MeGUI: bug reports and feature requests

[mini-moose]

The next problem is not MeGUI's fault as it happens when I extract the subtitles with vsrip myself, but after extracting subtitles from the extras on the DVD I used for testing they look like this. I re-authored a single DVD extra with DVDShrink so there's just a couple of ifo files and a vob file in the folder.

https://s27.postimg.org/prv5gta43/subs1.gif
You must have not chosen the correct PGC when you ran VSRip on your own. Judging by the 12mins start timecode, it appears you may have used the "play all" pgc, which led to your subs starting in the wrong time.

The video itself is less than twelve minutes long. When I convert it to an MKV with Handbrake and get Handbrake to mux the subtitles into an MKV, they look like this after I extract them:

https://s27.postimg.org/mjqo3rnub/subs2.gif

Which is correct. Subtitle Edit also rips them correctly from the ifo file. The only clue I have so far is MPC-HC displays 20 minutes as the duration of the video when opening the ifo or vob file, when it should be about twelve.
MPC-BE and VLC both display the correct duration when opening the ifo file, but they both display a 20 min duration when opening the vob file. Logically it appears the subtitle timing must be stored in two places and in this case one them is wrong, but does anyone know how I can investigate this further, or should I just file it under "one of those things"?
There might be some 15milisecond chapter at the end of the vob, I'm not entirely sure what those are for but they are very common. They could sometimes cause the video to appear longer than it is.
If you play it though with mpc, what happens after 12mins? does it keep playing or stops?

[hello_hello]

Zathor, I mentioned this in the text file included with the samples I uploaded here (https://sourceforge.net/p/megui/feature-requests/596/) for you, but in case you didn't get to that yet.....

I extract the audio from this MKV:

General
Unique ID : 207719713466905909570113506023949172250 (0x9C455C9329FE9F47B982E81989237E1A)
Complete name : E:\test.mkv
Format : Matroska
Format version : Version 4 / Version 2
File size : 550 MiB
Duration : 45 min
Overall bit rate : 1 703 kb/s
Encoded date : UTC 2017-01-01 20:17:11
Writing application : mkvmerge v9.7.1 ('Pandemonium') 32bit
Writing library : libebml v1.3.4 + libmatroska v1.4.5

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L3.1
Format settings, CABAC : Yes
Format settings, ReFrames : 9 frames
Codec ID : V_MPEG4/ISO/AVC
Duration : 45 min
Bit rate : 1 254 kb/s
Width : 720 pixels
Height : 480 pixels
Display aspect ratio : 4:3
Frame rate mode : Constant
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) : 0.151
Stream size : 394 MiB (72%)
Writing library : x264 core 104 r1681 84a051f
Encoding settings : cabac=1 / ref=9 / deblock=1:0:0 / analyse=0x3:0x133 / me=umh / subme=10 / psy=1 / psy_rd=1.00:0.00 / mixed_ref=0 / me_range=64 / chroma_me=1 / trellis=2 / 8x8dct=1 / cqm=0 / deadzone=21,11 / fast_pskip=0 / chroma_qp_offset=-2 / threads=12 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / constrained_intra=0 / bframes=9 / b_pyramid=2 / b_adapt=2 / b_bias=0 / direct=3 / weightb=1 / open_gop=0 / weightp=2 / keyint=240 / keyint_min=23 / scenecut=40 / intra_refresh=0 / rc_lookahead=40 / rc=2pass / mbtree=1 / bitrate=1254 / ratetol=1.0 / qcomp=0.60 / qpmin=10 / qpmax=51 / qpstep=4 / cplxblur=20.0 / qblur=0.5 / ip_ratio=1.40 / aq=1:1.00
Language : English
Default : Yes
Forced : No
Color range : Limited
Color primaries : BT.601 NTSC
Transfer characteristics : BT.709
Matrix coefficients : BT.601

Audio
ID : 2
Format : AC-3
Format/Info : Audio Coding 3
Mode extension : CM (complete main)
Format settings, Endianness : Big
Codec ID : A_AC3
Duration : 45 min
Bit rate mode : Constant
Bit rate : 448 kb/s
Channel(s) : 6 channels
Channel positions : Front: L C R, Side: L R, LFE
Sampling rate : 48.0 kHz
Frame rate : 31.250 FPS (1536 spf)
Compression mode : Lossy
Stream size : 145 MiB (26%)
Language : English
Default : Yes
Forced : No

Text
ID : 3
Format : UTF-8
Codec ID : S_TEXT/UTF8
Codec ID/Info : UTF-8 Plain Text
Language : English
Default : No
Forced : No

MeGUI gives it the following file name:

Test - [0] English.ac3

Is the number in the name [0] supposed to indicate the number of the audio stream in the MKV?
When I extract audio from an MKV with 2 audio streams, the file indexer shows them as stream numbers 2 and 3, but they're extracted as streams 0 and 1. Is that expected behaviour?

Cheers.

[hello_hello]

You must have not chosen the correct PGC when you ran VSRip on your own. Judging by the 12mins start timecode, it appears you may have used the "play all" pgc, which led to your subs starting in the wrong time.

After I re-authored with DVDShrink there was definitely only one PGC.

There might be some 15milisecond chapter at the end of the vob, I'm not entirely sure what those are for but they are very common. They could sometimes cause the video to appear longer than it is.
If you play it though with mpc, what happens after 12mins? does it keep playing or stops?

The video is exactly 12 minutes and 13 seconds long (I opened it with DGIndex to check) and the last chapter is at 00:12:12.520. Aside from being annoying, that's not caused a playback problem before.
When playing the vob file with MPC-HC it jumps straight from 12 minutes 13 seconds to the "end" at about the 20 minute mark.

[Zathor]

The point is, that if I want the subs to be extracted in OCE, I would have to set them to be muxed. So it's either:

1) File indexer (load source ifo-->pgcdemux-->indexing), then go back to source, load ifo again and run vobsubber.

2) Use OCE to do all the above in one go but will have to mux the subs.
Sorry, I still do not get it :confused: Why do you have to mux the subtitles manually in 2)? They will be automatically muxed by OCE into the output file. Please explain in more detail what 2) is about.


MeGUI gives it the following file name:

Test - [0] English.ac3

Is the number in the name [0] supposed to indicate the number of the audio stream in the MKV?
When I extract audio from an MKV with 2 audio streams, the file indexer shows them as stream numbers 2 and 3, but they're extracted as streams 0 and 1. Is that expected behaviour?
[0] is the track number. Depending on the tool it may be the overall track number (then usually video comes first, then audio then subtitles) or the number of the individual type only (e.g. audio) - also it can happen that the index starts at 0 or 1.
So yes, this is expected and should not cause any issues.

[Zathor]

After I re-authored with DVDShrink there was definitely only one PGC.

Out of interest could you try it please directly with the DVD source (not shrinked, remuxed, stripped, ...)? I would like to know if then the same (wrong) timestamps will appear.

[mini-moose]

The video is exactly 12 minutes and 13 seconds long (I opened it with DGIndex to check) and the last chapter is at 00:12:12.520. Aside from being annoying, that's not caused a playback problem before.
When playing the vob file with MPC-HC it jumps straight from 12 minutes 13 seconds to the "end" at about the 20 minute mark.

Best bet would be what I mentioned about the extra chapter at the end. Looks like this:

https://s24.postimg.org/e8929lac5/cell.png

[mini-moose]

Sorry, I still do not get it :confused: Why do you have to mux the subtitles manually in 2)? They will be automatically muxed by OCE into the output file. Please explain in more detail what 2) is about.

I don't want to mux them (at least not always).
What I meant is if I want to do the indexing and subextracting in one go, I could then use OCE, but in order to extract them with it, they must be set to be muxed.

Many thanks for your patience. I'm just used to my old routine. It's possible I'm the only one bothered with those changes.

[hello_hello]

As you do not have enabled the "single output file" option, the files Toast1_VTS_01_1.* are not needed. I assume they have not been deleted as you have not enabled the option to remove intermediate files, correct?

Yes, the option is disabled. It's not possible to extract them with the correct name in the first place and avoid the intermediate copy? It's no big deal. I just wondered why it was happening.

No, I have to use vsrip for that. Could you please share the files with me so that I can check why they are shown? If they are empty (= no timestamps) MeGUI should delete them.

I'll have to do it a bit later (the real world calls) but did you want the extracted idx/sub files or the original ifo/vob files?

They certainly appear to be empty. I opened the extracted idx file with subresync and it didn't display anything. When I open the original ifo/vob files with Subtitle Edit, the option to only show forced subtitles is unavailable, so it doesn't think there are any.

Actually, looking at the log file again, MeGUI agrees, but they're not being deleted. Are the empty forced subtitles counted as intermediate files?

-[Information] Log for job33 (sub, VTS_01_0.IFO -> Test2_VTS_01_1.idx)
--[Information] [05/01/17 8:38:45 AM] Started handling job
--[Information] [05/01/17 8:38:45 AM] Preprocessing
--[Information] [05/01/17 8:38:45 AM] VobSub configuration file
---[NoImage] E:\Test2\VTS_01_0.IFO
---[NoImage] D:\Test2_VTS_01_1
---[NoImage] 1
---[NoImage] 1
---[NoImage] 0
---[NoImage] RESETTIME
---[NoImage] CLOSEIGNOREERRORS
---[NoImage] CLOSE
--[Information] [05/01/17 8:38:45 AM] Job command line: "C:\Program Files\MeGUI\tools\vsrip\vsrip.exe" "D:\Test2_VTS_01_1.vobsub"
--[Information] [05/01/17 8:38:45 AM] Process started
--[Information] [05/01/17 8:38:45 AM] Standard output stream
--[Information] [05/01/17 8:38:45 AM] Standard error stream
--[Information] [05/01/17 8:38:50 AM] Subtitle file created: Test2_VTS_01_1_0_English.sub
--[Information] [05/01/17 8:38:50 AM] VobSub configuration file
---[NoImage] E:\Test2\VTS_01_0.IFO
---[NoImage] D:\Test2_VTS_01_1_forced
---[NoImage] 1
---[NoImage] 1
---[NoImage] 0
---[NoImage] FORCEDONLY
---[NoImage] RESETTIME
---[NoImage] CLOSEIGNOREERRORS
---[NoImage] CLOSE
--[Information] [05/01/17 8:38:50 AM] Job command line: "C:\Program Files\MeGUI\tools\vsrip\vsrip.exe" "D:\Test2_VTS_01_1.vobsub"
--[Information] [05/01/17 8:38:50 AM] Process started
--[Information] [05/01/17 8:38:50 AM] Standard output stream
--[Information] [05/01/17 8:38:50 AM] Standard error stream
--[Information] [05/01/17 8:38:54 AM] no forced subtitles found
--[Information] [05/01/17 8:38:54 AM] MediaInfo
---[Information] [05/01/17 8:38:54 AM] File: D:\Test2_VTS_01_1.idx
---[Information] General
----[Information] Format:
----[Information] FormatString:
----[Information] FileSize: 13719
----[Information] PlayTime:
--[Information] [05/01/17 8:38:55 AM] Postprocessing
--[Information] [05/01/17 8:38:55 AM] Job completed

[mini-moose]

Yes, the option is disabled. It's not possible to extract them with the correct name in the first place and avoid the intermediate copy?
I don't love deleting working files either, as I believe this also deletes video-only encoded files (the video.mkv/mp4 that was encoded and then muxed). Sometimes one would want to remux for some reason but files are gone.

They certainly appear to be empty. I opened the extracted idx file with subresync and it didn't display anything. When I open the original ifo/vob files with Subtitle Edit, the option to only show forced subtitles is unavailable, so it doesn't think there are any.

Actually, looking at the log file again, MeGUI agrees, but they're not being deleted. Are the empty forced subtitles counted as intermediate files?

Empty subs are deleted for me.
Keep in mind that Forced subs could be for any of the languages. It might appear empty when you view the first stream on idx, but there could be forced for another language.

[Zetti]

FLAC v1.3.2 is released.

https://xiph.org/flac/changelog.html

[hello_hello]

Out of interest could you try it please directly with the DVD source (not shrinked, remuxed, stripped, ...)? I would like to know if then the same (wrong) timestamps will appear.

This is where my understand of the DVD structure is lacking. In re-author mode DVDShrink just displays title numbers, as does MPC-HC when View/Information is enabled. That's how I know which titles to re-author, by playing each episode with MPC-HC and noting the title numbers. How does everybody else do it, if they want to rip just a single episode, for example?

Anyway, I was able to find the correct PCG due to it's duration (DVDShrink and MPC-HC display it as Title 27). I ripped the subtitles directly from the disc with MeGUI, Unfortunately the timings were still wrong, but it must be something peculiar to this DVD. I don't think I've ever had this problem before.

https://s30.postimg.org/m9ds2j0bl/PCG9.gif

https://s30.postimg.org/nzwt40hup/PCG9b.gif

[mini-moose]

Zathor, I'm trying File Indexer with pgcdemux now.

Couple of questions:

1) It doesn't extract the chapters (another thing I always do with pre-extracting vobs on my own). The chapters are extracted on OCE though.

2) There is an option with DGIndex to choose which audios to keep, but it's greyed out (I think it's been like that for a long time?) :

https://s23.postimg.org/899ffb20b/pgc.png

OCE and subs:

You said I can add edit a profile to include more/all subs.
From what I can see, if I want to include all subs I need to add all the languages on the langs menu?

[hello_hello]

Zathor, I'm trying File Indexer with pgcdemux now.

Couple of questions:

1) It doesn't extract the chapters (another thing I always do with pre-extracting vobs on my own). The chapters are extracted on OCE though.

Yes, now the File Indexer is PGC aware, it'd be nice if it could be configured to extract all the required streams. Although as it uses DGIndex to extract the audio and index, I don't know how hard it'd be to implement.

2) There is an option with DGIndex to choose which audios to keep, but it's greyed out (I think it's been like that for a long time?) :

It's been greyed out for quite a while. I've wondered about that myself, but because I only ever keep the required audio streams when ripping/re-authoring, I generally don't need to select specific ones, so I've kept forgetting to mention it.

[Zathor]

FLAC v1.3.2 is released.
Thanks, uploaded

What I meant is if I want to do the indexing and subextracting in one go, I could then use OCE, but in order to extract them with it, they must be set to be muxed.
Ah, so you area searching for a way to only extract with OCE. Yes, this is not directly possible. As you mentioned you have to create a profile where video encoding is disabled & all audio processing is set to "never encode". That way everything will be extracted and muxed again - if you do not run the mux steps you have your extracted files.

Yes, the option is disabled. It's not possible to extract them with the correct name in the first place and avoid the intermediate copy?
If you have selected to use independent files for each track: no


I'll have to do it a bit later (the real world calls) but did you want the extracted idx/sub files or the original ifo/vob files? They certainly appear to be empty. I opened the extracted idx file with subresync and it didn't display anything. When I open the original ifo/vob files with Subtitle Edit, the option to only show forced subtitles is unavailable, so it doesn't think there are any. Actually, looking at the log file again, MeGUI agrees, but they're not being deleted. Are the empty forced subtitles counted as intermediate files?
Correct, so I do not need the files anymore.

I'm trying File Indexer with pgcdemux now.
1) It doesn't extract the chapters (another thing I always do with pre-extracting vobs on my own). The chapters are extracted on OCE though.
Correct, you have to use either OCE or the ChapterCreator for that. The FileIndexer is not a "demux everything tool".

2) There is an option with DGIndex to choose which audios to keep, but it's greyed out (I think it's been like that for a long time?) :
Thanks, was a regression of a recent change. Will be changed in 2701.

OCE and subs:
You said I can add edit a profile to include more/all subs.
From what I can see, if I want to include all subs I need to add all the languages on the langs menu?
Yes, or no selection at all (no entry, no even [none]) and then the option "languages to select if selection does not match" to all. But yes, an [all] option would make more sense.

[mini-moose]

Ah, so you area searching for a way to only extract with OCE. Yes, this is not directly possible. As you mentioned you have to create a profile where video encoding is disabled & all audio processing is set to "never encode". That way everything will be extracted and muxed again - if you do not run the mux steps you have your extracted files.
Exactly.

Tried OCE with encode phases disabled. Sadly, it won't work for what I wanted.

[nandoom]

I wonder if a thumbnail/screenshot creation tool ( configurable ) could be integrated in MeGUI. It would be a nice feature to have :)

Thanks

[hello_hello]

Is anyone here using MeGUI for 2 pass x264 encoding on a 64bit Windows?

There's a log file attached to a post at VideoHelp here (http://forum.videohelp.com/threads/382119-MeGUI-Only-Running-First-Pass?p=2473967&viewfull=1#post2473967) which seems to indicate the command line is broken. If you're around Zathor, could you please take a quick look?

Thanks.

[Zathor]

Is anyone here using MeGUI for 2 pass x264 encoding on a 64bit Windows?

There's a log file attached to a post at VideoHelp here (http://forum.videohelp.com/threads/382119-MeGUI-Only-Running-First-Pass?p=2473967&viewfull=1#post2473967) which seems to indicate the command line is broken. If you're around Zathor, could you please take a quick look?

It does work for me.

Sadly the person there does not post a full log including the versions stuff. Also it seems to me that the first pass for him/her fails with
---[Error] [15/01/2017 7:47:47 AM] x264 [error]: can't parse qpfile for frame 0
Therefore the first pass is detected as with errors and the second job wont start. See here for a potential solution:
https://forum.doom9.org/showthread.php?t=167736

[Zathor]

2702 [x265] added the cutree file to the intermediate file set. Bug #869
2701 [FileIndexer] fixed unable to select specific audio tracks to be demuxed from a DVD source with DG Indexer (regression of 2668)

[hello_hello]

Sadly the person there does not post a full log including the versions stuff. Also it seems to me that the first pass for him/her fails with
---[Error] [15/01/2017 7:47:47 AM] x264 [error]: can't parse qpfile for frame 0
Therefore the first pass is detected as with errors and the second job wont start. See here for a potential solution:
https://forum.doom9.org/showthread.php?t=167736

Yeah it turned out to be caused by the qpfile error. I thought the first pass had run though and there was a problem with the second pass and it didn't appear to be related to the qpfile problem. I guess I was wrong about that.

Thanks.

[kuchikirukia]

Taking a look at the "Add pre-rendering job", it looks like ffmpeg is running the huffyuv encode in the same thread as Avisynth. (opening in process explorer, I don't see anything that looks like a separate thread.) Since Avisynth is single-thread limited, can the encode be split out into its own thread?

[LigH]

Only if ffmpeg can create separate threads for reading the input and encoding it. AviSynth will then be used by ffmpeg, not by MeGUI.

The ffmpeg option "-threads auto" should enable multithreading in general.

[kuchikirukia]

Only if ffmpeg can create separate threads for reading the input and encoding it. AviSynth will then be used by ffmpeg, not by MeGUI.

The ffmpeg option "-threads auto" should enable multithreading in general.

Just looked at the command line and it's set to -threads 0, which I take it should do it. So I guess it's unsupported?

[Information] [1/23/2017 1:02:48 AM] Job command line: "C:\Users\Documents\MeGUI\tools\ffmpeg\ffmpeg.exe" -y -i "C:\Users\Documents\4.avs" -c:v ffvhuff -threads 0 -sn -an -context 1 -vstrict -2 -pred 2 "C:\Users\Documents\hfyu_4.avi"

I guess I'll be running some test later to see if lossless x264 works better.

[kalehrl]

Has anyone noticed that MeGUI has become slower when creating and saving avisynth scripts?
There is a delay of 6, 7 seconds during which time MeGUI becomes unresponsive and you can't click any button.

[Taurus]

Has anyone noticed that MeGUI has become slower when creating and saving avisynth scripts?
There is a delay of 6, 7 seconds during which time MeGUI becomes unresponsive and you can't click any button.

No!:o
Just out of curiosity:
Which indexer did you use?
Tried all available, just the normal pause at the indexing stage.
Avisource, directshowsource and dss2 should load almost instantly.

[hello_hello]

Has anyone noticed that MeGUI has become slower when creating and saving avisynth scripts?
There is a delay of 6, 7 seconds during which time MeGUI becomes unresponsive and you can't click any button.

Yeah.... if you're referring to loading them into the video section and adding them to the job queue. At one stage Zathor removed some of the checking MeGUI does (I believe) and it became faster, but for some reason it appears to be getting progressively slower again, especially when there's slow filtering such as QTGMC in a script, and for whatever reason it seems to be slower when L-Smash is used for indexing than when ffms2 is used, which is kind of annoying given ffms2 and the script creator don't play nice together. Six or seven seconds is probably longer then the delay I experience, but it'd be several seconds.... however many that is. ;)

A couple of times I've loaded a script into the video section and clicked the Queue button before MeGUI had finished whatever checking it does, and ended up with the previous script added to the job queue only with the name of the current script as the output file.... or something like that.... but I learned to be a bit more patient after that did my head in.

[LigH]

The index files are in different formats, I believe (one binary, one in plain text), thus have different sizes. Also don't forget, one of them may only index audio streams when audio is indexed first, the other instead may always; but I can't keep in mind which does what.

[hello_hello]

MeGUI tells ffms2 to only index the video when there's no audio selected in the file indexer or when the audio is being extracted. L-Smash has no such option and always indexes each stream.

[LouieChuckyMerry]

A couple of times I've loaded a script into the video section and clicked the Queue button before MeGUI had finished whatever checking it does, and ended up with the previous script added to the job queue only with the name of the current script as the output file.... or something like that.... but I learned to be a bit more patient after that did my head in.

The same type of thing has happened to me in the past, so I was forced to learn (at least a bit of) patience. Thanks MeGUI!

[kalehrl]

The biggest "freezes" are when saving avisynth scripts and queueing video files to be encoded.

[StainlessS]

So long as AVS files exist, and nothing special is required in MeGUI, could try MeGUI AutoEncode Batcher,
just select a bunch of existing avs in FileSelector, click OK button and go to the pub, no patience required (unless you order Guiness).

here:- https://forum.doom9.org/showthread.php?t=163343&highlight=Megui+autoencode+Batcher

EDIT: In avs scripts, can use eg RT_Stats RT_SignalDar(4.0/3.0) or RT_SignalDar2(16,9) to auto set
Display aspect Ratio in avs and pass it on to MeGUI without having to set manually in MeGUI.

EDIT: From RT_Stats Docs

RT_SignalDAR2(int darx,int dary)
As for RT_SignalDAR() except it allows setting of numerator and denominator individually.
Signal DAR to MEGUI. Sets Global vars MeGUI_darX and MeGUI_darY which are read during MEGUI loading of AVS file.

[Zathor]

MeGUI has to check the AVS file and has therefore to open it in several internal steps. If you have a script which need a long time for that, you will have those delays.

[kalehrl]

There aren't any heavy filters, just source, resize, minimal noise denoise.
I remember MeGUI being faster in the recent past.

[LigH]

Well possible that MeGUI now ships more (and more complex) plugins which are scanned for functions each time a script is opened (and if your AV guard is active, it may scan their DLLs as well)...

[hello_hello]

(MeGUI 2702)
I disabled my Avisynth auto-loading plugin folder and indexed a 720p video twice (1280x536). Nothing fancy. Something I encoded myself with the x264 Slow preset and Tune Film.

Script one:
LoadPlugin("C:\Program Files\MeGUI\tools\lsmash\LSMASHSource.dll")
LWLibavVideoSource("D:\Video.mkv.lwi")

Script two:
LoadPlugin("C:\Program Files\MeGUI\tools\ffms\ffms2.dll")
FFVideoSource("E:\video.mkv", cachefile="D:\video.mkv.ffindex", fpsnum=24000, fpsden=1001, threads=1)

I didn't time it exactly..... 1 Mississippi, 2 Mississippi..... but the first script takes around three seconds to load into the video section (preview disabled) and it's repeatable. That's the time from double clicking on the script after opening the "Open Avisynth Script" window, to the cursor changing from busy back to a pointy thing. During that time Task Manager said MeGUI's memory usage went from just under 50000k to about 75000k and back (I'd been running some encodes, but they'd finished). CPU usage hits 25%, so I guess that's a single core of my old Q9450 working hard.

Script 2 takes CPU usage to 6%, if memory usage increases it decreases again too quickly for the human eye, and it loads too fast for me to time.

So at least I wasn't imagining the L-Smash thing, which gets noticeably worse with fancy Avisynth filtering, but maybe someone else who isn't running XP and the last XP friendly version of L-Smash can try the same thing to see if the result is similar. If it is, I don't know if MeGUI or L-Smash would be to blame. I'd assume the latter, but what do I know....

Edit: I enabled the auto preview when loading scripts and tried again. Around 5 seconds for the preview to display after loading the first script, and something approaching one second for the second script.

And one last look-see, a quick encode of the first 10000 frames using each script.

Script 1:
[Information] [08/02/17 9:02:13 PM] encoded 10000 frames, 40.57 fps, 1934.23 kb/s

Script 2:
[Information] [08/02/17 9:07:29 PM] encoded 10000 frames, 41.01 fps, 1934.23 kb/s

So it's possibly not a decoding speed issue, although I think it is time for a faster computer. Hurry up AMD, I want to see what Zen is like first.

[Zathor]

Do you have the same loading times with these scripts in e.g. VirtualDub?

[hello_hello]

I can't check that properly today as I'm in the middle of a run of encodes, but I'll test and report back tomorrow(ish),

[hello_hello]

Do you have the same loading times with these scripts in e.g. VirtualDub?

I guess nobody wants to test a newer L-Smash, but yes..... it's slower loading a script using LWLibavVideoSource with VirtualDub than with FFVideoSource. Maybe not quite as much difference in speed, but I assume VirtualDub doesn't have any checking to do.

[nandoom]

Forgive me if this has been answered before. I wonder if it is possible to add AvsP plugin in MeGUI so one can able to edit an avs? I think this will be an awesome addition.

Thanks

[hello_hello]

Forgive me if this has been answered before. I wonder if it is possible to add AvsP plugin in MeGUI so one can able to edit an avs? I think this will be an awesome addition.

What about copying and pasting the script into MeGUI's script creator?
You'll have to specify the input video first, but you can use Lsmash and DGindex index files for the input video, or ffms2 index files if they're in the same folder as the source, so you don't have to index again, and once the video has opened you can copy and paste the rest of the script into the script creator to edit it.

Another method is to load the script into the video section and open the preview. Edit the script with Notepad, save the changes, and click the reload button at the bottom of the preview window to see them. You can keep Notepad and the preview open as you do it. No need to close either till you're done.

[nandoom]

Yes, those are the ways that I know to edit an avs file. However it requires you to manually edit ( not bad thing at all ) and you got to know the right text for any new arguments you might want to add. The script creator doesn't populate the script parameters in the respective tabs after pasting the script. The AvsP offers more controls, options, tools. So I just thought the AvsP integration would be an awesome feature to add on. I understand that it's easier to say than done. Of course, the developers would better and would know the feasibility of the integration. So I thought to ask away.

[StainlessS]

Display Aspect Ratio bug.


Avisource("Test.avi")
Trim(1000,-200)
BilinearResize(720,480)
MeGUI_darX = 16 MeGUI_darY = 9
return last


MeGUI Development Output file DAR 16:10 (ERROR)
MeGUI Stable Output file DAR 16:9

Logs + MediaInfo for both:- http://www.mediafire.com/file/8fipgotylx1mp8w/MeGUI_DAR_Error.zip

EDIT: Both versions MeGUI current.

EDIT:


[Information] Log
...

----[Information] Format: AVS
----[Information] AspectRatio: 16:9
----[Information] AspectRatioString:

...

-[Information] Log for job2 (video, _TEST.avs -> _TEST-Video.264)
--[Information] [18/02/2017 09:32:02] Started handling job
--[Information] [18/02/2017 09:32:02] Preprocessing
--[Information] [18/02/2017 09:32:02] Avisynth input script
---[NoImage] Avisource("Test.avi")
---[NoImage] Trim(1000,-200)
---[NoImage] BilinearResize(720,480)
---[NoImage] MeGUI_darX = 16 MeGUI_darY = 9
---[NoImage] return last
--[Information] [18/02/2017 09:32:03] resolution: 720x480
--[Information] [18/02/2017 09:32:03] frame rate: 24/1
--[Information] [18/02/2017 09:32:03] aspect ratio: 16:9 (1.778)
--[Information] [18/02/2017 09:32:03] Job command line: "C:\NON-INSTALL\MeGUI\tools\x264\x264.exe" --profile main --preset fast
\ --crf 21.5 --keyint 240 --qpmin 10 --qpmax 51 --sar 16:15 --output "C:\Z\_TEST-Video.264" "D:\_TEST.avs"
--[Information] [18/02/2017 09:32:03] Process started
--[Information] [18/02/2017 09:32:03] Standard output stream
--[Information] [18/02/2017 09:32:03] Standard error stream
---[Information] [18/02/2017 09:32:04] avs [info]: 720x480p 16:15 @ 24/1 fps (cfr)
---[Information] [18/02/2017 09:32:04] x264 [info]: using SAR=16/15

...

----[Information] Width: 720
----[Information] Height: 480
----[Information] FrameCount: 200
----[Information] FrameRate: 24.000
----[Information] FrameRateOriginal:
----[Information] FrameRateMode:
----[Information] ScanType: Progressive
----[Information] Codec: AVC
----[Information] CodecString: AVC
----[Information] Bits Depth: 8
----[Information] Format: AVC
----[Information] AspectRatio: 1.600
----[Information] AspectRatioString: 16:10
----[Information] Delay:



EDIT: Can someone else reproduce and verify above results. (should get same with ColorBars.ConvertToYV12.BilinearResize etc).

EDIT: Eg

Global MeGUI_darX = 16 Global MeGUI_darY = 9
return ColorBars(Pixel_Type="YV12",Width=720,height=480).Trim(0,-200)

Should have set the MeGUI_darX and MeGUI_darY as Globals, but makes no difference to ERROR results.

[Zathor]

Thanks for the report, StainlessS. It should be fixed:

2706 [x264] fixed wrong SAR values in some cases (regression of 2683)
[x265] added 16:15 to the predefined --sar values
2705 [HD Streams Extractor] improved input/output folder selection
2704 [ChapterCreator] removed an internal md5 hash function to be FIPS compliant. #Bug 873

[hello_hello]

MeGUI 2706 is still displaying incorrect resizing in the script creator when anamorphic encoding is enabled.

- Open a 4:3 PAL DVD, Input DAR 4:3.
- Without cropping or resizing enabled, the resize section displays 720x576.
- Enable cropping and crop 20 pixels from both left and right, and 16 pixels from both top and bottom.
- The remaining video is 680x544
- Enable anamorphic encoding, and select "encode non-mod16" (silly name). The resizing now displays 640x512. It's not being added to the script incorrectly though.
- The "resize to selected mod" option at least lets you resize it back to 680x544, and the resizing doesn't change to red until it exceeds 680x544, so that's okay.

Oddly, if the "resize to selected mod" option is selected, and you switch to "encode non-mod16", the width in the resizing field very briefly displays 680, before it changes back to 640 again.

[Sharc]

Never mind, I am just occasionally asking myself: Is it really more difficult to understand resizing basics and do the resizing manually, writing your own simple script, or try to understand, hangle through and trust all the tickboxes, tabs, options and windows of advanced GUIs ..... rant closed :devil:

[hello_hello]

Never mind, I am just occasionally asking myself: Is it really more difficult to understand resizing basics and do the resizing manually, writing your own simple script, or try to understand, hangle through and trust all the tickboxes, tabs, options and windows of advanced GUIs ..... rant closed :devil:

I find it is, especially if you're encoding a bunch of related videos with different cropping required, but you want them all cropped neatly and resized to the same resolution and/or aspect ratio.
Often I'll crop a section of a video differently, and the aspect error calculator makes it easy to crop cleanly and resize to the same resolution as the rest of the video.

Plus.... for an anamorphic output, even if you don't resize, if you want exactly 4:3 or 16:9 you can switch to square pixel resizing, crop and resize to exactly 4:3 or 16:9, then switch to anamorphic encoding and you'll know the cropping is correct to still give you a 4:3 or 16:9 output. Saves having to do pixel aspect ratio calculations.
It's also easy to create custom input display aspect ratios and when you change the input DAR, all the usual calculations are based on that.

Having said all that, I recently created an Avisynth script function (my first) for auto-cropping and resizing based on the cleverness AutoGK was endowed with, only allowing full user control over the output aspect ratio. It not overly complicated and seems to work as intended so far.
CropResize Script (http://forum.videohelp.com/threads/382601-CropResize-Script)

MeGUI's cropping and resizing is working fine, but for some reason when resizing is disabled it displays an incorrect post cropping resolution in the resize fields when anamorphic encoding without resizing is enabled.

[hello_hello]

Someone found a bug in the latest flac. https://sourceforge.net/p/flac/bugs/453/

In case it'd be an idea to update MeGUI's version, there's a patch on the above sourceforge page, or there's a patched exe in the Foobar2000 Encoder Pack 2017-02-13.
http://www.foobar2000.org/encoderpack

[hello_hello]

There appears to be an odd encoding speed bug when encoding audio with the FDKAAC encoder. I haven't been able to find the cause as yet, but the details are in the post linked to below. I strongly suspect it's something to do with the encoder not handling 32 bit input well. So far, if I add ConvertAudioTo32bit() to the script and open it with foobar2000, FDKAAC encoding speed drops dramatically, but QAAC still encodes at the normal speed.

I'm not sure if the same applies to 32 bit float though because the foobar2000 AVS input plugin doesn't seem to accept it, but if I remember correctly, FDKAAC downsamples to 16 bit before encoding, in which case it'd probably be an idea to convert the audio to 16 bit for FDKAAC encoding anyway.

https://forum.doom9.org/showthread.php?p=1799699#post1799699

[dissory]

I've been using MeGUI and found the one-click profile really helpful. One thing I noticed was that when transcoding audio e.g. FLAC -> AAC within video files, it doesn't retain attributes such as audio title name or language in the new video file with the transcoded audio. I'm not sure if the same happens with video but it does with audio.

Is it possible to make it so the attributes/metadata of the original audio stream are also transferred to the transcoded stream?

[stax76]

This could be interesting: https://youtu.be/AEFQL1LM-wI?t=156