RipBot264 v1.18.3 - Simple and easy to use GUI -> IPOD . PSP . CONSOLES . BLURAY

[guest]

I simply edited the "default.style" file with notepad located in the ripbot program folder under "Tools" and then "srttossa", I changed the Fontsize from 18 if I remember correctly to 16 and now each new file what load in which I print the subtitles above, it gives them to me immediately with this default, I hope I have been of help.

Hi pinco1117,

OMG, that has been a LOT of help, thankyou.

Even Atak had forgotten about this, I think, that VideoHelp link where he mentions how to edit that "default.style" was slightly off in its location, but there you have it.

So now it's just a matter of what to change to get the desired result :)

These are the default settings :-

[V4 Styles]
Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, TertiaryColour, BackColour, Bold, Italic, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, AlphaLevel, Encoding
Style: Default,Arial,20,16777215,65535,65535,-2147483640,0,0,1,2,1,2,10,10,10,0,1

I would think that the red would be the main changes.

EDIT:-

I just had a play around with these settings, and as far as I could tell, it does not change anything...there must be some other subtitling going on somewhere else :(

I even removed the srttossa & ssatosrt folders, and I still got subtitles in the preview, when set to render.

[Guest]

You mean by adding fake frames?

Hi Atak,

I have run into a strange issue...

When using this call:-

video=FrameRateConverter(video,NewNum=60000,NewDen=1000)

When using DE, it doesn't create the chunks properly, and only processes 10 frames, and starts again.

However, in "single" mode, it works without issue.

Any ideas ??

https://i.imgur.com/KwT3MK0.jpg

[guest]

If I recall correctly font style and size was controlled by external file .style.

example

[V4 Styles]
Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, TertiaryColour, BackColour, Bold, Italic, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, AlphaLevel, Encoding

Style: Default,Tahoma,18,&Hffffff,&H00ffff,&H000000,&H000000,1,0,1,2,1,2,20,20,20,0,1

It should be probably in the same directory as subtitle file with name like this

YourSubtitle.srt.style or YourSubtitle.style

I do not remember now...

In reference to what pinco1117 has mentioned, would you consider adding some tweaking options to the subtitles ?

Like once enabled, some drop down lists appeared, and you could change the font name, size & colour.

EDIT:-

This post is probably irrelevant, now :(

[pinco1117]

Hi pinco1117,

OMG, that has been a LOT of help, thankyou.

Even Atak had forgotten about this, I think, that VideoHelp link where he mentions how to edit that "default.style" was slightly off in its location, but there you have it.

So now it's just a matter of what to change to get the desired result :)

These are the default settings :-

[V4 Styles]
Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, TertiaryColour, BackColour, Bold, Italic, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, AlphaLevel, Encoding
Style: Default,Arial,20,16777215,65535,65535,-2147483640,0,0,1,2,1,2,10,10,10,0,1

I would think that the red would be the main changes.

EDIT:-

I just had a play around with these settings, and as far as I could tell, it does not change anything...there must be some other subtitling going on somewhere else :(

I even removed the srttossa & ssatosrt folders, and I still got subtitles in the preview, when set to render.


I'm sorry it didn't work for you, I don't know if it's relevant but I use ripbot version 1.26.0, maybe there was some change to these settings with new versions.

[guest]

I'm sorry it didn't work for you, I don't know if it's relevant but I use ripbot version 1.26.0, maybe there was some change to these settings with new versions.

Interesting that you're still using an older version, is there a reason why you haven't updated ?

I would very much doubt that the subtitle function has been changed in a long time, but maybe I could try 1.26.0 for myself.

Thanks for the suggestion in the first place ;)

[pinco1117]

Interesting that you're still using an older version, is there a reason why you haven't updated ?

I would very much doubt that the subtitle function has been changed in a long time, but maybe I could try 1.26.0 for myself.

Thanks for the suggestion in the first place ;)

Let's say that for now I'm not interested in updating, I stayed with the 1080p resolution because the 4k format disappointed me especially for the choice of HDR which worsens the image for me, I did many tests, so for now the program is for me it works perfectly i have no DE problems, problems with resolutions and compression times and therefore i prefer not to mess with new problems, that's all so more for a personal choice.

[Guest]

I have now tried several FPS filters/scripts, whatever you want to call them, and DE just doesn't produce correct, usable chunks.

As per previous post.

The ONLY call that seems to work is:-

video=FrameRateConverter(video,FrameDouble=true)

[god_md5]

https://github.com/rigaya/NVEnc/releases
i know x264 or x265 is best compression , and i only need Interlaced to progressive file ,ripbot264 is best qtgmc deinterlace is gpu denoise,and file to video ai upscale,so big file ,fast encode is need.

[guest]

https://github.com/rigaya/NVEnc/releases
i know x264 or x265 is best compression , and i only need Interlaced to progressive file ,ripbot264 is best qtgmc deinterlace is gpu denoise,and file to video ai upscale,so big file ,fast encode is need.

This has been asked for so many times over the years, and it hasn't happened, and unfortunately probably won't, any time soon.

You will have to try another app, StaxRip, FastFlix, Hybrid, and maybe even Handbrake.

[guest]

I was asked why there weren't certain scripts/filters included in the last "Lite" build, so I have made available, some extra scripts that you can try, and see if they are more suitable for you..

Note, these are already included in the full build.

Simply click on the link in my signature...

[Atak_Snajpera]

I have now tried several FPS filters/scripts, whatever you want to call them, and DE just doesn't produce correct, usable chunks.

As per previous post.

The ONLY call that seems to work is:-

video=FrameRateConverter(video,FrameDouble=true)

Fixed. Enable Auto Update if you have disabled it in ripbot.

[Guest]

Fixed. Enable Auto Update if you have disabled it in ripbot.

Thankyou so very much for fixing this issue, it has just added another dimension to what RB can do :), even tho you didn't seem to be too keen on it.

You mean by adding fake frames? (https://forum.doom9.org/showthread.php?p=1985261#post1985261)

I have to mention, initially I was getting an FFmpeg stall @ 99.8%, but it turned out to be the clip I had chosen to test...seems good on a different, freshly imported Job.

[god_md5]

new update core 4.10 have error , lost frame ,sound not sync

[guest]

new update core 4.10 have error , lost frame ,sound not sync

Not sure what you're referring to...

The latest update was on 09-04-2023, and is version 1.26.3.0.

So what's core 4.10 ??

[Guest]

Atak,

Is there some way you could provide a Colourspace conversion from BT709 to BT2020.

Not sure in ColourMatrix can do this...

I tried a simple conversion with Handbrake, and it worked, and MediaInfo show that the BT709 clip had been changed to BT2020.

Some info from HB's log:-

"Settings": {
"desat": "0",
"matrix": "bt2020ncl",
"primaries": "bt2020",
"tonemap": "hable",
"transfer": "bt2020-10"
}

[Guest]

Any thoughts on what might cause this series of errors in DE ??

https://i.imgur.com/z7VRZEt.jpg

It only seems to happen with 4K HDR clips.

But I have to confess, I am using a custom script.

[Atak_Snajpera]

Not enough RAM or VRAM?

[Guest]

Not enough RAM or VRAM?

I have 32Gb DDR5...

I've only got 8Gb on the RTX 3060, could I use multiple GPU's

Forgot to mention, would it be the iGPU on the 7950X being used instead of the 3060 ??

[Atak_Snajpera]

You are using script which uses vulkan api so 8 GB is not enough.

[Guest]

You are using script which uses vulkan api so 8 GB is not enough.

Right you are...

So it's either not use it on 4K, or buy a better GPU, 12Gb or so.

NOTE:-

I forgot to mention, (unfortunately), that script does run in "single" encoding mode....so maybe it's just something about DE, again, like it was with the 60 FPS conversions.

If you wouldn't mind looking into that side of things :)

[Ryushin]

So I've been fighting with color banding on the Blu-ray U-571 underwater scenes. Problem really shows itself when using SMDegrain (hard) as it cleans up the source so much I've been getting banding.

I tried encoding to 12 bit color to but did not make a difference. So I tried using PD's full build which including ColorMatrix2.6 and using the custom script to convert BT709 to BT2020:
#ColorSpace
SetCacheMode(1)
SetMemoryMax(16384)
LoadPlugin("%AVISYNTHPLUGINS%\PD_TOOLS\ColorMatrix26\ColorMatrix64.dll")
video=ColorMatrix(video,mode="Rec.709->Rec.2020")

But this script did not work and it did not show any improvement. Mediainfo still showed BT709.
So U-571 was on the back burner for a few days as I put it aside until I could figure out the above how to fix the above.

So after that, I started working with Topaz Video AI in upscaling video. Oh, my word, Topaz Video AI is bloody amazing. Upscaled a few DVDs to HD using a two step process, DVD to 720p, 720p to 1080p, using the Artemis model. Results have been staggeringly good. Topaz can even clean up the remaining grain that SMDegrain (hard) can't remove. I was having issues with Moonstruck still having quite a bit of grain. Topaz removed the grain and upscaled it to 4K and the results were amazing!

I noticed my BT709 10 bit color was being converted back to 8 bit color. By chance I opened Handbrake and I saw that it had the option to convert the colorspace and it actually worked. So that sent me down a different path.

So I have a DVD series called Jericho that I wanted to redo in Topaz.
*) Extracted the episodes using MakeMKV
*) Dumped them into Handbrake, with BT709 color conversion as DVDs use BT601, upscaled to get rid of anamorphic (854x480 16:9), CQ14, no other filters.
*) Took the Handbrake output and pulled it into RB and SMDegrain (hard), CQ14
*) Pulled the RB output into Topaz and started a two step output, DVD to 720p, 720p to 1080p. I looked at the 720p last night, and the colors are richer compared to the DVD counterpart.

So converting DVDs BT601 to BT709 also has good benefits.

So back to U-571 and my banding problem. I pulled U-571 into Handbrake, set CQ14, set the cropped dimensions and the filter option to BT2020 and had it do an encode. I took the Handbrake output, pulled it into Ripbot and set the SMDegrain (hard) and low and behold, no color banding, at all. Even the sunlight coming down through the water is perfect. Perfect! Next I'm going to pull U-571 into Topaz and try upscaling to 4K 60fps. We we'll see how that turns out.

I see that @FTLOY just mentioned Handbrake could convert the colorspace as well, so at least two of us have found that.

So Atak, a feature request, please if possible, the ability to convert colorspaces. From BT601 to BT709, and BT709 to BT2020. My workflow process will always have this step now.

Also, something I noticed, Ripbot displays the colored HDR logo even though no HDR is used. I wonder if you are checking for BT2020 in your detection script and when it sees that it shows the HDR logo. It would be better to check for HDR entry in mediainfo instead:
HDR format : SMPTE ST 2086, HDR10 compatible

Thanks Atak.

[guest]

So I've been fighting with color banding on the Blu-ray U-571 underwater scenes. Problem really shows itself when using SMDegrain (hard) as it cleans up the source so much I've been getting banding.

That is a pretty intense workflow :(

Maybe you could try some de-banding filters..

f3kdb

neo-f3kdb

Gradfun3plus

Uband, and there might be some others out there...

Cheers

[Ryushin]

That is a pretty intense workflow :(

Maybe you could try some de-banding filters..

f3kdb

neo-f3kdb

Gradfun3plus

Uband, and there might be some others out there...

Cheers

I did try a couple of those but did not actually have much success. It introduced more artifacts than I liked. Actually changing the color space to provide more gradient colors seems to me to be more of a proper fix. We'll see as I go down this path further.

Looking at my post above, I left out the final step of pulling in the Topaz file back into RB and reencoding to CQ18 to give me the smaller file size.

[slalom]

new update core 4.10 have error , lost frame ,sound not sync
Same here, core 2023.04.09

[guest]

Same here, core 2023.04.09

I'm pretty sure I haven't had this issue....but if 2 users have reported it, then who knows....

But it's not core 4.10, tho :scared:

[slalom]

Maybe you didn't have a video that needed sync

[guest]

Maybe you didn't have a video that needed sync

I don't understand what you're saying...

All videos need the audio to be sync'ed...

[Ryushin]

I've seen this recently, sometimes, for some reason one of the chunk size from Frame Start to Frame End shows a -1.
https://i.postimg.cc/WFN4Pgwh/Broken-Chunk-Size.png (https://postimg.cc/WFN4Pgwh)

Problem only seems to be DVD episodes extracted using MakeMKV.

Media Info:
Video
ID : 1
ID in the original source medium : 224 (0xE0)
Format : MPEG Video
Format version : Version 2
Format profile : Main@Main
Format settings : CustomMatrix / BVOP
Format settings, BVOP : Yes
Format settings, Matrix : Custom
Format settings, GOP : Variable
Codec ID : V_MPEG2
Codec ID/Info : MPEG 1 or 2 Video
Duration : 49 min 46 s
Bit rate mode : Variable
Bit rate : 4 089 kb/s
Maximum bit rate : 9 800 kb/s
Width : 720 pixels
Height : 480 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Scan order : 2:3 Pulldown
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.494
Time code of first frame : 00:59:59;00
Time code source : Group of pictures header
Stream size : 1.42 GiB (95%)
Language : English
Default : No
Forced : No
Original source medium : DVD-Video

Source file: https://cloud.chrisdos.com/s/afdHAGNcaoH6EWK

When I encounter this, I dump the file into Handbrake converting it to x265 and then adding it into Ripbot to make it work.

[Atak_Snajpera]

I've seen this recently, sometimes, for some reason one of the chunk size from Frame Start to Frame End shows a -1.
https://i.postimg.cc/WFN4Pgwh/Broken-Chunk-Size.png (https://postimg.cc/WFN4Pgwh)

Problem only seems to be DVD episodes extracted using MakeMKV.

Media Info:
Video
ID : 1
ID in the original source medium : 224 (0xE0)
Format : MPEG Video
Format version : Version 2
Format profile : Main@Main
Format settings : CustomMatrix / BVOP
Format settings, BVOP : Yes
Format settings, Matrix : Custom
Format settings, GOP : Variable
Codec ID : V_MPEG2
Codec ID/Info : MPEG 1 or 2 Video
Duration : 49 min 46 s
Bit rate mode : Variable
Bit rate : 4 089 kb/s
Maximum bit rate : 9 800 kb/s
Width : 720 pixels
Height : 480 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Scan order : 2:3 Pulldown
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.494
Time code of first frame : 00:59:59;00
Time code source : Group of pictures header
Stream size : 1.42 GiB (95%)
Language : English
Default : No
Forced : No
Original source medium : DVD-Video

Source file: https://cloud.chrisdos.com/s/afdHAGNcaoH6EWK

When I encounter this, I dump the file into Handbrake converting it to x265 and then adding it into Ripbot to make it work.

Enable INVERSE TELECINE option under DEINTERLACE section.
29.97 fps NTSC video is just a telecined 23.976 fps.
If you were encoding those videos as 29.97 fps then you just wasted a lot of time because it is wrong!

[guest]

Enable INVERSE TELECINE option under DEINTERLACE section.
29.97 fps NTSC video is just a telecined 23.976 fps.
If you were encoding those videos as 29.97 fps then you just wasted a lot of time because it is wrong!

Can confirm, this definitely fixed the issue. :o

[Ryushin]

Enable INVERSE TELECINE option under DEINTERLACE section.
29.97 fps NTSC video is just a telecined 23.976 fps.
If you were encoding those videos as 29.97 fps then you just wasted a lot of time because it is wrong!

Thanks Atak. I will note that for the future. Going to be converting a lot of DVDs soon.

[guest]

Atak,

I have recently realised that my LG 4K TV supports VP9 HDR10.

So I have done a couple of tests using Handbrake to encode an episode of Game of Thrones 4K HDR10, and it looks very smooth & crisp (to my eyes).

From what I've read, VP9 is supposed to be good for high bitrate, high resolution video...

So, I have to ask, would you consider incorporating the option to encode using VP9 ?

I was going to ask about AV1 as well, but unfortunately the TV doesn't support it, and AV1 is also a very slow encoding process too, apparently,
but if AV1 can be used with DE, then that would relieve the pain a lot :)

[Atak_Snajpera]

In terms of quality VP9 is on par with HEVC (x265). Besides VP9 is already an outdated codec superseded by AV1. I'm more interested in x266 which is still in very early phase of development though.

[guest]

In terms of quality VP9 is on par with HEVC (x265). Besides VP9 is already an outdated codec superseded by AV1. I'm more interested in x266 which is still in very early phase of development though.

So despite that VP9 is "outdated" by AV1, would you consider adding them to RB ??

[Atak_Snajpera]

no. Syntax and output is too different. Too much work for outdated codec. For example adding x265 was simple because it was built from x264. I hope x266 will follow that pattern.

[guest]

no. Syntax and output is too different. Too much work for outdated codec. For example adding x265 was simple because it was built from x264. I hope x266 will follow that pattern.

OK, fair enough, at least I asked...so I guess if I want to pursue some VP9 encoding, I'll have to use another app...so many to choose from, but Handbrake does a pretty good job, and is pretty easy to use...some of the others are not.

[Atak_Snajpera]

Still no idea why you prefer VP9 over HEVC. Compression is more or less the same. Also Highly optimized x265 may be even faster than vp9 in terms of quality/compression.

[guest]

Still no idea why you prefer VP9 over HEVC. Compression is more or less the same. Also Highly optimized x265 may be even faster than vp9 in terms of quality/compression.

I'm not saying I prefer it....its just something I'd like to try.

And by highly optimised x265, I guess you mean customised commands ???

[Ryushin]

Hi Atak,

I thought that Ripbot support Prores files. I have an output from Topaz Video AI that is Prores LT and is a staggering 214GB. I tried encoding single thread and the job aborts after a couple of seconds and Distributed Encoding which causes the Encoding server just start the chunk over and over without processing any of it.

Is the issue that it's Prores LT or something else?

Media info:
General
Complete name : /netshares/movies/Incoming/AI_Upscale/Shop_Around_the_Corner_1940_HD_to_4K_amq13.mov
Format : MPEG-4
Format profile : QuickTime
Codec ID : qt 0000.02 (qt )
File size : 215 GiB
Duration : 1 h 38 min
Overall bit rate mode : Variable
Overall bit rate : 312 Mb/s
Encoded date : UTC 2023-04-17 12:32:38
Tagged date : UTC 2023-04-17 12:32:38
major_brand : qt
minor_version : 512
compatible_brands : qt
videoai : Enhanced using amq-13. Changed resolution to 2960x2160
title : Shop Around the Corner (1940) HD
creation_time : 2023-04-17T12:32:38.000000Z
encoder : Lavf60.4.100

Video
ID : 1
Format : ProRes
Format version : Version 0
Format profile : 422 LT
Codec ID : apcs
Duration : 1 h 38 min
Bit rate mode : Variable
Bit rate : 311 Mb/s
Width : 2 960 pixels
Height : 2 160 pixels
Display aspect ratio : 1.370
Frame rate mode : Constant
Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:2
Scan type : Progressive
Bits/(Pixel*Frame) : 2.027
Stream size : 214 GiB (100%)
Writing library : Apple
Encoded date : UTC 2023-04-17 12:32:38
Tagged date : UTC 2023-04-17 12:32:38
Color primaries : BT.2020
Transfer characteristics : BT.2020 (10-bit)
Matrix coefficients : BT.2020 non-constant
Menus : 3

First error I got in the log file was:
x265 [error]: main10 profile not compatible with i422 input chroma subsampling.

Doing a google found this:
https://stackoverflow.com/questions/66328252/lossless-video-to-h-265-error-main10-profile-not-compatible-with-i422-input
I manually added the "--pix_fmt yuv420p" to the job1_EncodeVideoPass1.cmd file.

Now the error is:
D:\Temp\RipBot264temp\job3>"C:\Program Files\RipBot264\Tools\ffmpeg\bin\ffmpeg.exe" -loglevel panic -i "D:\Temp\RipBot264temp\job1\job1.avs" -strict -1 --pix_fmt yuv420p -f yuv4mpegpipe - | "C:\Program Files\RipBot264\tools\x265\x265_x64.exe" --colorprim bt2020 --transfer bt2020-10 --colormatrix bt2020nc --crf 18 --fps 24000/1001 --min-keyint 24 --keyint 240 --frames 141961 --sar 1:1 --profile main10 --output-depth 10 --sar 1:1 --ctu 64 --merange 57 --y4m --output "D:\Temp\RipBot264temp\job1\video.265" -
x265 [error]: unable to open input file <->

So I'm not sure what the problem is at this point.

Any insight you can provide Atak I would appreciate it. Afraid I'm going to be doing quite a few more of these Prores LT files as it's the only output from Video AI for me that is BT2020.

[guest]

Hi Atak,

I thought that Ripbot support Prores files. I have an output from Topaz Video AI that is Prores LT and is a staggering 214GB. I tried encoding single thread and the job aborts after a couple of seconds and Distributed Encoding which causes the Encoding server just start the chunk over and over without processing any of it.

Is the issue that it's Prores LT or something else?

So I'm not sure what the problem is at this point.

Any insight you can provide Atak I would appreciate it. Afraid I'm going to be doing quite a few more of these Prores LT files as it's the only output from Video AI for me that is BT2020.

OMG & WTF....

That is ridiculously large...

As a casual Topaz user, surely there are other formats that can do BT2020, but if not, there's other ways to convert to BT2020!!

But I think it might be an x265 issue...

https://x265.readthedocs.io/en/3.4/cli.html

--input-csp <integer|string>
Chroma Subsampling (YUV only): Only 4:0:0(monochrome), 4:2:0, 4:2:2, and 4:4:4 are supported at this time. The chroma subsampling format of your input must match your desired output chroma subsampling format (libx265 will not perform any chroma subsampling conversion), and it must be supported by the HEVC profile you have specified.

i400 (4:0:0 monochrome) - Not supported by Main or Main10 profiles
i420 (4:2:0 default) - Supported by all HEVC profiles
i422 (4:2:2) - Not supported by Main, Main10 and Main12 profiles
i444 (4:4:4) - Supported by Main 4:4:4, Main 4:4:4 10, Main 4:4:4 12, Main 4:4:4 16 Intra profiles
nv12
nv16

You may have to create a custom x265 profile !

Just a thought, what if you removed this --profile main10 from the x265 profile... (might revert back to 8 bit)...it didn't !!!

However, a small sample of this HUGE file would be handy, and maybe even benefit Atak in a solution.

Good Luck.

UPDATE:-

OK, I created my own little 422 LT prores .mov sample, with Topaz 3.2.2, and proceeded to test in RB, and after just a couple of profile edits, this is the command line (or similar) you need to use.

--level 6.2 --hdr10 --output-depth 10 --ctu 64 --high-tier --repeat-headers --vbv-bufsize 800000 --vbv-maxrate 800000

x265 will create it's own 4:2:2 10 profile :scared::D, or just remove the --profile main10 from the profile you're using.

Easy peasy.

NOTE:- If possible, it seems that Topaz does a "nicer" job on a clip that has already been "dealt with" prior to using it. Meaning any filtering, denoising, deinterlacing, degraining, etc. And possibly saves a LOT of Topaz processing time.

[Ryushin]

OMG & WTF....

That is ridiculously large...

As a casual Topaz user, surely there are other formats that can do BT2020, but if not, there's other ways to convert to BT2020!!

But I think it might be an x265 issue...

https://x265.readthedocs.io/en/3.4/cli.html

--input-csp <integer|string>
Chroma Subsampling (YUV only): Only 4:0:0(monochrome), 4:2:0, 4:2:2, and 4:4:4 are supported at this time. The chroma subsampling format of your input must match your desired output chroma subsampling format (libx265 will not perform any chroma subsampling conversion), and it must be supported by the HEVC profile you have specified.

i400 (4:0:0 monochrome) - Not supported by Main or Main10 profiles
i420 (4:2:0 default) - Supported by all HEVC profiles
i422 (4:2:2) - Not supported by Main, Main10 and Main12 profiles
i444 (4:4:4) - Supported by Main 4:4:4, Main 4:4:4 10, Main 4:4:4 12, Main 4:4:4 16 Intra profiles
nv12
nv16

OK, I created my own little 422 LT prores .mov sample, with Topaz 3.2.2, and proceeded to test in RB, and after just a couple of profile edits, this is the command line (or similar) you need to use.

--level 6.2 --hdr10 --output-depth 10 --ctu 64 --high-tier --repeat-headers --vbv-bufsize 800000 --vbv-maxrate 800000

x265 will create it's own 4:2:2 10 profile :scared::D, or just remove the --profile main10 from the profile you're using.

Easy peasy.

NOTE:- If possible, it seems that Topaz does a "nicer" job on a clip that has already been "dealt with" prior to using it. Meaning any filtering, denoising, deinterlacing, degraining, etc. And possibly saves a LOT of Topaz processing time.

Thanks for this. I've been going down the rabbit hole and it seems that FFMPEG will do the chroma conversion before passing it through to x265 as 4:2:0.

I used Fastflix to start the conversion (it's running now) and it had the ffmpeg command being used. I looked at the FFMPEG command:
ffmpeg.exe" -y -i "M:\Incoming\AI_Upscale\Shop_Around_the_Corner_1940_HD_to_4K_amq13.mov" -metadata title="Shop Around the Corner (1940) 4K Upscale from HD" -max_muxing_queue_size 1024 -map 0:0 -c:v libx265 -pix_fmt yuv420p10le -profile:v main10 -x265-params "aq-mode=2:repeat-headers=0:strong-intra-smoothing=1:bframes=4:b-adapt=2:frame-threads=0:colorprim=bt2020:transfer=bt2020-10:colormatrix=bt2020nc:hdr10_opt=0:hdr10=0" -crf:v 18 -preset:v medium -metadata:s:v rotate=0 -map_metadata 0 -map_chapters 0 -map 0:1 -metadata:s:1 title="" -metadata:s:1 handler="" -metadata:s:1 language=eng -c:1 copy -default_mode infer_no_subs "M:\Incoming\AI_Upscale\Shop_Around_the_Corner_1940_4K_Upscale.mkv"

So I'm going to adapt that command to RB. Right now, the ffmpeg command being used by RB (does not work) is:
ffmpeg.exe" -loglevel panic -i "D:\Temp\RipBot264temp\job1\job1.avs" -strict -1 -f yuv4mpegpipe - | "C:\Program Files\RipBot264\tools\x265\x265_x64.exe" --colorprim bt2020 --transfer bt2020-10 --colormatrix bt2020nc --crf 18 --fps 24000/1001 --min-keyint 24 --keyint 240 --frames 141961 --sar 1:1 --profile main10 --output-depth 10 --sar 1:1 --ctu 64 --merange 57 --y4m --output "D:\Temp\RipBot264temp\job1\video.265" -

Once I figure out the correct syntax, I'll post it here.

[guest]

Thanks for this.

Once I figure out the correct syntax, I'll post it here.

I don't understand all this FFmpeg BS...all I did was create a similar Topaz output as you did, loaded that x265 profile, and it loaded, DE ran, looking at the details of the process in the encoders window, it clearly shows the 4.2.2.10 profile has been "created automatically".

And it simply worked !!!!

From what I can tell, it's an x265 problem, nothing to do with FFmpeg.

[Ryushin]

I don't understand all this FFmpeg BS...all I did was create a similar Topaz output as you did, loaded that x265 profile, and it loaded, DE ran, looking at the details of the process in the encoders window, it clearly shows the 4.2.2.10 profile has been "created automatically".

And it simply worked !!!!

From what I can tell, it's an x265 problem, nothing to do with FFmpeg.

What I'm aiming for is compatibility. If I can encode to Main10 I'll have the most compatibility between devices. So using FFMPEG allows me to convert to 4:2:0 so I can use Main10. I'd like the files to direct play on all modern'ish devices without transcoding.

So yes, the problem is with x265. Since ffmpeg is a swiss army knife for video, it can fix the problem that x265 can't and I get the best of both worlds.

I very much appreciate you leading me down the path to show me what the actual problem was. Should be a simple enough fix once I figure out the syntax.

[guest]

What I'm aiming for is compatibility. If I can encode to Main10 I'll have the most compatibility between devices. So using FFMPEG allows me to convert to 4:2:0 so I can use Main10. I'd like the files to direct play on all modern'ish devices without transcoding.

So yes, the problem is with x265. Since ffmpeg is a swiss army knife for video, it can fix the problem that x265 can't and I get the best of both worlds.

I very much appreciate you leading me down the path to show me what the actual problem was. Should be a simple enough fix once I figure out the syntax.

So why can't you go straight to x265 with Topaz ?

[Ryushin]

So why can't you go straight to x265 with Topaz ?

It seems the AMD H265 Encoder that Topaz has created does not support BT2020 output. So the only other option besides image sequence files is Prores LT. I think the nVidia Main 10 on Topaz does do BT2020. I requested that they add x265 as a encoder. Who knows if that will happen.

[slalom]

I don't understand what you're saying...

All videos need the audio to be sync'ed...
When there is a difference between video & audio duration, most likely the result will be out of sync

That's why you see a delay added to some filenames in job temp folder

like this one "1_audio_English_DEFAULT_DELAY6ms.dtsma"

[guest]

It seems the AMD H265 Encoder that Topaz has created does not support BT2020 output. So the only other option besides image sequence files is Prores LT. I think the nVidia Main 10 on Topaz does do BT2020. I requested that they add x265 as a encoder. Who knows if that will happen.

OK, so if that is the case, then I guess Topaz determines what hardware your system has, and uses the "matching" appropriate software...

So in your case, you have an AMD GPU, and I have nVidia GPU's.

What if you chose the CPU as the encoder...but that's AMD as well..

I might do a quick test, and see what happens..

It's just such a PITA that "we" discover the benefits of Topaz, but in turn, it creates a whole other set of problems :(

I thought I might have had a reply to the PM's I sent a few days ago.

This is the "list" of available encoders on my Topaz setup....what's yours??

https://i.imgur.com/OOp64pN.jpg

[ReinerSchweinlin]

Topaz Video Enhance AI does not offer CPU MPEG Encoding - depending on the hardware installed, the GPU Encoders are used for AVC and HEVC.... This is only usable for quick and dirty encodes or if bitrate is not important and you can crank it up..... to be bold - the implementations are not very good, only bitrate encoding is offered...

but you could bake your own json file and use any encoder setting you like...

most people output either prores or single images and encode later... in case of a crash, the single image method is the most robust....

[ReinerSchweinlin]

@Atak

I got three requests:

- Enable Ripbot to pass through attachements of MKV Containers. For example stored fonts for subtitles - these are left out at the moment.
- Enable Ripbot to process more audio-streams than the first one in batch mode (I get that if channel count is different, this is not possible in the current way, since only one profile can be chosen - but maybe the next one helps in this regard)
- Give the option of "bitrate per channel" in the batch mode - this way if a file has one 5.1 stream, one stereo and one 7.1 - choosing a profile like "AAC LC 64kbit/s per channel" would be choosable in batch mode.

Thanx for Ripbot once again. It really is a wonderful program :)

Regards

[ReinerSchweinlin]

So why can't you go straight to x265 with Topaz ?

Due to licence issues, x265/4 is not offered in TVAI. Only the GPU Encoders are usable (unless you tweak your JSON Files or compile your own ffmpeg - enabling avisynth and such...)...