MeGUI Bug-Report Thread - Page 63

foxyshadis · 24th June 2006, 04:14

unskinnyboy, you do know that 4.0a is a broken alpha and that 3.98a is the main current branch? (Though even 3.98 had minor quality bugs as of last month.) 4.0 has several known bugs and missing features that reduce quality at equivalent bitrates. 3.97 is the only version HA recommends, having been cleaned of all known bugs.

I presume you mean 3.98a instead of 3.8a, which is... ancient.

I just wanted to let you know so that you don't end up watching the backup a year from now and freaking out about audio blips. >.> If the speed is important enough, then it doesn't matter, I suppose.

unskinnyboy · 24th June 2006, 04:56

Sorry for the big-fat typo. Yes, I meant 3.98a and not 3.8a. HA recommended version is 3.97b2 to be precise, which I am aware of.

While I know that 4.0a is experimental and several things could be broken, I cannot see how it could be any more broken than 3.98a, or is it? As per here, all builds >= 3.98a are disclaimed. Nothing specific mentioned about 4.0a. But there could be other threads, I don't visit HA that much. If you know, please link me.

This makes me wonder, should MeGUI update LAME to 3.98a even? Considering HA recommended version is only 3.97b2? Was there a separate testing performed which determined that 3.98a was OK and usable? Are the test results available? Sharktooth did mention that auto-updater updated to only versions which were supported and tested.

Limobar · 24th June 2006, 05:04

Quote:

Originally Posted by unskinnyboy

Was there a separate testing performed which determined that 3.98a was OK and usable? [...] Sharktooth did mention that auto-updater updated to only versions which were supported and tested.

That's no fun. Asking questions and giving the answers right away.

unskinnyboy · 24th June 2006, 05:10

Quote:

Originally Posted by Limobar

That's no fun. Asking questions and giving the answers right away.

Why did you snip away the important middle part? --> "Are the test results available?"

Limobar · 24th June 2006, 06:35

Quote:

Originally Posted by unskinnyboy

Why did you snip away the important middle part? --> "Are the test results available?"

I did it, because it was not important for the point I was making, that you answered your own question.

I don't think that every update is fully tested. There are only a few developers, while there are hundreds of users that are provided with the latest version of the software. I strongly believe that when it comes to open source software, it is the users' task to find out whether there are problems with updates and report these, so the developers can fix these problems.

unskinnyboy · 24th June 2006, 07:09

Oh, for sure I agree, and if you had read the whole chain of posts, you would have seen that I wasn't thinking any differently - including my usage of the very very alpha version of LAME, [along with my usual usage of a slew of alphas and betas of every software]. I am using it mainly to test and report bugs.

What I was attempting was to get some clarification as to why MeGUI updated LAME to 3.98a since there is nothing special about it as far as I can see. If it updated to 3.98a, it might as well have updated to 4.0a, unless 4.0a is totally unusable & 3.98a is reasonably OK to use. It might be very true, but just wanted to know what the basis of the updates are, especially since Sharktooth said that all the components are tested. If 3.98a is tested and found OK, then why HA disclaims it? If HA recommends only 3.97b2, then why doesn't MeGUI adhere to it?

Everything will have very plausible answers, I am sure, but I just want to hear them out if possible.

EDIT: Oh, 3.98a is the current branch - foxyshadis already said that. Problem solved!

bkman · 24th June 2006, 12:37

Profile bug is indeed fixed, but I found another.

When a job errors out (like with an incorrect commandline arg), the log stalls and won't print any new output after that.

shon3i · 24th June 2006, 21:58

When doing 3pass encoding with clever TM, in second pass --sar x:y are not included in comandline, and i get very streched output, can i just fix this in mkvmerge or i must encode again because i lost in commpresion because picture is bigger than normal??

foxyshadis · 25th June 2006, 03:46

Well, I could only find 4.0 bug reports on quite earlier prereleases, which might well be fixed by now. For 3.98a the problem sample count seems to be about the same as 3.97b so... *shrug* I guess the warning was premature.

Couple threads:
http://www.hydrogenaudio.org/forums/...howtopic=44515
http://www.hydrogenaudio.org/forums/...howtopic=39314

Latexxx · 25th June 2006, 20:46

MeGUI wants to update my my Neroaac from 2006-05-08 to 2006-05-26 but

Updating neroaacenc. File 1/1.
Error: neroaacenc could not be found on the server
Update completed.
0 files were completed successfully
1 files had problems.

unskinnyboy · 25th June 2006, 21:13

From here:

Quote:

Originally Posted by Sharktooth

Keep in mind we had to remove the nero aac encoder from the auto-update server and it needs to be extracted manually until we manage to get it directly from the nero FTP.

Hesse · 25th June 2006, 21:24

There seems to be a bug in the AutoEncode section. In the add additonal content section, after I have added an audio track name and chapters file (ogg format), it does not get muxed into the output file. I'm using MKV container. I took a glance at the command line and it seems that those options were missing there as well.

The Link · 26th June 2006, 11:15

Since I cannot find it in the first post I describe my problem(s) here:

I encoded a movie to AVC (CE-Mainprofile) with 2 NeroAAC streams (Q0.2 resulting in AAC-SBR streams) in the MP4 container. I used the latest MEGui build + latest tools provided by the Autoupdater. The resulting MP4-file was corrupt in a way that seeking to the end of the file wasn't possible in MPC and VLC. Foobar2000 didn't even load the file at all saying it was corrupt.

Furthermore instead of displaying 2 audio streams i get 4 audio streams displayed with Haali's splitter (every audio stream is displayed two times). Since I accidently deleted the MP4 file I cannot provide more information but I hope someone knows about the problem or can spot it in any way.

Thanks in advance!

edit: I just encoded the same movie into the mkv container. This time the file isn't corrupt but I got 4 audio streams again though there should only be 2. Perhaps there's a problem with mp4 audio files created by the neroaac encoder?

edit2: The Ogg chapter file I pointed to use in the auto encoder wasn't used. I had to remux the file anyway because of the problem reported above but I wanted to mention it.

Rumbah · 26th June 2006, 12:53

I found a bug with deinterlace method selection.
After the analysis MeGUI detects that my source is interlaced. It automatically selects TDeint. But if I change the deinterlace method, the avs file is not updated properly. It is only changed if you select the method you want and then switch deinterlacing off and on again.

EDIT: I forgot to mention, this is in the Avisynth script creator in the filters section.

junglemike · 26th June 2006, 21:05

Megui completely ignores bitrate settings for small files.
Please anybody verify that. It takes just 1 minute, you don't need to encode anything.
With normal length files (30min+) there is no problem, But if I try to encode very small clips (like 1-2 minutes) - I press "Auto Encode", than i choose average bitrate. The bitrate setting is either wrong for 2-5min files (like 640 instead of 220) - I can see it in log files. And for very small files (like 30 sec) - No bitrate setting is used at all, so it x264 uses codec's default. (1000kbps)

chros · 26th June 2006, 23:32

v0.2.3.2177: at the Adaptive Muxer divmux.exe changes the fourcc from xvid to divx ...

berrinam · 27th June 2006, 06:02

Is that a problem?

woah! · 27th June 2006, 06:46

Quote:

Originally Posted by ChronoCross

hmm anyone know what this means? I encoded the 5.1 fine but the stereo won't encode.

did you find out what caused this error as i get it also

the ac3 file encodes with besweet ok but i get this error and thats it goodnight.

Code:

Log for job job2

Channels=2, BitsPerSample=16, SampleRate=48000Hz
D:\megui\tools\neroaacenc\neroAacEnc.exe -ignorelength -q 0.5 -if - -of "O:\INDEXED T01 2_0ch 384Kbps DELAY 0ms.mp4"Error:
MeGUI.AviSynthException: m2AudioAC3Source: error in file "O:\INDEXED T01 2_0ch 384Kbps DELAY 0ms.ac3"
   at MeGUI.AviSynthAudioEncoder.encode()

it encodes 5.1 correctly or at least a 5.1 ac3 like this:

INDEXED T01 3_2ch 384Kbps DELAY 0ms.ac3

weird and annoying at the same time...

chros · 27th June 2006, 08:07

Quote:

Originally Posted by berrinam

Is that a problem?

Yes, if the decoders are configured so that they decode theirs files. (divx-divx, xvid-xvid)
I have tried yesterday a small trailer with this, and there was no picture in the final muxed avi file ...

Sharktooth · 27th June 2006, 12:31

Well that's not a MeGUI problem, just a badly configured system on your side... dont you think?

24th June 2006, 04:56	#1242 \| Link
unskinnyboy Registered User Join Date: Feb 2004 Location: NTSC R1 Posts: 2,046	Sorry for the big-fat typo. Yes, I meant 3.98a and not 3.8a. HA recommended version is 3.97b2 to be precise, which I am aware of. While I know that 4.0a is experimental and several things could be broken, I cannot see how it could be any more broken than 3.98a, or is it? As per here, all builds >= 3.98a are disclaimed. Nothing specific mentioned about 4.0a. But there could be other threads, I don't visit HA that much. If you know, please link me. This makes me wonder, should MeGUI update LAME to 3.98a even? Considering HA recommended version is only 3.97b2? Was there a separate testing performed which determined that 3.98a was OK and usable? Are the test results available? Sharktooth did mention that auto-updater updated to only versions which were supported and tested. __________________ I encode, therefore I am. Doom9 Toolbar Extension for Firefox [v3.1]

24th June 2006, 07:09	#1246 \| Link
unskinnyboy Registered User Join Date: Feb 2004 Location: NTSC R1 Posts: 2,046	Oh, for sure I agree, and if you had read the whole chain of posts, you would have seen that I wasn't thinking any differently - including my usage of the very very alpha version of LAME, [along with my usual usage of a slew of alphas and betas of every software]. I am using it mainly to test and report bugs. What I was attempting was to get some clarification as to why MeGUI updated LAME to 3.98a since there is nothing special about it as far as I can see. If it updated to 3.98a, it might as well have updated to 4.0a, unless 4.0a is totally unusable & 3.98a is reasonably OK to use. It might be very true, but just wanted to know what the basis of the updates are, especially since Sharktooth said that all the components are tested. If 3.98a is tested and found OK, then why HA disclaims it? If HA recommends only 3.97b2, then why doesn't MeGUI adhere to it? Everything will have very plausible answers, I am sure, but I just want to hear them out if possible. EDIT: Oh, 3.98a is the current branch - foxyshadis already said that. Problem solved! __________________ I encode, therefore I am. Doom9 Toolbar Extension for Firefox [v3.1] Last edited by unskinnyboy; 24th June 2006 at 07:14.

25th June 2006, 20:46	#1250 \| Link
Latexxx Registered User Join Date: Nov 2001 Location: Tampere, Finland Posts: 618	MeGUI wants to update my my Neroaac from 2006-05-08 to 2006-05-26 but Updating neroaacenc. File 1/1. Error: neroaacenc could not be found on the server Update completed. 0 files were completed successfully 1 files had problems. __________________ A/V moderator @ hydrogenaudio.org My weird old sh*t: http://www.nic.fi/~lhahne/ http://last.fm/user/Latexxx/

25th June 2006, 21:24	#1252 \| Link
Hesse Registered User Join Date: Apr 2003 Posts: 32	Adding chapters, audiotrack name in AutoEncode There seems to be a bug in the AutoEncode section. In the add additonal content section, after I have added an audio track name and chapters file (ogg format), it does not get muxed into the output file. I'm using MKV container. I took a glance at the command line and it seems that those options were missing there as well.

26th June 2006, 11:15	#1253 \| Link
The Link Registered User Join Date: Nov 2001 Location: Germany Posts: 189	Since I cannot find it in the first post I describe my problem(s) here: I encoded a movie to AVC (CE-Mainprofile) with 2 NeroAAC streams (Q0.2 resulting in AAC-SBR streams) in the MP4 container. I used the latest MEGui build + latest tools provided by the Autoupdater. The resulting MP4-file was corrupt in a way that seeking to the end of the file wasn't possible in MPC and VLC. Foobar2000 didn't even load the file at all saying it was corrupt. Furthermore instead of displaying 2 audio streams i get 4 audio streams displayed with Haali's splitter (every audio stream is displayed two times). Since I accidently deleted the MP4 file I cannot provide more information but I hope someone knows about the problem or can spot it in any way. Thanks in advance! edit: I just encoded the same movie into the mkv container. This time the file isn't corrupt but I got 4 audio streams again though there should only be 2. Perhaps there's a problem with mp4 audio files created by the neroaac encoder? edit2: The Ogg chapter file I pointed to use in the auto encoder wasn't used. I had to remux the file anyway because of the problem reported above but I wanted to mention it. __________________ When birds burp, it must taste like bugs! (Calvin&Hobbes) Last edited by The Link; 26th June 2006 at 18:51.

24th June 2006, 04:14	#1241 \| Link
foxyshadis ангел смерти Join Date: Nov 2004 Location: Lost Posts: 9,558	unskinnyboy, you do know that 4.0a is a broken alpha and that 3.98a is the main current branch? (Though even 3.98 had minor quality bugs as of last month.) 4.0 has several known bugs and missing features that reduce quality at equivalent bitrates. 3.97 is the only version HA recommends, having been cleaned of all known bugs. I presume you mean 3.98a instead of 3.8a, which is... ancient. I just wanted to let you know so that you don't end up watching the backup a year from now and freaking out about audio blips. >.> If the speed is important enough, then it doesn't matter, I suppose.

24th June 2006, 12:37	#1247 \| Link
bkman Registered User Join Date: Jan 2006 Posts: 294	Profile bug is indeed fixed, but I found another. When a job errors out (like with an incorrect commandline arg), the log stalls and won't print any new output after that.

24th June 2006, 21:58	#1248 \| Link
shon3i BluRay Maniac Join Date: Dec 2005 Posts: 2,419	When doing 3pass encoding with clever TM, in second pass --sar x:y are not included in comandline, and i get very streched output, can i just fix this in mkvmerge or i must encode again because i lost in commpresion because picture is bigger than normal??

25th June 2006, 03:46	#1249 \| Link
foxyshadis ангел смерти Join Date: Nov 2004 Location: Lost Posts: 9,558	Well, I could only find 4.0 bug reports on quite earlier prereleases, which might well be fixed by now. For 3.98a the problem sample count seems to be about the same as 3.97b so... shrug I guess the warning was premature. Couple threads: http://www.hydrogenaudio.org/forums/...howtopic=44515 http://www.hydrogenaudio.org/forums/...howtopic=39314

26th June 2006, 12:53	#1254 \| Link
Rumbah Registered User Join Date: Mar 2003 Posts: 480	I found a bug with deinterlace method selection. After the analysis MeGUI detects that my source is interlaced. It automatically selects TDeint. But if I change the deinterlace method, the avs file is not updated properly. It is only changed if you select the method you want and then switch deinterlacing off and on again. EDIT: I forgot to mention, this is in the Avisynth script creator in the filters section. Last edited by Rumbah; 26th June 2006 at 13:04.

26th June 2006, 21:05	#1255 \| Link
junglemike Registered User Join Date: Dec 2002 Posts: 288	Megui completely ignores bitrate settings for small files. Please anybody verify that. It takes just 1 minute, you don't need to encode anything. With normal length files (30min+) there is no problem, But if I try to encode very small clips (like 1-2 minutes) - I press "Auto Encode", than i choose average bitrate. The bitrate setting is either wrong for 2-5min files (like 640 instead of 220) - I can see it in log files. And for very small files (like 30 sec) - No bitrate setting is used at all, so it x264 uses codec's default. (1000kbps)

26th June 2006, 23:32	#1256 \| Link
chros Registered User Join Date: Mar 2002 Posts: 2,323	v0.2.3.2177: at the Adaptive Muxer divmux.exe changes the fourcc from xvid to divx ...

27th June 2006, 06:02	#1257 \| Link
berrinam Registered User Join Date: Apr 2005 Posts: 1,740	Is that a problem? __________________ MeGUI wiki MeGUI bug tracker MeGUI feature request tracker

27th June 2006, 12:31	#1260 \| Link
Sharktooth Mr. Sandman Join Date: Sep 2003 Location: Haddonfield, IL Posts: 11,768	Well that's not a MeGUI problem, just a badly configured system on your side... dont you think? __________________ MPEG-4 ASP Custom Matrices: EQM V1(old), EQM AutoGK Sharpmatrix (aka EQM V2), EQM V3HR (updated 01/10/2004), EQM V3LR, EQM V3ULR (updated 04/02/2005), EQM V3UHR (updated 17/12/2004) and EQM V3EHR (updated 05/10/2004) Info about my ASP matrices. MPEG-4 AVC Custom Matrices: EQM AVC-HR Info about my AVC matrices My x264 builds. Mooo!!!

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode