MeGUI: bug reports and feature requests

[Anteys]

Tried to change back to the stable version, since the development server wasn't available recently and the updater wanted to downgrade a few things.

Switching back to the development server didn't help either, it still shows older versions as an update.

http://i.imgur.com/Dl7m4xH.jpg

[LigH]

The first development update server should instead be: http://megui.org/auto/

Maybe try to use this as custom update server once. Or wait for an affirmative suggestion by developers of MeGUI...

[hello_hello]

For Generic AR, like in MPEG-2 video, it is indeed relevant: If there is e.g. a 16:9 DAR flag (so the PAL target resolution will be 576 lines * 16:9 = 1024 pixels width per line, in any case), there are different PARs (display width : encoded width) depending on the encoded width (e.g. 1024:704 as well as 1024:720, both encoded widths are allowed on DVD). And that's only on DVD. For DVB (SD) there are also widths of 528 pixels.

You asked if it's relevant, now you're saying it is. When you've decided..... ;)

I don't understand how any of that makes a difference to what I originally said. MeGUI has a setting in it's options telling it whether to default to an ITU input aspect ratio for DVD video etc. When it's not checked, it defaults to a straight 16:9 or 4:3 aspect ratio, but it can be changed by the user in the script creator when setting up an encode. I have it set to non-ITU by default, but I change it to ITU if I think that's wrong.

All I'm wondering, is when MeGUI or the user decides the video in question is ITU, if it'd be better for MeGUI to use the mpeg4 PARs instead of the PARs it's using now. If it used the mpeg4 PARs, it'd be consistent with the PARs you can set in the x264 encoder configuration and it'd possibly be more accurate. Even if it's only a fraction of a pixel more accurate, please explain your objection to that, because so far I'm not understanding why you'd have one.

Even Wikipedia seems to be saying the ITU PARs MeGUI uses are a little wrong.
http://en.wikipedia.org/wiki/Pixel_aspect_ratio#Pixel_aspect_ratios_of_common_video_formats
I haven't checked lots of other encoder GUI's yet but I did look at StaxRip and it definitely defaults to the mpeg4 PARs for ITU resizing. I suspect MeGUI is one of the few remaining encoder GUI's which doesn't.

Cropping has nothing to do with it is as such. If you're using anamorphic encoding you pick ITU or non-ITU for the input aspect ratio, crop and encode. The PAR doesn't change from there (well there's one anamorphic option which also resizes but that aside).

After all, you are concerned about a difference of less than 4 pixels after resizing when upscaling (which has to be enabled manually). MOD 4 is already a minimum for some AviSynth filters due to SSE optimizations. The AR error for even trying to achieve a traditional MOD 16 resolution (for optimal macroblock encoding) is most probably worse.

None of which is an argument for not trying to use the correct input aspect ratio to begin with.
I don't know how long it's been since you've used MeGUI but it calculates the aspect error, so even when cropping and resizing to mod16 it's pretty easy to adjust the cropping and/or resizing for minimal aspect ratio distortion.
The idea was simply to use a PAR that might be more accurate and consistent with the PARs in the x264 encoder configuration as a starting point..... no matter what resizing, aspect ratio distortion or Avisynth filter pandering takes place in the script from there.

[Xennex83]

I want to express my anger at the creators of MeGUI and/or AVISynth.

This products hardly works out of the box.

* One Click Encode can only encode towards MKV. Most of the time, my source is MKV and I want to go somewhere else. MeGUI seems to be one of the few good programs that can even work with MKV.
* MKV to AVI using Xvid doesn't work. I am spending hours just researching a problem I can never tackle on my own, because the inner workings are too complicated.

One the second point, the video encoder will give me an error that the AVS script cannot be opened. When using H??? this never happens. When searching for solutions on the web, everything is way too hard to understand for a novice. Randomly trying suggestions is a real pain and never solves anything.

It didn't work when I used that "AVC Source" on an extracted h264 file, and it doesn't work when I use a FFVideoSource on the original MKV. This is not a troubleshooting request. Please don't respond with troubleshooting suggestions. I am going to be trying another, perhaps even commercial application, just so I can convert a MKV to something else.

Good riddance, I guess. Bye.

[AMED]

Before you rant like this how about asking for help first. There are many MeGUI users here and if you had posted the MeGUI log they would have helped pinpoint the problem.

There are many many other encoding GUIs out there just dig through this sub forum and try them out.

[LigH]

Ignoring all the anger, just a few points:

a) MeGUI, despite a long time of develoment, may still not be made for people with little patience and little background knowledge. In general.

b) If you expect One Click Encode to combine both a minimum of usage efforts and a maximum of configuration freedom, you certainly misunderstand its purpose. By the way, the output format is indeed configurable...

c) I remember another report that xvid_encraw is unable to process an AviSynth script which is working in other tools... so there must be a specific reason.

[Zathor]

Sorry, I will not spent more time with such a post then I already did. Wild guessing will not help here. Also the user does not want any help from us so let him/her go in peace.

Back to other topics:
- Yes, the second mirror of the MeGUI update server is offline for a while now (as it was me - sadly I am back with such a post directly thinking to go offline again). I will provide a "quick fix" with a small update only adressing that soon.
- There are still some other problems mentioned which I would like to address. If there are any important problems please mention them again here. Otherwise I possibly will not find them.

A very quick fix:
2513 [Update] removed update server megui.xvidvideo.ru completly
[Update] increased update data refresh interval to 2 days for the development server

[hello_hello]

- There are still some other problems mentioned which I would like to address. If there are any important problems please mention them again here. Otherwise I possibly will not find them.

Overdrive80 re-reported (http://forum.doom9.org/showthread.php?p=1699600#post1699600) the ffms2/AVI problem causing video motion not to be smooth.
While I think adding ffms2's frame rate conversion to the script is generally a good idea, if it causes a problem with AVIs......

Neither ffms2 or L-Smash seem to handle null frames in AVIs correctly (ffms2 without it's frame rate conversion added to the script). They both appear to ignore null frames and output an incorrect frame rate as a result.
AVISource handles them correctly, replacing null frames with duplicate frames and outputting the correct frame rate. Is AVISource not reliable? I'm not sure I've ever had an issue with it and at least the decoded video should be smooth. I guess the next best fix would be to not add ffms2's frame rate conversion to the script for AVIs, or make L-Smash the default indexer for AVIs instead.

[tebasuna51]

Please update the BassAudio package from BeHappy-0.2.6.34898e or here: https://www.sendspace.com/file/vwodyl

[Zathor]

Please update the BassAudio package from BeHappy-0.2.6.34898e or here: https://www.sendspace.com/file/vwodyl
Thx, updated.

Overdrive80 re-reported (http://forum.doom9.org/showthread.php?p=1699600#post1699600) the ffms2/AVI problem causing video motion not to be smooth.
While I think adding ffms2's frame rate conversion to the script is generally a good idea, if it causes a problem with AVIs......
I still have no clue what to do here. Sadly I do not have the time required for all the tests. I will compile a test build.

Neither ffms2 or L-Smash seem to handle null frames in AVIs correctly (ffms2 without it's frame rate conversion added to the script). They both appear to ignore null frames and output an incorrect frame rate as a result.
AVISource handles them correctly, replacing null frames with duplicate frames and outputting the correct frame rate. Is AVISource not reliable? I'm not sure I've ever had an issue with it and at least the decoded video should be smooth. I guess the next best fix would be to not add ffms2's frame rate conversion to the script for AVIs, or make L-Smash the default indexer for AVIs instead.

The reason to not use AviSource is that it relys on installed filters. In the past there where several issue reports where AviSource was not able to open a file.

[hello_hello]

I still have no clue what to do here. Sadly I do not have the time required for all the tests. I will compile a test build.

It's an ffms2 problem, not a MeGUI problem as such, so disabling ffms2's frame rate conversion (or not adding it to the script) when indexing AVIs might be the only solution. Assuming that's possible. Or, as much as I tend to think it's a good idea in theory, maybe not adding ffms2's frame rate conversion to any script might have to be the solution (if it can't be disabled only for AVIs).

I've been playing around with an old AVI (created by AutoGK). I opened it with ffms2 and MPC-HC and let it run. It's 25fps. Mostly it was fine, but in a couple of places motion gets really "jittery" for a few seconds. Stepping through the frames one at a time you can see frames are being repeated which means others are being dropped. Removing "fpsnum=25, fpsden=1" from the script fixes it. I've tried several ffms2 versions but it always happens in the same place. According to ffdshow the frame timestamps are all normal, all exactly 40ms apart. It's a mystery to me.

[Zathor]

Test build (http://megui.org/test/MeGUI.zip)

Really quick and dirty... open an ffindex file in th AVS Script Creator (or anywhere else) and in the log you will get in the "File Indexer" section a few new entries. This is from my sample file (regarding mediainfo 29.976fps & 10220 frames):

-[NoImage] LoadCPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
-[NoImage] FFVideoSource("D:\MeGUI\_SAMPLE_VIDEOS\VTS_01_5.VOB", fpsnum=30000, fpsden=1001, threads=1)
---[Information] [16.12.2014 23:17:03] num_frames:10226
---[Information] [16.12.2014 23:17:03] rated:1001
---[Information] [16.12.2014 23:17:03] raten:30000

-[NoImage] LoadCPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
-[NoImage] FFVideoSource("D:\MeGUI\_SAMPLE_VIDEOS\VTS_01_5.VOB", threads=1).AssumeFPS(30000,1001)
---[Information] [16.12.2014 23:17:04] num_frames:8266
---[Information] [16.12.2014 23:17:04] rated:1001
---[Information] [16.12.2014 23:17:04] raten:30000

-[NoImage] LoadCPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
-[NoImage] FFVideoSource("D:\MeGUI\_SAMPLE_VIDEOS\VTS_01_5.VOB", threads=1, rffmode=1).AssumeFPS(30000,1001)
---[Information] [16.12.2014 23:17:05] num_frames:20450
---[Information] [16.12.2014 23:17:05] rated:1001
---[Information] [16.12.2014 23:17:05] raten:30000

-[NoImage] LoadCPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
-[NoImage] FFVideoSource("D:\MeGUI\_SAMPLE_VIDEOS\VTS_01_5.VOB", threads=1)
---[Information] [16.12.2014 23:17:07] num_frames:8266
---[Information] [16.12.2014 23:17:07] rated:1001
---[Information] [16.12.2014 23:17:07] raten:24250

-[NoImage] LoadCPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
-[NoImage] FFVideoSource("D:\MeGUI\_SAMPLE_VIDEOS\VTS_01_5.VOB", threads=1, rffmode=1)
---[Information] [16.12.2014 23:17:08] num_frames:20450
---[Information] [16.12.2014 23:17:08] rated:1001
---[Information] [16.12.2014 23:17:08] raten:60000

First one is the currently used ffms command and 4 additional commands with the AVS report of frame numbers and FPS (raten/rated). Maybe my source is broken but only the first one is giving correct results (IVTC required).

[LigH]

Using a VOB file, and not the first in a VTS, makes it almost certain to start with a broken GOP. Almost, because, well, there may be DVD multiplexers which split a VTS at GOP limits on purpose. Many others try to split as close as possible below 1 GB.

In your case, I wonder if MediaInfo respected a "Soft Telecine" flag while reporting the frame rate and number. So the encoded frame rate is NTSC-Film, to be presented as NTSC.

[hello_hello]

Are we looking for the same problem? The AVI I was playing with yesterday..... it's 25fps and "fpsnum=25, fpsden=1" is added to the script. The frame count and video duration displayed in MeGUI's preview don't change when fpsnum and fpsden are removed.

What's happening, for example, when there's no frame rate conversion in the script:

frame 6234 = frame A
frame 6235 = frame B
frame 6236 = frame C

With the frame rate conversion:

frame 6234 = frame A
frame 6235 = frame A
frame 6236 = frame C

And the motion gets "jittery". It's repeatable in exactly the same place using the current ffms2 version and two older versions, and the same frames are incorrectly being replaced by a duplicate of the previous frame each time. It's kind of like ffms2 is sometimes incorrectly treating real frames as null frames, or something similar.

So far, I've only been able to reproduce it when the source file is AVI. I usually remux AVIs as MKVs before re-encoding and as it happens I've re-encoded a bunch of them over the last few days and not seen a similar problem, and I've watched quite a few after encoding. The codec or frame rate don't seem to be a factor. In the case of the AVI I referred to previously, probably 98% of it plays smoothly. It's only jittery in a few places. It's got to be some sort of ffms2 problem. I can try uploading a sample later if you need one, but when I tried saving the section of the AVI in question as a new AVI (VirtualDubMod again), the new sample played smoothly after indexing, so I might need to mess around a bit on order to create a small one.

Log file from the test build:
[Information] Log
-[Information] Versions
--[Information] [17/12/14 6:16:35 PM] MeGUI: 2513
--[Information] [17/12/14 6:16:35 PM] Operating System: Windows XP Professional x86 SP3 (5.1.196608.2600)
--[Information] [17/12/14 6:16:35 PM] .Net Framework: 2.2.30729
--[Information] [17/12/14 6:16:35 PM] .Net Framework: 4.0.30319
--[Information] [17/12/14 6:16:35 PM] AviSynth: 2.6.0.4 (18-09-2013) (inactive)
--[Information] [17/12/14 6:16:35 PM] AviSynth+ portable: 2.6.0.5 (02-01-2014) (active)
--[Information] [17/12/14 6:16:35 PM] AvisynthWrapper: (02-01-2009)
--[Information] [17/12/14 6:16:35 PM] Haali Matroska Splitter: 1.13.138.14 (14-04-2013)
--[Information] [17/12/14 6:16:35 PM] Haali DSS2: (14-04-2013)
--[Information] [17/12/14 6:16:35 PM] ICSharpCode.SharpZipLib: 0.85.5.452 (07-08-2008)
--[Information] [17/12/14 6:16:35 PM] LinqBridge: 1.0.0.0 (27-05-2009)
--[Information] [17/12/14 6:16:35 PM] MediaInfo: 0.7.71.0 (09-11-2014)
--[Information] [17/12/14 6:16:35 PM] MediaInfoWrapper: 0.7.61.0 (06-01-2013)
--[Information] [17/12/14 6:16:35 PM] MessageBoxExLib: 1.0.2218.28317 (19-12-2008)
--[Information] [17/12/14 6:16:35 PM] SevenZipSharp: 0.64.3890.29348 (02-01-2011)
--[Information] [17/12/14 6:16:35 PM] 7z: 9.20 (18-11-2010)
-[Information] Update detection
--[Information] [17/12/14 6:16:39 PM] Connected to server: http://megui.org/auto/
--[Information] [17/12/14 6:16:39 PM] No package requires an update
-[Information] FileIndexer
--[Information] [17/12/14 6:17:35 PM] MediaInfo
---[Information] [17/12/14 6:17:35 PM] File: E:\test.avi
---[Information] General
----[Information] Format: AVI
----[Information] FormatString: AVI
----[Information] FileSize: 387999744
----[Information] PlayTime: 00:23:39.040
---[Information] Video
----[Information] ID: 0
----[Information] StreamOrder: 0
----[Information] Width: 640
----[Information] Height: 480
----[Information] FrameCount: 35476
----[Information] FrameRate: 25.000
----[Information] FrameRateOriginal:
----[Information] FrameRateMode:
----[Information] ScanType: Progressive
----[Information] Codec: XVID
----[Information] CodecString: XviD
----[Information] Bits Depth: 8
----[Information] Format: MPEG-4 Visual
----[Information] AspectRatio: 1.333
----[Information] AspectRatioString: 4:3
----[Information] Delay: 0
----[Information] Title:
----[Information] Language:
----[Information] LanguageString:
----[Information] Default:
----[Information] DefaultString:
----[Information] Forced:
----[Information] ForcedString:
---[Information] Audio
----[Information] ID: 1
----[Information] StreamOrder: 1
----[Information] Format: AC-3
----[Information] FormatProfile:
----[Information] FormatSettingsSBR:
----[Information] FormatSettingsPS:
----[Information] SamplingRate: 48000
----[Information] SamplingRateString: 48.0 KHz
----[Information] Channels: 2
----[Information] ChannelsString: 2 channels
----[Information] ChannelPositionsString2: 2/0/0
----[Information] BitRateMode: CBR
----[Information] Delay: 0
----[Information] Title:
----[Information] Language: en
----[Information] LanguageString: English
----[Information] Default:
----[Information] DefaultString:
----[Information] Forced:
----[Information] ForcedString:
--[Information] [17/12/14 6:17:56 PM] AVS
---[NoImage] LoadPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
---[NoImage] FFVideoSource("E:\test.avi", cachefile="D:\test.avi.ffindex", fpsnum=25, fpsden=1, threads=1)
---[Information] [17/12/14 6:17:56 PM] num_frames:35476
---[Information] [17/12/14 6:17:56 PM] rated:1
---[Information] [17/12/14 6:17:56 PM] raten:25
--[Information] [17/12/14 6:17:57 PM] AVS
---[NoImage] LoadPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
---[NoImage] FFVideoSource("E:\test.avi", cachefile="D:\test.avi.ffindex", threads=1).AssumeFPS(25,1)
---[Information] [17/12/14 6:17:58 PM] num_frames:35476
---[Information] [17/12/14 6:17:58 PM] rated:1
---[Information] [17/12/14 6:17:58 PM] raten:25
--[Information] [17/12/14 6:17:59 PM] AVS
---[NoImage] LoadPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
---[NoImage] FFVideoSource("E:\test.avi", cachefile="D:\test.avi.ffindex", threads=1, rffmode=1).AssumeFPS(25,1)
---[Information] [17/12/14 6:17:59 PM] num_frames:35476
---[Information] [17/12/14 6:17:59 PM] rated:1
---[Information] [17/12/14 6:17:59 PM] raten:25
--[Information] [17/12/14 6:18:00 PM] AVS
---[NoImage] LoadPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
---[NoImage] FFVideoSource("E:\test.avi", cachefile="D:\test.avi.ffindex", threads=1)
---[Information] [17/12/14 6:18:00 PM] num_frames:35476
---[Information] [17/12/14 6:18:00 PM] rated:1
---[Information] [17/12/14 6:18:00 PM] raten:25
--[Information] [17/12/14 6:18:02 PM] AVS
---[NoImage] LoadPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
---[NoImage] FFVideoSource("E:\test.avi", cachefile="D:\test.avi.ffindex", threads=1, rffmode=1)
---[Information] [17/12/14 6:18:02 PM] num_frames:35476
---[Information] [17/12/14 6:18:02 PM] rated:1
---[Information] [17/12/14 6:18:02 PM] raten:25
--[Information] [17/12/14 6:18:04 PM] AVS
---[NoImage] LoadPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
---[NoImage] FFVideoSource("E:\test.avi", cachefile="D:\test.avi.ffindex", fpsnum=25, fpsden=1, threads=1)
---[Information] [17/12/14 6:18:04 PM] num_frames:35476
---[Information] [17/12/14 6:18:04 PM] rated:1
---[Information] [17/12/14 6:18:04 PM] raten:25
--[Information] [17/12/14 6:18:05 PM] AVS
---[NoImage] LoadPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
---[NoImage] FFVideoSource("E:\test.avi", cachefile="D:\test.avi.ffindex", threads=1).AssumeFPS(25,1)
---[Information] [17/12/14 6:18:05 PM] num_frames:35476
---[Information] [17/12/14 6:18:05 PM] rated:1
---[Information] [17/12/14 6:18:05 PM] raten:25
--[Information] [17/12/14 6:18:06 PM] AVS
---[NoImage] LoadPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
---[NoImage] FFVideoSource("E:\test.avi", cachefile="D:\test.avi.ffindex", threads=1, rffmode=1).AssumeFPS(25,1)
---[Information] [17/12/14 6:18:06 PM] num_frames:35476
---[Information] [17/12/14 6:18:06 PM] rated:1
---[Information] [17/12/14 6:18:06 PM] raten:25
--[Information] [17/12/14 6:18:08 PM] AVS
---[NoImage] LoadPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
---[NoImage] FFVideoSource("E:\test.avi", cachefile="D:\test.avi.ffindex", threads=1)
---[Information] [17/12/14 6:18:08 PM] num_frames:35476
---[Information] [17/12/14 6:18:08 PM] rated:1
---[Information] [17/12/14 6:18:08 PM] raten:25
--[Information] [17/12/14 6:18:09 PM] AVS
---[NoImage] LoadPlugin("D:\MeGUI\tools\ffms\ffms2.dll")
---[NoImage] FFVideoSource("E:\test.avi", cachefile="D:\test.avi.ffindex", threads=1, rffmode=1)
---[Information] [17/12/14 6:18:09 PM] num_frames:35476
---[Information] [17/12/14 6:18:09 PM] rated:1
---[Information] [17/12/14 6:18:09 PM] raten:25
-[Information] Log for job25 (idx, test.avi -> test.avi.ffindex)
--[Information] [17/12/14 6:17:52 PM] Started handling job
--[Information] [17/12/14 6:17:52 PM] Preprocessing
--[Information] [17/12/14 6:17:52 PM] Job commandline:
"D:\MeGUI\tools\ffms\ffmsindex.exe" -f "E:\test.avi" "D:\test.avi.ffindex"
--[Information] [17/12/14 6:17:52 PM] Process started
--[Information] [17/12/14 6:17:52 PM] Standard output stream
---[Information] [17/12/14 6:17:54 PM] Writing index... done.
--[Information] [17/12/14 6:17:52 PM] Standard error stream
--[Information] [17/12/14 6:17:54 PM] MediaInfo
---[Information] [17/12/14 6:17:54 PM] File: D:\test.avi.ffindex
---[Information] General
----[Information] Format:
----[Information] FormatString:
----[Information] FileSize: 75228
----[Information] PlayTime:
--[Information] [17/12/14 6:17:54 PM] Postprocessing
--[Information] [17/12/14 6:18:10 PM] Job completed


Edit: The location of the ffmsindex cache file relative to the source doesn't make a difference. I'm very sure I tested that yesterday.

[hello_hello]

Some miscellaneous feature requests.

AVS Script Creator video preview:
When changing some script creator options (ie de-interlacing) the script creator preview continues to display the current frame number. That drives me nuts..... "want to punch the monitor" nutty, at times, because often it means changing an option will cause the preview to jump to a different location.
For instance, I've selected Yadif de-interlacing and I'm previewing frame number 14386 at the 10 minute point. Now I try "Yadif with bob". I'm still previewing frame number 14386 but I'm now only five minutes into the video because there's twice as many frames.
If the preview could stick to the current location time-wise rather than by frame number, the amount of navigating back and forth it'd save....... if there could at least be an option......
It mightn't be possible to stay at the exact point in time when the frame rate changes, but within a frame or two would be nice.

Default frame rate:
I've re-encoded the odd VFR video lately and mostly MeGUI defaults to 23.976 when it can't decide, and mostly so far that seems likely to be wrong. MeGUI tells you in the log file when it can't determine the frame rate, but I have wondered if 29.970 would be a better default, at least for "NTSC". You might end up with video with repeated frames in places, but the alternative is possibly lots of dropped frames.

Speaking of repeated frames....
If you enable the "M-in-5 Decimation" option in the script creator and click on the Preview button you get an error along the lines of "CycleR must be greater than zero" so it seems a little silly for CycleR to default to zero. :)
Also.... the "M" value can be anything you like, but anything greater than "4" will result in "CycleR must to be less than Cycle", because the default value for Cycle is 5. So values outside the range of 1 thru 4 don't make sense for "M" in the script creator. Unless....
The script creator also provided a method for changing the "Cycle" value. Would it be possible to have an option to set the "N" (Cycle) value one day? Maybe it could then be renamed "M-in-N Decimation"?

Anamorphic encoding options:
The "Encode non-mod16" option is maybe counter-intuitively named for MeGUI newbies, given it can be any mod, including mod16. If it was up to me, I'd just borrow the Handbrake description of "Anamorphic Strict".

I'm not sure if I've mentioned it yet :), but I think the Script Creator should use the same mpeg4 PARs for ITU video as can be set when configuring the x264 encoder. Having non-matcing PARs doesn't seem logical and I think the mpeg4 PARs would probably be a tad more accurate than the ITU PARs currently being used. Time to go digital..... ;)

[Zathor]

Using a VOB file, and not the first in a VTS, makes it almost certain to start with a broken GOP. Almost, because, well, there may be DVD multiplexers which split a VTS at GOP limits on purpose. Many others try to split as close as possible below 1 GB.

In your case, I wonder if MediaInfo respected a "Soft Telecine" flag while reporting the frame rate and number. So the encoded frame rate is NTSC-Film, to be presented as NTSC.

The file has been provided by a user a long time ago. Nevertheless I have the same behaviour also with other telecined files.

[IbrahimKh]

http://puu.sh/dAMVF.png
now what :confused:

[LigH]

Thank you for posting your log.

How much shall I bet on your AviSynth script using DirectShowSource, and you having installed a mega super duper codec pack?

[bxyhxyh]

Avira antivirus is recognizing x264-2525 as a virus.
I'm not sure if it's a false report or not.

[LigH]

False positives are more probable than you may imagine. VirusTotal may make it a little more certain. In this case, Avira is only alarmed because the EXE is runtime-compressed.

[Vincent Vega]

http://forum.doom9.org/showthread.php?p=1704409#post1704409

[Kurtnoise]

Wondering if there have been any progress made regarding the adding of FDK-AAC into MeGUI.

I'm forking megui to github (https://github.com/Kurtnoise/megui/commits/master)...really annoyed by SF. Still, 2-3 things to add and that should be fine...:)


btw, with this new year, megui is 10 years old !!!! Damned :p

[Barough]

I'm forking megui to github (https://github.com/Kurtnoise/megui/commits/master)...really annoyed by SF. Still, 2-3 things to add and that should be fine...:)


btw, with this new year, megui is 10 years old !!!! Damned :p

Heard some others also say that they have become really annoyed with SF.

Looking forward 2 when ur finished with adding it :)

MeGUI 10 years old....... WOW :)

[hello_hello]

I'm forking megui to github (https://github.com/Kurtnoise/megui/commits/master)...really annoyed by SF. Still, 2-3 things to add and that should be fine...:)

What'd SF do?

So does that mean there'll be two versions of MeGUI?

[Kurtnoise]

It's mainly because I cannot commit anymore on SF even recreating a new password...and also how they manage/monetize the projects.

This doesnt mean that it will be 2 versions...it's just the hosting that's moving, at least for me.

[Betsy25]

KurtNoise,

Is Zathor still on board the MeGui ship ? Haven't seen anything regarding code chances recently.

[Kurtnoise]

He's still around here, don't worry...

[Zathor]

Yes, a little bit busy these days but still around.

[Zathor]

As I have received some messages about the update server, here are some details:

MeGUI is now down to one update server only. As this server is on its limit the update feature may have to be disabled in the future in total - therefore we ask for additional update servers.

The remaining update server was facing in December a download of ~500GB. I expect that when we release a new stable build or there are other bigger package changes that we will have a download volume up to ~2TB. This volume will be split over all available update servers so e.g. when there are 4 of them each will have to deal with ~500GB.

Requirements for a update server:

>= 500GB transfer volume available monthly
1GB+ webspace
available over DNS / HTTP
FTP accounts for Kurtnoise and me or sync script in place so that the content of megui.org will be synced automatically (if anything goes wrong a ftp access may be a good idea nevertheless)
bandwidth does not matter so much as the detection & download of new packages is split over a few days


UPDATE: Luckily an additional update server has been added so we are now at 2 update servers. Nevertheless we are still searching for new servers.

[LigH]

Zathor, Kurtnoise: Please check your PMs for an offer...

[hello_hello]

Zathor, Kurtnoise, or whoever's working on MeGUI these days.....

A fair while ago I asked about using frame rate conversion when L-Smash was indexing (http://forum.doom9.org/showthread.php?p=1660439#post1660439), and was told it doesn't have the ability. While that was no doubt true at the time, it appears that's no longer the case. I was looking through the readme file for the latest L-Smash version today and discovered it now has the same parameters for frame rate conversion as ffmsindex (ie fpsnum & fpsden).

I mention this due to the problem previously reported (http://forum.doom9.org/showthread.php?p=1679175#post1679175) with ffmsindex and AVIs when it's frame rate conversion is used, and the apparent lack of interest (http://forum.doom9.org/showthread.php?p=1679218#post1679218) in fixing it, so I thought I'd point out L-Smash can now do the same frame rate conversion, in case you weren't aware. It might be a good idea to switch to L-Smash as the default indexer, at least for AVIs.

I haven't tested L-Smash yet to see if it has the same issue, although I'd assume it'd be unlikely. I will try to test it at some stage fairly soon though and report back if I find a problem.

Cheers.

[Kurtnoise]

Sure...I will have a look but the sample provided by Overdrive is not available anymore. Could you upload another one please ?

[Overdrive80]

You have already avalaible a sample, regards sir.

https://dl.dropboxusercontent.com/u/19135067/Sample_Vertical_Scroll.m2v

[Betsy25]

I'm forking megui to github (https://github.com/Kurtnoise/megui/commits/master)...really annoyed by SF. Still, 2-3 things to add and that should be fine...:)


btw, with this new year, megui is 10 years old !!!! Damned :p

Fantastic move ! :thanks:

[Zathor]

2523 [Update] get update server configuration data from the update server itself
2522 [FDK-AAC Encoder] added FDK-AAC encoder support
due to the FDK-AAC license the encoder will not be available in the online updater
2521 [Audio Encoder] added custom command line support
2520 [Chapter Creator] fixed sometimes incorrect duration (regression of 2516)
2519 [Update] if an outdated package is enabled the update will be downloaded automatically (if auto update is not disabled)
[Update] reduced time after an inactive package will be disabled to 14 days
a lot of the last changes in the update handling are caused by the fact that the remaining update servers have to be protected
if you want to help please have a look here: http://forum.doom9.org/showpost.php?p=1706329&postcount=4410
2518 updated copyright year to 2015
2517 [HD Streams Extractor] cosmetics
2516 [Chapter Creator] added option to sort the title list by chapter count
[Chapter Creator] cosmetics
2515 [OneClick] improved handling of track information in case no language information is available (regression of 2512). Part II
2514 [OneClick] improved handling of track information in case no language information is available (regression of 2512)

Thanks to Tereay and LigH we have now a second update server. As the changes are in this build and the clients must adjust there is only the core package available currently. In the next days I will also upload other packages.

[hello_hello]

Sure...I will have a look but the sample provided by Overdrive is not available anymore. Could you upload another one please ?

It seems to happen with almost any AVI, although quite randomly. Attached is an example (about 10MB). When I index with ffms2 and the frame rate conversion is included in the script, frames 401 and 402 are the same. When I remove the frame rate conversion from the script the total frame count doesn't change but frames 401 and 402 are no longer the same, so ffms2 must be doing something odd.

So far I'm yet to see any oddness when L-Smash is indexing, with or without frame conversion in the script (except for the problem of ffms2 and L-Smash both treating AVIs with null frames as variable frame rate unless frame rate conversion is used).

The ffms2 oddness only seems to apply to AVI. Well.... it's fine for MKV but I've rarely used ffms2 with other file types myself.

http://ge.tt/6i3DZVA2/v/0

[hello_hello]

Thanks to Tereay and LigH we have now a second update server.

Thank you Tereay and LigH!!

[Kurtnoise]

2523 [Update] get update server configuration data from the update server itself
2522 [FDK-AAC Encoder] added FDK-AAC encoder support
due to the FDK-AAC license the encoder will not be available in the online updater
2521 [Audio Encoder] added custom command line support
2520 [Chapter Creator] fixed sometimes incorrect duration (regression of 2516)
2519 [Update] if an outdated package is enabled the update will be downloaded automatically (if auto update is not disabled)
[Update] reduced time after an inactive package will be disabled to 14 days
a lot of the last changes in the update handling are caused by the fact that the remaining update servers have to be protected
if you want to help please have a look here: http://forum.doom9.org/showpost.php?p=1706329&postcount=4410
2518 updated copyright year to 2015
2517 [HD Streams Extractor] cosmetics
2516 [Chapter Creator] added option to sort the title list by chapter count
[Chapter Creator] cosmetics
2515 [OneClick] improved handling of track information in case no language information is available (regression of 2512). Part II
2514 [OneClick] improved handling of track information in case no language information is available (regression of 2512)

Thanks to Tereay and LigH we have now a second update server. As the changes are in this build and the clients must adjust there is only the core package available currently. In the next days I will also upload other packages.
For those who would like to test, fdkaac-0.6.2 are available from here :

x86 (http://www.mediafire.com/download/wb2aoxm1ivcoe85/fdkaac-0.6.2_x86.7z) & x64 (http://www.mediafire.com/download/y72yqm9misgeay8/fdkaac-0.6.2_x64.7z)

[Barough]

For those who would like to test, fdkaac-0.6.2 are available from here :

x86 (http://www.mediafire.com/download/wb2aoxm1ivcoe85/fdkaac-0.6.2_x86.7z) & x64 (http://www.mediafire.com/download/y72yqm9misgeay8/fdkaac-0.6.2_x64.7z)


Thnx a lot :)

[Barough]

When trying to enable FDK-AAC so do i get the error message @ the same time i click on 'save' in the Settings

http://i.imgur.com/K0gxFQj.png


Log screenshot
http://i.imgur.com/OA936XF.png


Have tested it on both of my computers and it's the same issue on both of them

Win 7 Home x64 (Laptop)
Win 8.1 Pro x64 (Desktop)


Any clues what it can be due to because i dont have a clue myself

[hello_hello]

When trying to enable FDK-AAC so do i get the error message @ the same time i click on 'save' in the Settings

Same here (XP). The problem seems to go away if you change the location of fdkaac.exe to anywhere but the default location.

[Barough]

Same here (XP). The problem seems to go away if you change the location of fdkaac.exe to anywhere but the default location.

That worked here also. Thnx :)

[hello_hello]

Just wanting to report the fdkaac encoder configuration doesn't seem to be working correctly. I've only tested using LC-AAC mode.

When VBR mode is selected, the encoder configuration changes to letting you select a variable bitrate, although it probably should change to selecting a quality as the QAAC encoder configuration does. It doesn't really matter what bitrate you select though, it either changes to 64kbp/s or 127kbps once the encoder configuration is closed.

In VBR mode, MeGUI specifies both a quality and a bitrate in the command line. For example:
C:\Program Files\MeGUI\tools\fdk\fdkaac.exe --ignorelength -m 1 -b 128 -p 2 - -o "D:\audio.m4a"

Also, unlike the x264 encoder configuration, when adding something to the audio encoder command line manually, it doesn't replace or over-ride the GUI. For example, after adding -w 18500 -m 4 to the custom command line, the command line MeGUI uses is:
C:\Program Files\MeGUI\tools\fdk\fdkaac.exe --ignorelength -m 1 -b 128 -p 2 -w 18500 -m 4 - -o "D:\audio.m4a"

ffdkaac.exe seems to cope with a conflicting command line, but some encoders (such as QAAC) don't.
C:\Program Files\MeGUI\tools\qaac\qaac.exe --ignorelength --threading -V 91 -V 73 - -o "D:\audio.m4a"
[Error] [16/02/15 12:28:23 PM] An error occurred
-[Error] [16/02/15 12:28:23 PM] Exception message: The pipe has been ended.

Thanks.

[Zathor]

When trying to enable FDK-AAC so do i get the error message @ the same time i click on 'save' in the Settings

http://i.imgur.com/K0gxFQj.png


Log screenshot
http://i.imgur.com/OA936XF.png


Have tested it on both of my computers and it's the same issue on both of them

Win 7 Home x64 (Laptop)
Win 8.1 Pro x64 (Desktop)


Any clues what it can be due to because i dont have a clue myself

I cannot reproduce that :-(
EDIT: got it. will be fixed in the next days.

[Zathor]

2524 [FDK-AAC Encoder] fixed update handling (regression of 2522)
Also some updated packages are available now.

[Barough]

2524 [FDK-AAC Encoder] fixed update handling (regression of 2522)
Also some updated packages are available now.

Thnx Zathor. :)

Nice 2 see some updates starting 2 show up again

[hello_hello]

Zathor,
Sometimes I update MeGUI components manually. If I discover there's an update I might download it manually before it's updated on the MeGUI update server, so I was wondering, would it be easy to implement a way to tell MeGUI when something has been updated manually so it doesn't later want to unnecessarily update it again? I updated QAAC myself a day or two ago, but today MeGUI wants to update it again.

Or, for those up us enthusiastic enough, when an update is found maybe there could be an option to tell MeGUI to take the user to the appropriate download page in order to do it manually?

I ask due to the load on the update servers being an obvious problem, but maybe it'd not be worth implementing that, or not practical?

I was also wondering if maybe there'd be an easy way to sync multiple copies of MeGUI. I have two PCs (networked) with MeGUI installed on each and at the moment both are being updated individually. Would there be an easy way to update both without having to download all the updates twice?

[kuchikirukia]

If audio settings are being worked on, could my feature request be penciled in?

https://sourceforge.net/p/megui/feature-requests/559/

It would be nice not having to go through the command line to get 5.1 channel QAAC over 320kbps outside of TVBR.

[hello_hello]

If audio settings are being worked on, could my feature request be penciled in?

https://sourceforge.net/p/megui/feature-requests/559/

It would be nice not having to go through the command line to get 5.1 channel QAAC over 320kbps outside of TVBR.

kuchikirukia,
Some QAAC info for you:
https://sites.google.com/site/qaacpage/news/qaacrelease241refalac141
For CVBR/ABR/CBR mode, bitrate value less than 8 is now treated as "bits per sample". Bitrate is computed as the following:
Bitrate = bits_per_sample * number_of_channels * sample_rate
For example, --cvbr 2 is now equivalent for --cvbr 192 (=2*2*48000) for 2ch, 48kHz case. This can be useful when you want to use CVBR/ABR/CBR and want constant quality setting for varying number of channels or sample rate.

So the way I understand it, in theory setting a bitrate of 7 should give you 2016kbps for 5.1ch audio, although in reality it won't as I'm pretty sure 768kbps is the limit.

https://github.com/nu774/qaac/wiki/Command-Line-Options
-c, --cbr <bitrate> AAC CBR mode / bitrate
For -a, -v, -c, "0" as bitrate means "highest". Highest bitrate available is automatically chosen.

MeGUI lets you specify a bitrate of zero.

MeGUI version 2524 lets you add a custom command line to the audio encoder configuration. It's not clever enough to use the custom command line in replace of what's specified in the GUI where applicable (at least not yet) so it'll specify two bitrates. The Opus encoder doesn't seem to mind though and it uses (based on my limited testing) the second bitrate in the command line which is what you'd want. According to MediaInfo, that looks like this (I added --bitrate 1568 to MeGUI's custom command line):

General
Complete name : D:\test.opus
Format : OGG
File size : 417 MiB
Duration : 37mn 49s
Overall bit rate : 1 543 Kbps
Writing application : opusenc from opus-tools 0.1.8
ENCODER_OPTIONS : --ignorelength --hard-cbr --bitrate 512 --bitrate 1568
Audio
ID : 13072 (0x3310)
Format : Opus
Duration : 37mn 49s
Channel(s) : 6 channels
Channel positions : Front: L C R, Rear: L R, LFE
Sampling rate : 48.0 KHz
Compression mode : Lossy
Writing library : libopus 1.1

I didn't seem to be able to get any more than roughly 1543kbps out of it for 5.1ch audio, and I tried again with foobar2000 to make sure the double bitrate in the command line wasn't messing with the output. Is that the limit?
Edit: I just realised you mentioned a 256kbps channel limit here (http://sourceforge.net/p/megui/feature-requests/559/). That answers the question.

[kuchikirukia]

Yeah, I set up batch files to get around the issue. It's just it's a very noticeable gap in MeGUI's abilities.