eac3to - audio conversion tool

[tebasuna51]

Decoding your sample from frame 0 to 232 and from frame 234 to end work fine always with all decoders.

The frame 233 is corrupt and some decoders abort (ArcSoft, libav, dcadec) or output garbage (NicAudio).

I don't have Sonic to test, but check yourself if decoding your 10s sample there are a click (garbage) at 2,5 s

BTW eac3to can't do nothig to recover corrupt frames or change the decoders behaviour.

[Smithy]

the bitstream results in No Sound during this frame/runtime,
but the decoded file with sonic is fine @ this frame. ^^

DTS Decode (Sonic) Sample (10s .wav)
http://workupload.com/file/uh342a1s

[tebasuna51]

Well seems the Sonic decoder do the trick.

I tested some other decoders, to resume:
- Abort: ArcSoft, dcadec and libav
- Big click: Nicaudio (AviSynth), Foobar2000, Tranzcode and Valdec (ac3filter_tools_0_31b)

[hello_hello]

I tested some other decoders, to resume:
- Abort: ArcSoft, dcadec and libav
- Big click: Nicaudio (AviSynth), Foobar2000, Tranzcode and Valdec (ac3filter_tools_0_31b)

On the directshow front:
LavAudio decodes it.
ffdshow and libdts doesn't (big click).
ffdshow and libavcodec is fine.

[Smithy]

Yes LavAudio decodes playd fine, too.

[Elegant]

I've been noticing that when I create FLAC files I keep getting "VALID_BITS: 00", any chance on getting that to match the actual bit depth of the audio file? Or is this an issue with libFLAC?

[r0lZ]

I have noticed a problem with the AAC/M4A files generated by eac3to, when they are muxed with mkvmerge.

When I use eac3to to convert an audio track to AAC, it produces the AAC track within a M4A container. That M4A file can be muxed directly with the video track with MkvMerge. But often, the sound is choppy when I play the resulting MKV on my Samsung TV. (The same file can be played without problem with all software players I have tried so far.)

If I extract the AAC track from the M4A and then mux it with MkvMerge, there is no problem any more. The sound is perfect, including when it is played with my TV.

If that matters, the input files are a PCM WAV mono at 48 KHz, and a DTS 5.1 at 48 KHz 1509 Kbps. They are standard files, and there is no warning during the conversions.

To retrieve the AAC track from the M4A container, I use MkvMerge to mux the M4A to MKA, and MkvExtract to extract the AAC from the MKA. I have also used ffmpeg with "-acodec copy" with the same result: the AAC itself can always be successfully muxed.

I wonder several things:

- Is it a known bug of eac3to (or MkvMerge) ?

- Why eac3to creates the M4A container and not an elementary AAC stream ?

- Is it possible to force eac3to to create the AAC stream without the M4A container? (Specifying "output.aac" instead of "output.m4a" is accepted by eac3to, but it produces a M4A file anyway.)

Someone else has noticed similar problems?

Thanks in advance for any pointer!

[Music Fan]

If you analyze both mkv's with MediaInfo, are there differences in the audio descriptive ? MediaInfo is sometimes wrong but it may help.

[tebasuna51]

- Is it a known bug of eac3to (or MkvMerge) ?
eac3to work with NeroAacEnc.exe in the same folder. I only now this problem:
"- NeroAacEnc crash sometimes when piped directly by eac3to, work always when use the OS pipe.
Workaround: eac3to input stdout.wav | NeroAacEnc -q 0.5 -ignorelength -if - -of output.m4a"

But I never read a problem like you say: finish OK and " sound is choppy".

MkvMerge need know when a AAC is HE, but NeroAacEnc output always AAC-LC with quality greater 0.35, the default quality with eac3to is 0.5

MkvMerge, AFAIK, read m4a metadata to correct delay but store only the AAC streams in mkv. Don't have sense for me than extract the AAC and after remuxed change anything.

I read than some players can't play mono or 5.1 correctly (stereo always ok), but your player seems work with the AAC.

- Why eac3to creates the M4A container and not an elementary AAC stream ?
NeroAacEnc always output AAC in mp4 container (m4a)

- Is it possible to force eac3to to create the AAC stream without the M4A container? (Specifying "output.aac" instead of "output.m4a" is accepted by eac3to, but it produces a M4A file anyway.)
Not possible with NeroAacEnc, but you can use other encoders (if you have the required soft not included in eac3to) to output AAC with standard ADTS headers. For instance:

eac3to input stdout.wav | qaac -V 99 --ignorelength --adts --no-delay -o output.aac -

Someone else has noticed similar problems?
AFAIK not until now.

EDIT:
When Mkvmerge mux a m4a from NeroAacEnc always correct the delay cutting the first frames and add a +9ms of delay to track.
If AAC is extracted and after remuxed the delay +9ms dissapear.
For me that is the only difference between mkv's, maybe your player have problems with the delay.

[Music Fan]

eac3to input | qaac -V 99 --ignorelength --adts --no-delay -o output.aac -
Don't forget stdout.wav after input.

Don't have sense for me than extract the AAC and after remuxed change anything.
I already had the same kind of problem with videos : when I open a TS including interlaced H264 in MKVmerge, it creates a file considered as VFR by MediaInfo (while it's CFR).
But if I demux the video with TSMuxer then open the ES in MKVmerge, the video is correctly considered as CFR by MediaInfo (as the original TS, there is no error in the TS stream).

It seems MKVmerge has problems with some containers or headers.

[szabi]

Hi

I tried to extract AC3 from THD audio.
However eac3to always encoding instead of extraction.
What can I do to solve it?

bye
szabi

[LigH]

Start with posting your complete command line.

[szabi]

Hi

It is the log file:
eac3to v3.29
d:\eac\eac3to.exe" "E:\my video\movie.mkv" 2: "E:\my video\movie2eng.ac3"
analyze: 1%
.
.
analyze: 99%
analyze: 100%
MKV, 1 video track, 2 audio tracks, 3 subtitle tracks, 2:00:23, 24p /1.001
1: h264/AVC, 1080p24 /1.001 (16:9)
2: TrueHD (Atmos), English, 7.1 channels, 48kHz
.
.
a02 thd, 48000, 7.1
a02 AC3 encoding doesn't support back channels. Will mix them into the surround.
a02 Extracting audio track number 2...
a02 Decoding with libav/ffmpeg...
a02 Mixing surround channels...
a02 Remapping channels...
a02 Encoding AC3 <640kbps> with libAften...
a02 Creating file "E:\my video\movie.mkv_2eng.ac3"...
process: 1%
.
.

It is the first time I can not extract AC3.

bye
szabi

[r0lZ]

Thanks for your reply, tebasuna51.
MkvMerge need know when a AAC is HE, but NeroAacEnc output always AAC-LC with quality greater 0.35, the default quality with eac3to is 0.5
Do you mean that the encoding type changes when the quality is less than 0.35? Anyway, the advantage of the M4A container is that it has the correct info, and it should not be necessary to specify the HE flag to MkvMerge. BTW, I did my test with the default quality 0.5. Also, I've tried to change the flag in MkvMerge, for the M4A and AAC streams, but that doesn't change anything. The AAC sound is always perfect, and the M4A stream is always choppy.

Don't have sense for me than extract the AAC and after remuxed change anything.
It's also something I don't understand. It's why I have suspected a bug in the MkvMerge M4A demuxer.

... you can use other encoders (if you have the required soft not included in eac3to) to output AAC with standard ADTS headers. For instance:

eac3to input | qaac -V 99 --ignorelength --adts --no-delay -o output.aac -
That might be a good alternative. What is the quality of the QAAC encoder? Is it comparable to the Nero encoder?

EDIT:
When Mkvmerge mux a m4a from NeroAacEnc always correct the delay cutting the first frames and add a -9ms of delay to track.
If AAC is extracted and after remuxed the delay -9ms dissapear.
For me that is the only difference between mkv's, maybe your player have problems with the delay. Well, the sound plays in sync with the video, so I guess it's not a delay problem. I don't understand why that -9 ms delay is necessary, but obviously it's also a problem related to the M4A container.

[r0lZ]

Don't forget stdout.wav after input.Thanks for the precision.

I already had the same kind of problem with videos : when I open a TS including interlaced H264 in MKVmerge, it creates a file considered as VFR by MediaInfo (while it's CFR).
But if I demux the video with TSMuxer then open the ES in MKVmerge, the video is correctly considered as CFR by MediaInfo (as the original TS, there is no error in the TS stream).

It seems MKVmerge has problems with some containers or headers.I mux a h264 elementary stream (encoded with x264) and a SRT stream with the M4A or AAC stream. IMO, it's something that MkvMerge should handle without problem. And anyway, if it's a problem with the video, that doesn't explain why the audio container causes the problem.

[tebasuna51]

Don't forget stdout.wav after input.
My fault, I forget the stdout.wav. Corrected in the post.

Do you mean that the encoding type changes when the quality is less than 0.35?
Yep, in all my test using 0.34 or less the output is HE, using 0.35 or higer the output is LC.

Anyway, the advantage of the M4A container is that it has the correct info, and it should not be necessary to specify the HE flag to MkvMerge. BTW, I did my test with the default quality 0.5.
Then your problem can't be about that.

That might be a good alternative. What is the quality of the QAAC encoder? Is it comparable to the Nero encoder?
Of course, QAAC is, for now, considered the best AAC encoder at low bitrates, for high bitrates we can't make test because the output can't be differentiated.

I don't understand why that -9 ms delay is necessary, but obviously it's also a problem related to the M4A container.
I'm not sure is this can cause the problem and how MkvMerge mux the m4a.

I will make some test about that.

[tebasuna51]

It is the log file:
eac3to v3.29
d:\eac\eac3to.exe" "E:\my video\movie.mkv" 2: "E:\my video\movie2eng.ac3"
...
2: TrueHD (Atmos), English, 7.1 channels, 48kHz
...

It is the first time I can not extract AC3.

In a mkv the THD can't have a embedded AC3, that is only possible in m2ts (bluray) container. Then the AC3 can't be extracted, only recoded.
Check if the mkv have the AC3 in other track.

[LeXXuz]

eac3to writes MKV files through the Haali MKV Muxer DirectShow filter. However, all other files are created and written to directly by eac3to. Replacing the Haali MKV Muxer would be quite a lot of work, so don't expect that anytime soon.

It's been 2 years. Any news on this issue? Any plans to ditch the old mkv muxer for ffmpeg or others?

[r0lZ]

Of course, QAAC is, for now, considered the best AAC encoder at low bitrates, for high bitrates we can't make test because the output can't be differentiated.If it is so good, why is it not distributed with eac3to? Rights problem?

I will make some test about that.Don't waste your time. I can do the tests myself and anyway I think I'll use QAAC (at least if I can freely distribute it with BD3D2MK3D).

Thanks for your help.

[tebasuna51]

If it is so good, why is it not distributed with eac3to? Rights problem?

Yep, the dll's needed can be obtained freely (see makeportable.zip in https://sites.google.com/site/qaacpage/cabinet ) but not distributed.

[Thunderbolt8]

would it then be possible to add a download dialogue in which the user is asked if eac3to should download the files now?

[LigH]

Adding a download routine like wget may be beyond the purpose of a CLI audio converter; it could at most call a ShellExecute of the URL to spawn a browser, IMHO.

But the QAAC download URL could certainly be readable in the documentation (maybe added to "\legal stuff\qaac" then, with a brief guide how to use the CoreAudioToolbox without actually installing QuickTime or even iTunes).

I believe many authors prefer a link to a download site over a direct download link; the latter may change sometimes, already depending on the version (well, eac3to is a partial exception here, always having the latest version at the same URL).

[r0lZ]

Yep, the dll's needed can be obtained freely (see makeportable.zip in https://sites.google.com/site/qaacpage/cabinet ) but not distributed.Damn! When I have seen that QAAC requires the libraries from the iTune package, I was about to abandon the idea to support it in BD3D2MK3D. I can't force the users to install that junk just to be able to convert to AAC. I will see what I can do with the makeportable archive. If it's easy enough, I'll add an option to use either Nero or QAAC and if QAAC is not already installed when the user selects it, I'll display a dialog with the link and instructions to download and install the required archives.
Thanks to everybody here!

[tebasuna51]

... I'll display a dialog with the link and instructions to download and install the required archives.
1) You can distribute qaac.exe (now v2.55) and makeportable.cmd in ...\BD3D2MK3D\toolset subfolder (like eac3to), notes:

- There are a 64 bit versión but you can use the 86 version to avoid problems.
- The files libsoxconvolver.dll, libsoxr.dll and refalac.exe are not needed to AAC conversión, only to advanced parameters not used automatically or ALAC conversión.

2) Then you need download iTunes6464Setup.exe (now iTunes 12.3) in the same subfolder. There are the required dll's for 32 and 64 bits.

3) Run makeportable.cmd than create QTfiles and QTfiles64 subfolders.
EDIT: To run makeportable.cmd you need 7z (http://www.7-zip.org/) decompressor installed in your sytem.

4) You can delete iTunes6464Setup.exe, and the QTfiles64 folder if don't want use the 64 bits version. And you have qaac ready to work with eac3to:

"[path]eac3to" "input" stdout.wav | "[path]qaac" -V 99 --ignorelength --adts --no-delay -o "[path]output.aac" -

About qaac quality you can see:
HydrogenAudio test
-----------------------
V 0 - V 4 = ~ 40 kb/s
V 5 - V 13 = ~ 45 kb/s
V 14 - V 22 = ~ 75 kb/s
V 23 - V 31 = ~ 80 kb/s
V 32 - V 40 = ~ 95 kb/s
V 41 - V 49 = ~105 kb/s
V 50 - V 58 = ~115 kb/s
V 59 - V 68 = ~135 kb/s
V 69 - V 77 = ~150 kb/s
V 78 - V 86 = ~165 kb/s
V 87 - V 95 = ~195 kb/s
V 96 - V104 = ~225 kb/s
V105 - V113 = ~255 kb/s
V114 - V122 = ~285 kb/s
V123 - V127 = ~320 kb/s
The V values have ranges, is the same put V 96 than V104.
The test was made with stereo music sources, with audio movies tracks the bitrate can be less than that.

I my test V 99 is, more or less, equivalent to Nero q 0.5, V 90 -> q 0.46, V 82 -> q 0.42

----------------------
A workaround to extract the aac from Nero .m4a can be:

MP4Box -out adts.aac -raw 1 nero.m4a

But the aac need a -55ms delay, like a aac (48 KHz) frame duration is 21.333 ms you can cut the first 3 frames with:

eac3to adts.aac adtsDelayed.aac -64ms

And now need a +9ms delay. This is the procedure than use MkvMerge when mux the .m4a.
You can forget the +9ms delay because can't be distinguished (less than a video frame duration)

[Thunderbolt8]

it would at least be nice to give the user a note when trying to get some .aac output that for best .aac quality he needs to install some additional stuff. so that way the user knows there is still room for improvement. I guess most people have the assumption that the best possible way to deal with .aac is already that which is eac3to doing/that which already comes with eac3to.

[r0lZ]

Thanks for the additional precisions, tebasuna51. I think I'll implement that (perhaps with a step by step wizard to help the user install the required stuff without problem).

@Thunderbolt8: I agree.
I was convinced that Nero was the best AAC encoder (at least among the free solutions). IMO, by default eac3to should always use the free solution that it can distribute, but it should also show in his help screen the best commercial or not freely distributable solution, perhaps with short instructions on how to obtain, install and use them or with a link to an online help page with complete instructions. This thread here is way too long and complicated to constitute a valuable help for the complete installation of eac3to.

[LigH]

NeroAacEnc is still one of the better AAC encoders, certainly convenient for most cases; but fdkaac and Apple CoreAudioToolbox surpassed it already in "best of" listening tests. Differences are marginal. They all can compete with Vorbis and surpass LAME, no reason to mourn.

A major disadvantage of fdkaac is the width of gaps between bitrate target presets. Tuning is better with qaac and free with Nero.

[tebasuna51]

I forget a requirement to decompress iTunes6464Setup.exe with makeportable.cmd. You need 7z installed (always recommended) in your system (added to post).
----------------

Like I say before the NeroAacEnc quality (with q >= 0.4) is equivalent to other encoders.
The question here is the problem detected by r0lZ when mux m4a with MkvMerge.

Anybody experiment the same problem (choppy sound with mono or 5.1 from .m4a)?

Is a MkvMerge problem (seems solved using .aac)?

[sneaker_ger]

When I use eac3to to convert an audio track to AAC, it produces the AAC track within a M4A container. That M4A file can be muxed directly with the video track with MkvMerge. But often, the sound is choppy when I play the resulting MKV on my Samsung TV. (The same file can be played without problem with all software players I have tried so far.)

If I extract the AAC track from the M4A and then mux it with MkvMerge, there is no problem any more. The sound is perfect, including when it is played with my TV.
Only "often", not always? Is it reproducible or erratic even with the same file?
Can you test these samples:
https://www.sendspace.com/file/lcnqy5

[Motenai Yoda]

I forget a requirement to decompress iTunes6464Setup.exe with makeportable.cmd. You need 7z installed (always recommended) in your system (added to post).

Why not the portable version?

[LigH]

IMHO, it should be possible to ship the static CLI version with other projects, if I'm not wrong. Was it the 7za.exe in the extras archive? May have to read the docs texts again...

[tebasuna51]

IMHO, it should be possible to ship the static CLI version with other projects, if I'm not wrong. Was it the 7za.exe in the extras archive?...
Seems than 7za.exe can extract AppleApplicationSupport*.msi from iTunes6464Setup.exe but not QTfiles* from AppleApplicationSupport*.msi.

Tested stable v9.20 and last beta v15.07, maybe because:
7za.exe - standalone console version of 7-Zip with reduced formats support.

EDIT: In my system (7z installed but not in PATH) work with only 7z.exe and 7z.dll (from 7z920.exe) in the same folder.
But I can't guarantee than work in all system's.

[r0lZ]

Only "often", not always? Is it reproducible or erratic even with the same file?
The problem occurs systematically when I mux the same video and audio files. When I've noticed the problem (with the mono WAV and 5.1 DTS streams), there is a gap in the sound every 2 seconds or so, regardless of the state of the HE flag in MkvMerge. (I know that it should be ignored when the input file is a M4A, but I've tried anyway with the flag in automatic, on and off modes, just to be sure).

I have then made another test, with a short clip (faster to mux) and another M4A, and the problem was still present, but much less evident. There was a gap only every 15 seconds or so, and it was less audible. But again, when muxing the AAC extracted from the M4A, there was no problem at all.

Therefore, it seems that the problem can be reproduced, but with different results. I have written "often" because usually I'm not interested in converting the original DTS or AC3 audio to AAC, and therefore I have not often the occasion to test that problem with my TV. But I'm sure I have already played movies with AAC audio without the problem. Maybe the AAC stream have been muxed directly, or another version of MkvMerge has been used. Or perhaps it's also related to the video? I don't know. I can't be sure that the problem occurs always, but since I've discovered it, I can confirm that it occurs systematically.
Strangely, the choppy sound problem occurs only when I play the movie with my Samsung TV. I suppose that it's a combination of 2 little bugs (one in the M4A when it is created by eac3to or when it is muxed by MkvMerge, and one in my TV).
Can you test these samples:
https://www.sendspace.com/file/lcnqy5
I'll do it when I'll have some time.
I will also try to mux it with different video streams.

[kypec]

Strangely, the choppy sound problem occurs only when I play the movie with my Samsung TV. I suppose that it's a combination of 2 little bugs (one in the M4A when it is created by eac3to or when it is muxed by MkvMerge, and one in my TV).
:thanks:
I noticed the audio choppiness in one of my remuxes as well. I'm using serviio (http://serviio.org) media streamer from my Windows 7 desktop PC to Samsung TV (model UE40D6000) over cabled LAN (not wi-fi). I assumed that choppiness was due to extremely high bitrate of video+audio stream since it's a straight remux of Disney's Frozen BD but maybe it was indeed due to some strange bug that involves specific MkvMerge / M4A audio file / TV firmware / whatever. I'll recheck when I'll find some spare time...

[Music Fan]

I have a problem with eac3to that often happens, I had never thought to mention it here ; with TS recordings (done with my STB), eac3to only detects video but not sound. And there are apparently no differences between files whose sound is detected and others.

[Xorp]

The latest eac3to build now automatically switches to dcadec for all DTS 7.x tracks. However, because the new decoder isn't that well tested yet, for all other channel configurations ArcSoft is currently still the default decoder option (if it's available, otherwise dcadec becomes default).

So it's been more than 6 months since this post. Has anyone discovered any really bad bugs with -dcadec? Should I force it's use with all DTS now?

[Sparktank]

So it's been more than 6 months since this post. Has anyone discovered any really bad bugs with -dcadec? Should I force it's use with all DTS now?

Still issues with DTS Express formats.
https://github.com/foo86/dcadec

Features not implemented:

Decoding of DTS Express streams
Applying dynamic range compression and dialog normalization

[r0lZ]

:thanks:
I noticed the audio choppiness in one of my remuxes as well. I'm using serviio (http://serviio.org) media streamer from my Windows 7 desktop PC to Samsung TV (model UE40D6000) over cabled LAN (not wi-fi). I assumed that choppiness was due to extremely high bitrate of video+audio stream since it's a straight remux of Disney's Frozen BD but maybe it was indeed due to some strange bug that involves specific MkvMerge / M4A audio file / TV firmware / whatever. I'll recheck when I'll find some spare time...
My Samsung TV is an UE40D6500, but I play most movies directly with the TV (HDD connected via USB, not streaming).
I will try with a straight remux to see if I have the same problem. Can you describe exactly what you did to obtain the "straight remux"? Demuxing with eac3to or tsMuxeR? What streams have you kept? All audio streams? And the AVC video has not been re-encoded, right?

[r0lZ]

Has anyone discovered any really bad bugs with -dcadec?It is now used by default by BD3D2MK3D, and nobody has reported problems. But since BD3D2MK3D processes only 3D movies and DTS-Express is not allowed as primary audio, BD3D2MK3D should never need to convert DTS-Express tracks. IMO, eac3to should use DCADEC by default for all DTS tracks except DTS-Express.

[tebasuna51]

Like Sparktank, and r0lZ, say dcadec work fine (even better than ArcSoft) for all DTS except DTS Express.

"Applying dynamic range compression and dialog normalization" is a not desired feature, eac3to always try not apply dynamic range compression and dialog normalization to decode sources.

[Xorp]

It is now used by default by BD3D2MK3D, and nobody has reported problems. But since BD3D2MK3D processes only 3D movies and DTS-Express is not allowed as primary audio, BD3D2MK3D should never need to convert DTS-Express tracks. IMO, eac3to should use DCADEC by default for all DTS tracks except DTS-Express.

Thanks for the info guys. I can't ever see myself needing DTS-Express decoding since they are usually for the PiP audio, so sounds like dcadec is the winner.

[gabbett1]

So I've been trying to convert my Bluray Insurgent to an MP4 and the old version of Ripbot I had been using wouldn't find any audio tracks other than 2.0 versions. I decided to upgrade to this newest version, also updating all the components needed (AviSynth , Haali Media Spliter , ffdshow). I'm running windows 7 btw. So anyhow, now I can't demux my streams. I keep getting a demuxing error:

eac3to v3.29
command line: "J:\Blue Ray Ripping\Tools\eac3to\eac3to.exe" "J:\Rip\INSURGENT\" 1) 2: "J:\Temp\RipBot264temp\job1\video.mkv" -seekToIFrames 3: "J:\Temp\RipBot264temp\job1\audio_1_English.thd.w64" -down16 -down6 8: "J:\Temp\RipBot264temp\job1\8_subtitles_English_1080.sup" 9: "J:\Temp\RipBot264temp\job1\9_subtitles_English_1080.sup" 10: "J:\Temp\RipBot264temp\job1\10_subtitles_Spanish_1080.sup" 1: "J:\Temp\RipBot264temp\job1\chapters.txt" -progressnumbers -log="J:\Temp\RipBot264temp\job1\demuxlog.txt"
------------------------------------------------------------------------------
M2TS, 1 video track, 5 audio tracks, 3 subtitle tracks, 2:00:00, 24p /1.001
1: Chapters, 16 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: TrueHD (Atmos), English, 7.1 channels, 48kHz
4: AC3 Surround, English, 2.0 channels, 224kbps, 48kHz
5: AC3 Surround, English, 2.0 channels, 224kbps, 48kHz
6: AC3, Spanish, 5.1 channels, 640kbps, 48kHz
7: AC3 Surround, English, 2.0 channels, 224kbps, 48kHz
8: Subtitle (PGS), English
9: Subtitle (PGS), English
10: Subtitle (PGS), Spanish
[a03] thd, 48000, 7.1
Creating file "J:\Temp\RipBot264temp\job1\chapters.txt"...
[s10] Extracting subtitle track number 10...
[a03] Extracting audio track number 3...
[s08] Extracting subtitle track number 8...
[s09] Extracting subtitle track number 9...
[v02] Extracting video track number 2...
[a03] Decoding with libav/ffmpeg...
[a03] Mixing surround channels...
[a03] Reducing depth from 24 to 16 bits...
[a03] Writing W64...
[a03] The libav decoder reported error -1094995529 while decoding. <ERROR>
Aborted at file position 1048576. <ERROR>

I need help resolving this issue.

[73ChargerFan]

I think the Insurgent BD is engineered to break eac3to. It appears to use playlist spoofing, so that the playlist which eac3to identifies as the main movie playlist " 1) " is likely incorrect.

To confirm, run:
"J:\Blue Ray Ripping\Tools\eac3to\eac3to.exe" "J:\Rip\INSURGENT\"

and it will tell you which playlist "xxxxx.mpls" it associates with 1)

Next play that playlist using mpc-hc or mpc-be to determine if it is valid or garbage.

[Thunderbolt8]

in case of such multiple potentially correct playlists it has always been the case that the correct one does not necessarily have to be 1)

[gabbett1]

I think the Insurgent BD is engineered to break eac3to. It appears to use playlist spoofing, so that the playlist which eac3to identifies as the main movie playlist " 1) " is likely incorrect.

To confirm, run:
"J:\Blue Ray Ripping\Tools\eac3to\eac3to.exe" "J:\Rip\INSURGENT\"

and it will tell you which playlist "xxxxx.mpls" it associates with 1)

Next play that playlist using mpc-hc or mpc-be to determine if it is valid or garbage.

Hmm, well originally I used Process Monitor to find the correct playlist. It just happened to be #1 on the list. Maybe I'll have to try it again to make sure I found the correct mpls the first time.

As far as this goes:
To confirm, run:
"J:\Blue Ray Ripping\Tools\eac3to\eac3to.exe" "J:\Rip\INSURGENT\"

When I look in that folder the only things inside are:
yr_eac3to_more_gui.exe
yr_eac3to_more_gui.ini

[gabbett1]

Ran the Process Monitor again and realized I made a mistake. Now the .mpls I'm coming up with is 407. However, when I tried to run it, I still get this:

eac3to v3.29
command line: "J:\Blue Ray Ripping\Tools\eac3to\eac3to.exe" "J:\Rip\INSURGENT\" 110) 2: "J:\Temp\RipBot264temp\job2\video.mkv" -seekToIFrames 3: "J:\Temp\RipBot264temp\job2\audio_1_English.thd.w64" -down16 8: "J:\Temp\RipBot264temp\job2\8_subtitles_English_1080.sup" 9: "J:\Temp\RipBot264temp\job2\9_subtitles_English_1080.sup" 10: "J:\Temp\RipBot264temp\job2\10_subtitles_Spanish_1080.sup" 1: "J:\Temp\RipBot264temp\job2\chapters.txt" -progressnumbers -log="J:\Temp\RipBot264temp\job2\demuxlog.txt"
------------------------------------------------------------------------------
M2TS, 1 video track, 5 audio tracks, 3 subtitle tracks, 1:59:00, 24p /1.001
1: Chapters, 16 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: TrueHD (Atmos), English, 7.1 channels, 48kHz
4: AC3 Surround, English, 2.0 channels, 224kbps, 48kHz
5: AC3 Surround, English, 2.0 channels, 224kbps, 48kHz
6: AC3, Spanish, 5.1 channels, 640kbps, 48kHz
7: AC3 Surround, English, 2.0 channels, 224kbps, 48kHz
8: Subtitle (PGS), English
9: Subtitle (PGS), English
10: Subtitle (PGS), Spanish
[a03] thd, 48000, 7.1
Creating file "J:\Temp\RipBot264temp\job2\chapters.txt"...
[s10] Extracting subtitle track number 10...
[s08] Extracting subtitle track number 8...
[s09] Extracting subtitle track number 9...
[v02] Extracting video track number 2...
[a03] Extracting audio track number 3...
[a03] Decoding with libav/ffmpeg...
[a03] Reducing depth from 24 to 16 bits...
[a03] Writing W64...
[a03] The libav decoder reported error -1094995529 while decoding. <ERROR>
Aborted at file position 1048576. <ERROR>

[Thunderbolt8]

maybe its because the decoder doesnt know how to process the atmos track, how to downsample all its information.

[ScottJ]

[a03] The libav decoder reported error -1094995529 while decoding. <ERROR>
Aborted at file position 1048576. <ERROR>

I got the same issue today with Insurgent Blu-ray (US retail disc) ripped through ClownBD. Playlist 407, as determined by AnyDVD HD.

[r0lZ]

Have you tried to demux with tsMuxeR?

[jpsdr]

maybe its because the decoder doesnt know how to process the atmos track, how to downsample all its information.

If i remember properly, when "recently" madshi update eac3to to fix some little issues and add the new dts decoder, he didn't update the ffmpeg part of eac3to, which begin to be a little old. So there is a big chance that this statement is indeed the right answer.