eac3to - audio conversion tool

[asc28]

No. Because volume change is done in floating point, while PCM is integer. Conversion from floating point to integer is not lossless.Thanks for your help! I think I'll convert to floating-point PCM to apply the volume change, and then feed the floating-point directly to Nero AAC encoder. Hopefully, if I understand these concepts correctly, this should not result in much quality loss, had I just encoded an originally correct-volume PCM to AAC.

I'll also try to contact ArcSoft about the incorrect decoding.

[madshi]

Thanks for your help! I think I'll convert to floating-point PCM to apply the volume change, and then feed the floating-point directly to Nero AAC encoder. Hopefully, if I understand these concepts correctly, this should not result in much quality loss, had I just encoded an originally correct-volume PCM to AAC.
The quality loss (compared to correct-volume PCM to AAC) should be minimal, most probably not audible for any normal person.

I'll also try to contact ArcSoft about the incorrect decoding.
If you get any useful reply, please let us know.

[Boulder]

A (yet another) feature request: ability to use the SoundTouch library for audio speedup or slowdown without changing the audio pitch.

[asc28]

I'll convert to floating-point PCM to apply the volume change, and then feed the floating-point directly to Nero AAC encoder. Hopefully, if I understand these concepts correctly, this should not result in much quality loss, had I just encoded an originally correct-volume PCM to AAC.I actually might instead change the volume by applying gain on the AAC post-encode. I believe this would be a lossless process if the gain can be applied the same way mp3gain does it (by modifying the frame headers). Will eac3to do it this way (change volume losslessly by modifying ADTS frame headers on the AAC) if I do: eac3to input.aac output.aac -Xdb?

[edit] Also here are some tests I did w/ the latest WaveGain:wavegain sin51.C.wav sin71.C.wav sin51.LFE.wav sin71.LFE.wav sin51.L.wav sin71.L.wav sin51.R.wav sin71.R.wav

Analyzing...

Gain | Peak | Scale | New Peak |Left DC|Right DC| Track
| | | |Offset | Offset |
--------------------------------------------------------------
-1.16 dB | 32767 | 0.87 | 28671 | 0 | 0 | sin51.C.wav
+2.19 dB | 21538 | 1.29 | 27715 | 0 | 0 | sin71.C.wav
-0.00 dB | 32767 | 1.00 | 32766 | 0 | 0 | sin51.LFE.wav
+3.64 dB | 21538 | 1.52 | 32765 | 0 | 0 | sin71.LFE.wav
+0.15 dB | 32209 | 1.02 | 32766 | 0 | 0 | sin51.L.wav
+3.64 dB | 21538 | 1.52 | 32765 | 0 | 0 | sin71.L.wav
+0.02 dB | 32681 | 1.00 | 32766 | 0 | 0 | sin51.R.wav
+3.64 dB | 21538 | 1.52 | 32765 | 0 | 0 | sin71.R.wav

WaveGain Processing completed normallyLooks like the volume difference is not exactly the same for every channel (varies from 3.35 to 3.64), though this might be explained by how the extra 2 channels are mapped / DTS compression.

[asarian]

Oh, I thought that message could've been included in the new eac3to versions. Is there any alternative to decoding this track (Sonic doesn't do 7.1, right)? Would love to have Sin City in beautiful 7.1.

Also, was this problem discussed in this thread/elsewhere? What specifically is wrong w/ the decoded track? My interpretation is that the "strange channel setup" messes w/ the volume? Thanks for the help

I have had both Sin-City Blu-Rays. The Japanese actually has TrueHD 5.1 audio (instead of just DD 5.1), but the track is simply broken: if you try to extract it, you'll find it to be corrupted, and eac3to aborted at the time, so I could only salvage the AC3 core. The Scandinavian Blu-Ray comes with an odd reddish tint in the video, but has a flawsless DTS-HD MA 7.1 track. So, I finally remuxed the video from the Japanese disc with the audio from the Scandinavian one, threw out the Japanese disc, and now all is fine again. My longwinded point being: I had no trouble whatsoever converting the DTS-HD MA 7.1 track from the Nordic disc to LPCM, and experienced no volume issue or conversion hicup of any kind. This was done with an older version of eac3to at the time (eac3to v2.78), and, of course, the Arcsoft DTS Decoder.

[nautilus7]

What you're saying about flawless decoding of the DTS-HD track is simply not possible. If you're talking about the same disc, then eac3to (+arcsoft) should have produced the same warning (unless that eac3to version didn't warn for this).

What do you mean the TrueHD from the JAP disc is corrupted? Do you have a log or something. (if you don't have don't mind, i have the disc so i can test in a few days.)

[Paddy97]

eac3to is not limited to 8 titles. It lists 500 titles, if there are so many on the Blu-Ray disk. However, only titles with a runtime of 15 minutes or longer are listed. The reason for that is that there are quite a lot of Blu-Ray discs which have literally hundreds of titles. If I'd list all of them, the command prompt memory would run over so even if you scrolled up, you wouldn't see the movie's main title, anymore. So I *have* to set a runtime limit somewhere. Or else we'll be flooded with unneeded titles.

Thanks for the update on how the logic works. I can relate to this design choice.

As allways there is a way around everything even though you need some more tools. What I do now is that I use BDInfo to pinpoint the correct playlist ( using the length as a way of picking the right one ) and then I go from there.

[asarian]

What do you mean the TrueHD from the JAP disc is corrupted? Do you have a log or something. (if you don't have don't mind, i have the disc so i can test in a few days.)

Actually, I still have a log. :)

eac3to 1) 3: c:\video\sincity.pcm
M2TS, 1 video track, 4 audio tracks, 5 subtitle tracks, 2:04:01
1: Chapters, 29 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: TrueHD/AC3, English, 5.1 channels, 48khz, dialnorm: -27dB
(embedded: AC3, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB), -13ms
4: TrueHD/AC3, Japanese, 5.1 channels, 48khz, dialnorm: -27dB
(embedded: AC3, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB), -13ms
5: AC3, English, 2.0 channels, 256kbps, 48khz, dialnorm: -27dB, -13ms
6: AC3, English, 2.0 channels, 256kbps, 48khz, dialnorm: -27dB, -13ms
7: Subtitle (PGS), Japanese
8: Subtitle (PGS), Japanese
9: Subtitle (PGS), English
10: Subtitle (PGS), Japanese
11: Subtitle (PGS), Japanese
[a03] Extracting audio track number 3...
[a03] This track is not clean.
[a03] Extracting TrueHD stream...
[a03] Decoding with libav/ffmpeg...
[libav] Stream parameters not seen; skipping frame
[a03] The libav decoder output an unexpected bitdepth (1).
Aborted at file position 49152.

The Nordic disc decoded without any warning or noticeable lower volume, near as I can tell.

[madshi]

A (yet another) feature request: ability to use the SoundTouch library for audio speedup or slowdown without changing the audio pitch.
As far as I know, SoundTouch is not the preferred choice for best quality processing. Commercial competing algorithms are supposed to sound noticeably better, right? I don't really like the idea to implement something and say: eac3to can do it, but better don't use it, cause it's quality isn't really good.

Actually, I still have a log. :)

eac3to 1) 3: c:\video\sincity.pcm
M2TS, 1 video track, 4 audio tracks, 5 subtitle tracks, 2:04:01
1: Chapters, 29 chapters
2: h264/AVC, 1080p24 /1.001 (16:9)
3: TrueHD/AC3, English, 5.1 channels, 48khz, dialnorm: -27dB
(embedded: AC3, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB), -13ms
4: TrueHD/AC3, Japanese, 5.1 channels, 48khz, dialnorm: -27dB
(embedded: AC3, 5.1 channels, 640kbps, 48khz, dialnorm: -27dB), -13ms
5: AC3, English, 2.0 channels, 256kbps, 48khz, dialnorm: -27dB, -13ms
6: AC3, English, 2.0 channels, 256kbps, 48khz, dialnorm: -27dB, -13ms
7: Subtitle (PGS), Japanese
8: Subtitle (PGS), Japanese
9: Subtitle (PGS), English
10: Subtitle (PGS), Japanese
11: Subtitle (PGS), Japanese
[a03] Extracting audio track number 3...
[a03] This track is not clean.
[a03] Extracting TrueHD stream...
[a03] Decoding with libav/ffmpeg...
[libav] Stream parameters not seen; skipping frame
[a03] The libav decoder output an unexpected bitdepth (1).
Aborted at file position 49152.
This is probably with a rather old eac3to build. Try a new one.

[Boulder]

As far as I know, SoundTouch is not the preferred choice for best quality processing. Commercial competing algorithms are supposed to sound noticeably better, right? I don't really like the idea to implement something and say: eac3to can do it, but better don't use it, cause it's quality isn't really good.What program would you recommend for such operation? Preferably something that accepts 32bit float input:)

[ggking7]

I'm using eac3to-3.03 via wine and I'm trying to get info about a title before I rip the disc. 'eac3to /mnt/cdrom' displays the titles, but 'eac3to /mnt/cdrom "1)"' doesn't show info on the title. Instead I get:

"Please specify the source and dest files first and then the options."

If I switch it to 'eac3to "1)" /mnt/cdrom' I get:

"HD DVD / Blu-Ray disc structure not found."

Does anyone know how to get this to work?

[jj666]

The quotation marks around the 1 are not necessary i think.

-jj-

[J0nThn]

I'm using eac3to-3.03 via wine and I'm trying to get info about a title before I rip the disc. 'eac3to /mnt/cdrom' displays the titles, but 'eac3to /mnt/cdrom "1)"' doesn't show info on the title. Instead I get:

"Please specify the source and dest files first and then the options."

If I switch it to 'eac3to "1)" /mnt/cdrom' I get:

"HD DVD / Blu-Ray disc structure not found."

Does anyone know how to get this to work?

Try a path without a "slash in front". So for example if your current directory is /home/me

you have to use eac3to like this :

wine eac3to ../../mnt/cdrom "1)"

[ggking7]

The quotation marks around the 1 are not necessary i think.

I actually do need those quotes in order to use the ')' character in Linux.

Try a path without a "slash in front".

Weird, I get:

$ ls ../../../mnt/cdrom
AACS BDMV CERTIFICATE
$ wine eac3to ../../../mnt/cdrom "1)"
The format of the source file could not be detected.

'eac3to /mnt/cdrom' and all other eac3to commands I've tried work perfectly.

[Esurnir]

Could we have an option to use a .mpls file as a source so that we get the -right- playlist (on blu-rays such as wall-e spreaded on tons of files with multiple files with video both in french english and spanish, the biggest mpls isnt the one containing the english video.

[lexor]

Could we have an option to use a .mpls file as a source so that we get the -right- playlist (on blu-rays such as wall-e spreaded on tons of files with multiple files with video both in french english and spanish, the biggest mpls isnt the one containing the english video.

Did you try doing that? I'm pretty sure you can already do that.

[Esurnir]

Did you try doing that? I'm pretty sure you can already do that.

*facepalm* it indeed does... (maybe it should be put in the instructions as a possibility).

[hubblec4]

hello madshi

i have a question to demuxing. the sups of some blurays are not correct demuxed. (the same result when i demux with TSMuxer).

the sup-file are very large and contains too much pictures (in this case over 2800).
subtitle 1 till 5 shows the same picture but subtitle 1 till 4 have a starttime only. subtitle 5 is ok. and than subtitle 6 till 8 have the same fault and so on...

if you open the file (http://rapidshare.com/files/187355972/00003_-_6_-_Subtitle__PGS___English.rar.html) in supread0.52b2 you will see what i mean.

is it possible to fix that?


hubble

[ggking7]

Any other ideas with my issue above? Does it work for anyone else on Linux?

[kypec]

Any other ideas with my issue above? Does it work for anyone else on Linux?
:search:
Did you try aforementioned trick (http://forum.doom9.org/showthread.php?p=1237848#post1237848) with escaping the parenthesis character by backslash?
eac3to /mnt/cdrom 1\)

[ggking7]

Did you try aforementioned trick with escaping the parenthesis character by backslash?


I get:


$ wine eac3to /mnt/cdrom 1\)
Please specify the source and dest files first and then the options.


I'm not having trouble getting 1) recognized. For example, the following works perfectly:


wine eac3to video "1)" 2: video.* 3: audio.*


It's just this one command:


wine eac3to /mnt/cdrom "1)"


Does that work for anyone else on Linux?

[honai]

eac3to 3.04 crashes when I start processing a demuxed AVC stream.

http://www.sendspace.com/file/wwsa9t

[madshi]

What program would you recommend for such operation? Preferably something that accepts 32bit float input:)
Don't know, maybe TimeFactory?

i have a question to demuxing. the sups of some blurays are not correct demuxed. (the same result when i demux with TSMuxer).

the sup-file are very large and contains too much pictures (in this case over 2800).
subtitle 1 till 5 shows the same picture but subtitle 1 till 4 have a starttime only. subtitle 5 is ok. and than subtitle 6 till 8 have the same fault and so on...

if you open the file (http://rapidshare.com/files/187355972/00003_-_6_-_Subtitle__PGS___English.rar.html) in supread0.52b2 you will see what i mean.

is it possible to fix that?
I'll have a look at this. Can you try creating a 50MB sample and check whether the problem can be reproduced with that sample? If so, such a 50MB chunk would be helpful for me to figure out what's going on...

eac3to 3.04 crashes when I start processing a demuxed AVC stream.

http://www.sendspace.com/file/wwsa9t
That crash seems to be caused by a framerate format eac3to doesn't like. I'm confused why that crash happens - it shouldn't. Can I have the first 10MB of that AVC stream, please?

[honai]

Thanks for looking into this.

http://www.sendspace.com/file/yy1tf6

[Thunderbolt8]

is it a stream demuxed by eac3to or was it demuxed by another program? in case of the latter it might be possible that this other program doesn't treat the stream the right for eac3to to continue working with it.

[madshi]

eac3to 3.04 crashes when I start processing a demuxed AVC stream.
The very first sequence header doesn't contain framerate information in this AVC stream, while all other sequence headers do. That made eac3to crash. This problem will be fixed in the next eac3to build. BTW, the "video_full_range_flag" is set in this stream, which is probably incorrect. So you should wait for the next eac3to build, anyway, which will then automatically clear that flag (and crop the stream to 1920x1080 and add timing information to the first sequence header).

[honai]

Excellent, many thanks!

[ggking7]

Does dialog normalization and pulldown exist on DVD like it does on Blu-Ray? If so, can eac3to operate on a DVD or DVD image? If not, is there another program that can operate on a DVD or image and will remove dialog normalization and pulldown?

[lexor]

Does dialog normalization and pulldown exist on DVD like it does on Blu-Ray? If so, can eac3to operate on a DVD or DVD image? If not, is there another program that can operate on a DVD or image and will remove dialog normalization and pulldown?

DialNorm exists on DVD, since it's in the AC3 stream and it doesn't matter what medium that is on. eac3to should remove it, because it also doesn't really care what medium it is from. ac3 is ac3.

I don't know about pulldown.

[madshi]

Does dialog normalization and pulldown exist on DVD like it does on Blu-Ray? If so, can eac3to operate on a DVD or DVD image? If not, is there another program that can operate on a DVD or image and will remove dialog normalization and pulldown?
DVD video encoding is normally not stable. In other words: There may be some passages where pulldown could technically be removed, but most often there are other passages in the stream where it's not possible or where removing pulldown would even result in a worse stream. So eac3to does not even try to remove pulldown from any SD sources. If you see eac3to listing a DVD as "480p with pulldown flags" you can try using "-stripPulldown", but most probably processing will not run through successfully.

DVD file structure is not supported yet. You have to manually tell eac3to which VOBs to handle. And you shouldn't feed seamless branching or multi angle DVDs to eac3to cause those aren't handled correctly yet. Full DVD structure support might come in a later version (or not).

[hubblec4]

I'll have a look at this. Can you try creating a 50MB sample and check whether the problem can be reproduced with that sample? If so, such a 50MB chunk would be helpful for me to figure out what's going on...



hi madshi

i have no idea how can i split the m2ts-file. with tsmuxer i can split it but after demuxing you can't load the sups.

is there another way so i can do that?
the movie is "NARC"

hubble

[madshi]

i have no idea how can i split the m2ts-file. with tsmuxer i can split it but after demuxing you can't load the sups.

is there another way so i can do that?
You can use TsSplitter or a hexeditor.

[hubblec4]

You can use TsSplitter or a hexeditor.

ok thanks. it works. after spliting (http://rapidshare.com/files/188091100/00003.TSSplit.17-181.rar.html) and demuxing you can load the sup-file in supread and you see the mistake.


hubble

[J0nThn]

I get:

It's just this one command:


wine eac3to /mnt/cdrom "1)"


Does that work for anyone else on Linux?

Did you try this :

cd /mnt && wine eac3to cdrom "1)"

[ggking7]

cd /mnt && wine eac3to cdrom "1)"

Thanks for the idea, but I get:


$ ls cdrom
AACS BDMV CERTIFICATE
$ wine /home/me/eac3to cdrom "1)"
The format of the source file could not be detected.


I should have mentioned that 'wine eac3to videofolder "1)"' works perfectly after the disc is decrypted and ripped to the hard disk.

[Chumbo]

I wanted to report an issue I found. Looks like the number of frames is not reported when eac3to does a 2nd pass. I did not test this with 3.04 as I didn't see the problem until I no longer have the source. So if someone can verify if this problem still exists in 3.04 please. I'll see if I have anything else that may need a 2nd pass.

Here's a one pass:eac3to v3.03
command line: eac3to BDSource 3) 1: "H:\Media\chapters1-3.txt" 2: "H:\Media\video1-3.mkv" 3: "H:\Media\audio1-3.ac3" -log="H:\Media\process1-3.txt"
------------------------------------------------------------------------------
M2TS, 1 video track, 1 audio track, 2 subtitle tracks, 0:29:33
1: Chapters, 2 chapters
2: MPEG2, 480i60 /1.001 (4:3)
3: AC3, English, 2.0 channels, 192kbps, 48khz
4: Subtitle (PGS), Spanish
5: Subtitle (PGS), Portuguese
Creating file "H:\Media\chapters1-3.txt"...
[v02] Extracting video track number 2...
[a03] Extracting audio track number 3...
[v02] Muxing video to Matroska...
[a03] Creating file "H:\Media\audio1-3.ac3"...
Added fps value to MKV header.
Video track 2 contains 53136 frames.
eac3to processing took 1 minute, 3 seconds.
Done.
This is the two-pass that does not contain the frame counteac3to v3.03
command line: eac3to BDSource 2) 1: "H:\Media\chapters1-2.txt" 2: "H:\Media\video1-2.mkv" 3: "H:\Media\audio1-2.ac3" -log="H:\Media\process1-2.txt"
------------------------------------------------------------------------------
M2TS, 1 video track, 1 audio track, 2 subtitle tracks, 0:56:05
1: Chapters, 13 chapters
2: MPEG2, 480i60 /1.001 (4:3)
3: AC3, English, 2.0 channels, 192kbps, 48khz
4: Subtitle (PGS), Spanish
5: Subtitle (PGS), Portuguese
Creating file "H:\Media\chapters1-2.txt"...
[a03] Extracting audio track number 3...
[v02] Extracting video track number 2...
[v02] Muxing video to Matroska...
[a03] Creating file "H:\Media\audio1-2.ac3"...
[a03] Audio overlaps for 7ms at playtime 0:04:33. <WARNING>
[a03] Audio overlaps for 18ms at playtime 0:09:52. <WARNING>
[a03] Audio overlaps for 29ms at playtime 0:15:23. <WARNING>
[a03] Audio overlaps for 16ms at playtime 0:19:55. <WARNING>
[a03] Audio overlaps for 30ms at playtime 0:23:57. <WARNING>
[a03] Audio overlaps for 6ms at playtime 0:28:56. <WARNING>
[a03] Audio overlaps for 10ms at playtime 0:33:18. <WARNING>
[a03] Audio overlaps for 9ms at playtime 0:38:05. <WARNING>
[a03] Audio overlaps for 7ms at playtime 0:42:39. <WARNING>
[a03] Audio overlaps for 23ms at playtime 0:46:56. <WARNING>
[a03] Audio overlaps for 6ms at playtime 0:51:22. <WARNING>
[a03] Starting 2nd pass...
[a03] Realizing (E-)AC3 gaps...
[a03] Creating file "H:\Media\audio1-2.ac3"...
Added fps value to MKV header.
number of frames found would normally go here
eac3to processing took 2 minutes, 9 seconds.
Done.

[madshi]

Thanks for the idea, but I get:


$ ls cdrom
AACS BDMV CERTIFICATE
$ wine /home/me/eac3to cdrom "1)"
The format of the source file could not be detected.


I should have mentioned that 'wine eac3to videofolder "1)"' works perfectly after the disc is decrypted and ripped to the hard disk.
Ehm, you are aware that eac3to does not remove copy protection? eac3to can only work on HD DVD and Blu-Ray discs which have no copy protection on them.

[ggking7]

DialNorm exists on DVD, since it's in the AC3 stream and it doesn't matter what medium that is on. eac3to should remove it, because it also doesn't really care what medium it is from. ac3 is ac3.

Thank you, does DialNorm only exist on ac3?

[ggking7]

Ehm, you are aware that eac3to does not remove copy protection? eac3to can only work on HD DVD and Blu-Ray discs which have no copy protection on them.
Of course. Does that mean it can't provide title details from the (encrypted) disc? I thought it might since 'wine eac3to /mnt/cdrom' works straight off the disc.

[jamos]

hello great tool. I am using this with scenarist and the dolby lossless profile (True-HD). I am extracting the .ac3 core track extracted from the original true-hd stream and using that as input then choosing Dolby Lossless from the pull down in scenarist then click create files. Then I choose the .thd extracted track instead of the mlp file. seems to work perfectly but am I doing it correct?

[madshi]

I wanted to report an issue I found. Looks like the number of frames is not reported when eac3to does a 2nd pass.
Will check this out. Probably a bug.

Thank you, does DialNorm only exist on ac3?
AC3, TrueHD and DTS, but it's rarely used on DTS.

Of course. Does that mean it can't provide title details from the (encrypted) disc? I thought it might since 'wine eac3to /mnt/cdrom' works straight off the disc.
I'm surprised that even "wine eac3to /mnt/cdrom" works. eac3to was never intended to work with copy protected content at all and it never will do that.

I am using this with scenarist and the dolby lossless profile (True-HD). I am extracting the .ac3 core track extracted from the original true-hd stream and using that as input then choosing Dolby Lossless from the pull down in scenarist then click create files. Then I choose the .thd extracted track instead of the mlp file. seems to work perfectly but am I doing it correct?
What do I know? This is not the Scenarist support thread.

[Kurtnoise]

A sample (http://kurtnoise.free.fr/index.php?dir=samples/&file=04%20-%20Le%20dernier%20clope_6ch.wav) not detected properly with the latest build...:) Should be AC3-in-WAV

[madshi]

A sample (http://kurtnoise.free.fr/index.php?dir=samples/&file=04%20-%20Le%20dernier%20clope_6ch.wav) not detected properly with the latest build...:) Should be AC3-in-WAV
That is very strange. I have two AC3WAV samples now in SPDIF format. And the SPDIF header "framesize" field is big endian in one of the files and small endian in the other. I don't really know how it's supposed to be. At least one of the AC3WAV samples I have is clearly incorrect.

Anyway, I've implemented a workaround that makes both samples be handled correctly. Will be fixed in the next build.

[madshi]

i have a question to demuxing. the sups of some blurays are not correct demuxed. (the same result when i demux with TSMuxer).

the sup-file are very large and contains too much pictures (in this case over 2800).
subtitle 1 till 5 shows the same picture but subtitle 1 till 4 have a starttime only. subtitle 5 is ok. and than subtitle 6 till 8 have the same fault and so on...
That seems to be a bug in SUPread and not in eac3to or tsMuxeR. Try SupRip instead of SUPread. The next eac3to version will output how many subtitles are contained in the demuxed SUP file. It counts 7 subtitles for the sample you've uploaded. SupRip also shows 7 subtitles. However, SUPread shows 26 subtitles and many are duplicates. So SUPread is buggy...

[jamos]

What do I know? This is not the Scenarist support thread.

well I thought you may have insight as previously you stated this:

If you want to make a demuxed HD DVD TrueHD track (or a Blu-Ray TrueHD track demuxed by an older eac3to version) fit for Scenarist Blu-Ray muxing you can do this:

Code:
eac3to source.thd dest.thd+ac3This will automatically create an AC3 track and interweave it with the TrueHD track. Of course if your source is an HD DVD folder,
you can directly do e.g. "eac3to hddvdMovie 1) 1: video.mkv 3: audio.thd+ac3". That will demux the TrueHD track and interweave an AC3 track at the same time.


but anyways from scenarist forums it is the right way to do what I described. Extract the core ac3 using eac3to.exe from the original true hd stream and also the thd file from the same stream. .thd is the same as .mlp in scenarist. using dolby lossless in scenarist import the ac3 core then it will prompt you for the mlp file just use the .thd one instead. thanks for the great tool.

[mrr19121970]

@madshi, I just demuxed 1 vid & 2 DTS MA (with conversion to AC3):

eac3to v3.04
command line: "E:\TVIX\eac3to\eac3to.exe" "R:\" 1) 2: "D:\DEMUX\Tortured\Video_2.*" 3: "D:\DEMUX\Tortured\Audio_3_German.AC3" 4: "D:\DEMUX\Tortured\Audio_4_English.AC3" -LOG="D:\DEMUX\Tortured\eac3to_PASS3_LOG.LOG"
------------------------------------------------------------------------------
M2TS, 1 video track, 2 audio tracks, 1 subtitle track, 1:47:09, 24p /1.001
1: Chapters, 12 chapters
2: VC-1, 1080p24 /1.001 (16:9)
3: DTS Master Audio, German, 5.1 channels, 16 bits, 48khz
(core: DTS, 5.1 channels, 16 bits, 1509kbps, 48khz)
4: DTS Master Audio, English, 5.1 channels, 16 bits, 48khz
(core: DTS, 5.1 channels, 16 bits, 1509kbps, 48khz)
5: Subtitle (PGS), Undetermined
[a03] The ArcSoft and Sonic decoders don't seem to work, will use libav instead.
[a03] The libav DTS decoder doesn't decode the full DTS-HD information. <WARNING>
[a04] The ArcSoft and Sonic decoders don't seem to work, will use libav instead.
[a04] The libav DTS decoder doesn't decode the full DTS-HD information. <WARNING>
[v02] Extracting video track number 2...
[a03] Extracting audio track number 3...
[a04] Extracting audio track number 4...
[a03] Extracting DTS core...
[a04] Extracting DTS core...
[a03] Decoding with libav/ffmpeg...
[a04] Decoding with libav/ffmpeg...
[a03] Remapping channels...
[a04] Remapping channels...
[a03] Encoding AC3 <640kbps> with libAften...
[a04] Encoding AC3 <640kbps> with libAften...
[v02] Creating file "D:\DEMUX\Tortured\Video_2.vc1"...
[a03] Creating file "D:\DEMUX\Tortured\Audio_3_German.AC3"...
[a04] Creating file "D:\DEMUX\Tortured\Audio_4_English.AC3"...
[a03] Clipping detected, a 2nd pass will be necessary. <WARNING>
[a03] The original audio track has a constant bit depth of 64 bits.
[a04] The original audio track has a constant bit depth of 64 bits.
[a03] The libav DTS decoder doesn't decode the full DTS-HD information. <WARNING>
[a03] Starting 2nd pass...
[a03] Extracting audio track number 3...
[a03] Extracting DTS core...
[a03] Decoding with libav/ffmpeg...
[a03] Remapping channels...
[a03] Encoding AC3 <640kbps> with libAften...
[a03] Creating file "D:\DEMUX\Tortured\Audio_3_German.AC3"...
[a03] The original audio track has a constant bit depth of 64 bits.
[a03] The processed audio track has a constant bit depth of 64 bits.
eac3to processing took 34 minutes, 43 seconds.
Done.



The 1st AC3 track was just fine (as always), but the second had that distinct 'underwater' bubble, crackle very loudly in the background.

I redid singly, and it worked ok.


eac3to v3.04
command line: "E:\TVIX\eac3to\eac3to.exe" "R:\" 1) 4: "D:\DEMUX\Tortured 2nd Try\Audio_4_English.AC3" -LOG="D:\DEMUX\Tortured 2nd Try\eac3to_PASS3_LOG.LOG"
------------------------------------------------------------------------------
M2TS, 1 video track, 2 audio tracks, 1 subtitle track, 1:47:09, 24p /1.001
1: Chapters, 12 chapters
2: VC-1, 1080p24 /1.001 (16:9)
3: DTS Master Audio, German, 5.1 channels, 16 bits, 48khz
(core: DTS, 5.1 channels, 16 bits, 1509kbps, 48khz)
4: DTS Master Audio, English, 5.1 channels, 16 bits, 48khz
(core: DTS, 5.1 channels, 16 bits, 1509kbps, 48khz)
5: Subtitle (PGS), Undetermined
[a04] The ArcSoft and Sonic decoders don't seem to work, will use libav instead.
[a04] The libav DTS decoder doesn't decode the full DTS-HD information. <WARNING>
[a04] Extracting audio track number 4...
[a04] Extracting DTS core...
[a04] Decoding with libav/ffmpeg...
[a04] Remapping channels...
[a04] Encoding AC3 <640kbps> with libAften...
[a04] Creating file "D:\DEMUX\Tortured 2nd Try\Audio_4_English.AC3"...
[a04] The original audio track has a constant bit depth of 64 bits.
Video track 2 contains 154148 frames.
eac3to processing took 7 minutes, 14 seconds.
Done.


regards.

[hubblec4]

That seems to be a bug in SUPread and not in eac3to or tsMuxeR. Try SupRip instead of SUPread. The next eac3to version will output how many subtitles are contained in the demuxed SUP file. It counts 7 subtitles for the sample you've uploaded. SupRip also shows 7 subtitles. However, SUPread shows 26 subtitles and many are duplicates. So SUPread is buggy...

thank you madshi.

thats really a bug of SupRead. damn. only this programm can convert bluray.sup to hd-dvd.sup.

hubble

[mbcd]

thank you madshi.

thats really a bug of SupRead. damn. only this programm can convert bluray.sup to hd-dvd.sup.

hubble

OffTopic:

You know that there is an option to delete doubled Entries automaticly !

Thats an good working Workaround for that case ... not always perfect but an good option.

[Chumbo]

Will check this out. Probably a bug.
...
Looks like it is. The log (top) in this post (http://forum.doom9.org/showthread.php?p=1241440#post1241440) above done in 3.04 exhibits the same issue since it's a two-pass process too.

[madshi]

@madshi, I just demuxed 1 vid & 2 DTS MA (with conversion to AC3).

The 1st AC3 track was just fine (as always), but the second had that distinct 'underwater' bubble, crackle very loudly in the background.

I redid singly, and it worked ok.
Can I have a sample of the original m2ts file, please? Maybe 50MB or something like that? I'd like to find out why there is a problem when doing it all in one step. That is supposed to work just fine.