:
*fixed*

Vapoursynth: script creation during script creation
GUI: save icon/text selection in GlobalDataModel


*removed*

misc.ini: disableIcons option


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: @locotus: If you still have issues, share a debug output.

[mparade]

Hello,

In Hybrid is it possible to convert only TrueHD Atmos audio track of a container file keeping all other tracks intact then mux everything back to an mkv container? I stream content to my OLED from NAS but there used to be a compatibility issue when an Atmos track exists in the container too. At the same time I would like to take the advantage of having an Atmos track, that is that I would like to encode it using Hybrid keeping most of the quality.

Thank you for the help.

[Selur]

You can specify for each track whether it should be passed through or reencoded in some way, so yes what you describe should be possible assuming you configure it manually. You can't tell Hybrid to automatically ony convert the audio streams that match a specifc format.

Cu Selur

[Selur]

Small heads-up to anyone using Hybrid with svt-encoders:
I'll remove support for those encoders for the time be, since:

encoding does not work with latest version
looking at https://github.com/AOMediaCodec/SVT-AV1/blob/dae3a5e62ea5244ab628f7f212c80019cd32f8f1/Docs/svt-av1_encoder_user_guide.md integration will have to be completely rewritten.


-> Update: I adjusted the SVT-AV1 which wasn't that much work (~10hrs) since most of the interessting stuff doesn't really work and thus doesn't need support.

Cu Selur

[solidservo]

Hi guys!

I recently started using Hybrid for converting some of my DVDs. Quality is really awesome! But I have a question about batch processing.

Obviously I want to use the same settings on all episodes of a tv show, so i created a profile. However I found that even with a profile I still have to manually set the following 2 settings for every single episode I want to run through Hybrid.

- Convert output to PAR: Square Pixel
- target resolution: 1024

Is there a way how I can avoid having to set these two settings for every single episode I run through hybrid?

Or is there a way how I can tell hybrid to simply process all files (episodes) inside a folder using the same settings?

Thank you! :)

[Selur]

regarding keeping the target resolution: enable "Crop/Resize->Misc->Misc->Keep resize for new Source"
No clue why 'Convert output to PAR' shouldn't be saved, but I'll check later.

Or is there a way how I can tell hybrid to simply process all files (episodes) inside a folder using the same settings?
Either select all the files as input or drag&drop the folder into Hybrids input line. Configure what you want, create jobs, start jobs.

Cu Selur

[solidservo]

No clue why 'Convert output to PAR' shouldn't be saved, but I'll check later.
Yeah these two always reset when I load a new input videofile.


Either select all the files as input
Oh god I am so stupid.
I did not think that I could just select all files by holding down shift or ctrl. doh
Thanks!

[Selur]

Yeah these two always reset when I load a new input videofile.
I checked default behaviour is: reset on source change unless it's a multi-input source (DVD, Blu-ray, list of files)
Enabling "Crop/Resize->Misc->Keep output PAR" should help.


Cu Selur

[Sharc]

Sorry if this has been addressed before, but could the deinterlace filter BWDIF() be added to the Avisynth filter selection in Hybrid? It is currently available for Vapoursynth only if I get it right.

[Selur]

(not using Avisynth nowadays) I'll look into it.
=> send you a link to a dev version via pm.

Cu Selur

[nel-son]

when loading a qp-file hybrid is searching the file in the folder $USER/.hybrid (on linux). i think it would be easier to search the file in the same folder where the input file is.

[Selur]

I'll look into it shouldn't be hard to change.

[Yups]

Can you add AV1 format which is supported by QSVEnc? Also you have to increase the constant quantizer limits and ICQ as well for AV1, they differ.

H265 8bit= 51
H265 10bit= 63 (51+12)
AV1= 255 (ICQ same limit 255)

You can also increase the gop-ref-dist/bframes limit to 33 for AV1 CQP, this is the maximum at the moment.

[Selur]

Haven't looked ad AV1 in QSVEnc since I have no hardware which supports it. (None of the ARC gpus are available here.)
-> it's planned, but will have to wait till I have a way to actually use and test it. :)

[Selur]

Hybrid rev 2022.11.26.1:
Tons of changes, see: https://selur.de/changelog
Make sure to either uninstall previous installations or reset defaults and load, check, resave every config&co you use.

-> downloads: http://www.selur.de/downloads

@Linux-Users: note that this is the last 5.95 Qt build will switch to Qt6 and a newer Ubuntu for building in the future.
@MacOS-Users: Sorry, no update of Hybrid atm. building all the tools is just too much trouble.

Cu Selur

[locotus]

Having this crash: helper exit code -1073741795.
After audio conversion finish.

Win 10, lastest updates, Using NVenc to x265

Bak to prior version, working fine.

[Selur]

@locotus: No debug output, no details. -> Can't do anything about it. Either stick with the old version or share details.

"helper exit code -1073741795" usually is caused by some bug in the decoder call or used scripts.

Cu Selur

[Selur]

Hybrid rev 2022.11.28.1:

Video: x264 range mixup
Decode: Mencoder -> FFMpeg pixFormat adjustement
Avisynth: Custom additions only working for first 50 filters


-> downloads: http://www.selur.de/downloads


Cu Selur

[Selur]

2023.01.22.1:
*fixed*

Analysis: mkv attachement type detection
Video: NVEncC disable psnr/ssim when input is 4:4:4
Vapoursynth: width&height mixup in SSIMD checks
Vapoursynth: missing matrix info on separate field handling
Video: x264: temporalsublayer adjustment
Video: x265: --sbr mixup
Video: aomenc: decouple auto-alt-ref and some other options
Vapoursynth: vsYadifMod, EEDI3 only supports YUVXXXP8
Jobs: ffmpeg muxing deleted audio stream even if it was an input file
Jobs: chuncked encoding
Avisynth: applying Toon, VagueDenoiser, AddGrain, GrainFactory3 during filter order handling
Synth: fix DG PullDown handling
Muxing: avi handling
Vapoursynth: vsFilterQueue fmtconv handling
Jobs: temp-folder per job handling for qp&stats files.
Analysis: understand "Mastering display color primaries: BT.709"
Avisynth: fix filterQueue
Video: NVencC references gui glitches
Video: NVEncC: weightedp not added to command line
Video: NVEncC: remove xxxhq modes
Video: fix chapter point calculation
Audio: audio only handling
Video: SVT-AVI options
Vapoursynth: Gradfun3d+Preview+Masking
Vapoursynth: fixed - save Masking&Co in Vapoursynth profiles
Vapoursynth: FilterQueue + MAsking and onRegionOnly</öo>
Vapoursynth: Santiang call
Filtering: AutoCrop adjustment
Avisynth: fix disabled controls
Vapoursynth: eedi3m vs eedi3m_opencl (Windows only)
NVEncC: no profile&level&tier control for AV1, only 4:2:0 supported.

*changed*

Qt Updates: Changed to Qt 6.4.2 on Windows and Qt 5.15.3 on Linux
Vapoursynth: more options for MVToolsFPS
Video: aomenc cosmetics
Analysis: scan order check adjusted
Synth: libavsource: repeat=true handling
Video: aomenc adjust golden frame max, min values
Video: NVEncC support Lut interp option
Avisynth: FFT3DGPU add bitdepth option to force input depth
Synth: forceFilm through DGDecode, thanks to DG!
Vapoursynth: updated TemporalDegrain2
Jobs: additional user prompt when clearing temp folder
Avisynth: add PreRoll when using DGDecNV
Video: NVEncC: support aq independent of encoding mode
NVEncC: disable multiRef when it's not supported
Jobs: chunking: restrict used filters during scene change detection
Vapoursynth: vsRealESRGAN, Fusion, num_streams support
Vaporusynth: vsRetinex: increase sigma max
Vapoursynth: increase dehalo_alpha and finedehalp rx/ry max values

*added*

Audio: 'Ignore reencode and passthrough if format is ...'-option
Vapoursynth: support for NeoF3kD
Vapoursynth: weighted resizing
Vapoursynth: torchAddon vsSCUNet, vs-femasr, AnimeSR support
Video: aomenc '--gf-max-pyr-height'-option

*removed*

Video: SVT-XXX removed 'Create qp file for chapters', see: https:// forum.doom9.org/showthread.php?p=1981213#post1981213
Video: cuda nvencu encoder support

-> downloads: http://www.selur.de/downloads


Cu Selur

[OrangeColaJuice]

Just installed the latest version.
Loaded a clip, changed no option but clicked on preview avisynth script and I get this error:
https://files.catbox.moe/l828ey.png
I cannot do anything but accept it at which point the program shuts down.

[Selur]

Since I can't reproduce this here at all my guess is that you had an older Hybrid version installed before, with incompatible defaults.
-> try resetting your defaults etc.

Cu Selur

[Selur]

Hybrid rev 2023.03.17.1:
*changed*

Muxing: FFmpeg can't handle ttxt subtitle
Avisynth: use vsDeblockP77 for DeblockPP7
Vapoursynth: missing some needed checks
Vapoursynth: re-enable RealCUGAN resizer
Tools: set priority through niceness on vsViewer&avsViewer
Torch: adjust to SwinIR v3, DPIR v3.0.0, vsFeMaSR updated, vsGMFSS_union
Video: disable NVEncC/QSVEncC 'encoder only' for 'DVD input

*fixed*

Avisynth: mclean not called with default filter order
Avisynth: SelectRangeEvery
Vapoursynth: mod requirement handling for some filters
Vapoursynth: crop with interlaced fields handling
Vapoursynth: a few issues around masking
Vapoursynth: restrict Santiag color spaces
Vapoursynth: (normal) TimeCube call
Vapoursynth: restrict color spaces for NVEncCL
Vapoursynth: convert RGBH to RGBS for fmtc
Vapoursynth: vsPreview handling non-square aspect ratio
Vapoursynth: loading BilateralGPU filter in vsGradFun
Synth: letterbox issue where letterbow width/height addition was uneven
Jobs: typo in job generation for blu-ray input
Jobs: index file path for 'Create sub-folder ...'
Video: SVT-AV1 denoise option rennamed
Video: x264 only add rc-lookahead when mbtree is used
Video: SVT-AV1 HDR10 signaling
Video: x265 signaling luma always as limited
Audio: multi-input raw audio reencode

*added*

Vapoursynth: mode vsSpotLess options
Vapoursynth: support for vsDGDenoise, vsDGSharpen
Video: NVEncC npp support
Filtering: overwrite input color matrix and luminance value


-> downloads: http://www.selur.de/downloads

Cu Selur

[ReinerSchweinlin]

2023.01.22.1:

*removed*

Video: cuda nvencu encoder support



Could you clayrify on this one? Is this the driver build in encoder?

[Selur]

It's this encoder: https://github.com/Selur/CudaEncoder for which NVIDIA dropped support with 340.52 which were released 2014.7.29. ;)
Hybrid still had support for it if a specific misc.ini option was added.

[OrangeColaJuice]

I think that some AVS scripts are missing.
https://files.catbox.moe/whs8pv.png

[Selur]

I'll look into it, no clue when that script was lost.
I remember that script was used when downsizing more than a factor of 2, it tried to preserve more of the chroma info.
=> this will take me a while to find that script :(

[Selur]

Found it:

function force444DownSize(clip v1, int "width", int "height")
{
Y = v1.ConvertToY8().Spline36Resize(width, height, src_left=-0.5)
U = v1.UToY8()
V = v1.VToY8()
return YToUV(U, V, Y)
}

Uploaded it to my GoogleDrive (https://drive.google.com/file/d/1vvrROdwCMpS7wmbHxkP77Xcd3J0rv5kn/view?usp=share_link)
Simply save a copy of the file inside Hybrid/32bit/avisynthPlugins and Hybrid/64bit/Avisynth/avisynthPlugins and it should work.

Cu Selur

[VoodooFX]

Hi Selur, can Hybrid do one click batch encoding like MeGUI?

[Selur]

Don't remember what MeGUI does in that mode, but you can do batch encoding (drag&drop a folder or tons of files as input, adjust configuration and create tons of jobs).

[Selur]

@OrangeColaJuice: Had some time to look at the 'force 444 downsample'-option and I noticed that I never properly implemented support for it in Vapoursynth and Avisynth.
I fixed that (hopefully) and send you a link to a dev version via pm. (link will be active for ~24hrs)

Cu Selur

[OrangeColaJuice]

Thank you @Selur
everything works on my machine as intended, sorry for answering so late, downloaded the build within the 24 hours but forgot to come back here

[Selur]

Thanks, for letting me know that it works now. :)

[juanitogan]

Hybrid 2023.03.17.1

I just finished my first test of Hybrid (handy tool, btw, for one-off jobs when I'm not in Avisynth mode) and found that the x265 option does not set PAR properly (well, not with an ordinary NTSC DVD 720x480i 4:3 source anyhow). The current workaround is to set PAR manually by adding it to the x265 command line with `--sar 8:9` for example.

The weird thing is, when packaging to MKV, the PAR appears to be set correctly according to both MediaInfo "DAR 4:3" and ffprobe "SAR 8:9 DAR 4:3". VLC also plays it correctly. Roku, however, does not play it correctly but instead plays it with 1:1 PAR (DAR 3:2). (I didn't dig deeper with ffprobe to try to find the exact missing data point.)

I've learned over the years that whenever I have trouble with an MKV, I should first test it as an MP4. MediaInfo gives more/better details on MP4s and ffmpeg somehow works better--or gives better messages--with MP4s. (MakeMKV, for example, does not build MKVs from DVDs that trim well because it fails to do the equivalent of `ffmpeg -fflags +genpts` -- which does not always work when transcoding MKV-to-MKV for some reason.)

Thus, when I tried packing with Hybrid to MP4 instead, nothing liked the result. MediaInfo showed DAR 3:2, ffprobe showed SAR and DAR metadata to be absent, and both VLC and Roku played it in 3:2 instead of 4:3.

I poked around at various setting for a few hours running many tests, trying to find some conflict somewhere, but found nothing that helped the PAR setting and ultimately looked up x265 switches and tried adding `--sar x:y`. Oh, yes, I also tested x264 and that worked fine as well.

This is using primarily all the default settings as installed. I didn't change anything surprising (except during troubleshooting). That is, I set video to x265, audio to passthrough, default container to MKV; x265 CRF to 19 (and others during testing), preset to Slow (and others), profile to Main (8-bit); left deint at GTGMC (Vapoursynth), set its preset to Very Slow and checked Bob. That's it.

All the boxes that show detected input PAR (such as in Crop/Resize) correctly show 8x9. The main Base tab correctly shows PAR 0.889 in the video summary.

The only other thing I found was that videos from Hybrid don't set the progressive metadata flag like ffmpeg does, but forcing that didn't change anything either since progressive is the default for x265 anyhow.

[Selur]

I'll look into the 8:9 sar issue.

[Selur]

Send you a pm with a link to a dev version which should fix the issue.

Cu Selur

[juanitogan]

Auto deint: VIVTC > VDecimate tab > Cycle

Changing the Cycle does not change the output frame rate or count, like it does with TIVTC > TDecimate > Cycle. Assuming, that is, cycle is number of frames out of which to drop 1 frame.

[Selur]

Will look at it. (Haven't used VIVTC for ages. :))

[Selur]

Send you a pm with a link to a dev version which should fix the bug.

Cu Selur

[Selur]

Hi, sadly the Linux dependency builds are broken atm.
(see: https://github.com/Selur/hybrid-vapoursynth-addon)
If someone gets neo-vague-denoise (vsvaguedenoise is gone) and vslsmashsource to compile, please let me know.
=> if someone can help with the Linux builds, please let me know. (Don't have the time to look into it really myself and without these, I probably will remove the .deb package)

Cu Selur

[Selur]

Small heads-up for the Linux users: I probably won't build any new Linux builds since it's more and a hassle to build all the dependencies (especially VS plugins), Hybrid then will be Windows only.

[ReinerSchweinlin]

Small heads-up for the Linux users: I probably won't build any new Linux builds since it's more and a hassle to build all the dependencies (especially VS plugins), Hybrid then will be Windows only.

Thanx for the info. Totaly understandeable...

Sorry I can´t help, I am just not competent enough for this level..

[Selur]

2023.12.02.1:
*fixed*

Subs: subtile sorting in SubtitleQueue
Mux: tsMuxeR job optimization fixed
Vapoursynth: fixed vsTemporalDegrain2 with FilterQueue
Vapoursynth: vsQTGMCFilter lossless + masking
Vapoursynth: resize before resize
Vapoursynth: descale before deinterlacing
Vapoursynth: YUV420PH und YUV444PH handling
Vapoursynth: vsGradFun,vsFlash3DB supported color spaces
NVEncC: weightP is only available when no b-frames are used
NVEnc: MREF bool<>int mixup
Mux: mkv subtitle support XX-XX language tags
Muxt: tag handling with mp4box
NVEnc: nvenc hardware deinterlacing
NVEnc: qsvenc hardware deinterlacing
Mux: sub language handling
Avisynth: addition handling
Audio: 24bit audio to flac handling
Avisynth: GenericScriptFilter path handling fix
Torch/MLRT: Resizer with interlaced content
Avisynth: deinterlace RGB handling fixed in compare view
Vapoursynth: vsRGBAdjust small fixes
Jobs: adjust chapters and length to cut
Vapoursynth: soft telecine handling
Jobs: mkvversion check
Vapoursynth: ReplaceBlackFrames cpu mode fixed
Vapoursynth: VIVTC frame rate calculation fixed
MLRT: vsRIFEmlrt requires mod32
MLSRT: vs-mlrt + masking issues
Output: mp4/mkv container par signaling
Torch: and CodeFormer
Video: hdr setting for svt hevc
Vapoursynth: RealCUGAN fix force tile enabling
Torch: vsDPIR FilterQueue
Vapoursynth: vsTemporalDegrain2 meBlksz vsTemporalDegrain manual overlap size
Vapoursynth: gmfss_union with gmfss_fortuna
Jobs: chapter handling
Avisynth: 444 downsize
Video: H.264 check width/height restriction

*changed*

small cosmetic changes
HEVC interlaced handling
Avisynth: LSFMod, LSF allow YUV420, YUV422, YUV444
Vapoursynth: RIFE/RIFEmlrt up to 4.12_lite
vsRorsch: vsTorchRIFE up to 4.11
Mux: tsMuxer calls
Vapoursynth: enabled vsKillerSpots advanced
Jobs: allow thd for mp4 muxing
Vapoursynth: VSGAN detection
Vapoursynth: sRestore havsfunc -> core =1, muvsfunc => linear
Input: MakeMKV interlaced handling
Vapoursynth: vsVagueDenoise -> vsNeoVagueDenoise
Video: avc support 6.0-6.1 levels
Vapoursynth: vsMCTemporalDenois no YUV444PS
vsmlrt: updated to lates vsmlrt
UI: switch to Base-tab if something is dragged into Hybrid
Avisynth: add Original/Filtered subtitle when comparing
UI: switch to Qt6.6.1 (Windows)
Vapoursynth: VSGAN: Allow to apply multiple models, one after the other.
DVD: disable DVD reencoding unless Synth is used, due to a bug in mencoder.
Vapoursynth: make sure to set scanorder to progressive after deinterlacing/IVTC/ fieldmatching
Video: allow aomenc lag-in-frames not just during 2pass encoding

*added*

x264: added 'auto color space / bit depth'
Vapoursynth: stepped resize: added "CAS(sharpness=X), "aWarpSharp2(depth=X)"
Vapoursynth: vsMotionMask, vsDeScratch, vsRemoveDirtMC, vsReplaceSingle, vsChormNR, vsChormNRcuda,VsFillDuplicateFrames, vsReplaceBlackFrames, vsSoftLight
Vapoursynth: vsBwdif NNEDI3,edeint-parameters
Vapoursynth: experimental support for vsBestSourc
NVEncC: 'auto' for b-frame reference
NVEncC: ffplayPreview
FFvHuff: interlaced support
Filtering: crop&resize stepping control
Vapoursynth: custom adds 'requiresColorMatrix'
Mux: subtitle passthrough title/name ffmpeg
MLRT: DPIRmlrt denoise/deblock
Filtering: added 'overwrite input frame rate' option
Video: DirectSynthLoading for NVEnc&QSVEnc
Vapoursynth: more general version of LimitMask through 'Merge'-option
Torch: vsGRLIR support
Vapoursynth: TFMBobQ/N


Note: Linux is no longer being supported in Hybrid.


-> downloads: http://www.selur.de/downloads

Cu Selur

[kedautinh12]

With VapoursynthR65_torch_2023.12.02-001.7z i got error when decompress
! C:\Users\ADMIN\Downloads\VapoursynthR65_torch_2023.12.02-001.7z: Unknown method in Vapoursynth\Lib\site-packages\pip\_vendor\distlib\t64-arm.exe
! C:\Users\ADMIN\Downloads\VapoursynthR65_torch_2023.12.02-001.7z: Unknown method in Vapoursynth\Lib\site-packages\pip\_vendor\distlib\w64-arm.exe
! C:\Users\ADMIN\Downloads\VapoursynthR65_torch_2023.12.02-001.7z: Unknown method in Vapoursynth\Lib\site-packages\setuptools\cli-arm64.exe
! C:\Users\ADMIN\Downloads\VapoursynthR65_torch_2023.12.02-001.7z: Unknown method in Vapoursynth\Lib\site-packages\setuptools\gui-arm64.exe

[Selur]

I don't know why this sometimes happens for some users (does not happen here and happened with previous versions for other users).
You can simply ignore those files.

[kedautinh12]

I think it's relate "arm" file

[Selur]

2023.12.20.1:
*fixed*

UI: multi-selection-comboboxes signal handling
Vapoursynth: fixed apply-range
UI: missing 'SpeedChange'-tab

*changed*

mlrt: supporting SAFA in vs-mlrt
mlrt: RIFE fraction support
mlrt: add tile_size to VSMLRT
Video: SVT-AV1 progress indication
Vapoursynth: added vsDestripe to ApplyOnlyTo, Only on region, Merge
Vapoursynth: revert to old vsDeSpot
UI: rewrote cropView


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

2023.12.21.1:
*fixed*

torch: SAFA mod restriction
Vapoursynth: AutoWhite no RGBH

*added*

Linux version is back


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

2023.12.22.1:
*fixed*

Vapoursynth: Always set Primaries&Transfer info
Vapoursynth: Mpeg2Source version problem
Vapoursynth: activationAttributnameFilter-ErrorPopUps


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

2024.01.12.1:
*fixed*

Vapoursynth: Letterbox not working
Synth: TIVTC frame rate/count calculation
Vapoursynth: sRestore default 25fps when frate not set
Vapoursynth: merge&masked filtering at the same time
Analyse: dts-hd detection
Torch: remove DDColor when not installed.
Audio: update audioQueue
Vapoursynth: frame props checks
Video: do not use file cache while during job processing svt-av1 progress
Jobs: delaycut on reencode needs output name update to avvoid confllicts
Vapoursynth: QTGMCFilter useEZDenoise UI

*changed*

Tags: 'und' language handling
Torch: RIFE v4.14 support
Audio: do not add 'downsampleTrueHDtoAC3' when input is an .mkv file
Container: mkv header compression handling</li>
Torch: support other network types than ESRGAN by name in VSGAN
Audio: add option to set 'default=true' in AudioQueue
Container: mp4 subtitle tagging; set empty title to empty string
Video: adjusted svtav1 defaults
Container: add warning for avi and interlace; stream vs container tag mismatch due to ffmpeg

*added*

Vapoursynth: vsTDecimateFilter
Automation: 'und' language handling


-> downloads: http://www.selur.de/downloads

Cu Selur

[weaker]

Hi Selur,

thank you for Hybrid!

Upon startup, Hybrid 2024.01.12.1 tells me "No card found that's supported by nvenc, the CUDA encoding library. -> Removing CUDA based encoding support. Card name was compared against: https://developer.nvidia.com/cuda-gpus"

However, the Titan V card I use is listed on the given nVidia-Website (with compute capabilities v7.0). On my PC the nvidia-smi tool also lists CUDA version 12.3 installed, driver 546.33 and I definitely used CUDA-acceleration (and nvenc) in the past.

Probably just a bug :)

Best,
weaker

[Selur]

2024.02.19.1:
*fixed*

Avisynth: YUY2 -> YUV422PX for high bit depth input
Vapoursynth: FillDrops ui fix
Vapoursynth: DotKill interlaced handling
Vapoursynth: fix crop&letterbox preview for 422 content
Video: x264 crash with 'always create stats file'
ui: svt-av1: wrong ui element spinbox vs combobox
synth: fixed bug in CustomSynthScriptLoader
Vapoursynth: QTGMC-DeNoise with FilterQueue
Vapoursynth: FilterQueue and YAHR and FineDeHalo
Vapoursynth: vsChromaNR Distance setting
Vapoursynth: TemporalDeGrain2 loading fftw (Linux)
Avisynth: (32bit) Spline100&Spline144

*changed*

synth: cropPreview handling
filtering: Deinterlacer none => force scan type overwrite to 'progressive'
ffmpeg: don't use zscale to set matrix
Vapoursynth: adjust, so that progressive 4:2:2 does allow uneven resolutions
Vapoursynth: added LevelsLumaRangeSignaling to vsSmooth and vsLimiter
mux: MP4Box can handle eac3 nowadays -> not forcing ffmpeg for muxing
linux: ~/.config/hybrid as folder-path to search for misc.ini
Vapoursynth: disabled anything related to KNLMeansCL and vs-placebo (i.e. GLSL) (Linux)

*added*

torch: CodeFormer supports background resize
mux: always set mkv ColorSignaling option
subtitle: 'cut by chapter/time'-option *experimental*
Video: svt-av1 custom addition


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: @weaker: I need to know the exact name Hybrid names in it's log file to add your card to the list. (Hybrid is probably expecting 'GeForce GTX TITAN V', which was how it was reported a while back,..)

[Selur]

2024.02.19.2:
*fixed*

Ui: double initialization of vsUseGLSLCAS

*changed*

Vapoursynth: enable KNLMeansCL related filters again (Linux)
Vapoursynth: search GLSL folder in filter path


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

2024.02.21.1:
*fixed*

ffmpeg: missing whitespace before '-color_primaries'
tool: Linux package using wrong fdkaac source


-> downloads: http://www.selur.de/downloads

Cu Selur

[locotus]

Using Hybrid last couple of versions:

Adding delay to audio or subtitles after conversion is over, Mkvtoolnix reported this problem and do not mux the file:

Color range parameters: not given in the form <TID>:n (argument was '0:18446744073709551615').

File seems to be Ok, can be muxed with other tools.

[Selur]

@locotus: without a debug output, it's hard to know who to reproduce, I fixed another bug with mkv muxing, which might also help with yout problem if not, please share proper details ( see: https://www.selur.de/support), so I can reproduce and fix the problem.

@all:
2024.02.22.1:
*fixed*

ffmpeg: removed unnecessary/wrong ticks around some color spaces in ffmpeg call
ffmpeg: wrong color space due to height vs width mixup on untagged content
Mux: mkv muxing on passthrough

-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

2024.02.23.1:
*fixed*

filter view: crop only got applied to filtered half not both views, which causes issues on resize and aspect ration changes
Vapoursynth: ignore 'Compare all' when 'Compare View' is enabled
ent


-> downloads: http://www.selur.de/downloads

Cu Selur

[locotus]

@locotus: without a debug output, it's hard to know who to reproduce, I fixed another bug with mkv muxing, which might also help with yout problem if not, please share proper details ( see: https://www.selur.de/support), so I can reproduce and fix the problem


After finishing remuxing the allready converted files, I noticed that the audio
conversion added 21ms to all audios even if I didn't set that value. Perhaps
this may be part of the mux problem in Mkvtoolnix.

Sorry, I don't have any other file to convert up to now.

[Selur]

After finishing remuxing the allready converted files, I noticed that the audio
conversion added 21ms to all audios even if I didn't set that value
there should be no audio conversion during remuxing,...
-> not enought details to understand or reproduce

[weaker]

Hi Selur,

sorry for the late reply. The card is reported as "NVIDIA TITAN V" in the Hybrid logfile. I think the card wasn't marketed as GTX at any time, as the Volta chip is basically the first RTX-like chip.

Best,
weaker

[Selur]

Send you a link to a dev version which should detect your card.

Cu Selur

[ReinerSchweinlin]

Hi Selur,

thank you for Hybrid!

Upon startup, Hybrid 2024.01.12.1 tells me "No card found that's supported by nvenc, the CUDA encoding library. -> Removing CUDA based encoding support. Card name was compared against: https://developer.nvidia.com/cuda-gpus"

However, the Titan V card I use is listed on the given nVidia-Website (with compute capabilities v7.0). On my PC the nvidia-smi tool also lists CUDA version 12.3 installed, driver 546.33 and I definitely used CUDA-acceleration (and nvenc) in the past.

Keep in mind, that Volta Video Encoding wasn´t very good back then - if it works for you - fine. But the turing and later encoders are much better. Or recent quicksync would be a cheaper option if you are in for HW-Encoding.

All other stuff about the Titan V is very nice :) Would love to get my hands on one (not for gaming, of course... compute stuff)

[Selur]

2024.03.03.1:
*fixed*

Avisynth: (32bit) SplineResize100/144 calls
Vapoursynth: custom section hinting accidentally required '# # ...' instead of '# ...'
Vapoursynth: allow lowering frame rate with rife
Vapoursynth: detect '2020 non .. ' as '2020cl'
Vapoursynth: crop in compare view
NoXSynth: fix subtitle embedding by changing local.conf handling
ui: subtitle controls enabled, even when no subititles are available

*added*

Subtitle: Support embedding subtitle when QSVEncC/NVEncC with 'only use encoder' is used.
Torch: added DeOldify through vs-deoldify

*changed*

Muxing/Extraction: Force FFmpeg for dts/dts-hd audio when dealing with mp4 containers.
Subtitle: Subtitle language passthrough handling.
Input: Support Gray8 input images.


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

2024.03.10.1:
*fixed*

FFmpeg: fixed color transfer signaling
Vapoursynth: GMFSS_fortuna model model index
Jobs: minimize command line glitch
Analysis: YUV jpeg handling

*added*

Vapoursynth: add custom CropView pad color

*changed*

Muxing/Extraction: Force FFmpeg for dts/dts-hd audio when dealing with mp4 containers.
Subtitle: Subtitle language passthrough handling.
Input: Support Gray8 input images.


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

2024.03.18.1:
*fixed*

UI: 'alwaysOnTop' caused Window to disappear
Vapoursynth: TFMBob fps calculation
Vapoursynth: CodeFormer in CompareView with 'undo upscale' disabled.

*changed*

Vapoursynth: updated RIFE models
Vapoursynth/Torch: update vs-deoldify


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

2024.03.26.1:
*fixed*

Vapoursynth: settings used for DeOldify
Vapoursynth: update to RIFE r9mod17
Vapoursynth: updated (mlrt)RIFE models

*changed*

Vapoursynth: settings used for DeOldify
Vapoursynth: update to RIFE r9mod17
Vapoursynth: updated (mlrt)RIFE models


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

2024.04.06.1:
*fixed*

Vapoursynth: VIVTC 'fix combed with deinterlacer'
Analysis: sox volume analysis
Audio: delaycut problems
Muxing: Crash with video pass-through
x264: disable 'adjust letterbox to crop' when Blu-ray/AVCHD restriction is enabled

*changed*

Vapoursynth: BasicVSR++ added 'wanted width'
Vapoursynth: adjusted to vsDeOldify v3.0.1
Video: x265 adjusted to v3.6


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: also updated torch-addon

[Selur]

2024.05.05.1:
*changed*

Subtitle: handling IETF BCP 47 languages
torch-addon: adjusted to vsDeOldify
addon: updated and mlrt
Vapoursynth: experimental script highlighting option
Vapoursynth: removed vsQTGMC order; always set to -1
Vapoursynth: SoftLight 10bit support
mlrt-addon: using 'engine_folder' to make sure .engine files are created in temp folder

*fixed*

Avisynth: TemporalDegrain2 postFFT handling
Vapoursynth: RGBH handling for CartoonEdges&Bloated-Edgemask
Crop: crop with multi-input
Muxing: mkv color signatling
Vapoursynth: use old RIFE version for FillDrops, FillDuplicates, ReplaceSingle due to: https://github.com/styler00dollar/VapourSynth-RIFE-ncnn-Vulkan/ issues/16
Vapoursynth: Merge not working with FilterQueue
Vapoursynth: TemporalDegrain2 extraSharp=True crash

*added*

mlrt-addon: SCUNet (mlrt)
Jobs: Vapoursynth chunking call
Avisynth: fixed TemporaDegrain(2) script/call
Vapoursynth: option for diff to increase difference visibility
torch-addon: VSGAN (filter only)

*removed*

torch-addon: VSGAN parameters


-> downloads: http://www.selur.de/downloads

Cu Selur

[FredThompson]

How does Hybrid convert the single setting for AAC bitrate into sample rate and depth? Does Hybrid have a preset sample rate?

[Selur]

Independent of the target audio format when reencoding.
If possible, bit depth is kept and sample rate is handled as set under Audio->Base->Audio Encoding Settings->Filtering.
By default Hybrid will resample to 48kHz, but you can enable 'Keep input sample rate' (if possible this will be kept).

[juanitogan]

Bug report for
Hybrid v2024.03.18.1:
Hybrid v2024.04.06.1:

Avisynth related:

Avisynth > Misc > Preview > [x] Show Frame number Positions: top right
This adds frame numbers to both preview and encoding (unlike what the hover help says).

Avisynth Preview window always closes and reopens when using the Refresh Preview button, losing frame number, window position and size.
Turnaround can be near a minute, depending on filters in use. The main Preview button does the same thing.
In case it helps, the log is filled with these messages from each use of the Preview:
> Couldn't find 'DePulse64' in the filter order, adding 'KillPulse32' at the end.
> using 32bit AvisynthViewer: C:\Program Files\Hybrid\32bit\avsViewer.exe

Avisynth Preview window does not remember window position and size.

Avisynth Preview window does not remember setting: [x] Scrolling. (Which can take several seconds to enable depending on filters.)

AutoAdjust > Gamma mode > hover shows TV output help, not Gamma mode help
AutoAdjust > Savety (misspelled Safety)

AutoAdjust (and similar auto tools) should probably go much further down the filter order by default. I can't imagine ever wanting to auto level before removing noise and artifacts.

Avisynth > DeNoise > QTGMC > enabling this breaks the ability to preview and encode (frame 0 only, stops responding, etc).

Avisynth > DeNoise > I can't get several filters to appear in the script: FFT3DFilter, FFT3DGPU, KNLMeansCL; but turning on mClean loads 8 dlls including FFT stuff.


Vapoursynth related:

(De-)Interlace/Telecine > VIVTC (Vapoursynth) > Misc > Fix combed with deinterlacer > Deinterlacer: QTGMC(preset="fast") > Preview
> Error: VsFixCombed - Unknown Deinterlacer: QTGMC(preset="fast")
Actually, none of the Deinterlace settings works here; not just QTGMC fast.

(De-)Interlace/Telecine > VIVTC (Vapoursynth) > VFM
Errors only with telecined source, I believe:
> File "C:\Program Files\Hybrid\64bit\vsscripts\havsfunc.py", line 2271, in QTGMC_Interpolate
> eedi3 = partial(core.eedi3m.EEDI3CL, field=field, planes=planes, mdis=EdiMaxD, device=device, **eedi3_args)
> ^^^^^^^^^^^^^^^^^^^
> File "src\\cython\\vapoursynth.pyx", line 2691, in vapoursynth.Plugin.__getattr__
> AttributeError: There is no function named EEDI3CL
Changing Deinterlacer to something besides one of the QTGMC options works. Same issue with TIVTC

(De-)Interlace/Telecine > VIVTC (Vapoursynth) > VDecimate
The Block (X/Y) fields have no effect. The script always shows a value of 16:
> clip = core.vivtc.VDecimate(clip=clip, blockx=16, blocky=16)# new fps: 20.2336

Vapoursynth > Misc > Filter Order/Queue
Turn on QTGMC but mClean gets bolded instead.

(De-)Interlace/Telecine > VIVTC (Vapoursynth) > VFM
Errors only with telecined source, I believe:
> File "C:\Program Files\Hybrid\64bit\vsscripts\havsfunc.py", line 2271, in QTGMC_Interpolate
> eedi3 = partial(core.eedi3m.EEDI3CL, field=field, planes=planes, mdis=EdiMaxD, device=device, **eedi3_args)
> ^^^^^^^^^^^^^^^^^^^
> File "src\\cython\\vapoursynth.pyx", line 2691, in vapoursynth.Plugin.__getattr__
> AttributeError: There is no function named EEDI3CL
Changing Deinterlacer to something besides one of the QTGMC options works. Same issue with TIVTC
[[ UPDATE: Still seeing this after loading a progressive and turning on QTGMC in the VS filters tab (after doing a telecine job).
In the deint tab, Auto-deint is set to None and [x] Overwrite... is greyed out (both checkbox and dropdown) and set to "telecine".
Fiddle with the deint settings until it shows [x] Overwrite... greyed out and set to "progressive" and it works again; very slow now.
And why is Auto-deint changing to match each new title load? I thought it usually just stayed at GTGMC. ]]
--> Restarting Hybrid now has Auto-deint back to QTGMC, even on progressive material, and [_] Overwrite... enabled but the dropdown disabled and set to "progressive".
This looks more normal and turning on the QTGMC filter in the VS tab now works fine.


Other:

Filter Config Save saves/sets individual Deinterlace/Telecine settings... but not which Auto Deint Handling was selected.
Very weird contradiction. Don't know how to save this per job... and would prefer it saved/set with Filter Config.

More of a feature add, I'm guessing: VIVTC/TIVTC/etc input and output frame counts are not correct for VFR sources (and could add a hint to manually set Change Speed [and/or fix that too]). Don't see a VFRtoCFR filter (or `ffmpeg -vsynch cfr` equivalent). Somehow this works without a CFR filter; not sure how. Oh, wait, I see AssumeFPS() in the script... but, no, that's not it since it contains the average VFR value. Anyway, I'm running into some confusion here that may or may not need work. Decimating telecined and interlaced VFR material _usually_ works well when I get the settings right... but it feels a bit like magic.

FDKAAC progress bar/text does not become visible until more than half done. This 20-second delay makes me poke around to make sure the job is running.

Hybrid writes APPDATA\hybrid\fontconfig\... folders in my working folder.

Audio cut short during detelecine (from a source flagged as telecine -- not an issue when decimating a source flagged as TFF). When setting audio to "custom" and encoding AC-3 to AAC FDK (no other codec tested), it cuts the audio short as if the length was changed by the number of frames the detelecine drops -- that is, it cuts off the last 1/5 of the audio track (presumably because it calcs the frames at 30fps still instead of 24fps). Tested "auto add (all)" and results were the same (but file size was 3kb [out of 50mb] larger somehow). Tested "auto add (all)" with multiple input files and IT WORKS!... very weird workaround, if you can call it that (I stumbled upon this by accident). I checked the temp *.aac file and it is also short, so not a muxing problem.

The completion bar for the video during detelecine (including decimating while deinterlacing) hits 100% 4/5 of they way through and keeps going to 125%. Not sure if this is intentional or not. I can't see the audio bar due to the other issue that hides it for the first 20 seconds or so, at which point it is already done with this material.

1920x1080 PAR 0.75 stretches Preview wrong way to 2560x1080 instead of 1440x1080 (and, yes, it was a weird to find such a source). I didn't try encoding it to see what happens. Instead, I forced a resize to 1440x1080 with 1:1 pixels before encode.

Bookmarks are lost whenever I reset the VS Preview. I would prefer it renamed the last tempPreviewVapoursynthFile*.vpy.bookmarks file to match the new tempPreviewVapoursynthFile*.vpy on refresh to keep the bookmarks.

Also, I saw you saw the report I made on DeScratch's memory leak and the new version released. Just a reminder here to put it in (it needs a new version of VS [API 4.1 minimum] and Python, etc.). Update: I see you talked him into rebuilding for API 4.0. Awesome!


I like the new Artifacts tab!


Suggestions:

- Add a field to the VS preview that shows the current frame's time.

- Add similar TV/PC presets to Vapoursynth > Color > SmoothLevels that I see on AutoAdjust.
That, or at least remind us in the hover help to set Input or Output to 16/235 for TV.

- Put a yes/no warning message box on Defaults > Reset All.
I've made the mistake more than once now (when I get tired, I guess) thinking it will reset to my last save. Ack!
Maybe add another option to reset to last save (same as a Restart, I suppose).

- Under custom, add an optional interface for setting bits and/or colorspace (that adds the appropriate line), so Hybrid can track it and doesn't have to do unnecessary conversions after it. Yeah, that's a big one, I'm sure.

- Don't add a Start chapter if there is a chapter less than 1 second from start. MakeMKV rarely (ever?) creates a chapter at 0.000, it seems. (Could maybe even reduce that to less than 0.1 seconds, if you want, from what I've seen so far from MakeMKV.)


Thanks again for this excellent tool!

Matt

[Selur]

Avisynth related:
Avisynth > Misc > Preview > [x] Show Frame number Positions: top right
This adds frame numbers to both preview and encoding (unlike what the hover help says).
hopefully fixed now

Avisynth Preview window always closes and reopens when using the Refresh Preview button, losing frame number, window position and size.
Can't reproduce here.
Turnaround can be near a minute, depending on filters in use.
Yes, depending on the filters preview is slow.
The main Preview button does the same thing.
yes, that is as expected.

In case it helps, the log is filled with these messages from each use of the Preview:
> Couldn't find 'DePulse64' in the filter order, adding 'KillPulse32' at the end.
> using 32bit AvisynthViewer: C:\Program Files\Hybrid\32bit\avsViewer.exe

My guess is that you are running out of RAM that Avisynth can use.
Try lowering the thread count and if that does not help switch to 64bit.

The 'DePulse64' and 'KillPulse32' info should not cause problems.
This probably happens since you haven't: properly uninstalled Hybrid with settings, migrated your settings, reset your defaults when it was mentioned in the changelog


Avisynth Preview window does not remember window position and size.
Avisynth Preview window does not remember setting: [x] Scrolling. (Which can take several seconds to enable depending on filters.)[/quote]
Correct. It never did. Patches to avsViewer (https://github.com/Selur/avsViewer) are welcome.

AutoAdjust > Gamma mode > hover shows TV output help, not Gamma mode help
hopefully fixed now, also needs to be renamed, it's not 'Gamma mode' but 'Gain mode' :)

AutoAdjust > Savety (misspelled Safety)
hopefully fixed now

AutoAdjust (and similar auto tools) should probably go much further down the filter order by default. I can't imagine ever wanting to auto level before removing noise and artifacts.
I'm not planning to change the default order.

Avisynth > DeNoise > QTGMC > enabling this breaks the ability to preview and encode (frame 0 only, stops responding, etc).
Works fine here, I suspect you are running out of RAM

Avisynth > DeNoise > I can't get several filters to appear in the script: FFT3DFilter, FFT3DGPU, KNLMeansCL; but turning on mClean loads 8 dlls including FFT stuff.
hopefully fixed now

Vapoursynth related:
(De-)Interlace/Telecine > VIVTC (Vapoursynth) > Misc > Fix combed with deinterlacer > Deinterlacer: QTGMC(preset="fast") > Preview
> Error: VsFixCombed - Unknown Deinterlacer: QTGMC(preset="fast")
Actually, none of the Deinterlace settings works here; not just QTGMC fast.
Works fine here, fixed it a while ago.

(De-)Interlace/Telecine > VIVTC (Vapoursynth) > VFM
Errors only with telecined source, I believe:
> File "C:\Program Files\Hybrid\64bit\vsscripts\havsfunc.py", line 2271, in QTGMC_Interpolate
> eedi3 = partial(core.eedi3m.EEDI3CL, field=field, planes=planes, mdis=EdiMaxD, device=device, **eedi3_args)
> ^^^^^^^^^^^^^^^^^^^
> File "src\\cython\\vapoursynth.pyx", line 2691, in vapoursynth.Plugin.__getattr__
> AttributeError: There is no function named EEDI3CL
Changing Deinterlacer to something besides one of the QTGMC options works. Same issue with TIVTC
Yeah, that's a error I know about, log should probably show something like:
Not loading, 'EEDI3m_opencl' for vsFixCombed since 'EEDI3m' is already loaded.
Wrote a workaround, but not sure it catches all cases. (this will leat to that whatever gets first loaded EEDI3m or EEDI3CL get always used)

(De-)Interlace/Telecine > VIVTC (Vapoursynth) > VDecimate
The Block (X/Y) fields have no effect. The script always shows a value of 16:
> clip = core.vivtc.VDecimate(clip=clip, blockx=16, blocky=16)# new fps: 20.2336
hopefully fixed now

Vapoursynth > Misc > Filter Order/Queue
Turn on QTGMC but mClean gets bolded instead.
Can't reproduce here. You might need to reset the 'Filter Order' in your defauts/configs.

(De-)Interlace/Telecine > VIVTC (Vapoursynth) > VFM
Errors only with telecined source, I believe:
> File "C:\Program Files\Hybrid\64bit\vsscripts\havsfunc.py", line 2271, in QTGMC_Interpolate
> eedi3 = partial(core.eedi3m.EEDI3CL, field=field, planes=planes, mdis=EdiMaxD, device=device, **eedi3_args)
> ^^^^^^^^^^^^^^^^^^^
> File "src\\cython\\vapoursynth.pyx", line 2691, in vapoursynth.Plugin.__getattr__
> AttributeError: There is no function named EEDI3CL
Changing Deinterlacer to something besides one of the QTGMC options works. Same issue with TIVTC

hopefully fixed now, same as the EEDI3CL problem above

In the deint tab, Auto-deint is set to None and [x] Overwrite... is greyed out (both checkbox and dropdown) and set to "telecine".
Fiddle with the deint settings until it shows [x] Overwrite... greyed out and set to "progressive" and it works again; very slow now.

hopefully fixed now, Overwrite should be enabled and set to 'progressive' is 'none' is selected.
If you do not want Hybrid to deinterlaced interlaced content, configure the encoder to output interlaced content.

And why is Auto-deint changing to match each new title load? I thought it usually just stayed at GTGMC. ]]
--> Restarting Hybrid now has Auto-deint back to QTGMC, even on progressive material, and [_] Overwrite... enabled but the dropdown disabled and set to "progressive".
This looks more normal and turning on the QTGMC filter in the VS tab now works fine.
Hybrid will not apply a deinterlacer if the scan type is not progressive and the encoder ist set to output interlaced content.

Other:
Filter Config Save saves/sets individual Deinterlace/Telecine settings... but not which Auto Deint Handling was selected.
Very weird contradiction. Don't know how to save this per job... and would prefer it saved/set with Filter Config.
Yes, this is as intended, depending on your source the auto deinterlace handling can change.

More of a feature add, I'm guessing: VIVTC/TIVTC/etc input and output frame counts are not correct for VFR sources (and could add a hint to manually set Change Speed [and/or fix that too]). Don't see a VFRtoCFR filter (or `ffmpeg -vsynch cfr` equivalent). Somehow this works without a CFR filter; not sure how. Oh, wait, I see AssumeFPS() in the script... but, no, that's not it since it contains the average VFR value. Anyway, I'm running into some confusion here that may or may not need work. Decimating telecined and interlaced VFR material _usually_ works well when I get the settings right... but it feels a bit like magic.
vfr content gets decoded as cfr when deinterlacing or frame reduction or interpolation gets applied
all too vage. no details -> nothing to look at

FDKAAC progress bar/text does not become visible until more than half done. This 20-second delay makes me poke around to make sure the job is running.
do vague

Hybrid writes APPDATA\hybrid\fontconfig\... folders in my working folder.
can't reproduce, need details

Audio cut short during detelecine (from a source flagged as telecine -- not an issue when decimating a source flagged as TFF). When setting audio to "custom" and encoding AC-3 to AAC FDK (no other codec tested), it cuts the audio short as if the length was changed by the number of frames the detelecine drops -- that is, it cuts off the last 1/5 of the audio track (presumably because it calcs the frames at 30fps still instead of 24fps). Tested "auto add (all)" and results were the same (but file size was 3kb [out of 50mb] larger somehow). Tested "auto add (all)" with multiple input files and IT WORKS!... very weird workaround, if you can call it that (I stumbled upon this by accident). I checked the temp *.aac file and it is also short, so not a muxing problem.
I can ivtc content without sync problems.
Share proper details. (read: https://www.selur.de/support)

The completion bar for the video during detelecine (including decimating while deinterlacing) hits 100% 4/5 of they way through and keeps going to 125%. Not sure if this is intentional or not. I can't see the audio bar due to the other issue that hides it for the first 20 seconds or so, at which point it is already done with this material.
Share proper details. (read: https://www.selur.de/support)

1920x1080 PAR 0.75 stretches Preview wrong way to 2560x1080 instead of 1440x1080 (and, yes, it was a weird to find such a source). I didn't try encoding it to see what happens. Instead, I forced a resize to 1440x1080 with 1:1 pixels before encode.
Share proper details and a source sample. (read: https://www.selur.de/support)

Bookmarks are lost whenever I reset the VS Preview. I would prefer it renamed the last tempPreviewVapoursynthFile*.vpy.bookmarks file to match the new tempPreviewVapoursynthFile*.vpy on refresh to keep the bookmarks.
patches for vsViewer (https://github.com/Selur/vsViewer) are welcome (I have never used the bookmarks in vsViewer, no clue what those are for)


Cu Selur

Ps.: send you a link to a dev version (via pm) for testing, link will be available ~24hrs.

[juanitogan]

My guess is that you are running out of RAM that Avisynth can use.
Try lowering the thread count and if that does not help switch to 64bit.

The 'DePulse64' and 'KillPulse32' info should not cause problems.
This probably happens since you haven't: properly uninstalled Hybrid with settings, migrated your settings, reset your defaults when it was mentioned in the changelog


The first version of Hybrid I used was 2023.03.17.1 and I do uninstall between versions. I do not, however, uninstall the settings between versions. I read the changelog for all versions between my last and latest, but I guess I missed that somewhere -- that and resetting defaults. I might try all that when I get a chance, just to see what happens. The Avisynth interface is now a low priority since I now have AutoAdjust working in VS scripts. Also, I generally run Hybrid with 5-7gb of free memory, so I'm doubtful about RAM issues, but will double check that (and threads) regardless.


Audio cut short during detelecine (from a source flagged as telecine -- not an issue when decimating a source flagged as TFF). When setting audio to "custom" and encoding AC-3 to AAC FDK (no other codec tested), it cuts the audio short as if the length was changed by the number of frames the detelecine drops -- that is, it cuts off the last 1/5 of the audio track (presumably because it calcs the frames at 30fps still instead of 24fps). Tested "auto add (all)" and results were the same (but file size was 3kb [out of 50mb] larger somehow). Tested "auto add (all)" with multiple input files and IT WORKS!... very weird workaround, if you can call it that (I stumbled upon this by accident). I checked the temp *.aac file and it is also short, so not a muxing problem.
I can ivtc content without sync problems.
Share proper details. (read: https://www.selur.de/support)


It's not a sync problem but a truncation problem. I'll do a proper debug on it and get back. It should be an obvious difference in the aac encode command line from one job versus another. I don't run into bonafide, non-interlaced, telecine DVD sources much, and the workaround to encode video separate from audio is obvious, but I'll try to get to this soon.

[Selur]

Also, I generally run Hybrid with 5-7gb of free memory,...
Avisynth 32bit is still limited to somewhat 3.5-4GB of RAM, so with a complex enough script and threads even with SD content you can hit that limit.

It's not a sync problem but a truncation problem.
Okay, I will need a debug output and probably a sample which allows to reproduce the problem.

Cu Selur

[Selur]

2024.05.11.1:
*changed*

Vapoursynth: adjusted to DeOldify 0.3.5.2

*fixed*

Vapoursynth: fixed RIFE + FillDrops, ReplaceSingle, FillDuplicates, ReplaceBlackFrames
Vapoursynth: EEDI3/EEDI3CL handling
Avisynth: Denoise filter not applied
Avisynth: some typos in gui
Avisynth: FrameNumber should not be in encode
ProRes: progressive to interlaced encoding when no synth is us

-> downloads: http://www.selur.de/downloads

Cu Selur

[NoX1911]

I have Hybrid/20240511 and Vapoursynth installed (ArchLinux) but its not detected in 'Config/Tools/Tool Versions'. 'vspipe' is in $path. I cannot set it manually in 'Config/Tools' (there is no 'vspipe' slot). What has to be done here?

[Selur]

If you start Hybrid from a terminal, what is the output?
Is Vapoursynth support enabled?
Are vsPluginsPath,vsPluginsPath, vsScriptPath set inside a misc.ini ? (misc.ini can be a. next to Hybrid, b. inside ~/.hybrid c. inside .~/./config/hybrid)
Can vspipe be found via 'which vspipe' ?`

Cu Selur

[NoX1911]

Looks like the plugins folder doesn't get picked up so i've put it into ~/.hybrid/misc.ini.

[General]
vsPluginsPath=/usr/lib/vapoursynth
vsScriptPath=/home/user/ffmpeg/vapoursynth

After that, vapoursynth is enabled but doesn't run all too well (job crashes). Loading the script in vsedit shows:

2024-06-08 20:50:29.241
Failed to evaluate the script:
Python exception: module 'validate' has no attribute 'transferIsInvalid'

Traceback (most recent call last):
File "src/cython/vapoursynth.pyx", line 3365, in vapoursynth._vpy_evaluate
File "src/cython/vapoursynth.pyx", line 3366, in vapoursynth._vpy_evaluate
File "/tmp/encodingTempSynthSkript_2024-06-08@20_44_40_4310_0.vpy", line 20, in
if validate.transferIsInvalid(clip):
^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: module 'validate' has no attribute 'transferIsInvalid'

Python version is 3.12.3.

The script is:

# Imports
import vapoursynth as vs
# getting Vapoursynth core
import sys
import os
core = vs.core
# Import scripts folder
scriptPath = '/home/user/ffmpeg/vapoursynth'
sys.path.insert(0, os.path.abspath(scriptPath))
# Import scripts
import validate
# Source: '/tmp/video.mp4'
# Current color space: YUV420P8, bit depth: 8, resolution: 480x640, frame rate: 29.848fps, scanorder: progressive, yuv luminance scale: limited, matrix: 470bg
# Loading /tmp/video.mp4 using LibavSMASHSource
clip = core.lsmas.LibavSMASHSource(source="/tmp/video.mp4")
frame = clip.get_frame(0)
# Setting detected color matrix (470bg).
clip = core.std.SetFrameProps(clip=clip, _Matrix=1)
# setting color transfer (709), if it is not set.
if validate.transferIsInvalid(clip):
clip = core.std.SetFrameProps(clip=clip, _Transfer=1)
# setting color primaries info (to 709), if it is not set.
if validate.primariesIsInvalid(clip):
clip = core.std.SetFrameProps(clip=clip, _Primaries=1)
# setting color range to TV (limited) range.
clip = core.std.SetFrameProps(clip=clip, _ColorRange=1)
# making sure frame rate is set to 29.848fps
clip = core.std.AssumeFPS(clip=clip, fpsnum=3731, fpsden=125)
# making sure the detected scan type is set (detected: progressive)
clip = core.std.SetFrameProps(clip=clip, _FieldBased=0) # progressive
# adjusting color space from YUV420P8 to RGB24 for vsLevels
clip = core.resize.Bicubic(clip=clip, format=vs.RGB24, matrix_in_s="470bg", range_s="limited")
# adjusting color using Levels on RGB24 (8 bit)
clip = core.std.Levels(clip=clip, min_in=16, max_in=235, min_out=16, max_out=100)
# adjusting output color from: RGB24 to YUV420P8 for x264Model
clip = core.resize.Bicubic(clip=clip, format=vs.YUV420P8, matrix_s="470bg", range_s="limited")
# set output frame rate to 29.848fps (progressive)
clip = core.std.AssumeFPS(clip=clip, fpsnum=3731, fpsden=125)
# output
clip.set_output()

My plugins are:

$ ls -1 /usr/lib/vapoursynth
pkgconfig
bestsource.so
libbestsource.so
libf3kdb.so
libffms2.so
libvslsmashsource.so

I guess this is not hybrid ground anymore and i should better take this to Vapoursynth thread instead.

[Selur]

Not answering my questions, isn't really helping. :(
On Linux Hybrid does not use the pluginsPath in the script, since Python and Vapoursynth should be installed systemwide.
Python exception: module 'validate' has no attribute 'transferIsInvalid'

scriptPath = '/home/user/ffmpeg/vapoursynth'
sys.path.insert(0, os.path.abspath(scriptPath))
should add '/home/user/ffmpeg/vapoursynth' to the system path used to look for scripts.
Assuming that path includes https://github.com/Selur/VapoursynthScriptsInHybrid/blob/master/validate.py Vapoursynth should find 'transferIsInvalid' (https://github.com/Selur/VapoursynthScriptsInHybrid/blob/bbe79881f9e1202bb518687678de6e435421860f/validate.py#L11C5-L11C22).
Since Vapoursynth does not fail directly on 'import validate' the validate.py should be present.
=> One reason for the error could be that you got a different file named validate.py in that folder.

Cu Selur

[Selur]

2024.06.09.1:
*aded*

Vapoursynth: torch, ProPainter

*changed*

Vapoursynth: torch, adjusted to DeOldify 0.3.5.3
Filtering: remove 'none' as auto deinterlacer handling option
Video: NVEnc support --tflevel --lookahead-level

*fixed*

Vapoursynth: another EEDI3/EEDI3CL handling
Vapoursynth: Masking for RGBH filters
QSVEnc: remove '--tier' during hevc encoding
QSVEnc: set '--output-csp' (this is broken for most choices)
Jobs: QSVEnc av1 encoding in webm
Vapoursynth: torch, DeOldify fix FP16
Vapoursynth: TFM/VFM, fps refresh
Avisynth: custom with dual view
Avisynth: interlaced output + cropView
x265: 3.0 and 4.0 vbv values not set

-> downloads: http://www.selur.de/downloads

Cu Selur

[NoX1911]

Not answering my questions, isn't really helping. :(
Sorry, here it is: hybrid-stdout.txt (https://0x0.st/Xbt9.txt/hybrid-stdout.txt)
That's from a pure AUR version (https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=hybrid-encoder) that pulls the minimal version only (i didn't know that at that time). After realizing it needs more stuff (like the scripts) from the base/deps image (https://www.selur.de/downloads) it works properly now.
'vsPluginsPath' has nevertheless to be defined though. Hybrid doesn't look up /usr/lib/vapoursynth (https://gitlab.archlinux.org/archlinux/packaging/packages/vapoursynth-plugin-bestsource/-/blob/main/PKGBUILD?ref_type=heads#L50) by itself and keeps VS disabled otherwise.

[Selur]

'vsPluginsPath' has nevertheless to be defined though.
Correct.

After realizing it needs more stuff (like the scripts) from the base/deps image it works properly now.
side note: the scripts Hybrid uses are present in hybrid\vsscripts
Sadly I have no clue how to make a full portable version with all the Python, scripts, filters included.
iirc, if you don't set vsScriptPath, Hybrid should find the scripts in (hybrid\vsscripts). :)
Hybrid also does try to find vsPluginsPath, but it does not look for it under ' /usr/lib/vapoursynth' but '/usr/local/lib/vapoursynth'. => all add ' /usr/lib/vapoursynth' to the default search path for the next release.
'./64bit/vsfilters' doesn't exist
'./vsfilters' doesn't exist
'/home/user/.hybrid/vsfilters' doesn't exist
'/home/user/opt/vapoursynth/vsplugins' doesn't exist
'/home/user/opt/vapoursynth/lib/vapoursynth/' doesn't exist
'/usr/local/lib/vapoursynth' doesn't exist
'/home/user/.hybrid/vsscripts' doesn't exist,..
'./vsscripts' doesn't exist,..
'/usr/share/vsscripts' doesn't exist,..
'/home/user/opt/vapoursynth/vsscripts' doesn't exist,..


Cu Selur

[Selur]

2024.06.15.1:
*added*

Vapoursynth: support CPreview from hello_hello

*changed*

Vapoursynth: torch, adjust to DeOldify 4.0.0
Vapoursynth: torch, adjust to vsRealESRGAN 5.0.0
Vapoursynth: reworked FillDuplicateFrames, FillDrops, ReplaceSingle into one class for stability
Vapoursynth: vs-mlrt, RIFE scale support optional requires 'onnx' in Vapoursynth/Lib/site-packages, see: https://github.com/AmusementClub/vs-mlrt/issues/94

*fixed*

Vapoursynth: do not change color matrix on RGB content
Vapoursynth: fixed some unnecessary, popups during startup
Video: NVEnc support --tflevel --lookahead-level

-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: deinstall old version before installing this one.

[Selur]

2024.06.16.1:
*changed*

Video: NVEnc, support '--vpn-nlmeans'

*fixed*

Vapoursynth: initialization error on Windows

-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: This is a WIndows only release.

[Selur]

@Linux users:
Vapoursynth plugins build script is broken atm., got no time to look into it.
=> no new Hybrid Linux release until I find time to look into it (probably end of the year)

Cu Selur

Ps.: patches are welcomed https://github.com/Selur/hybrid-vapoursynth-addon/

[Selur]

*2024.09.29.1:
*changed*

Vapoursynth: SoftLight require PC scale when mode != 8
Vapoursynth: updated RIFE, RIFE(mlrt)
Vapoursynth: use constants for Matrix, Primaries, Chroma, Luma
Video: NVEnc/QSVEnc support --vpy-mt
Jobs: add abort-option to size- and audio-warning onjo create&start
Jobs: added scripts to report file
Video: removed NVEnc input type checks
Filtering: hide deinterlace control on progressive input, since it confuses users
Avisynth: FFmpeg2Source use YUV422PX instead of YUY2
Avisynth: increased suported memory max to 65536MB
Avisynth: Hybrid now comes without Avisynth 32bit support, it is an addon now
Video: do not check for UT Video, assume its present
Vapoursynth: prefer BestSource as default for file input
Vapoursynth: torch - added 'Requires'-controls to VSGAN and VSGAN Filter
Input: Switch from Vapoursynth to Avisynth (when available) on .avs input
Vapoursynth: TemporalDegrain2: degrainTR default and crash with degrainTR=0
Video: x264, disable and deactivate 'auto' for color space on profile 'None'
Video: x264, x264 - added 'adjust gop max to output fps'
Ui: changed maximize handling
Ui: option to set debug output path
Vapoursynth: add masking for frame matching
Vapoursynth: added FTF to TIVTC
FFmpeg: don't set color signaling for pipes
Avisynth: added 'Never use AviSource'

*fixed*

Avisynth: enable avsToon for 32bit
Video: x264 minimizeCL with 'superfast'
Vapoursynth: FileQueue and Masking issues
Vapoursynth: ReplaceSince
Vapoursynth: torch - remove unnecessary resize call after VSGAN
Extraction: use ffmpeg for ProRes from mkv extraction
Video: SVT-AV1 typo 'enableResotration' vs 'enableRestoration'
Vapoursynth: SMDegrain smdegrain.py script interlaced handling fixed
Vapoursynth: mlrt - DPIR FP16 inverted
Video: NVEnc - max bitrate initial value
Video: NVEnc - EdgeLevel not added to command line
Input: Zip format handling (ffmpeg and analysis)
Synh: fix typo in CustomSynthScript QQDFPlus and abcxyz YUVALLX -> ALLYUVX
Vapoursynth: QTGMC in CompareView with 'custom', use Preset='Fast' for original side.
Vapoursynth: torch, VSGAN (Filter), width/height mixup
Vapoursynth: torch, DeOldify 'render_vivid' and 'dark' passed as string instead of bool
Vapoursynth: SmoothGrad fix min value 1
Vapoursynth: Frame->Resize UI scaling
Vapoursynth: torch, ProPainter 'Inference'-CheckBox not working
Vapoursynth: cfhd handling
Vapoursynth: SSIMD: fix output color
Vapoursynth: letterboxing YUV422 mod2
Vapoursynth: vsRemoveDirt defaults
Video: FFmpegs zscale-call for PNG output
Video: x264 slow preset using b-adapt 2 instead of 1
FFmpeg: 2020 signaling
Extract: vp9 video
Model: loadfromJson respect skipList
Avisynth: TIVTC+QTGMC
Avisynth: fix histogram color restriction check, no YUY2 support
Avisynth: QTGMC+Custom+UltraFast missing YadifMod2 dependency
Avisynth: DeblockPP7 and aSharp were unnecessarily disabled for 64bit Avisynth
Video: ProRes encoding when mencoder is used as decoder
Vapoursynth: DecimateMode set output fps for mode 2, which is (still) broken (https://github.com/dubhater/vapoursynth-tivtc/issues/4)
Vapoursynth: RainbowSmooth require mod16
Vapoursynth: don't try to set ColorMatrix on RGB content
Vapoursynth: DGSharpen requires mod16
Vapoursynth: rectangular mask for high bit depth
Decoder: changed Avisynth input handling
Muxing: mkv helper for ffmpeg color matrix and vfr
Jobs: fix cleanup job handling on raw audio only output
Jobs: fix warning handling

*added*

Vapoursynth: GLSL Gamma, RGBAdjust filter ports(Levels broken&disabled)
Vapoursynth: proToon
Vapoursynth: vs_temporalfix (https://github.com/pifroggi/vs_temporalfix)
Vapoursynth: torch/vmlrt option to set (trt) 'engine path'
Avisynth: added CustomSynthScripts for 32bit and 64bit proToon-v0.75.1
Generate output name suffix added '%DATE%','%TIME%','%TIMEMS%' replacements
Vapoursynth: torch - add vs_colorfix (https://github.com/pifroggi/vs_colorfix) for VSGAN and VSGANFilter

-> downloads: http://www.selur.de/downloads

Cu Selur

[Leo 69]

Just installed the last version of Hybrid, loaded a movie (regular H.264 video in .mkv container) for re-encoding, then navigated to "Filtering" section and without enabling any Vapoursynth filters, pressed the "Preview" button ========> Nothing happened, and no preview window appeared. When a couple of filters are enabled, the preview window doesn't appear either.

Is this a designed behavior? How to fix it and make the preview window appear so that I could actually see the effect of applied filters? I've attached the debug log.

https://drive.filen.io/d/228e5b58-7aff-4fee-9023-67a43dfa9171#HBNawAYfeBTOuG2joQAeq7LGIrMkRXjS

[Selur]

Probably a problem with BestSource, disabling "Filtering->Vapoursynth->Misc->Source->Prefer BestSource" or replacing the Hybrid\64bit\vsfilters\SourceFilter\BestSource\BestSource.dll with R6 from https://github.com/vapoursynth/bestsource/releases should help.
Would need a sample of your source to say for sure where the problem is.

Cu Selur

[Leo 69]

Selur,

Thank you very much. Eventually I found that ffms2 is working better for my use case and I can see the preview window now.

Could I ask you some questions?

How to make this work in Hybrid? => https://github.com/JingyunLiang/RVRT

Is converting from .pth to .onnx with Chainner sufficient to make this work? How to import these models into Hybrid? Are there models similar but better that these?

And also does Hybrid support importing .webp or .avif image sequences as input?

Also, I have say that Hybrid is the most complex and comprehensive encoding tool I've used. The amount of mental effort that was put into it by just one person is incredible.

[Selur]

How to make this work in Hybrid?
No clue how to make RVRT work in Hybrid, never tried it since iirc. BasicVSR++ produce better results.
Is converting from .pth to .onnx with Chainner sufficient to make this work?
You can try converting the pretrained models to .onnx try whether VSLMLRT can load them.
How to import these models into Hybrid?
Filtering->Vapoursynth->Resizer->VSMLRT has a file open dialog which allows selecting any .onnx-models.
Note that neither does VSGAN support all .pth files nor does VSMLRT support all .onnx files.
Are there models similar but better that these?
Can't say for sure since I never used RVRT and thus can't compare it to anything I know.

And also does Hybrid support importing .webp or .avif image sequences as input?
No, atm. only png, tga, tiff, tif, exr and jpg are supported. I haven't even tested whether webp or avif are supported by Avisynth, Vapoursynth and/or FFmpeg, so no clue whether it could be done.

Cu Selur

[Leo 69]

Hi Selur!

Do you have plans to implement the BM3D denoiser for Vapoursynth? Or maybe even BM3DCUDA (https://github.com/WolframRhodium/VapourSynth-BM3DCUDA)?

[Selur]

Short: No, got no plans to support either.

Long:
I thought about using it, thru the wrapper for it in muvsfunc (https://github.com/HomeOfVapourSynthEvolution/mvsfunc/blob/865c7486ca860d323754ec4774bc4cca540a7076/mvsfunc/mvsfunc.py#L673), but even that wrapper seems rather complex with tons of parameters where I don't know the range of allowed values and that use 'None' as default. BM3DCUDA refers to the BM3D documentation.
So it's not planned, since I have no clue how to use it (properly), what the parameters are meant to do and what the allowed values etc. are, what dependencies there are etc.

If you know how to use it, you could:
a. add it to the gui yourself using a Custom filter (https://forum.selur.net/thread-3231.html) addition.
b. use it through the custom section
Judging by the examples, this seems like something that might be best used in a custom section by users themselves.

=> Don't know enough about it to implement support for it.

Cu Selur

[Z2697]

Animated WebP is somewhat tricky, FFmpeg doesn't support it, dwebp can't directly decode it
>dwebp asd.webp
Error! Decoding of an animated WebP file is not supported.
Use webpmux to extract the individual frames or
vwebp to view this image.
Decoding of a failed.
Status: 4(UNSUPPORTED_FEATURE)

Animated HEIF (HEVC/AV1) is supported by FFmpeg, actually it's pretty much like a plain MP4 file with a video track, the difference is there's a image item stored as "thumbnail" (I assumed based on observation, I didn't read the actual specification)

The method I chose to handle animated WebP was to extract timing info with webpmux, then extract individual frames as PNG with ImageMagick (more on that later) and use FFmpeg concat demuxer to convert them to APNG and move on...

Why ImageMagick? webpmux can extract frames from animated WebP files and it's "demuxing" while ImageMagick is decoding/re-encoding. Because webpmux only extracts one frame a time, it's not a big deal but I just prefer to do it in one command at the moment.
Why extract individual frames instead of let ImageMagick output APNG directly? Because when doing animated WebP to APNG conversion, ImageMagick used LOSSY WEBP as intermediate. (I'm not sure if it's still the case now)

A VapourSynth source filter, imwri, uses ImageMagick to read images, but it looks like it only reads a single frame, it's meant for image sequences I guess.

[Selur]

okay, so probably no need to look into supporting webp any time stone.

[juanitogan]

I'm stumped.

I just uninstalled 2024.04.06.1 completely including removing settings (backing up scripts first). Reboot, uninstall, wipe all remaining folders (in Program Files and Roaming), reboot.

I then installed the latest 2024.09.29.1 and found that VS' FilterOrder is not working. Uninstalled Hybrid completely and tried everything again.

Before setting any defaults or anything else, Filtering>Vapoursynth>Misc>FilterOrder does not appear to work:
- Drag/drop moves item (but does not change script).
- Both Up and Down buttons simply move the highlight bar down instead of moving the item.
- Reset button does nothing.

Tried loading a file and turning on a few filters.
- Nothing gets bolded like it used to.
- Drag/drop moves item but does not change script.

Tried setting paths, tried Defaults>ResetAll, tried restarting. Still can't change filter order.

Is there something else I need to reset?

[Selur]

No, works fine here (Win11). Haven't changed that code for ages.
What OS are you using? Did you modify the Windows themes or similar through some tool?

[juanitogan]

Weird... but I didn't think this was a feature you would miss testing so I figured I must be missing something (or some other local conflict). Win 10 Home 64bit. I don't load unusual things into the OS like themes and such. I'm a dev myself and so I keep things as generic as possible. I'll check earlier versions and/or go back to the version I was using and see what happens. BTW, I should have mentioned that jobs otherwise run fine... I just can't change the order of the filters with this feature.

[juanitogan]

Oh, yes, the one suspicious line from the log:

Encounter XML error reading 'readGeneral': Opening and ending tag mismatch.

[Selur]

Can't reproduce it on Win11, but I could reproduce the problem on MacOS. :)
For some reason the ui <> mediator connections are not working,.. => I'll try to look into it, but probably won't really have much time before Wends day evening after work.
But I can reproduce it. Might be related to so Qt version change. => nope its a problem with the FilterOrder not properly populated.

[Selur]

Think I fixed it, uploaded a new dev (to the 'experimental'-folder of my GoogleDrive, which is linked on the download page)

Cu Selur

[juanitogan]

Wow, that is funny! That's a first for me: First time I've ever seen a bug report from a Windows user confirmed on a Mac because it couldn't be reproduced on Windows.

Anyways... YES! Your latest dev build fixes it. I've only tested the UI and haven't run a job yet... but I doubt any further issue with it. The bolding and script response is all there. Up/down/reset buttons work fine too (although, if you care [I don't], the bolding disappears on reset, but all the bolding comes back if you tick a filter off and on).

[[Yeah, my first thought was Qt as well since I've had minor issues with it in the past with Hybrid. But my second thought was "not likely" since the overall UI seemed fine and it looked to be more of a communication issue. So I didn't bother poking at Qt and just left it to you to find.]]

Thank you for the fix and, as always, thank you for the tool!

[Selur]

Different environment setups on different machine. ;)

[Selur]

2024.12.21.1:
*changed*

Analysis: extended workaround for MakeMKV muxed content
Muxing: support FLAC in mp4
Vapoursynth: moved EdgeFixer and BalanceBorder to Frame->Edge
Vapoursynth: vs-mlrt/torch - removed mod restrictions for TensorRT when using RIFE
Vapoursynth: ColorFix add wavelet mode
Input: support dpx as image sequence
Vapoursynth: torch - DeOldify 4.5.0
UI: make Hybrid Window resizable
Video: svtav1 support quant matrices
Video: x264 - set '--slices 4'
Avisynth: amDCT 64bit support
Vapoursynth: changed filter specifc 'flip vertical' into generic gimmick option
Hardware: support more data server graphic cards

*fixed*

Vapoursynth: torch - VSGAN(Filter) + ColorFix
Jobs: Chunked encoding crash with whitespace in file path
Decode: FFmpeg color matrix handling
Vapoursynth: torch - machine learning resizers and FilterQueue
Vapoursynth: don't use stepped resize when Vapoursynth->Frame->Resize->Resizer is disabled)
Decode: AutoCrop for DVD with spaces in the path; missing quotes
Video: FFVNEnc presets
Filtering: disable 'Filtering->FFmpeg' for 'Only use encoder'
Decoder: set vui params for input on FFmpeg
Avisynth: Separate Fields filtering
Avisynth: typo in ProToon 32bit custom addition

*added*

Vapoursynth: torch - added ColorFix for SCUNet
Vapoursynth: torch - added vs-mfdin
Vapoursynth: vsmlrt - added ColorFix for SCUNetmlrt (requires torch addon)
Vapoursynth: vsmlrt - added ColorFix for VSMLRT (requires torch addon)
Vapoursynth: vsmlrt - fix 'Multi' for VSMLRT
Vapoursynth: torch - fix 'Multi' for VSGAN
Vapoursynth: support Bore, FillBorders


-> downloads: http://www.selur.de/downloads

Cu Selur

[ReinerSchweinlin]

2024.12.21.1:
Vapoursynth: torch - added vs-mfdin


NICE!

thanks !

[Leo 69]

Hi Selur,

Do you know what's going on here? As soon as I specify tiling in BasicVSR++, I'm getting this error. I'm using your latest development build from Google Drive (from February 2025). The input resolution of the video is 2560x1440, but it doesn't really matter, tiling fails with any resolution.

EDIT: I downgraded to the stable build and the error is gone.

https://i.postimg.cc/y6SpLyN5/2025-02-22-061715.png

[Selur]

Looks like you updated Hybrid to the dev build, but you did not update the torch add-on. (from the same folder)
The BasicVSR++ parameter call changed in https://github.com/HolyWu/vs-basicvsrpp/releases/tag/v3.0.0 :)

Cu Selur

[Leo 69]

Looks like you updated Hybrid to the dev build, but you did not update the torch add-on. (from the same folder)
The BasicVSR++ parameter call changed in https://github.com/HolyWu/vs-basicvsrpp/releases/tag/v3.0.0 :)

Cu Selur

Thank you and you're right. It took me a while to figure out how to make BasicVSR++ work in Hybrid. There are no guides anywhere and I only found some bits of information on your forum and had to put them all together. But it was worth it. I have a question though:

https://i.postimg.cc/BQT9cqJb/2025-02-23-164757.png

Could you explain what this function is for and how to use it?

[Selur]

To see the 'Merge' and other, usually not shown controls, enable the corresponding options under 'Filtering->Vapoursynth->Misc->UI'.
The Merge option, simply applies the Merge (https://www.vapoursynth.com/doc/functions/video/merge.html)-filter between the version of the clip before and after the current filter.
A value of Y/U/V = 1 would fully apply the effect of the filter to the output ( = it normally applies the filter).
A value of Y/U/V = 0.5 would mix the previous and filtered version with a weight of 0.5.
A value of Y/U/V = 0/1/1 would keep the luma plane as it is and take the chroma planes from the filtered version, so it would only apply the filter to the chroma planes.
A value of Y/U/V = 1/0/0 would keep the chroma planes and only take the effect of the filter for the luma plane.

There are no guides anywhere and I only found some bits of information on your forum and had to put them all together.
There is a README.md in the GoogleDrive which explains how to install the torch add-on.

Cu Selur

[Selur]

2025.03.09.1:
*changed*

Video: NVEncC - support tune uhq
Vapoursynth: torch updated RealESRGAN, RIFE
Vapoursynth: torch update DeOdify (old saved settings might not work as expected in new version)
Monitor: monitoring folder "-autoAdd addAndStart" case insensitive.
Vapoursynth: switch RemoveDirt port by pinterf
Vapoursynth: torch - disable DDD 'DfConvEkSA' due to dependency problems (openmim + newer Pytorch)
Vapoursynth: torch - renamed vsDeoldify to HAVC, removed 'custom' preset/speed option
Vapoursynth: added Rotate, PreColor to 'Gimmick'-controls
Vapoursynth: removed dependency to librife_r9_mod_v6

*fixed*

Vapoursynth: torch - vsCodeFormer with FilterQueue set num_batches
Vapoursynth: reinterlace indention
Vapoursynth: make sure to adjust resolution for preview when custom section changed it
Vapoursynth: fixed BinarMask GrayMask bit depth calculation
Video: NVEncC - H.265&H.264 lossless support detection
Video: FFNVEnc - H.265 profile
Vapoursynth: mlrt resizers were not removed from the resizer list, even when vs-mlrt wasn't installed
Vapoursynth: output luma singaling updateing
Analysis: audio detection problems when reported order of streams were different in mediainfo and mplayer
Muxing: added workaround for mp4fpsmod issue
Avisynth: d2vSource can't use 'fieldop'

*added*

Vapoursynth: torch - RealESRGAN 1x scale filter
Vapoursynth: vs-mlrt - vsMLRTFilter


-> downloads: http://www.selur.de/downloads

Cu Selur

[Leo 69]

To see the 'Merge' and other, usually not shown controls, enable the corresponding options under 'Filtering->Vapoursynth->Misc->UI'.
The Merge option, simply applies the Merge (https://www.vapoursynth.com/doc/functions/video/merge.html)-filter between the version of the clip before and after the current filter.
A value of Y/U/V = 1 would fully apply the effect of the filter to the output ( = it normally applies the filter).
A value of Y/U/V = 0.5 would mix the previous and filtered version with a weight of 0.5.
A value of Y/U/V = 0/1/1 would keep the luma plane as it is and take the chroma planes from the filtered version, so it would only apply the filter to the chroma planes.
A value of Y/U/V = 1/0/0 would keep the chroma planes and only take the effect of the filter for the luma plane.


There is a README.md in the GoogleDrive which explains how to install the torch add-on.

Cu Selur

Thank you very much, your software is the best! I have another question. I would like to output a png image sequence with Hybrid instead of video. What should I configure so that it would start from, let's say, 000000.png or 000001.png and then continue outputting every frame separately onwards? Which container should I choose?

[Selur]

To output PNGs, you should:

Set 'Base->Processing->Video' to PNG
Adjust the settings in the 'PNG'-tab.
If the output names of the output PNGs should just be numbers, you should enable 'Number as name' here.
Then you should either set a file name with .raw extension, or use 'Generate' with 'raw' selected under 'Default container'.

Side note about the numbering: You can't set the leading number of zeros in the output name. Hybrid will adjust the leading number of zeros to the frame count.

Cu Selur

[Selur]

The selur.de Homepage is down atm. It will probably take some time before it's up again.

[Leo 69]

Hi Selur!

Is it possible to use ffmpeg x264 and x265 encoders instead of the dedicated ones? I'd like to write some metadata into the streams, which I don't now know to do with dedicated encoders, I make it like this in ffmpeg:

ffmpeg -i input.mp4 -c:v libx265 -metadata:s:v comment="encoded by Leo 69" -c:a copy output.mkv

So this writes the metadata directly into the h.265 stream instead of the mkv container.

And also is there a way to encode x264 and x265 videos directly into the mkv container instead of creating temporary .264 and .265 files, like ffmpeg can do? All in all, I would prefer to use ffmpeg internal encoders, since they have a bit more flexibility. Is it possible in Hybrid?

[Z2697]

So this writes the metadata directly into the h.265 stream instead of the mkv container.

It's not.

[Leo 69]

It's not.

And what exactly does that command do then, if "it's not"?

Download this file and tell me where the metadata is stored, in the container or in the video stream? => https://app.filen.io/#/d/8a32b38c-468f-4155-b64e-367df4f38ddc%23cfJEsa435CkHygQcJBprozleS3KUBYfS

[Z2697]

And what exactly does that command do then, if "it's not"?

Download this file and tell me where the metadata is stored, in the container or in the video stream? => https://app.filen.io/#/d/8a32b38c-468f-4155-b64e-367df4f38ddc%23cfJEsa435CkHygQcJBprozleS3KUBYfS

It's in the container.
https://files.catbox.moe/t1zsq6.png
https://files.catbox.moe/hit22c.png
It's somthing you can choose to discard when remuxing.

[Selur]

Is it possible to use ffmpeg x264 and x265 encoders instead of the dedicated ones?
No, that is not supported in Hybrid.

I'd like to write some metadata into the streams, which I don't now know to do with dedicated encoders
Enabling "Base->Tagging". The 'Tag'-tab allows setting container and stream tags. (both are part of the container)
For mkv, there is extended container tag support under Config->MKV Tagging. (usually the official 'ENCODED_BY' (https://www.matroska.org/technical/tagging.html)-tag should be uses to tag, who encoded the stream)

So this writes the metadata directly into the h.265 stream instead of the mkv container.
No. I agree with Z2967, that command is setting container streams tags.


=> If you really want/need to use FFmpeg for encoding x264/x265, Hybrid is the wrong tool.

Cu Selur

Ps.: to modify stream specific meta data see: https://ffmpeg.org/ffmpeg-bitstream-filters.html

[Selur]

2025.05.17.1:
*fixed*

Vapoursynth: SMDegrain with (constrast) sharpen
Vapoursynth: vsnlm_cuda handling (Windows only)
Vapoursynth: Gimmick undo resize
Vapoursynth: vsmlrt/vsgan - zoom handling
Vapoursynth: torch - ddColor miscfilter import
Vapoursynth: SMDegrain in FilterQueue
Vapoursynth: torch - CodeFormer output resolution; padding calculation
Vapoursynth: custom - handle 'scantype' change properly
Vapoursynth: profiles include merge-, mask-,... -options
Jobs: progress indication with FFmpeg as encoder
Vapoursynth: mlrt - vsMLRTFilter fix undo resize when combined with mod
Vapoursynth: FilterPresets saving
Vapoursynth: EdgeFixer with BoxBlur
Jobs: SvpAv1nc/SvpHevc temp output file detection
Vapoursynth: Gimmick Resize + required mod
Vapoursynth: vsLimiter: std vs vszip
Vapoursynth: torch/vsmlrt - force logging to stderr when
Vapoursynth: better Preview Histogram handling
Video: pc luma range not properly set with out filtering
Video: SVT-AV1 master display import

*changed*

Analysis: try using ffmpeg to get frame count on raw input
Vapoursynth: vsmlrt(filter) overlap
Tools: simplify cuda/nvenc checks
Video: NVEnc split, parallel encoding, luma signaling
Vapoursynth: vsSpotLess: interations supported
Vapoursynth: vsPreview: 'original|filtered|diff'
Vapoursynth: using zsmooth instead of RemoveGrainVS
Vapoursynth: use more akarin
Link to version.txt changed to https://www.selur.de/files/hybrid_downloads/version.txt

*added*

Vapoursynth: torch - vsDRBA support
Vapoursynth: Checkmate support (Windows only)
Vapoursynth: ChromaShiftF shift chroma by field/frame
Vapoursynth: option to force bit depth for filters (Gimmick)
Vapoursynth: option compare before/after current filter (Gimmick)
Vapoursynth: AutoGain, UnsharpMask, ReplaceMultipleFrames

*removed*

Video: SVT-AV1: '--enable-hdr'-option


-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

2025.05.18.1:

*changed*

audio: support FLAC multi-threaded encoding
Avisynth: moved default Pluging to avisynthPlugins folder since the default folder keeps chaning the name
Avisynth: updated to Avisynth+ 3.7.5

*fixed*

Avisynth: loading VDubFilter.dll before LoadVirtualDubPlugin.
Vapoursynth: multiple missing definitions in the scripts



-> downloads: http://www.selur.de/downloads

Cu Selur

[Selur]

[b]2025.05.18.2[b]:

*fixed*

Vapoursynth: scale vs scale_value mess-up

-> downloads: http://www.selur.de/downloads

Cu Selur

[valnar]

I started using Hybrid instead of Handbrake for NTSC DVD's because I have some that need deinterlacing (well, I guess all DVD's do). This one in particular is not simply telecined.

The one setting I'm not clear on is the Crop/Resize. I want to maintain the 740x480 pixels for maximum quality, but keep the 4:3 aspect ratio. In Handbrake, that's:
* No cropping
* Resolution limit 480p NTSC SD
* Automatic Anamorphic (pixel aspect 8:9)

Dimensions show Storage size is 720x480 and display is 640x480 AR 4:3

What are the equivalent settings in Hybrid?

[Selur]

No clue, don't use handbrake.
But if you don't resize Hybrid will keep the detected input par, if you don't resize or letterbox the output dar should also stay the same.

[Leo 69]

Hi Selur,

Could you add this encoder in the future build? Thanks.

https://www.phoronix.com/news/FFmpeg-Faster-FFV1-Vulkan-Enc

[Selur]

Does MBAS (https://github.com/m-ab-s/media-autobuild_suite) support this? If yes, what ffmpeg_options need to be enabled to build ffmpeg with it?

[Leo 69]

I don't know, not a developer. But Shutter Encoder for example has included that encoder in the latest build, and it's also built around ffmpeg AFAIK.

[Selur]

Okay, that doesn't help.
No clue about shutter encoder, I tried building ffmpeg with:
configuration: --pkg-config=pkgconf --cc='ccache gcc' --cxx='ccache g++' --ld='ccache g++' --extra-cxxflags=-fpermissive --extra-cflags=-Wno-int-conversion --disable-autodetect --enable-amf --enable-bzlib --enable-cuda --enable-cuvid --enable-d3d11va --enable-dxva2 --enable-iconv --enable-lzma --enable-nvenc --enable-zlib --enable-sdl2 --enable-ffnvcodec --enable-nvdec --enable-cuda-llvm --enable-libmp3lame --enable-libopus --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libx265 --enable-libdav1d --enable-libaom --disable-debug --enable-fontconfig --enable-libass --enable-libbluray --enable-libfreetype --enable-libharfbuzz --enable-libmfx --enable-libmysofa --enable-libopenjpeg --enable-libsoxr --enable-libspeex --enable-libtheora --enable-libtwolame --enable-libvidstab --enable-libwebp --enable-libxml2 --enable-libzimg --enable-libshine --enable-gpl --enable-avisynth --enable-libxvid --enable-libopenmpt --enable-librav1e --enable-libsrt --enable-libgsm --enable-libvmaf --enable-libsvtav1 --enable-vulkan --extra-cflags=-DLIBTWOLAME_STATIC --extra-cflags=-DLIBXML_STATIC --extra-libs=-liconv --disable-w32threads
but that does not add ffv1_vulkan as encoder.

Configure also doesn't list it:
$ ./configure --list-encoders
a64multi dvvideo libmp3lame pcm_alaw_at s302m
a64multi5 dxv liboapv pcm_bluray sbc
aac eac3 libopencore_amrnb pcm_dvd sgi
aac_at exr libopenh264 pcm_f32be smc
aac_mf ffv1 libopenjpeg pcm_f32le snow
ac3 ffv1_vulkan libopus pcm_f64be sonic
ac3_fixed ffvhuff librav1e pcm_f64le sonic_ls
ac3_mf fits libshine pcm_mulaw speedhq
adpcm_adx flac libspeex pcm_mulaw_at srt
adpcm_argo flashsv libsvtav1 pcm_s16be ssa
adpcm_g722 flashsv2 libtheora pcm_s16be_planar subrip
adpcm_g726 flv libtwolame pcm_s16le sunrast
adpcm_g726le g723_1 libvo_amrwbenc pcm_s16le_planar svq1
adpcm_ima_alp gif libvorbis pcm_s24be targa
adpcm_ima_amv h261 libvpx_vp8 pcm_s24daud text
adpcm_ima_apm h263 libvpx_vp9 pcm_s24le tiff
adpcm_ima_qt h263_v4l2m2m libvvenc pcm_s24le_planar truehd
adpcm_ima_ssi h263p libwebp pcm_s32be tta
adpcm_ima_wav h264_amf libwebp_anim pcm_s32le ttml
adpcm_ima_ws h264_mediacodec libx262 pcm_s32le_planar utvideo
adpcm_ms h264_mf libx264 pcm_s64be v210
adpcm_swf h264_nvenc libx264rgb pcm_s64le v308
adpcm_yamaha h264_omx libx265 pcm_s8 v408
alac h264_qsv libxavs pcm_s8_planar v410
alac_at h264_v4l2m2m libxavs2 pcm_u16be vbn
alias_pix h264_vaapi libxeve pcm_u16le vc2
amv h264_videotoolbox libxvid pcm_u24be vnull
anull h264_vulkan ljpeg pcm_u24le vorbis
apng hap magicyuv pcm_u32be vp8_mediacodec
aptx hdr mjpeg pcm_u32le vp8_v4l2m2m
aptx_hd hevc_amf mjpeg_qsv pcm_u8 vp8_vaapi
ass hevc_d3d12va mjpeg_vaapi pcm_vidc vp9_mediacodec
asv1 hevc_mediacodec mlp pcx vp9_qsv
asv2 hevc_mf movtext pfm vp9_vaapi
av1_amf hevc_nvenc mp2 pgm wavpack
av1_mediacodec hevc_qsv mp2fixed pgmyuv wbmp
av1_mf hevc_v4l2m2m mp3_mf phm webvtt
av1_nvenc hevc_vaapi mpeg1video png wmav1
av1_qsv hevc_videotoolbox mpeg2_qsv ppm wmav2
av1_vaapi hevc_vulkan mpeg2_vaapi prores wmv1
avrp huffyuv mpeg2video prores_aw wmv2
avui ilbc_at mpeg4 prores_ks wrapped_avframe
bitpacked jpeg2000 mpeg4_mediacodec prores_videotoolbox xbm
bmp jpegls mpeg4_omx qoi xface
cfhd libaom_av1 mpeg4_v4l2m2m qtrle xsub
cinepak libcodec2 msmpeg4v2 r10k xwd
cljr libfdk_aac msmpeg4v3 r210 y41p
comfortnoise libgsm msrle ra_144 yuv4
dca libgsm_ms msvideo1 rawvideo zlib
dfpwm libilbc nellymoser roq zmbv
dnxhd libjxl opus roq_dpcm
dpx libjxl_anim pam rpza
dvbsub libkvazaar pbm rv10
dvdsub liblc3 pcm_alaw rv20


$ ./configure --list-hwaccels
does list it:

av1_d3d11va h264_dxva2 mjpeg_nvdec mpeg4_vdpau vp9_dxva2
av1_d3d11va2 h264_nvdec mjpeg_vaapi mpeg4_videotoolbox vp9_nvdec
av1_d3d12va h264_vaapi mpeg1_nvdec prores_videotoolbox vp9_vaapi
av1_dxva2 h264_vdpau mpeg1_vdpau vc1_d3d11va vp9_vdpau
av1_nvdec h264_videotoolbox mpeg1_videotoolbox vc1_d3d11va2 vp9_videotoolbox
av1_vaapi h264_vulkan mpeg2_d3d11va vc1_d3d12va vvc_vaapi
av1_vdpau hevc_d3d11va mpeg2_d3d11va2 vc1_dxva2 wmv3_d3d11va
av1_videotoolbox hevc_d3d11va2 mpeg2_d3d12va vc1_nvdec wmv3_d3d11va2
av1_vulkan hevc_d3d12va mpeg2_dxva2 vc1_vaapi wmv3_d3d12va
ffv1_vulkan hevc_dxva2 mpeg2_nvdec vc1_vdpau wmv3_dxva2
h263_vaapi hevc_nvdec mpeg2_vaapi vp8_nvdec wmv3_nvdec
h263_videotoolbox hevc_vaapi mpeg2_vdpau vp8_vaapi wmv3_vaapi
h264_d3d11va hevc_vdpau mpeg2_videotoolbox vp9_d3d11va wmv3_vdpau
h264_d3d11va2 hevc_videotoolbox mpeg4_nvdec vp9_d3d11va2
h264_d3d12va hevc_vulkan mpeg4_vaapi vp9_d3d12va


No clue how to build ffmpeg to support ffv1_vulkan, atm.
With '--enable-vulkan', hevc_vulkan and h264_vulkan are in the encoders list, but ffv1_vulkan is not there, only the standard ffv1 encoder is listed for ffv1.

=> Can't/Won't add support unless I can compile an ffmpeg version which supports it.

Cu Selur

Ps.: I created an issue entry (https://github.com/m-ab-s/media-autobuild_suite/issues/2944) over at mbas, maybe someone there knows how to build ffmpeg with ffv1 support.

[Jamaika]

I created something but I don't know if it works correctly.
https://forum.doom9.org/showthread.php?p=2019185#post2019185

V....D a64multi Multicolor charset for Commodore 64 (codec a64_multi)
V....D a64multi5 Multicolor charset for Commodore 64, extended with 5th color (colram) (codec a64_multi5)
V....D alias_pix Alias/Wavefront PIX image
V....D amv AMV Video
V....D apng APNG (Animated Portable Network Graphics) image
V....D liboapv liboapv APV (codec apv)
V....D asv1 ASUS V1
V....D asv2 ASUS V2
V....D libaom-av1 libaom AV1 (codec av1)
V....D avrp Avid 1:1 10-bit RGB Packer
V..X.D avui Avid Meridien Uncompressed
VF...D bitpacked Bitpacked
V....D bmp BMP (Windows and OS/2 bitmap)
VF...D cfhd GoPro CineForm HD
V....D cinepak Cinepak
V....D cljr Cirrus Logic AccuPak
V.S..D vc2 SMPTE VC-2 (codec dirac)
VFS..D dnxhd VC3/DNxHD
V....D dpx DPX (Digital Picture Exchange) image
VFS..D dvvideo DV (Digital Video)
VFS..D dxv Resolume DXV
V....D libxeve libxeve MPEG-5 EVC (codec evc)
VF...D exr OpenEXR image
V.S..D ffv1 FFmpeg video codec #1
V....D ffv1_vulkan FFmpeg video codec #1 (Vulkan) (codec ffv1)
VF...D ffvhuff Huffyuv FFmpeg variant
V....D fits Flexible Image Transport System
V....D flashsv Flash Screen Video
V....D flashsv2 Flash Screen Video Version 2
V....D flv FLV / Sorenson Spark / Sorenson H.263 (Flash Video) (codec flv1)
V....D gif GIF (Graphics Interchange Format)
V....D h261 H.261
V....D h263 H.263 / H.263-1996
V.S..D h263p H.263+ / H.263-1998 / H.263 version 2
V....D h264_vulkan H.264/AVC (Vulkan) (codec h264)
V....D libopenh264 OpenH264 H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10 (codec h264)
V....D libx264 libx264 H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10 (codec h264)
V....D libx264rgb libx264 H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10 RGB (codec h264)
V.S..D hap Vidvox Hap
VF...D hdr HDR (Radiance RGBE format) image
V....D hevc_d3d12va D3D12VA hevc encoder (codec hevc)
V....D hevc_vulkan H.265/HEVC (Vulkan) (codec hevc)
V....D libx265 libx265 H.265 / HEVC (codec hevc)
V....D libkvazaar libkvazaar H.265 / HEVC (codec hevc)
VF...D huffyuv Huffyuv / HuffYUV
VF...D jpeg2000 JPEG 2000
VF.... libopenjpeg OpenJPEG JPEG 2000 (codec jpeg2000)
VF...D jpegls JPEG-LS
V....D libjxl libjxl JPEG XL (codec jpegxl)
V....D libjxl_anim libjxl JPEG XL animated (codec jpegxl_anim)
V..... libsvtjpegxs SVT JPEG XS(Scalable Video Technology for JPEG XS) encoder (codec jpegxs)
VF...D ljpeg Lossless JPEG
VFS..D magicyuv MagicYUV video
VFS..D mjpeg MJPEG (Motion JPEG)
V.S..D mpeg1video MPEG-1 video
V.S..D mpeg2video MPEG-2 video
V.S..D mpeg4 MPEG-4 part 2
V....D libxvid libxvidcore MPEG-4 part 2 (codec mpeg4)
V....D msmpeg4v2 MPEG-4 part 2 Microsoft variant version 2
V....D msmpeg4 MPEG-4 part 2 Microsoft variant version 3 (codec msmpeg4v3)
V....D msrle Microsoft RLE
V..... msvideo1 Microsoft Video-1
V....D pam PAM (Portable AnyMap) image
V....D pbm PBM (Portable BitMap) image
V....D pcx PC Paintbrush PCX image
V....D pfm PFM (Portable FloatMap) image
V....D pgm PGM (Portable GrayMap) image
V....D pgmyuv PGMYUV (Portable GrayMap YUV) image
V....D phm PHM (Portable HalfFloatMap) image
VF...D png PNG (Portable Network Graphics) image
V....D ppm PPM (Portable PixelMap) image
VF...D prores Apple ProRes
VF...D prores_aw Apple ProRes (codec prores)
VFS... prores_ks Apple ProRes (iCodec Pro) (codec prores)
VF...D qoi QOI (Quite OK Image format) image
V....D qtrle QuickTime Animation (RLE) video
V....D r10k AJA Kona 10-bit RGB Codec
V....D r210 Uncompressed RGB 10-bit
VF...D rawvideo raw video
V....D roqvideo id RoQ video (codec roq)
V....D rpza QuickTime video (RPZA)
V....D rv10 RealVideo 1.0
V....D rv20 RealVideo 2.0
V....D sgi SGI image
V....D smc QuickTime Graphics (SMC)
V....D snow Snow
V....D speedhq NewTek SpeedHQ
V....D sunrast Sun Rasterfile image
V....D svq1 Sorenson Vector Quantizer 1 / Sorenson Video 1 / SVQ1
V....D targa Truevision Targa image
V....D libtheora libtheora Theora (codec theora)
VF...D tiff TIFF image
VF...D utvideo Ut Video
VF...D v210 Uncompressed 4:2:2 10-bit
V....D v308 Uncompressed packed 4:4:4
V....D v408 Uncompressed packed QT 4:4:4:4
V....D v410 Uncompressed 4:4:4 10-bit
V.S..D vbn Vizrt Binary Image
V..... vnull null video
V....D libvpx-vp9 libvpx VP9 (codec vp9)
V....D libvvenc libvvenc H.266 / VVC (codec vvc)
V....D libuvg266 libuvg266 H.266 / VVC (codec vvc)
VF...D wbmp WBMP (Wireless Application Protocol Bitmap) image
V....D libwebp_anim libwebp WebP image (codec webp)
V....D libwebp libwebp WebP image (codec webp)
V....D wmv1 Windows Media Video 7
V....D wmv2 Windows Media Video 8
V..... wrapped_avframe AVFrame to AVPacket passthrough
V....D xbm XBM (X BitMap) image
V....D xface X-face image
V....D xwd XWD (X Window Dump) image
V....D y41p Uncompressed YUV 4:1:1 12-bit
V....D yuv4 Uncompressed packed 4:2:0
VF...D zlib LCL (LossLess Codec Library) ZLIB
V....D zmbv Zip Motion Blocks Video
A....D aac AAC (Advanced Audio Coding)
A....D libfdk_aac Fraunhofer FDK AAC (codec aac)
A....D ac3_fixed ATSC A/52A (AC-3) (codec ac3)
A....D adpcm_adx SEGA CRI ADX ADPCM
A....D adpcm_argo ADPCM Argonaut Games
A....D g722 G.722 ADPCM (codec adpcm_g722)
A....D g726 G.726 ADPCM (codec adpcm_g726)
A....D g726le G.726 little endian ADPCM ("right-justified") (codec adpcm_g726le)
A....D adpcm_ima_alp ADPCM IMA High Voltage Software ALP
A....D adpcm_ima_amv ADPCM IMA AMV
A....D adpcm_ima_apm ADPCM IMA Ubisoft APM
A....D adpcm_ima_qt ADPCM IMA QuickTime
A....D adpcm_ima_ssi ADPCM IMA Simon & Schuster Interactive
A....D adpcm_ima_wav ADPCM IMA WAV
A....D adpcm_ima_ws ADPCM IMA Westwood
A....D adpcm_ms ADPCM Microsoft
A....D adpcm_swf ADPCM Shockwave Flash
A....D adpcm_yamaha ADPCM Yamaha
A....D alac ALAC (Apple Lossless Audio Codec)
A..... anull null audio
A....D aptx aptX (Audio Processing Technology for Bluetooth)
A....D aptx_hd aptX HD (Audio Processing Technology for Bluetooth)
A....D libcodec2 codec2 encoder using libcodec2 (codec codec2)
A....D comfortnoise RFC 3389 comfort noise generator
A....D dfpwm DFPWM1a audio
A..X.D dca DCA (DTS Coherent Acoustics) (codec dts)
A....D eac3 ATSC A/52 E-AC-3
A....D flac FLAC (Free Lossless Audio Codec)
A....D g723_1 G.723.1
A....D libgsm libgsm GSM (codec gsm)
A....D libgsm_ms libgsm GSM Microsoft variant (codec gsm_ms)
A....D liblc3 LC3 (Low Complexity Communication Codec) (codec lc3)
A..X.D mlp MLP (Meridian Lossless Packing)
A....D mp2 MP2 (MPEG audio layer 2)
A....D mp2fixed MP2 fixed point (MPEG audio layer 2) (codec mp2)
A....D libtwolame libtwolame MP2 (MPEG audio layer 2) (codec mp2)
A....D libmp3lame libmp3lame MP3 (MPEG audio layer 3) (codec mp3)
A....D libshine libshine MP3 (MPEG audio layer 3) (codec mp3)
A....D nellymoser Nellymoser Asao
A..X.D opus Opus
A....D libopus libopus Opus (codec opus)
A....D pcm_alaw PCM A-law / G.711 A-law
A....D pcm_bluray PCM signed 16|20|24-bit big-endian for Blu-ray media
A....D pcm_dvd PCM signed 16|20|24-bit big-endian for DVD media
A....D pcm_f32be PCM 32-bit floating point big-endian
A....D pcm_f32le PCM 32-bit floating point little-endian
A....D pcm_f64be PCM 64-bit floating point big-endian
A....D pcm_f64le PCM 64-bit floating point little-endian
A....D pcm_mulaw PCM mu-law / G.711 mu-law
A....D pcm_s16be PCM signed 16-bit big-endian
A....D pcm_s16be_planar PCM signed 16-bit big-endian planar
A....D pcm_s16le PCM signed 16-bit little-endian
A....D pcm_s16le_planar PCM signed 16-bit little-endian planar
A....D pcm_s24be PCM signed 24-bit big-endian
A....D pcm_s24daud PCM D-Cinema audio signed 24-bit
A....D pcm_s24le PCM signed 24-bit little-endian
A....D pcm_s24le_planar PCM signed 24-bit little-endian planar
A....D pcm_s32be PCM signed 32-bit big-endian
A....D pcm_s32le PCM signed 32-bit little-endian
A....D pcm_s32le_planar PCM signed 32-bit little-endian planar
A....D pcm_s64be PCM signed 64-bit big-endian
A....D pcm_s64le PCM signed 64-bit little-endian
A....D pcm_s8 PCM signed 8-bit
A....D pcm_s8_planar PCM signed 8-bit planar
A....D pcm_u16be PCM unsigned 16-bit big-endian
A....D pcm_u16le PCM unsigned 16-bit little-endian
A....D pcm_u24be PCM unsigned 24-bit big-endian
A....D pcm_u24le PCM unsigned 24-bit little-endian
A....D pcm_u32be PCM unsigned 32-bit big-endian
A....D pcm_u32le PCM unsigned 32-bit little-endian
A....D pcm_u8 PCM unsigned 8-bit
A....D pcm_vidc PCM Archimedes VIDC
A....D real_144 RealAudio 1.0 (14.4K) (codec ra_144)
A....D roq_dpcm id RoQ DPCM
A..X.D s302m SMPTE 302M
A....D sbc SBC (low-complexity subband codec)
A..X.D sonic Sonic
A..X.D sonicls Sonic lossless
A....D libspeex libspeex Speex (codec speex)
A..X.D truehd TrueHD
A....D tta TTA (True Audio)
A..X.D vorbis Vorbis
A....D libvorbis libvorbis (codec vorbis)
A....D wavpack WavPack
A....D wmav1 Windows Media Audio 1
A....D wmav2 Windows Media Audio 2
S..... ssa ASS (Advanced SubStation Alpha) subtitle (codec ass)
S..... ass ASS (Advanced SubStation Alpha) subtitle
S..... dvbsub DVB subtitles (codec dvb_subtitle)
S..... dvdsub DVD subtitles (codec dvd_subtitle)
S..... mov_text 3GPP Timed Text subtitle
S..... srt SubRip subtitle (codec subrip)
S..... subrip SubRip subtitle
S..... text Raw text subtitle
S..... ttml TTML subtitle
S..... webvtt WebVTT subtitle
S..... xsub DivX subtitles (XSUB)

Decoders
.. lcevc V->V LCEVC
D avisynth AviSynth script +plugins
V....D librsvg Librsvg rasterizer (codec svg)

[Selur]

I'd prefer to use MBAS, since I build most of the Windows stuff with it. But thanks for the link.

[Selur]

Thanks to the suggestion of L4cache using:
--enable-libglslang
--enable-vulkan
did the trick.

F:\media-autobuild_suite\local64\bin-video>ffmpeg -h encoder=ffv1_vulkan
ffmpeg version N-119813-g6ad95be306 Copyright (c) 2000-2025 the FFmpeg developers
built with gcc 15.1.0 (Rev5, Built by MSYS2 project)
configuration: --pkg-config=pkgconf --cc='ccache gcc' --cxx='ccache g++' --ld='ccache g++' --extra-cxxflags=-fpermissive --extra-cflags=-Wno-int-conversion --disable-autodetect --enable-amf --enable-bzlib --enable-cuda --enable-cuvid --enable-d3d11va --enable-dxva2 --enable-iconv --enable-lzma --enable-nvenc --enable-zlib --enable-sdl2 --enable-ffnvcodec --enable-nvdec --enable-cuda-llvm --enable-libmp3lame --enable-libopus --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libx265 --enable-libdav1d --enable-libaom --disable-debug --enable-fontconfig --enable-libass --enable-libbluray --enable-libfreetype --enable-libharfbuzz --enable-libmfx --enable-libmysofa --enable-libopenjpeg --enable-libsoxr --enable-libspeex --enable-libtheora --enable-libtwolame --enable-libvidstab --enable-libwebp --enable-libxml2 --enable-libzimg --enable-libshine --enable-gpl --enable-avisynth --enable-libxvid --enable-libopenmpt --enable-librav1e --enable-libsrt --enable-libgsm --enable-libvmaf --enable-libsvtav1 --enable-libglslang --enable-vulkan --extra-cflags=-DLIBTWOLAME_STATIC --extra-cflags=-DLIBXML_STATIC --extra-libs=-liconv --disable-w32threads
libavutil 60. 3.100 / 60. 3.100
libavcodec 62. 3.101 / 62. 3.101
libavformat 62. 0.102 / 62. 0.102
libavdevice 62. 0.100 / 62. 0.100
libavfilter 11. 0.100 / 11. 0.100
libswscale 9. 0.100 / 9. 0.100
libswresample 6. 0.100 / 6. 0.100
Encoder ffv1_vulkan [FFmpeg video codec #1 (Vulkan)]:
General capabilities: dr1 delay hardware
Threading capabilities: none
Supported hardware devices: vulkan
Supported pixel formats: vulkan
ffv1_vulkan AVOptions:
-slicecrc <int> E..V....... Protect slices with CRCs (from -1 to 2) (default -1)
-context <int> E..V....... Context model (from 0 to 1) (default 0)
-coder <int> E..V....... Coder type (from -2 to 2) (default range_tab)
rice 0 E..V....... Golomb rice
range_def -2 E..V....... Range with default table
range_tab 2 E..V....... Range with custom table
-qtable <int> E..V....... Quantization table (from -1 to 2) (default default)
default -1 E..V.......
8bit 0 E..V.......
greater8bit 1 E..V.......
-slices_h <int> E..V....... Number of horizontal slices (from -1 to 1024) (default -1)
-slices_v <int> E..V....... Number of vertical slices (from -1 to 1024) (default -1)
-force_pcm <boolean> E..V....... Code all slices with no prediction (default false)
-rct_search <boolean> E..V....... Run a search for RCT parameters (level 4 only) (default true)
-async_depth <int> E..V....... Internal parallelization depth (from 1 to INT_MAX) (default 1)

=> I'll look at it. :)

Cu Selur

[Selur]

Can't get ffv1_vulkan working at all. :(
=> can someone share an example on how to use this?
Got a working command line:
ffmpeg -y -hide_banner -an -init_hw_device vulkan=vkdev:0 -filter_hw_device vkdev -i "g:\TestClips&Co\test.avi" -vf "format=yuv444p,hwupload" -c:v ffv1_vulkan -pix_fmt vulkan "g:\Output\test.mkv"

Tested:
gbrp10le:
"No accelerated colorspace conversion found from yuv420p to gbrp10le (encoding works), output is

Color space : RGB
Bit depth : 10 bits

yuv420p10 => "Encoding frames with subsampling and unaligned dimensions is only supported in version 4 (-level 4)" adding '-level 4' fails, additing "-level 4 -strict -2" fails
-vf "format=yuv422p,hwupload" -level 4 -strict -2 works
-vf "format=yuv420p,hwupload" -level 4 -strict -2 works
Does anyone know what other formats are supported atm. ?

[Jamaika]

You can use bgr0

[Selur]

Using:
"F:\Hybrid\64bit\ffmpeg.exe" -y -noautorotate -nostdin -threads 8 -color_primaries bt470bg -color_trc smpte170m -colorspace bt470bg -color_range tv -i "G:\TestClips&Co\files\test.avi" -map 0:0 -an -sn -color_primaries bt470bg -color_trc smpte170m -colorspace bt470bg -color_range tv -fps_mode passthrough -init_hw_device vulkan=vkdev:0 -filter_hw_device vkdev -vf "format=yuv420p,hwupload" -level 4 -strict -2 -vcodec ffv1_vulkan -context 0 -coder 0 -pix_fmt vulkan -metadata encoding_tool="Hybrid 2025.05.29.1" "g:\Output\test.mkv"
The encoding works without remporting an error message, but the output is broken.
Same with:
ffmpeg -y -noautorotate -nostdin -threads 8 -color_primaries bt470bg -color_trc smpte170m -colorspace bt470bg -color_range tv -i "G:\TestClips&Co\files\test.avi" -map 0:0 -an -sn -color_primaries bt470bg -color_trc smpte170m -colorspace bt470bg -color_range tv -fps_mode passthrough -init_hw_device vulkan=vkdev:0 -filter_hw_device vkdev -vf "format=yuv444p,hwupload" -level 4 -strict -2 -vcodec ffv1_vulkan -context 0 -coder 0 -pix_fmt vulkan -metadata encoding_tool="Hybrid 2025.05.29.1" "g:\Output\test.mkv"

=> seems to me that this does not work atm.

[Selur]

Same when using bgr0, can't produce any working file with ffv1_vulkan. (also tried with different graphic chips; Intel, NVIDIA, AMD)

[Leo 69]

Hi Selur!

I experimented a bit with the ffv1_vulkan encoder and was able to make it work. Here's the command that produces readable files, we just need to disable the slice CRC checking:

ffmpeg -init_hw_device vulkan -i "D:\sample.mp4" -vf "hwupload=derive_device=vulkan,format=vulkan" -c:v ffv1_vulkan -level 4 -strict experimental -slicecrc 0 -c:a copy "D:\sample.mkv"

[Selur]

Can't reproduce this here.
ffmpeg -init_hw_device vulkan -i "G:\TestClips&Co\files\test.avi" -vf "hwupload=derive_device=vulkan,format=vulkan" -c:v ffv1_vulkan -level 4 -strict experimental -slicecrc 0 -c:a copy "D:\Output\sample.mkv"
creates a broken file here.
same with:
ffmpeg -init_hw_device vulkan -i "G:\TestClips&Co\files\test.avi" -vf "format=yuv420p,hwupload=derive_device=vulkan,format=vulkan" -c:v ffv1_vulkan -level 4 -strict experimental -slicecrc 0 -c:a copy "G:\Output\sample.mkv"
:(

[Leo 69]

Really? I tried just now again with another file and it works. Please try with the ffmpeg version I'm using:

https://app.filen.io/#/d/1c647093-0cba-4d35-9e0a-90e949e433ae%23kt4h55jH7MYH9vahZ88AcC9O5cR7b8b0

[Selur]

With your ffmpeg version, I get a playable fine, but when I play the file with mpc-hc I get:

---------------------------
MPC-HC
---------------------------
Timeout when closing filter graph.

Click YES to terminate player process. Click NO to wait longer (up to 15 seconds).
---------------------------
Ja Nein
---------------------------

after closing MPC-HC.

So, seems like the way I build ffmpeg with MBAS doesn't work as expected and even when I use your version the created files are not totally fine.
=> Sorry, the whole thing just seems to be still too experimental for it to be supported in Hybrid. (or for it to be really useful)

Cu Selur

[Leo 69]

You're right, I don't understand why they ship half-baked encoders.

[NoX1911]

Any idea how to fix this in Linux? When i preview my D2V/QTGMC VapourSynth script in Hybrid i get this error:

Failed to load /usr/lib/vapoursynth/bestsource.so. Error given: /usr/lib/libgobject-2.0.so.0: undefined symbol: g_pointer_bit_unlock_and_set
Failed to load /usr/lib/vapoursynth/libd2vsource.so. Error given: /usr/lib/libgobject-2.0.so.0: undefined symbol: g_pointer_bit_unlock_and_set

The same script runs properly on local vsedit. glib2 v2.84.2 was installed here on Wed May 21 20:40:26 2025. Maybe Hybrid is not yet updated to this?

Arch Linux 6.14.10
Hybrid 2025.05.17.1
vapoursynth-git 72.11.g99aa2a08-1
glib2 2.84.2


# Imports
import vapoursynth as vs
# getting Vapoursynth core
import sys
import os
core = vs.core
# Import scripts folder
scriptPath = '/home/user/opt/hybrid/vsscripts'
sys.path.insert(0, os.path.abspath(scriptPath))
# Import scripts
import qtgmc
import validate
# Source: '/tmp/JD2/parkrun1280_12mbps.ts'
# Current color space: YUV420P8, bit depth: 8, resolution: 1280x1080, frame rate: 29.97fps, scanorder: top field first, yuv luminance scale: limited, matrix: 709, format: mpeg-2
# Loading /tmp/JD2/parkrun1280_12mbps.ts using D2VSource
clip = core.d2v.Source(input="/tmp/JD2/ts_71a9607d01917b18f0481787a70185ed_853323747.d2v")
frame = clip.get_frame(0)
# setting color matrix to 709.
clip = core.std.SetFrameProps(clip, _Matrix=vs.MATRIX_BT709)
# setting color transfer (vs.TRANSFER_BT709), if it is not set.
if validate.transferIsInvalid(clip):
clip = core.std.SetFrameProps(clip=clip, _Transfer=vs.TRANSFER_BT709)
# setting color primaries info (to vs.PRIMARIES_BT709), if it is not set.
if validate.primariesIsInvalid(clip):
clip = core.std.SetFrameProps(clip=clip, _Primaries=vs.PRIMARIES_BT709)
# setting color range to TV (limited) range.
clip = core.std.SetFrameProps(clip=clip, _ColorRange=vs.RANGE_LIMITED)
# making sure frame rate is set to 29.97fps
clip = core.std.AssumeFPS(clip=clip, fpsnum=30000, fpsden=1001)
# making sure the detected scan type is set (detected: top field first)
clip = core.std.SetFrameProps(clip=clip, _FieldBased=vs.FIELD_TOP) # tff
# Deinterlacing using QTGMC
clip = qtgmc.QTGMC(Input=clip, Preset="Fast", TFF=True) # new fps: 29.97
# Making sure content is preceived as frame based
clip = core.std.SetFrameProps(clip=clip, _FieldBased=vs.FIELD_PROGRESSIVE) # progressive
clip = clip[::2] # selecting previously even frames
# adjusted resizing to achieve PAR 1:1 for preview
# resizing for preview to square pixel to 1920x1080
clip = core.resize.Bicubic(clip=clip, width=1920, height=1080)
# set output frame rate to 29.97fps (progressive)
clip = core.std.AssumeFPS(clip=clip, fpsnum=30000, fpsden=1001)
# output
clip.set_output()

[Selur]

Hybrid itself needs no changes to work with Vapoursynth R72. (works fine here on Windows, Ubuntu Linux and MacOS)

To me, it looks like there is something wrong with the aur package through which you installed bestsource.so and libd2vsource.so, but then it should not work in vsedit too.
If you compiled bestsource.so and libd2vsource yourself, re-compiling might help.
No clue, why the script would work in vsedit and not in vsviewer.
Since I don't use archlinux myself, I can't really comment on it.


Cu Selur

[NoX1911]

When doing 'ldd Hybrid' there are many libs pulled from hybrid/lib folder. One of these might collide. I've renamed/disabled the folder and Hybrid still loads up (pulling only system libs now) but fails when analyzing video files and aborts ("Found no video or audio data in input"). I will try to selectively inject the shipped libs to see if it leads to something.

Edit: Now it works. 'HybridDebugOutput.txt' showed the culprits. I've removed all libs but these 4 that are needed by hybrids ffmpeg and mplayer bins.
libicudata.so.70
libicuuc.so.70
libudfread.so.0
libxml2.so.2

My local 'libxml2 2.13.5' has this files (so its not picked up by hybrid/ffmpeg):
libxml2 /usr/lib/libxml2.so
libxml2 /usr/lib/libxml2.so.16
libxml2 /usr/lib/libxml2.so.16.0.3

Edit2: A probably easier way is to rename/disable 'ffmpeg' and 'mplayer' in the first place. I guess this is the reason the AUR version just comes with the single 'hybrid' binary. Its probably better to opt-in than opt-out files.

[Selur]

Yes, the package is only tested for Ubuntu. :)
If someone knows his way around creating .app images, patches for https://github.com/Selur/HybridDeploy are welcome. :)

[NoX1911]

Your link is 404. Private?

The current deployment method is correct but incomplete. You would need to provide all dependent shared libs (but libc) not just some of them.
Here are three ways to gather all shared libs: Link (https://stackoverflow.com/questions/43063709/how-to-copy-an-executable-with-all-needed-libraries#comment89283269_43063709) (two bash solutions and 'linuxdeploy').
Edit: Another bug: The custom path field to 'd2vwitch' is missing.

$ ldd ffmpeg
linux-vdso.so.1 (0x00007f24ad671000)
libopenal.so.1 => /home/user/opt/hybrid/./lib/libopenal.so.1 (0x00007f24a8b16000)
libxcb.so.1 => /usr/lib/libxcb.so.1 (0x00007f24ad5fc000)
libxcb-shm.so.0 => /home/user/opt/hybrid/./lib/libxcb-shm.so.0 (0x00007f24ad5f6000)
libxcb-shape.so.0 => /home/user/opt/hybrid/./lib/libxcb-shape.so.0 (0x00007f24ad5f0000)
libxcb-xfixes.so.0 => /home/user/opt/hybrid/./lib/libxcb-xfixes.so.0 (0x00007f24ad5e4000)
libasound.so.2 => /usr/lib/libasound.so.2 (0x00007f24a8a2e000)
libm.so.6 => /usr/lib/libm.so.6 (0x00007f24a8936000)
libpulse.so.0 => /home/user/opt/hybrid/./lib/libpulse.so.0 (0x00007f24a88e1000)
libfontconfig.so.1 => /usr/lib/libfontconfig.so.1 (0x00007f24a8891000)
libfreetype.so.6 => /usr/lib/libfreetype.so.6 (0x00007f24a87c7000)
libz.so.1 => /usr/lib/libz.so.1 (0x00007f24ad5c9000)
libva.so.2 => /home/user/opt/hybrid/./lib/libva.so.2 (0x00007f24a8795000)
libbz2.so.1.0 => /home/user/opt/hybrid/./lib/libbz2.so.1.0 (0x00007f24a8781000)
libxml2.so.2 => /home/user/opt/hybrid/./lib/libxml2.so.2 (0x00007f24a8592000)
liblzma.so.5 => /home/user/opt/hybrid/./lib/liblzma.so.5 (0x00007f24a8565000)
libudfread.so.0 => /home/user/opt/hybrid/./lib/libudfread.so.0 (0x00007f24a8559000)
libmvec.so.1 => /usr/lib/libmvec.so.1 (0x00007f24a8460000)
libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x00007f24a8000000)
libnuma.so.1 => /home/user/opt/hybrid/./lib/libnuma.so.1 (0x00007f24a8450000)
libva-drm.so.2 => /home/user/opt/hybrid/./lib/libva-drm.so.2 (0x00007f24a844a000)
libva-x11.so.2 => /home/user/opt/hybrid/./lib/libva-x11.so.2 (0x00007f24a8441000)
libvdpau.so.1 => /home/user/opt/hybrid/./lib/libvdpau.so.1 (0x00007f24a843a000)
libX11.so.6 => /usr/lib/libX11.so.6 (0x00007f24a82f9000)
libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x00007f24a82ca000)
libc.so.6 => /usr/lib/libc.so.6 (0x00007f24a7e10000)
libsndio.so.7 => /home/user/opt/hybrid/./lib/libsndio.so.7 (0x00007f24a82b5000)
libXau.so.6 => /usr/lib/libXau.so.6 (0x00007f24a82b0000)
libXdmcp.so.6 => /usr/lib/libXdmcp.so.6 (0x00007f24a82a8000)
/lib64/ld-linux-x86-64.so.2 => /usr/lib64/ld-linux-x86-64.so.2 (0x00007f24ad673000)
libpulsecommon-15.99.so => /home/user/opt/hybrid/./lib/libpulsecommon-15.99.so (0x00007f24a7d83000)
libdbus-1.so.3 => /home/user/opt/hybrid/./lib/libdbus-1.so.3 (0x00007f24a7d2f000)
libexpat.so.1 => /usr/lib/libexpat.so.1 (0x00007f24a7d05000)
libpng16.so.16 => /usr/lib/libpng16.so.16 (0x00007f24a7ccb000)
libharfbuzz.so.0 => /usr/lib/libharfbuzz.so.0 (0x00007f24a7b9f000)
libbrotlidec.so.1 => /usr/lib/libbrotlidec.so.1 (0x00007f24a8297000)
libicuuc.so.70 => /home/user/opt/hybrid/./lib/libicuuc.so.70 (0x00007f24a7800000)
libdrm.so.2 => /usr/lib/libdrm.so.2 (0x00007f24a7b88000)
libXext.so.6 => /home/user/opt/hybrid/./lib/libXext.so.6 (0x00007f24a7b71000)
libXfixes.so.3 => /home/user/opt/hybrid/./lib/libXfixes.so.3 (0x00007f24a7b68000)
libbsd.so.0 => /home/user/opt/hybrid/./lib/libbsd.so.0 (0x00007f24a7b4e000)
libsndfile.so.1 => /home/user/opt/hybrid/./lib/libsndfile.so.1 (0x00007f24a7acb000)
libX11-xcb.so.1 => /usr/lib/libX11-xcb.so.1 (0x00007f24a7ac6000)
libsystemd.so.0 => /home/user/opt/hybrid/./lib/libsystemd.so.0 (0x00007f24a7733000)
libasyncns.so.0 => /home/user/opt/hybrid/./lib/libasyncns.so.0 (0x00007f24a7abd000)
libapparmor.so.1 => /home/user/opt/hybrid/./lib/libapparmor.so.1 (0x00007f24a7aa7000)
libglib-2.0.so.0 => /usr/lib/libglib-2.0.so.0 (0x00007f24a75dd000)
libgraphite2.so.3 => /usr/lib/libgraphite2.so.3 (0x00007f24a7a83000)
libbrotlicommon.so.1 => /usr/lib/libbrotlicommon.so.1 (0x00007f24a7a60000)
libicudata.so.70 => /home/user/opt/hybrid/./lib/libicudata.so.70 (0x00007f24a5800000)
libmd.so.0 => /home/user/opt/hybrid/./lib/libmd.so.0 (0x00007f24a7a52000)
libFLAC.so.8 => /home/user/opt/hybrid/./lib/libFLAC.so.8 (0x00007f24a759d000)
libvorbis.so.0 => /home/user/opt/hybrid/./lib/libvorbis.so.0 (0x00007f24a756e000)
libvorbisenc.so.2 => /home/user/opt/hybrid/./lib/libvorbisenc.so.2 (0x00007f24a74c2000)
libopus.so.0 => /home/user/opt/hybrid/./lib/libopus.so.0 (0x00007f24a7463000)
libogg.so.0 => /home/user/opt/hybrid/./lib/libogg.so.0 (0x00007f24a7a44000)
libzstd.so.1 => /home/user/opt/hybrid/./lib/libzstd.so.1 (0x00007f24a572c000)
liblz4.so.1 => /home/user/opt/hybrid/./lib/liblz4.so.1 (0x00007f24a7442000)
libcap.so.2 => /home/user/opt/hybrid/./lib/libcap.so.2 (0x00007f24a7a36000)
libgcrypt.so.20 => /home/user/opt/hybrid/./lib/libgcrypt.so.20 (0x00007f24a55ec000)
libpcre2-8.so.0 => /usr/lib/libpcre2-8.so.0 (0x00007f24a5540000)
libgpg-error.so.0 => /usr/lib/libgpg-error.so.0 (0x00007f24a5515000)

[Selur]

Thought the repository was public,... (should be public now)

[Leo 69]

Hi Selur!

Is there a way to optimize the file indexing in Hybrid? So I work with large ffv1 mkv files a lot (100 Gb - 200 Gb) and I prefer to use FFMpegSource2 with the files.

The problem is that the program seems to always need to index the file every time I do certain actions.

For example, I open a file, it gets indexed by FFMpegSource2 after hitting the "Preview", I make adjustments in Vapoursynth and then I add the job and hit "Start". At this point FFMpegSource2 starts to index the file AGAIN and since the files are large, it takes a considerable amount of time. The file has already been indexed, so is there a way to use the previous .ffindex file and not analyze the file again after adding a job?

The second scenario is: 1) I open a file in Hybrid => it gets indexed (after hitting the "Preview") 2) I close Hybrid 3) Open Hybrid and the same file again 4) The file gets indexed again after clicking "Preview", disregarding the fact that the previous index file already exists.

How do I avoid this excessive indexing with FFMpegSource2 in Hybrid? The goal is to index the file just once and use that index file for all future manipulations.

[Selur]

Is there a way to optimize the file indexing in Hybrid?
a. Hybrid deletes generated index files (for preview) on closing to avoid abandoned files.
b. The preview file generated for preview and the file generated for job processing have different names. If for example, you create two jobs with the same source file you wouldn't want the one job to delete the shared index file.
Hybrid would need to:
a. generate unique named index files (to not have conflicts for index files for inputs with the same name)
b. copy existing index files for job processing, if already an index file for the input exists. (this would also require adjusting generated sub-jobs during processing)
c. some way to know that index files are okay to be deleted
So in theory yes index handling could be changed, but any method that pops into my mind atm. has it's downsides. (i.e. abandoned files, potential conflicts, early deletions,..)

How do I avoid this excessive indexing with FFMpegSource2 in Hybrid? The goal is to index the file just once and use that index file for all future manipulations.
That would require Hybrid to know that it should:
a. not delete the index file
b. re-use the index file
c. delete it once it's not needed any more

Cu Selur

[Leo 69]

Well, if those ffindex files are kept in temporary directory, I think there shouldn't be much worry about what happens to them. I believe the logic should be: if the ffindex file exists - use it, if it doesn't exist - generate a new one. Simple.

After all, I'm also using the TensorRT inference within your app and those .engine files are basically kept forever in the temporary directory until I delete them manually so we can also classify them as 'abandoned', right?

So yeah, if the index file can be created for a specified file only once and then reused for all future jobs without the need to generate it every time again and again, that would be great. At least if you could include this as an option to choose from - that would be superb!

[Selur]

Well, if those ffindex files are kept in temporary directory, I think there shouldn't be much worry about what happens to them. I believe the logic should be: if the ffindex file exists - use it, if it doesn't exist - generate a new one. Simple.
That logic already exists. Otherwise, when closing and reopening a preview window, indexing would always take place. :)
After all, I'm also using the TensorRT inference within your app and those .engine files are basically kept forever in the temporary directory until I delete them manually so we can also classify them as 'abandoned', right? Wrong. Those 'engine'-files are not file but resolution and setting dependant, so you can use the same 'engine'-file for different inputs. If you for example convert 100 files where the resolution&filter settings are the same this file would just be used once. You don't want to reuse an index file for different inputs, even if they have the same length.

I will think about how this could be done, without having to modify tons of options, but it probably won't be soon, since I'm rather busy this week, next week I'm at RockHarz (https://www.rockharz-festival.com/en). :)

Cu Selur

[Selur]

Uploaded a new dev with a 'keep index files' option under Config->Internals.
Read the tool-tip, test it and let me know whether it works as expected. :)

Cu Selur

[Leo 69]

Uploaded a new dev with a 'keep index files' option under Config->Internals.
Read the tool-tip, test it and let me know whether it works as expected. :)

Cu Selur

Thank you so much, I'll work with some files in the following weeks and let you know what happens.

[Selur]

2025.07.27.1:
*changed*

Vapoursynth: changed to Vapoursynth R72
Vapoursynth: more zsmooth when available
Vapoursynth: prefer BestSource for image sequences by default
Vapoursynth: TorchRIFE scene change not used
Vapoursynth: TorchRIFE 4.0 and 4.1 require mod16
Vapoursynth: Finegrain masking (experimental)
Vapoursynth: simplified masking script code
Vapoursynth: FixChromaBleeding h&v controls switched
Vapoursynth: Merge on SeparateFields
Vapoursynth: vs-mlrt, bf16 support
Synth: support 'keep index files'
Vapoursynth: adding 'duplicate aware' gimmick option to some filters
Vapoursynth: vsCCD through zsmooth (for AVX2 capable CPUs)
Vapoursynth: make sure Hybrid comes with a vsViewer.config on Windows
Vapoursynth: max_cache_size auto adjust options

*fixed*

Avisynth: there's no 64bit nnedi2
Avisynth: don't use YUY2 for QTGMC
Vapoursynth: loading vsnlm_cuda instead of KNLMeansCL on non-NVIDIA systems
ProRes: VUI controls
Vapoursynth - progressive filters + tfm before deinterlacer

*added*

Vapoursynth: core.std.FreezeFrames


-> downloads: http://www.selur.de/downloads

Cu Selur

[takla]

Can't get ffv1_vulkan working at all. :(
=> can someone share an example on how to use this?
Got a working command line:


Level 4 requires frames to be a multiple of 64 pixels (https://trac.ffmpeg.org/ticket/11352). So you'd need to add padding.

Edit: Yeah nvm. I've tried with padding and this thing is broken. Why tf do they ship this...

[Leo 69]

Hi Selur!

I wanted to let you know since I wasn't happy with certain AI upscaling models, I've been training my own models for upscaling and removing various artifacts and I thought you'd like them:

https://github.com/limitlesslab/AI-upscaling-models

If you want, feel free to include them in the future builds of Hybrid. I think many users could benefit from the great visual quality that I'm after.

[Selur]

busy the next few days, but I'll try to give these models a test.

[Selur]

For those interested, here a few before/after for 1x_Purevison on random SD content, I had flying around
https://imgsli.com/NDIzNDUx
https://imgsli.com/NDIzNDUy
https://imgsli.com/NDIzNDUz
https://imgsli.com/NDIzNDU0
https://imgsli.com/NDIzNDU1
https://imgsli.com/NDIzNDU3
https://imgsli.com/NDIzNDU4
https://imgsli.com/NDIzNDU5
https://imgsli.com/NDIzNDYw
https://imgsli.com/NDIzNDYx
https://imgsli.com/NDIzNDYy
https://imgsli.com/NDIzNDYz
seems to be okay for noncommercial content.

[Selur]

@Hybrid Linux users:
I'm planning to reduce the Linux versions to a single AppImage (https://drive.google.com/drive/folders/17HHkTMaZiHWo00OyAcHrLmwF2DmAcpjO?usp=drive_link) would be nice to get some feedback about:
1. does the AppImage work for you?
2. is a separate Hybrid binary only download still necessary?

Cu Selur

[Selur]

AppImage seems to work, so it'll replace the current package in the future,..

[Selur]

2025.11.08.1:
*changed*

Vapoursynth: allow multiple in-/outputs in custom sections
Vapoursynth: adjusted to Anime4k v3
Vapoursynth: SCUNet FP16 support
Vapoursynth: luma scaling handling
Vapoursynth: FilterQueue + Deinterlace
Vapoursynth: filter support simplification
Vapoursynth: vsMLRTFilter/vsVSGANFilter: configure undo-resizer (torch/mlrt)
Vapoursynth: vsDPID: support RGBS
FFmpeg: speedChange set '-r' before and after input
Vapoursynth: adjust to HAVC update (torch)
Vapoursynth: support dynamic shapes (mlrt)
Vapoursynth: ResampleHQ HDR option
Vapoursynth: torch BasicVSR++ filter scene_aware
tons of code refactored

*fixed*

Vapoursynth: FilterQueue + Interpolate or TDecimate filter
Vapoursynth: vs-mlrt. VSMLRT-Filter call
NVEnC: maxFALL not properly read from input
Monitoring: handling of monitored inputs
Vapoursynth: mclean tcanny dependency
Vapoursynth vsSantiagAA limit range 0-128
Vapoursynth: vsCCD typo temporal_radius
Audio: limitAudioAll lang/type controls
Vapoursynth: converting to RGB force pc scale
Vapoursynth: SSMD on HDR content


-> downloads: http://www.selur.de/downloads

Cu Selur

Ps.: Dropped Ubuntu package release, created an AppImage which hopefully works.

[Selur]

2025.11.09.1:
*fixed*

Vapoursynth: masking not working
Vapoursynth: DGDenoise luma minimum
Vapoursynth: MClean vcm dependency for veed


-> downloads: http://www.selur.de/downloads

Cu Selur

[ReinerSchweinlin]

As always: Thank You !

I need to find time to update my hybrid install :)

Someone porting seedvr2 to vapoursynth ? (long shot, I know...)

[Selur]

Not that I heard of it. ��

[Leo 69]

Hi Selur! What is the new "SceneAware" feature for BasicVSR++ supposed to do? I wasn't able to find any documentation. Could you explain its parameters as well (thresh/min). Thanks!

[Selur]

Ah, I forgot to write some tool tips. :)
The SceneAware options I added to a few filters, simply adds a wrapper (https://github.com/Selur/VapoursynthScriptsInHybrid/blob/f5b39fe88a06c2481082e0debe2abd99ba8b0450/misc.py#L222):
clip = misc.scene_aware(clip, filter_func=BasicVSRPP, sc_threshold=0.100, min_scene_len=5, color_matrix="470bg", model=4)
What that wrapper does is it automatically splits a clip by scene changes (sc_threshold = how much change before splitting, min_sc_len = ignore split if resulting scene would be X or less frames in size) and apply a filter separately per scene.

This helps with some filters where I encountered issues around scene changes otherwise.

Cu Selur

[Selur]

2026.03.21.1:
*changed*

Vapoursynth: enable Vapoursynth script highlighting
Vapoursynth: support SNEEDIF , EEDI3VK
UI: scale icons depending on font size ('+/-'-buttons allow scaling now)
Video: SVTAV1 progress, encoder uses new reporting format
Vapoursynth: use fillBorders when padding HAVC or ColorAdjust(HAVC)
Video: SVTAV1 a few new options
Vapoursynth: 'adjust PAR before resize'-option
Vapoursynth: support cranexpr
Vapoursynth: add 'DeltaRestore' to SpotLess
Avisynth: explicitly set 'AssumeFrameBased()' after deinterlacers

*fixed*

Vapoursynth: SMDegrain with RefineMotion and Plane != all
Vapoursynth: STPresso, CNR + cuda
Vapoursynth: fix dither type when changing to higher bit depth
Vapoursynth: MDegrain dependencies
Muxing: mkvmerge color tags on passthrough
UI: ComboBox transparent background in some styles.
AVX2 detection

*added*

Vapoursynth: DeJitter, deblendS6

*Notes*

On Windows: Deinstall old version, before installing, since the Vapoursynth version changed


-> downloads: http://www.selur.de/downloads

Cu Selur