eac3to - audio conversion tool

[Yraen]

And my second stupid question is - how can I get from eac3to "WAVs (multiple mono WAV files, PCM only)"? (I gave up playing with eac3to options).

You actually answered your own question there, wavs is the right answer.

eac3to input.file output.wavs

[monohouse]

-----

[jchappo]

Don't have much time right now, but the "WAV" file posted by jchappo is neither WAV nor RAW. It seems to contain blocks of LPCM data with a 4 byte header for each block. It seems to me that either tsMuxer uses its own private data format for storing LPCM data, or there's a bug in tsMuxer's demuxing code, or there's a new "container" format I've never heard about yet.

I demuxed with tsremux, and muxed with tsmuxer so I don't think it's related to a specific program?

tsmuxer won't accept the pcm output from eac3to, so that extra header is what it needs. could the header be related to the m2ts container it came from?

[jchappo]

why swapping endian is necessary for raw/pcm to flac conversion ?
is that process looses data ?

why does it remap channels ? I wasn't awear that remapping channels was required, is the process losses data ? is it possible to avoid both of the processes ?

endian swap is just a re-ordering of the bits different people use different endian, so it's needed to swap between them.

same thing with channel mapping, some people use different order.

[monohouse]

-----

[jchappo]

well you have the exact same number of bits, they are just in different order.

[Thunderbolt8]

the AVC video track of the 'things we lost in the fire' is indicated as a 16:9 track with pulldown flags. is this of any importance for me when I just want to remux the movie? so far I just remuxed it as usual and it also looks normally, but I just want to be sure.

[n0mag!c]

You actually answered your own question there, wavs is the right answer.
eac3to input.file output.wavs
Wow! Thanks a lot! I would never figure it out by myself! Well it's obvious and not obvious at the same time.

[Rectal Prolapse]

Well crap to that then. Do the newer versions (of Scenarist) 4.3 or 4.5 fix any of this?

I think it was fixed in 4.3, but I'm not 100% sure.

[EPiPH0NE]

I think it was fixed in 4.3, but I'm not 100% sure.


I just got 4.3 and I know you can do multi mono WAVs -> LPCM for sure.

[Momber]

Hey madshi!

Thanks for your continuing work on eac3to. You have developed it into one helluva tool!

This week I've been working on converting the Galactica Season 1 HD DVDs to BluRay, using eac3to for audio and EVOdemux + vc1conv for the video.
The VC-1 video processed like this is recognized by tsMuxeR as 1080p @ 23.976 fps and the remuxed BluRays play back perfectly smooth in PDVD 8 at a refresh rate of 2x 23.976 Hz.

So I thought I'd give eac3to a go on the video front, too, and had it demux and remove the pulldown flags. The resulting vc-1 file is recognized by tsMuxeR as 1080i @ 23.976 fps and what's worse, playback got really jerky about 17 minutes into the episode I tried.

Is it so that eac3to only removes the pulldown=1 flag but leaves the interlaced=1 flag in place? And if that is the case, would you please be so kind as to correct it?

TIA!
S.

[dorati]

I have a question:
Eac3To reportet this:

1) 00120.mpls (angle 1), 00152.m2ts+00153.m2ts+00154.m2ts, 1:26:15
- h264/AVC, 1080p24 /1.001 (16:9)
- AC3, English, multi-channel, 48khz
- AC3, German, multi-channel, 48khz
- RAW/PCM, German, multi-channel, 48khz
- AC3, English, stereo, 48khz

2) 00120.mpls (angle 2), 00152.m2ts+00155.m2ts+00154.m2ts, 1:26:15
- h264/AVC, 1080p24 /1.001 (16:9)
- AC3, English, multi-channel, 48khz
- AC3, German, multi-channel, 48khz
- RAW/PCM, German, multi-channel, 48khz
- AC3, English, stereo, 48khz

3) 00037.mpls, 00004.m2ts+00005.m2ts, 0:22:50
- MPEG2, 480i30 /1.001 (16:9)
- AC3, English, stereo, 48khz

4) 00004.mpls, 00004.m2ts, 0:15:14
- MPEG2, 480i30 /1.001 (16:9)
- AC3, English, stereo, 48khz

What I must use in the commandline, to use: 2) 00120.mpls (angle 2) ??

[nautilus7]

Type the command you typed to list the above plus 2) at the end.

See my example here (http://forum.doom9.org/showpost.php?p=1123939&postcount=4226)for details.

[Yraen]

So I thought I'd give eac3to a go on the video front, too, and had it demux and remove the pulldown flags. The resulting vc-1 file is recognized by tsMuxeR as 1080i @ 23.976 fps and what's worse, playback got really jerky about 17 minutes into the episode I tried.


Here (http://forum.doom9.org/showthread.php?p=1110490#post1110490) you can see where madshi says that vc1conv is better than eac3to for removing pulldown and temporarily disabled -stripPulldown. You could go eac3to > vc1conv > tsMuxer and save a step. This will get you 1080p @ 23.976.

[BlackJack1]

I can not drag'n'drop eac3to app. and hddvd/bd files to cmd window on Vista system on my laptop. Do you know how to fix it? In desktop machine with WinXP - no problem...

[dorati]

@nautilus7:
Thx !!

Somebody has succesfully converted the Movie "Open Season"?

The Movie is 00152.m2ts + 00153.m2ts + 00154.m2ts.
or in German 00152.m2ts + 00155.m2ts + 00154.m2ts.

But the change from 00152.m2ts to 00153/00154.m2ts is not clean. My standaloneplayer (popcorn) hangs at the change.
At Windows-PC with Powerdvd is not better. It stops a short Moment and than goes on with playback.

I testet all:
- copy /b 00152.m2ts + 00153.m2ts + 00154.m2ts
- demux and mux with TSMuxer
- demux with eac3to (playlist) and Mux with TSMuxer
- convert to MKV

Somebody can help?

[nautilus7]

Can you post the eac3to log? Re ripping the disc could make the trick.

[Momber]

here you can see where madshi says that vc1conv is better than eac3to for removing pulldown and temporarily disabled -stripPulldown.
Thank you. That was an older version and I thought perhaps in the meantime things were improved.
You could go eac3to > vc1conv > tsMuxer and save a step. This will get you 1080p @ 23.976.
Right - that's what I did in the end. It's just that with eac3to -removePulldown we would be able to save another step. When I'm converting episode after episode of TV series I'm grateful for every chance to streamline my workflow.
The batch creation of eac3to_more_gui is heaven-sent in that regard btw.

Ta
S.

[Sephiroth0000]

Hiya everyone,

I am currently using EAC3TO GUI (Keymakers version) and noticed that one of the options is video quality which spans from quality 0 to quality 4....can anyone tell me the differences as I am trying to make the mkvs smaller but do not want to sacrifice too much quality.

[nautilus7]

From 1st post:

-quality=4 slowdown/speedup/resampling quality (0 = low; 4 = very high)

So, nothing to do with video. Probably a bug. My advice is to use the new eac3to gui (http://forum.doom9.org/showthread.php?t=135095), since it's up-to-date.

[mic64]

Hi All

I got a problem with all FLAC Files.
its not the encoding into it or playing them back.
Its something different.
I wanted to extract some flac from my MKVs.
mkvextract gives you this
Extracting track 2 with the CodecID 'A_FLAC' to the file 'z:\xxxxx.flac'. Container format: Ogg (FLAC in Ogg)

Well ok so far I thought. But..
those flac files are not being recognized by eas3to at all, so no way to convert them into something else.

Any idea? or solution?

[Snowknight26]

mkvextract tracks source.mka --no-ogg 1:dest.flac

[EPiPH0NE]

@nautilus7:
Thx !!

Somebody has succesfully converted the Movie "Open Season"?

The Movie is 00152.m2ts + 00153.m2ts + 00154.m2ts.
or in German 00152.m2ts + 00155.m2ts + 00154.m2ts.

But the change from 00152.m2ts to 00153/00154.m2ts is not clean. My standaloneplayer (popcorn) hangs at the change.
At Windows-PC with Powerdvd is not better. It stops a short Moment and than goes on with playback.

I testet all:
- copy /b 00152.m2ts + 00153.m2ts + 00154.m2ts
- demux and mux with TSMuxer
- demux with eac3to (playlist) and Mux with TSMuxer
- convert to MKV

Somebody can help?

If you have a PCH why not read the forums? ES streams in MKV is NOT supported, you need to keep it in TS/M2TS. The answer to AVC + seamless branching is right here:

http://www.networkedmediatank.com/viewtopic.php?t=2921

Scroll down to where it says 'Seamless Branching (AVC)'. You can replace TSSplitter with tsMuxeR for doing the first join which I will do when I update the guide.

[mic64]

mkvextract tracks source.mka --no-ogg 1:dest.flac

forgot to say..the --no-ogg option didn't help to get a pure flac from the mkv.
But I've found a way.
mplayer -dumpaudio -dumpfile works.

[madshi]

I've got something new going on with a seamless branching disc. Since the latest versions of eac3to support seamless branching I decided to give Nine Inch Nails: Beside You in Time another try.
Could I have a sample, please, with which this problem occurs?

I have used this tool to rip my BD and HDDVDs. Two days ago, I tried to rip Chris Botti Live with friends, but for some reason, I can't seem to play this after the conversion. I have tried this on 2.44 and 2.45 version. Both of them produces the gap information.
For video or audio or for both?

Can I have a sample of the m2ts file, please?

I've decoded E-AC3 1536kbps to RAW file with GraphEdit and its Dump filter. By the way, I did it with Nero(8!) Audio Decoder 2, so how it correlates with yours "Nero 8 doesn't allow its DirectShow filters to be used from outside of Nero ShowTime"?
Huh!? Maybe Nero changed something in the later Nero 8 builds? Can somebody double check this?

I demuxed with tsremux, and muxed with tsmuxer so I don't think it's related to a specific program?

tsmuxer won't accept the pcm output from eac3to, so that extra header is what it needs. could the header be related to the m2ts container it came from?
It seems to me that TsRemux outputs a specific LPCM format which TsMuxer understands and likes. People always complained about that TsRemux' LPCM demuxing produces files with garbage in them. Maybe the headers in the demuxed files are kept there intentional? But I don't think this is any "official" format, at least I've never heard about that yet.

so the flac "multichannel standard" has different positions for every channel than the track in question ? so eac3to can actually tell which channel is what ?

and the endian is also because flac uses a different endian ?
Each codec has its own private channel ordering. E.g. LPCM data stored in m2ts has a different channel ordering than WAV files have. eac3to knows the channel ordering of all supported input formats/codecs and remaps the channels to the WAV channel order in the first step. Furthermore LPCM data is stored in big endian while WAV files are usually stored in little endian. So eac3to changes that, too.

But don't worry, these are all totally lossless modifications.

the AVC video track of the 'things we lost in the fire' is indicated as a 16:9 track with pulldown flags. is this of any importance for me when I just want to remux the movie? so far I just remuxed it as usual and it also looks normally, but I just want to be sure.
All HD DVD movies have pulldown flags, older eac3to versions just didn't show this information. So there's nothing you need to do differently than before.

You have developed it into one helluva tool!
Thanks... :) It gets a bit better every week...

So I thought I'd give eac3to a go on the video front, too, and had it demux and remove the pulldown flags. The resulting vc-1 file is recognized by tsMuxeR as 1080i @ 23.976 fps and what's worse, playback got really jerky about 17 minutes into the episode I tried.

Is it so that eac3to only removes the pulldown=1 flag but leaves the interlaced=1 flag in place? And if that is the case, would you please be so kind as to correct it?
eac3to's current pulldown removal is based on an outdated vc1conv version. At some time I had even removed it (because the latest vc1conv is better), but put it back in on request. I will probably include the latest vc1conv logic in a future eac3to build. But I've only limited programming time for eac3to, so I can only do so much every week. For now if you need to have the pulldown completely removed please use eac3to for demuxing and then vc1conv afterwards.

Somebody has succesfully converted the Movie "Open Season"?

The Movie is 00152.m2ts + 00153.m2ts + 00154.m2ts.
or in German 00152.m2ts + 00155.m2ts + 00154.m2ts.

But the change from 00152.m2ts to 00153/00154.m2ts is not clean. My standaloneplayer (popcorn) hangs at the change.
At Windows-PC with Powerdvd is not better. It stops a short Moment and than goes on with playback.
That's strange! Did eac3to report any *video* gaps? How long is that 00152.m2ts file? It would help if you could send me a the last 20MB of the 00152.m2ts file and the first 20MB of the 00153.m2ts file.

[Encoder888]

@madshi

eac3to 2.45 successfully extracted the correct chapters from The Terminator, but now I'm having another chapter issue with Terminator 3. eac3to reports and demuxes 97 chapters, almost one every 1-2 minutes, which I'm sure is not the correct chapter structure. Here's a link to the .mpls file:

http://www.mediafire.com/?n2cxkcm9amv

Please, let me know if you were able to fix it, so I know whether to wait for the next eac3to release before I mux my chapter file :) Thanks.

[Atak_Snajpera]

Madshi
Could you add --demuxaudioonly switch? Or something similar which will extract all audio streams without video.

[dorati]

@epiphone:
If you have a PCH why not read the forums? ES streams in MKV is NOT supported, you need to keep it in TS/M2TS. The answer to AVC + seamless branching is right here

I read - but this way don't work - Sorry.

1. Joining with TSSplitter or copy /b (i tested both - same result)
2. Demux AVC-Videostream and AC3-Audiostream with TsRemux
3. Mux with tsMuxer

Later the join-point the video ist stuttering and later 20 Sec. hangs. The stuttering you can see in the short clips, I uploaded.

@madshi:
Audiogaps are reported - but no Videogaps

http://rapidshare.de/files/39377863/00152.m2ts.html
http://rapidshare.de/files/39377791/00155.m2ts.html

[madshi]

@tebasuna51, do you happen to have a sample for acmod 1 with LFE ("1/0.1")? I don't understand why it shows as "2.0" for you. Looking at my source code it should show as "1.1".

[EPiPH0NE]

@epiphone:


I read - but this way don't work - Sorry.

1. Joining with TSSplitter or copy /b (i tested both - same result)
2. Demux AVC-Videostream and AC3-Audiostream with TsRemux
3. Mux with tsMuxer

Later the join-point the video ist stuttering and later 20 Sec. hangs. The stuttering you can see in the short clips, I uploaded.

@madshi:
Audiogaps are reported - but no Videogaps

http://rapidshare.de/files/39377863/00152.m2ts.html
http://rapidshare.de/files/39377791/00155.m2ts.html

How are you playing these files cause I've personally tested this method several times with flawless playback from internal HDD.

[dorati]

At PC (PowerDVD) it works - at Standalone not.
I use NMT (http://www.syabas.de/)

[madshi]

eac3to 2.45 successfully extracted the correct chapters from The Terminator, but now I'm having another chapter issue with Terminator 3. eac3to reports and demuxes 97 chapters, almost one every 1-2 minutes, which I'm sure is not the correct chapter structure.
BDEdit outputs the same chapters. Why are you sure that this is not correct?

The problem is this: There are 2 different types of marks in the playlist files: "entry points" and "link points". Originally I only used "entry points" for the chapter export. This resulted in the problem you experienced with "The Terminator". The first half of the chapters in "The Terminator" were "entry points" and the 2nd half were "link points". Now with Terminator 3 there are about 3 "link points" for every "entry point". If I'd ignore the "link points", the chapter list would be more reasonable for Terminator 3. However, if I ignore the "link points" then The Terminator will only have chapters in the first half of the movie. BDEdit exports both "entry points" and "link points".

[madshi]

I read - but this way don't work - Sorry.

1. Joining with TSSplitter or copy /b (i tested both - same result)
2. Demux AVC-Videostream and AC3-Audiostream with TsRemux
3. Mux with tsMuxer

Later the join-point the video ist stuttering and later 20 Sec. hangs. The stuttering you can see in the short clips, I uploaded.
Thanks for the upload. Next time if possible please use a different server, though. rapidshare is the worst of all. I had to wait a minute to download the first sample and than over an hour to download the 2nd one. Anyway...

I've checked the samples you uploaded and they work beautifully with eac3to v2.45 for me. I've simply done this:

"eac3to 00152.m2ts+00155.m2ts movie.mkv" and the resulting MKV played perfectly fine with no stuttering at all on my HTPC with Media Player Classic (Haali Media Splitter + Cyberlink h264 decoder). The join point between the two m2ts parts was not visible at all.

Have you tried this command line? Are you using eac3to v2.45?

[madshi]

Could you add --demuxaudioonly switch? Or something similar which will extract all audio streams without video.
Of course I could. But I don't really like adding and documenting a lot of options because every option makes the help text more complicated. Maybe some day I'll split the help into "main options" and "expert options" or something like that. Then it wouldn't hurt as much if I add some more options. But right now I prefer to keep the option list as short as possible. Of course you can do the same as "-demuxaudioonly" would do by typing each audio track into the command line. So such an option would just save you a few key presses. Because of that I don't consider it very important at this point in time. I'm more worried about making everything work well right now...

[jchappo]

madshi, any chance of getting a switch -tsmuxer to output PCM with the extra data so I can remux these PCM tracks with tsMuxer?

or, do you know of a program that could create this extra data?

[tebasuna51]

@tebasuna51, do you happen to have a sample for acmod 1 with LFE ("1/0.1")? I don't understand why it shows as "2.0" for you. Looking at my source code it should show as "1.1".
You are right, is my fault, sorry. :confused:
I edited my post (http://forum.doom9.org/showthread.php?p=1134394#post1134394)

[Snowknight26]

[a03] This TS/M2TS file seems to be damaged (sync byte missing).
Aborted at file position 11482513408.
Blu-ray.. need a sample?

[madshi]

madshi, any chance of getting a switch -tsmuxer to output PCM with the extra data so I can remux these PCM tracks with tsMuxer?
I don't know exactly what tsMuxer expects. E.g. does it want to have a different block size for 2.0 tracks compared to 5.1 tracks? I don't feel like experimenting. If there's an official documentation/explanation somewhere which format tsMuxer exactly wants to have then I can add support for that. But actually I think Roman should really add support for big WAV files and/or multi mono WAV files. Because that's a standard format. That "LPCM with headers" format is totally non-standard, as far as I can say...

or, do you know of a program that could create this extra data?
No.

[a03] This TS/M2TS file seems to be damaged (sync byte missing).
Aborted at file position 11482513408.
Blu-ray.. need a sample?
I don't think a sample will help. This looks like a real damage in the source file. A missing sync byte is a very simple check. There's not really a big chance that eac3to has a bug in this situation.

Is this a one part movie? Or does the movie consist of multiple m2ts files? In the latter case check the file size of all m2ts parts. Are they all divisible by 192?

[lexor]

Hi guys, I have a couple of question on the tool use.

1) How do we use the following command
-seekToIFrames make all h264/AVC "I" frames seekable
It says that audio formats are supported only in demuxed form, so how would it know anything about avc i-frames?

2) What does this line (from the help file) do?
eac3to source.evo 1: chapters.txt 2: video.mkv 3: audio.flac 5: subtitle.sup

[madshi]

eac3to v2.46 released

http://madshi.net/eac3to.zip

* MPEG2 muxing now fully supports streams with mixed 23.976 and 29.970 content
* mixed video/movie MPEG2 streams are now always muxed with 29.970 timestamps
* if a movie MPEG2 stream goes 29video, processing is automatically restarted
* MPEG2 pulldown is now automatically removed whenever an MPEG2 stream is read
* new option "-keepPulldown" can be used to disable MPEG2 pulldown removal
* corrected default WAV channel masks for 4.0, 6.1 and 7.1
* added proper channel remaps for libav AC3 decoding of "funny" channel formats
* added general channel mask support
* WAV parser reads channel mask from extensible header
* (E-)AC3 parser sets correct channel mask
There are 2 bigger changes in this release:

(1) MPEG2 handling is noticably improved. Especially MPEG2 streams which contain a mixture of movie content (24p or 48i with pulldown flags) and video content (60i) should be handled perfectly now. Also the pulldown is now automatically removed for movie content which has pulldown flags.

(2) The audio processing chain got full support for custom channel masks. This is necessary to handle funny channel combinations like e.g. "5.0" correctly. However, the channel mask support is not complete yet. Especially the DTS, TrueHD, MLP and MP2 parsers do not always properly set the correct channel masks yet. So if you guys have any DTS/TrueHD/MP2 samples with funny channel masks, please upload samples for me. Thanks!

@tebasuna51, I think I implemented all the changes from your detailed AC3 testing post. But I couldn't really test it because I don't have AC3 samples other than 1.0, 2.0 and 5.1. If you have time, maybe you could check whether my fixed work correctly? Thanks!

[madshi]

1) How do we use the following command
-seekToIFrames make all h264/AVC "I" frames seekable
It says that audio formats are supported only in demuxed form, so how would it know anything about avc i-frames?
I'm confused. First of all AVC is video and not audio. Furthermore neither audio nor video are supported only in demuxed form. EVO/VOB/TS/M2TS containers are fully supported with most video and audio codecs.

P.S: I just noticed that the help text still stated "only in demuxed form" for the audio formats. This is outdated. I just removed it from the first page and from my source code. However, it didn't apply to AVC, cause AVC is video and not audio.

2) What does this line (from the help file) do?
eac3to source.evo 1: chapters.txt 2: video.mkv 3: audio.flac 5: subtitle.sup
It reads the source EVO file and creates all the other files listed in the command line. Of course the track numbers ("1:" etc) vary, depending on the source file. So you first have to do "eac3to sourcefile" to get a proper track listing. Then you can build the correct command line from there by using the correct track numbers.

[lexor]

P.S: I just noticed that the help text still stated "only in demuxed form" for the audio formats. This is outdated. I just removed it from the first page and from my source code. However, it didn't apply to AVC, cause AVC is video and not audio.


Yeah that's the part that got me, since if audio is in the demuxed form it wouldn't know anything about the video. With the update about container support, it makes sense now, thank you.

Also while there are recommended DTS en/decoder, there is only recommended AC3 decoder, what about encoder? Aften as good as it gets?

[Thunderbolt8]

thanks for the improvements again! :)
will test my tricky videos with the new version again, but I wont be home before 1 week, so I guess it will take 1-2 weeks until then.

[tebasuna51]

eac3to v2.46 released

@tebasuna51, I think I implemented all the changes from your detailed AC3 testing post. But I couldn't really test it because I don't have AC3 samples other than 1.0, 2.0 and 5.1. If you have time, maybe you could check whether my fixed work correctly?

Thanks madshi.
All work fine, maskchannels and channel map, with ac3 sources and libav decoder.

Using a 3/3.1 wav still:
eac3to v2.46
command line: "D:\Test\AudioN\eac3to\eac3to.exe" "E:\Test\7_61.wav" "E:\Test\z331.wav" -extensible
------------------------------------------------------------------------------
WAV, 6.1 channels, 0:00:20, 16 bits, 48khz
Doubling 7th channel...
Reading WAV...
Writing WAV...
Creating file "E:\Test\z331.wav"...
eac3to processing took 1 second.
Done.
And the final wav is 8 channels with the default channel mask from 7 channels.

[Snowknight26]

I don't think a sample will help. This looks like a real damage in the source file. A missing sync byte is a very simple check. There's not really a big chance that eac3to has a bug in this situation.

Is this a one part movie? Or does the movie consist of multiple m2ts files? In the latter case check the file size of all m2ts parts. Are they all divisible by 192?

The whole film is one m2ts. Wouldn't ignoring the missing sync byte help? I know xport can demux it though, so I guess it must be ignoring it.

[STaRGaZeR]

Ey madshi, thanks for this program :)

I'm having a weird problem I also posted here: http://forum.doom9.org/showpost.php?p=1136225&postcount=1461

The sample in that post has 4 audio tracks: 1-DTS 2-LPCM 3-AC3 4-AC3

tsMuxer only detects 1, 3 and 4, and eac3to only 3 and 4. Can you help me with this? I just want to demux that LPCM track, it plays fine but I have this problem.

[MichalHabart]

eac3to v2.46 released

http://madshi.net/eac3to.zip

* MPEG2 muxing now fully supports streams with mixed 23.976 and 29.970 content
* mixed video/movie MPEG2 streams are now always muxed with 29.970 timestamps
* if a movie MPEG2 stream goes 29video, processing is automatically restarted
* MPEG2 pulldown is now automatically removed whenever an MPEG2 stream is read
* new option "-keepPulldown" can be used to disable MPEG2 pulldown removal
* corrected default WAV channel masks for 4.0, 6.1 and 7.1
* added proper channel remaps for libav AC3 decoding of "funny" channel formats
* added general channel mask support
* WAV parser reads channel mask from extensible header
* (E-)AC3 parser sets correct channel mask
There are 2 bigger changes in this release:

(1) MPEG2 handling is noticably improved. Especially MPEG2 streams which contain a mixture of movie content (24p or 48i with pulldown flags) and video content (60i) should be handled perfectly now. Also the pulldown is now automatically removed for movie content which has pulldown flags.

(2) The audio processing chain got full support for custom channel masks. This is necessary to handle funny channel combinations like e.g. "5.0" correctly. However, the channel mask support is not complete yet. Especially the DTS, TrueHD, MLP and MP2 parsers do not always properly set the correct channel masks yet. So if you guys have any DTS/TrueHD/MP2 samples with funny channel masks, please upload samples for me. Thanks!

@tebasuna51, I think I implemented all the changes from your detailed AC3 testing post. But I couldn't really test it because I don't have AC3 samples other than 1.0, 2.0 and 5.1. If you have time, maybe you could check whether my fixed work correctly? Thanks!

Thanks Madshi for excellent work. I have just one question. Will extraction of subtitles from bluray be supported in any future release?

[Yraen]

Thanks Madshi for excellent work. I have just one question. Will extraction of subtitles from bluray be supported in any future release?

It already is. I just did one about an hour ago.

[MichalHabart]

It already is. I just did one about an hour ago.

And how did you use eac3to to export subtitles from m2ts file? What command did you use?

[bmnot]

TS, 1 video track, 1:55:47
1: MPEG2, 1080p24 /1.001 (16:9) with pulldown flags
[v01] Extracting video track number 1...
[v01] Removing MPEG2 pulldown...
[v01] Muxing video to Matroska...
[v01] The MPEG2 stream is a mixture of video and movie content.
[v01] This type of MPEG2 stream cannot be muxed with 24p timestamps.
[v01] Will have to abort processing and redo everything.
[v01] Extracting video track number 1...
[v01] Muxing video to Matroska...
[v01] The MPEG2 stream is a mixture of video and movie content.

So it includes the pulldown flags in the 2nd try? Should I hold onto the original ts in case 24p will work on it in future eac3to revisions?