eac3to - audio conversion tool

[geminigod]

I don't think so.

Play with DRC enabled lose the original quality of audio.

Was a help for old amplifiers/receivers without the actual process capacity than can replace DRC with Night Mode for all audio formats, not only AC3.

BTW, if you play with PC you can use Ac3Filter with a dynamic DRC function, or you can recode the audio with BeLight with a Boost function.

I wish my receiver had a "night mode". Even some of the new receivers out there are retarded about the signal processing options. My father just bought a new Denon receiver that sucks. There are a bazillion useless ways to screw with the EQ, all of which make the sound worse, and no way to process dynamic range.

At any rate, no worries. I just have an ac3 audio track that I have since deleted the source material for. I encoded it with a Film Light DRC setting and decided that Film Standard would be preferrable. It isn't worth transcoding to change it though.

[An IP BreAKDoWN]

Are mono audio files still not supported? I'm trying to rip a mono audio file that is in thd to a dts. Does anyone know how to do this if eac3to can't?

[kalston]

I converted a mono DTS file to FLAC without any problem.

[An IP BreAKDoWN]

How about a mono THD to DTS (FLAC, or WAV for that matter)?

[geminigod]

Are mono audio files still not supported? I'm trying to rip a mono audio file that is in thd to a dts. Does anyone know how to do this if eac3to can't?

I can't remember the specifics of whether surecode supports mono dts encoding or not, assuming you even have this $250 codec setup in your eac3to. Try converting to wav first and then to dts. See if it works then.

More likely your eac3to is not setup to properly decode or encode dts. Type eac3to -test at the command prompt and see what it says.

I could ask why you are trying to make a mono dts file... :scared: Stereo acoustics will sound much better, even if you are just duplicating the mono track.

If you are determined to work with this challenging format (due to commercial copyright issues), you can play around with this brand new experimental open source dts codec that folks here have been working on.

http://forum.doom9.org/showthread.php?p=1552620#post1552620

[kws53]

I can't get eac3to to downconvert a 3 channel [L/R/C] DTS audio stream. Using -down2 switch does not solve it. The output is still 3 channels.

Kurt

[An IP BreAKDoWN]

I can't remember the specifics of whether surecode supports mono dts encoding or not, assuming you even have this $250 codec setup in your eac3to. Try converting to wav first and then to dts. See if it works then.

More likely your eac3to is not setup to properly decode or encode dts. Type eac3to -test at the command prompt and see what it says.

I could ask why you are trying to make a mono dts file... :scared: Stereo acoustics will sound much better, even if you are just duplicating the mono track.

If you are determined to work with this challenging format (due to commercial copyright issues), you can play around with this brand new experimental open source dts codec that folks here have been working on.

http://forum.doom9.org/showthread.php?p=1552620#post1552620

Yes I have Surecode, it set me back but I have it installed. The reason I'm trying to convert to mono is because the source is mono in THD format.

Here is that command you told me to use:

eac3to v3.24
command line: eac3to -test
------------------------------------------------------------------------------
eac3to (v3.24) is up to date
Nero Audio Decoder (Nero 6 or older) doesn't seem to be installed
http://www.nero.com/eng/store-blu-ray.html
CAUTION: You need Nero 7. Nero 8 won't work with eac3to.
ArcSoft DTS Decoder (1.1.0.1) works fine
Sonic Audio Decoder (3.24.0.0) doesn't seem to be installed
Haali Matroska Muxer (2011-09-08) is up to date
Nero AAC Encoder could not be located
http://www.nero.com/eng/nero-aac-codec.html
Copy NeroAacEnc.exe to the eac3to or to the Windows folder.
Surcode DTS Encoder (1.0.29.0) is installed
MkvToolnix (2.5.3.0, release version) is installed
There's a new release version (5.2.1.0) available
http://www.bunkus.org/videotools/mkvtoolnix
There's a new beta version (5.2.1.0, 2012-01-13) available
http://www.bunkus.org/videotools/mkvtoolnix/win32/pre


And here is the error I am having when I try to rip it to .dts

eac3to v3.24
command line: eac3to 00001.m2ts 4: japanese-1.0.dts
------------------------------------------------------------------------------
M2TS, 1 video track, 3 audio tracks, 2 subtitle tracks, 0:24:13, 24p /1.001
1: h264/AVC, 1080p24 /1.001 (16:9)
2: TrueHD/AC3, 5.1 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 5.1 channels, 448kbps, 48kHz, dialnorm: -25dB)
3: TrueHD/AC3, 2.0 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 2.0 channels, 192kbps, 48kHz, dialnorm: -25dB)
4: TrueHD/AC3, 1.0 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 1.0 channels, 96kbps, 48kHz, dialnorm: -25dB)
5: Subtitle (PGS)
6: Subtitle (PGS)
[a04] Extracting audio track number 4...
[a04] Extracting TrueHD stream...
[a04] Removing TrueHD dialog normalization...
[a04] Decoding with libav/ffmpeg...
[a04] [libav] Substream min channel cannot be greater than max channel. <WARNING>
[a04] The libav decoder reported error -1 while decoding. <ERROR>
Aborted at file position 1048576. <ERROR>


Here is the error message when I rip it to .wav

eac3to v3.24
command line: eac3to 00001.m2ts 4: japanese-1.0.wav
------------------------------------------------------------------------------
M2TS, 1 video track, 3 audio tracks, 2 subtitle tracks, 0:24:13, 24p /1.001
1: h264/AVC, 1080p24 /1.001 (16:9)
2: TrueHD/AC3, 5.1 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 5.1 channels, 448kbps, 48kHz, dialnorm: -25dB)
3: TrueHD/AC3, 2.0 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 2.0 channels, 192kbps, 48kHz, dialnorm: -25dB)
4: TrueHD/AC3, 1.0 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 1.0 channels, 96kbps, 48kHz, dialnorm: -25dB)
5: Subtitle (PGS)
6: Subtitle (PGS)
[a04] Extracting audio track number 4...
[a04] Extracting TrueHD stream...
[a04] Removing TrueHD dialog normalization...
[a04] Decoding with libav/ffmpeg...
[a04] [libav] Substream min channel cannot be greater than max channel. <WARNING>
[a04] The libav decoder reported error -1 while decoding. <ERROR>
Aborted at file position 1048576. <ERROR>


And here is the error message when I rip to .wavs

eac3to v3.24
command line: eac3to 00001.m2ts 4: japanese-1.0.wavs
------------------------------------------------------------------------------
M2TS, 1 video track, 3 audio tracks, 2 subtitle tracks, 0:24:13, 24p /1.001
1: h264/AVC, 1080p24 /1.001 (16:9)
2: TrueHD/AC3, 5.1 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 5.1 channels, 448kbps, 48kHz, dialnorm: -25dB)
3: TrueHD/AC3, 2.0 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 2.0 channels, 192kbps, 48kHz, dialnorm: -25dB)
4: TrueHD/AC3, 1.0 channels, 48kHz, dialnorm: -25dB
(embedded: AC3, 1.0 channels, 96kbps, 48kHz, dialnorm: -25dB)
5: Subtitle (PGS)
6: Subtitle (PGS)
[a04] Extracting audio track number 4...
[a04] Extracting TrueHD stream...
[a04] Removing TrueHD dialog normalization...
[a04] Decoding with libav/ffmpeg...
[a04] [libav] Substream min channel cannot be greater than max channel. <WARNING>
[a04] The libav decoder reported error -1 while decoding. <ERROR>
Aborted at file position 1048576. <ERROR>


I would like to try the experimental version, however I can't get the wav so I'm stuck till I can get the wav. :(

[b66pak]

first you need to demux the thd from the m2ts:

eac3to 00001.m2ts 4:japanese.thd

get the latest ffmpeg (http://ffmpeg.zeranoe.com/builds/)...and use this:

ffmpeg -i japanese.thd japanese.wav
_

[An IP BreAKDoWN]

@b66pak Thanks that worked. Except the eac3to complained about it being mono on the conversion from wav to dts, however I used the SureCode GUI and that worked! Thanks again!

[TDiTP_]

ffmpeg -i japanese.thd japanese.wav
As far as i remember, with this command line you'll be obtain 16-bit PCM always, even if TrueHD is 24-bit.
You need:
ffmpeg -i input.thd -acodec pcm_s24le output.wav
and then, if you need it, you can check the true bitdepth in audioeditor or by -analyzebitdepth in eac3to:
eac3to input.wav -analyzebitdepth

[An IP BreAKDoWN]

As far as i remember, with this command line you'll be obtain 16-bit PCM always, even if TrueHD is 24-bit.
You need:
ffmpeg -i input.thd -acodec pcm_s24le output.wav
and then, if you need it, you can check the true bitdepth in audioeditor or by -analyzebitdepth in eac3to:
eac3to input.wav -analyzebitdepth
Thats very useful thanks!

One question though, how can I find the bit depth of the THD since the libav in eac3to can't open it since its a mono?

[TDiTP_]

how can I find the bit depth of the THD since the libav in eac3to can't open it since its a mono?
If eac3to isn't able to decode lossless-stream, then it can't check his bitdepth, so in this case -analyzebitdepth doesn't work.

[An IP BreAKDoWN]

If eac3to isn't able to decode lossless-stream, then it can't check his bitdepth, so in this case -analyzebitdepth doesn't work.
Right, do you know a program that can analyze the bit depth in THD format?

[Rodeo]

Thats very useful thanks!

One question though, how can I find the bit depth of the THD since the libav in eac3to can't open it since its a mono?

Please try to follow the provided instructions. First, b66pak's:

eac3to 00001.m2ts 4:japanese.thd

japanese.thd becomes input.thd in TDiTP_'s instructions:

ffmpeg -i input.thd -acodec pcm_s24le output.wav

So here, the TrueHD file is now 24-bit PCM in WAV…

eac3to input.wav -analyzebitdepth

…thus the old, buggy libavcodec TrueHD decoder is no longer involved.

[TDiTP_]

Right, do you know a program that can analyze the bit depth in THD format?
i know only eac3to with his -analyzebitdepth. This command can be applied to any lossless-stream or PCM (in WAV/W64).
BTW, BDInfo can check bitdepth of THD incorrectly: look known issues (http://www.cinemasquid.com/blu-ray/tools/bdinfo).
Of course, if libav of eac3to can't decode THD, then -analyzebitdepth does not make sense (because for determine bitdepth you need to decode). In this case you should try fresh ffmpeg.

[Floatingshed]

I've been using eac3to to join various audio files together. I have been doing it by hand as a commandline and it works well.
However I have a directory containing 37 mp3 files (all same params) that I want to join. A batch file would be nice, perhaps using IN DO, but my brain won't work, please somebody take pity...
Thanks.

[Sparktank]

I've been using eac3to to join various audio files together. I have been doing it by hand as a commandline and it works well.
However I have a directory containing 37 mp3 files (all same params) that I want to join. A batch file would be nice, perhaps using IN DO, but my brain won't work, please somebody take pity...
Thanks.

>.< multiple file batching can be the morning-coffee-buzzkill.

I really haven't done anything like joining files in eac3to, but I do do some batch files.

For doing multiple files, I just open a text file with the "eac3to" commands, copy and paste them several times.
Going back and modifying only the input/output file names.
And then exit at the very bottom.

Note: I also have windows configured to open CMD in a folder, using the context menu: "Open Comand Promt here"

"path\eac3to.exe" "sourcefile[+sourcefile2]" [trackno:] "[destfile|stdout]" [-options]
"path\eac3to.exe" "sourcefile[+sourcefile2]" [trackno:] "[destfile|stdout]" [-options]
"path\eac3to.exe" "sourcefile[+sourcefile2]" [trackno:] "[destfile|stdout]" [-options]
exit



Then I just copy/paste that into the CMD of the destination folder.
I use "" if the path of my eac3to is in a folder with 2 words including a space "App Bin", and also some of the source files are in folders that contain spaces as well.

I also add in custom logs.
-log=custom_name_codec__etc_with_no_spaces_to_reduce_amount_of_quotes.txt

As for creating a batch script, I'm not that far in batching.
Yet :D

[Floatingshed]

OK. I've done it. I'm not a whizz at programming so it may not be very elegant but it works. If you want to use it just copy and paste into a batch file (set path to eac3to). Drop into a folder of mp3 files, double click and a single file of them all nicely joined will appear!


FOR %%a IN (*.mp3) DO echo/|set /p =" "%%~NXa"+" + >> temp.txt
set /p inlist= <temp.txt
set newlist=%inlist:~0,-1%
path.to\eac3to.exe" %newlist% "Joined mp3 files.mp3"
del temp.txt

[nibus]

OK. I've done it. I'm not a whizz at programming so it may not be very elegant but it works. If you want to use it just copy and paste into a batch file (set path to eac3to). Drop into a folder of mp3 files, double click and a single file of them all nicely joined will appear!


FOR %%a IN (*.mp3) DO echo/|set /p =" "%%~NXa"+" + >> temp.txt
set /p inlist= <temp.txt
set newlist=%inlist:~0,-1%
path.to\eac3to.exe" %newlist% "Joined mp3 files.mp3"
del temp.txt

:goodpost: excellent!

[LeKouz]

Hi.

Does anybody know how to set SoX to handle the 64 floating point output from eac3to?
Or in an other question, what kind of 64bit-Stream is eac3to giving out?
I've tried a lot of commands but all I've got is noise and it sounds like playing a CD-ROM in an audio-player.

This is the last command-line i've tried to pipe from eac3to to SoX:
eac3to "test.ac3" stdout.wav -full -simple | sox -V3 --type .f64 --rate 48000 --encoding floating-point --bits 64 --channels 6 --ignore-length - -t wav --encoding floating-point --bits 32 result.wav



"--type raw --rate 48000 --encoding floating-point --bits 64 --channels 6" -> Doesn't works too. Lot of noise.

If i set "--type" to wav, SoX says:
sox WARN wav: wave header missing FmtExt chunk
sox FAIL formats: can't open input `-': Sorry, don't understand .wav size

And if I command eac3to to write the 64bit output directly to a WAV-File, Mediainfo means:
General
Complete name : E:\Tools\eac3to\test.wav
Format : Wave
File size : 668 MiB
Duration : 5mn 4s
Overall bit rate mode : Constant
Overall bit rate : 18.4 Mbps

Audio
ID : 0
Format : PCM
Format profile : Float
Format settings, Endianness : Float
Codec ID : 00001000-0000-0300-8000-00AA00389B71
Codec ID/Hint : IEEE
Duration : 5mn 4s
Bit rate mode : Constant
Bit rate : 18.4 Mbps
Channel(s) : 6 channels
Channel positions : Front: L C R, Side: L R, LFE
Sampling rate : 48.0 KHz
Bit depth : 64 bits
Stream size : 668 MiB (100%)

[nu774]

sox WARN wav: wave header missing FmtExt chunk

This is due to SoX wave parser bug. It currently always shows this strange warning to IEEE float wave file with WAVE_FORMAT_EXTENSIBLE header.


sox FAIL formats: can't open input `-': Sorry, don't understand .wav size

This is another thing. Currently SoX cannot read 64bit float wave files. Though one line of modification to the wave parser source code can enable it.
However, since SoX internally treats audio samples in the form of 32bit float, feeding 64bit float PCM to SoX is pointless anyway.


"--type raw --rate 48000 --encoding floating-point --bits 64 --channels 6" -> Doesn't works too. Lot of noise.

You cannot treat wave file as raw PCM. In this case, length of wave header is not divisible by the length of double PCM frame (= 8 * number of channels).
Therefore, access to each PCM samples are not aligned, and produces completely wrong result.
Even if it is aligned, since wave header is treated as audio, you will get some garbage at the beginning.

[LeKouz]

THX nu774 for the infos :)

[geminigod]

Nu774, you can go here. https://sourceforge.net/projects/sox/files/

The latest release candidate that just came out addresses some of your concerns. Not sure about handling 64 bit float, but that is a giant waste anyway. You can check the tracker bug status at that website as well.

[nu774]

AFAIK, this issue (http://sourceforge.net/tracker/?func=detail&aid=3481510&group_id=10706&atid=110706) is not yet addressed. Am I missing something?

[Floatingshed]

I have a problem getting BluRay info from eac3to. When I run:

eac3to.exe E:\ >text.txt

to generate a text file of disc contents I get a lot of wierd characters (as well as the required data). e.g.

 - DTS Master Audio, English, multi-channel, 48kHz

Different text editors display this very differently, so clearly none of them understand it!

Does anyone here? Thanks.

[kypec]

Different text editors display this very differently, so clearly none of them understand it!

Does anyone here? Thanks.
That's because eac3to outputs special (non-printable) characters into console (to change fore/background colors for example).
Why not use other tool for getting information on Blu-ray disc, BDInfo (http://www.cinemasquid.com/blu-ray/tools/bdinfo#about) comes to mind...

[NanoBot]

Hi,

I just did a quick test trying to reproduce your problem, but without one special case I can't reproduce it. The textfile produced by

eac3to m: > D:\tmp\test.txt

is nearly perfect readable when I load it into either the standard XP text editor or into Ultraedit.

Nearly means, that the german umlauts and e.g. the french circumflex accent or grave accent are not displayed correctly. But the reason for this behavior is simple: Since I am in germany, my command line uses codepage 850 by default ( should be cp 437 in english language countrys ? ), while Windows uses codepage 1252 by default. So the only thing I have to do to solve the problem is to execute "chcp 1252" on my cli, and all characters are displayed correctly within the Windows editor.

Perhabs your problem has a similar cause ?

[skeeve1000]

Hi,
since i updated to the latest version of popcorn 1.85.1, of mkvtoolnix 5.3 as well as eac3to i always get an eac3to crash when eac3to starts to work within popcorn.
You can see the errormessage here:
http://www.networkedmediatank.com/showthread.php?tid=20887&page=99
(i asked the popcorn developement first).

May this could be a problem with MS.NET? I´m not sure but it can be possible, that i killed a version of .NET on my computer.

Help would be great.
Thanks in advance.

Skeeve

[tebasuna51]

Hi, since i updated to the latest version of popcorn 1.85.1,...

Like eac3to version don't change seems a bug in AudioConverter.exe 1.85.1 (maybe capturing the eac3to progress) then we can't help you here.

Try execute the command line out of AudioConverter (in a cmd window) and see if work.

I don't know eac3to dependencies with .NET

[skeeve1000]

Hi again,

eac3to crashes directly while starting in comand line (e.g. eac3to d:\tmß\xyz.mkv).
It asks me to send a bug report, but i am not sure, if i should do this?
Any further advice?

Skeeve

[Floatingshed]

That's because eac3to outputs special (non-printable) characters into console (to change fore/background colors for example).
Why not use other tool for getting information on Blu-ray disc, BDInfo (http://www.cinemasquid.com/blu-ray/tools/bdinfo#about) comes to mind...


I am using eac3to to extract files, therefore I wanted to use the info in the text file as a means of identifying and selecting the required tracks. Does anyone know how to get a clean feed of the track info?

Found a workaround, the info is nice and clean in the eac3to log file...

[tebasuna51]

eac3to crashes directly while starting in comand line (e.g. eac3to d:\tmß\xyz.mkv)

Sorry but I can't reproduce your problem, work fine here. :)

[Midzuki]

Hi again,

eac3to crashes directly while starting in comand line (e.g. eac3to d:\tmß\xyz.mkv).
It asks me to send a bug report, but i am not sure, if i should do this?
Any further advice?

Skeeve

1) :logfile:

2) Wild guess:
d:\tmß\xyz.mkv

Try renaming that folder to an "ASCII-friendly" :) string.

[tebasuna51]

...
Try renaming that folder to an "ASCII-friendly" :) string.

Yeah, I think skeeve1000 (from germany) know the problem.

In the command line I say to execute (see http://www.networkedmediatank.com/showthread.php?tid=20887&page=99 ) don't exist non-ASCII chars, for that my :)

[Midzuki]

Hummm you're right, certainly I need more buckets of coffee per day.

[ Midzuki :uglylol:s himself ]

[VincAlastor]

hi,
is it possible to use flaccl.dll instead libflac.dll directly in eac3to? i've read flaccl -8 provides better compression than flac -8 - besides it is faster, too.
http://www.cuetools.net/wiki/FLACCL
i know, i could use it with the command line, but channel mapping for each different codec is confusing for me. eac3to gives the right channel map to libflac even the source has a strange 7.1 surround setup, isn't it?
i could not found any limitations for flaccl, so i think it supports 24bit and 8 channels.

[mindbomb]

is it ever a good idea to update the avcodec.dll that comes with eac3to?

my friend ran into a decoding problem, i thought about suggesting this to him, not really sure if it uses a special one or something.

[kypec]

is it ever a good idea to update the avcodec.dll that comes with eac3to?

my friend ran into a decoding problem, i thought about suggesting this to him, not really sure if it uses a special one or something.
I'm afraid not. I'd tried that last year and replaced both avcodec.dll and avutil-50.dll with some April-ish 2011 versions -> this prevented eac3to to decode/process even some of the previously working formats :( Unless madshi himself recompiles against more recent ffmpeg/libav libraries (or releases the source code in the wild) then we're out of luck. :mad:

[Jeff B]

If I run a wav or PCM file through eac3to and output as .lpcm, will the result be both Blu-ray and DVD compliant?

[06_taro]

VincAlastor
stable released FlacCL#0.3 does not support 24bit, only FlacCL#0.4 in CUETools 2.1.2a does. But it's a bit experimental, and there's at least one known (if rare) bug in it (http://www.hydrogenaudio.org/forums/index.php?showtopic=64628&view=findpost&p=763244).


madshi
eac3to internally encodes flac using max compression level as you mentioned here (http://forum.doom9.org/showthread.php?p=1222771#post1222771). Is the encoded flac file within the flac subset (http://flac.sourceforge.net/format.html#subset) like libflac's -8(without --lax), or non-subset like flacCL's -9~11/flake's -11~12?

edit: re-encoding eac3to generated flac file with flac.exe -8 gives only difference in header, so it seems that eac3to is using exactly the same setting as libflac's -8....

[shh]

Where's madshi? Is eac3to dead?

[tebasuna51]

madshi is still active here http://forum.doom9.org/showthread.php?t=146228

I can't answer to second question.

[szabi]

Hi

I tried this:
Extract the DTS core of a DTS-HD track:
eac3to input.dtshd output.dts -core

Why is it not working?
IAW wikibooks it should: click (http://en.wikibooks.org/wiki/Eac3to/How_to_Use)

bye
szabi

[madshi]

eac3to is not dead, it's just kinda "on hold" for the moment, because I'm busy working on madVR. Once madVR has reached v1.0, I'll get back to eac3to development.

[szabi]

I tried using gui's, but no luck as well.
Eac3to and More GUI (http://forum.doom9.org/showthread.php?t=135095)
This one got crash all the time.
Another mod of this is Eac3to and More GUI v.1.10.01.
Even the input file is given, it is asking to give again and again, it can not be started.

eac3to µGUI (http://forum.doom9.org/showthread.php?t=156762)
Third one i tried is this.
But the needed button is inactive in the gui.

Is that possible it is just a planned feature which does not work momentarily?

bye
szabi

[the_weirdo]

@szabi: You may try this GUI: UsEac3to (http://forum.doom9.org/showthread.php?t=145574).

[szabi]

It worked, many thanks. :thanks:
I would reccommend to take this tool to initial post.
Out of 4 gui this was the only one which could work.

bye
szabi

[jpsdr]

eac3to is not dead, it's just kinda "on hold" for the moment, because I'm busy working on madVR. Once madVR has reached v1.0, I'll get back to eac3to development.

Thanks for the information !

[setarip_old]

@szabi

If you can find it, you might also try the MUCH older "EAC3toGUI.exe"...

[Jeff B]

If I run a wav or PCM file through eac3to and output as .lpcm, will the result be both Blu-ray and DVD compliant?

Any ideas about this?

[nevcairiel]

Any ideas about this?

Blu-ray and DVD use different types of LPCM encoding, so with some confidence i can say that it won't be compatible with both at the same time.

[Jeff B]

Blu-ray and DVD use different types of LPCM encoding, so with some confidence i can say that it won't be compatible with both at the same time.

Thanks for the info. In that case, I need to know what commands to input for Blu-ray and for DVD.

[tebasuna51]

Thanks for the info. In that case, I need to know what commands to input for Blu-ray and for DVD.
That depend of the soft you want use to mux video and audio.
For instance tsMuxeR accept .wav (<4GB) and .w64 (>4GB) to obtain .m2ts files for BD

[Jeff B]

That depend of the soft you want use to mux video and audio.

Really? I understood that muxers don't actually change files: they just, well, mux them. I don't see how muxing .w64 can be BD compliant, unless tsmuxer changes the file in some way.

[mindbomb]

so, im not completely clear on the use of flac for things above 6 channels.
WAVEFORMATEXTENSIBLE_CHANNEL_MASK is supposed to be used? and eac3to does this by default?
And decoders should be able to use that for to figure out the channels?

If my ultimate goal is to use this in a matroska file, does the channel mask remain through the muxing process?

[tebasuna51]

Really? I understood that muxers don't actually change files: they just, well, mux them. I don't see how muxing .w64 can be BD compliant, unless tsmuxer changes the file in some way.

Of course, a muxer can modify the data order (unique diference between lpcm and wav/w64 files)

so, im not completely clear on the use of flac for things above 6 channels.
WAVEFORMATEXTENSIBLE_CHANNEL_MASK is supposed to be used? and eac3to does this by default?
And decoders should be able to use that for to figure out the channels?

If my ultimate goal is to use this in a matroska file, does the channel mask remain through the muxing process?

AFAIK, flac can't store channel mask.
There are a unique channel mask supported for each num_channels type, and undefined for 7 and 8 channels.

[me7]

eac3to doesn't seem to like 5.0 content. I have a 5.0 DTS Master Audio tracks that I want to encode to 5.0 AC3, but eac3to crashes with an error:

DTS Master Audio, 5.0 channels, 24 bits, 48kHz
(core: DTS, 5.0 channels, 24 bits, 1509kbps, 48kHz)
Decoding with ArcSoft DTS Decoder...
The AC3 encoder received a non-supported data format (pcm, 5, 24, -).

Apparently the included AC3 encoder doesn't like 5 channel content.
If I use eac3to to create a 5.0 wav file and encode it with a standalone AC3 encoder, do I need to worry about channel order?

[rapscallion]

Audio question about my logic :

If I extract a 385 Mb AC-3, 5.1, 640 Kbps track to 6 wavs, the results are 975 Mb wavs @1152 Kbps each.

Enter the wavs into DTS-MA suite, encode them into DTS-High Res audio track format with a bitrate @ 3840 Kbps.
The result is still lossy ( I know they can't be converted to DTS-HD) but much higher bitrate and, I would think, higher quality than DTS @ 1509 Kbps and the original AC-3 track.

Is my reasoning correct or am I out in left field somewhere ?

[Asmodian]

... higher quality than DTS @ 1509 Kbps and the original AC-3 track.

You can never get higher quality than the original, a lossy reencode will always be lower quality than the source.

How much lower is dependent on the settings and in your case I would assume not much. Still it is better to just leave it as the AC3 if you can.

[Jeff B]

Of course, a muxer can modify the data order (unique diference between lpcm and wav/w64 files)

Ah, so the only difference between those types of files is data order, and as long as the muxer accepts the input the result should be compliant. Thanks. That helps.

[rapscallion]

You can never get higher quality than the original, a lossy reencode will always be lower quality than the source.

How much lower is dependent on the settings and in your case I would assume not much. Still it is better to just leave it as the AC3 if you can.

However, the AC-3 isn't the original, the wavs are. Thus, a re encode of the extracted wavs should result in better quality at the higher bitrate, no?

Edit: This stuff really does make my head hurt.
BTW, I did the encode and the resulting DTS-HR file is ~ 3.2 gb

[tebasuna51]

...
If I use eac3to to create a 5.0 wav file and encode it with a standalone AC3 encoder, do I need to worry about channel order?

No problem.
You can use also eac3to with 'pipe' if channels aren't 2.0 or 5.1:
eac3to input stdout.wav | Aften -b 640 -readtoeof 1 output.ac3

However, the AC-3 isn't the original, the wavs are.

The decoded ac3 wav's aren't the originals, have the same quality than the ac3.

[Asmodian]

If I extract a 385 Mb AC-3, 5.1, 640 Kbps track to 6 wavs

However, the AC-3 isn't the original, the wavs are.


If you started with it as your source of the audio the AC3 is your "original". You cannot recover the real original audio; some information was lost when it was compressed to AC3.

As tebasuna51 said the wav files contain the same information as the AC3, just no longer in a compressed form.

[rapscallion]

If you started with it as your source of the audio the AC3 is your "original". You cannot recover the real original audio; some information was lost when it was compressed to AC3.

As tebasuna51 said the wav files contain the same information as the AC3, just no longer in a compressed form.
Yes, what I should have said is that the wav files are the originals minus the "unnecessary" data that the compression expunges.

So, am I to understand that a 5.1 AC-3 640 Kbps audio is the same quality as it's 1150 Kbps extracted wavs? If that's the case, then shouldn't a 440 Kbps be the same quality as a 640 Kbps if compressed form the same source ?
I just can't get my head around this, so sorry for dwelling on it.

[Asmodian]

It is lossy compression, an AC3 is not like a zip file where after decompression you get exactly the same data back. Lossy compression drops information it thinks you will not miss (much); the 440 Kbps file dropped more information than the 640 Kbps file. Both dropped some information compared to the original.

It isn't really "unnescessary" data it is just less noticable.

I should also mention that re-encoding your extracted wavs to a 640 Kbps AC3 would lose information again; your new AC3 would be lower quality compaired to the original AC3.

[rapscallion]

Ok, thanks Asmodian. No how much I googled this, I couldn't find a concise answer.
Especially re the greater the compression, the more information dropped. So the answer to my previous question "5.1 AC-3 640 Kbps audio is the same quality as it's 1150 Kbps extracted wavs? " is YES. Very interesting.

[Asmodian]

So the answer to my previous question "5.1 AC-3 640 Kbps audio is the same quality as it's 1150 Kbps extracted wavs? " is YES. Very interesting.

You can trust tebasuna51. ;)
We both answered this in the previous two posts.

When you play an AC3 (or any audio file) it is converted to raw digital audio data and sent to your sound card for digital to analogue conversion. "Extracting" to a wav file is like playing the AC3 but saving the digital data to the hard drive instead of sending it to the sound card.

[rapscallion]

I know you did, I was just confirming that I now get it.

[Asmodian]

Sorry, I thought you were annoyed I didn't answer your obvious question. I hate ignored questions but left that one out as it was already addressed.

Glad it makes sense now. :)

[odin24]

Are there any issues converting a 7.1 DTS-HD MA to DD5.1, in regards to the back channels mixing into the surround channels? I use Arcsoft to decode the audio, output to .w64, encode to DD5.1 (libav).

[Sparktank]

Are there any issues converting a 7.1 DTS-HD MA to DD5.1, in regards to the back channels mixing into the surround channels? I use Arcsoft to decode the audio, output to .w64, encode to DD5.1 (libav).

According to tebasuna51, there shouldn't be any problems downmixing 7.1 DTS-HD MA to 5.1 (http://forum.doom9.org/showthread.php?p=1529424#post1529424).
As long as you're using Arcsoft 1.1.0.0 (25/04/2008), it can downmix 7.1 with either proper channel setup or strange setup.
Post 1 (http://forum.doom9.org/showthread.php?p=1519949#post1519949); Post 2 (http://forum.doom9.org/showthread.php?p=1519992#post1519992); Arcsoft version info (http://forum.doom9.org/showthread.php?p=1266679#post1266679)

Not quite sure if I missed anything... :scared:
There is this post (http://forum.doom9.org/showthread.php?p=1520511#post1520511) of some interest.

[taiyoyuden]

For best DTS decoding you need:
(1) ArcSoft DTS Decoder - version 1.1.0.0 or newer

Is the ArcSoft DTS Decoder 1.1.0.0+ the recommended/default DTS-HD decoder now? I'm confused because here (http://en.wikibooks.org/wiki/Eac3to/In_Depth_Technical_Explanation#Evaluation_of_available_decoders), the default says Sonic and no mention of the Arcsoft. Thus, I've been using the Sonic 4.3.0.169 decoder for numerous DTS-HD 5.1 tracks.

Is there any difference in sound quality, channel mapping, etc.? Should I have them redone with the ArcSoft? Which version of the ArcSoft DTS Decoder is the best?

[tebasuna51]

Please read the first post in this thread:

"The Sonic DTS decoder is very good for DTS, DTS-ES, DTS-96/24, DTS-HD Master Audio and DTS-HD High Resolution tracks. The only problem is that it decodes DTS-HD 7.1 tracks only as 5.1.
...
The ArcSoft DTS decoder seems to be perfect for DTS and DTS-HD decoding. It supports every format and channel configuration that exists including 6.1 and 7.1."

I recommend ArcSoft DTS Decoder 1.1.0.0

[Jeff B]

Tebasuna, taiyoyuden also asked if there is any difference in sound quality between ArcSoft and Sonic when DTS-HD 5.1 tracks are being decoded. I am interested in knowing the answer to this too. The only limitation for Sonic mentioned on the first page is that 7.1 tracks are decoded as 5.1.

[tebasuna51]

Decoding lossless (DTS-MA) don't exist differences.
Decoding lossy DTS all decoders can have very little differences but is dificult to say what is better.
Even free decoders are good for standard DTS.

[Jeff B]

Thank you.

[Marin85]

I am wondering about the precise meaning of the following type of log messages from eac3to: "A remaining delay of +1ms could not be fixed." Does it mean that the audio track still has a residual delay of +1ms and hence I hypothetically need to somehow apply -1ms delay to fix it completely if I want to? Or does it mean that eac3to managed to apply only (X-1)ms delay, where -X is the full audio delay determined by eac3to, and I still need to apply +1 ms (hypothetically) delay to the audio track to fix it completely?

This is just a theoretical question. I am well aware that in general I should not be worrying about 1ms audio delays, nor that I would be able to fix such a small value via eac3to (say, for an AC3 audio track). I would be thankful for a straight and clear answer. I have searched the web and there appear to be some controversial opinions on this matter.

[pandv2]

AC3 is a packed format. What it means is the samples are grouped and compressed. The packet lenght and duration is fixed and depends on the samplerate (44.1 KHz, 48 KHz...). If you want to do a lossless operation with a AC3 stream you need to do at packet level. Supose (I don't remember the real number) a packet is 16 ms, you want to delay the stream 20 ms, eac3to only can add a complete ac3 silence packet (16 ms) or two (32 ms). In the first case the error is -4ms and in the second +12ms.

[Marin85]

AC3 is a packed format. What it means is the samples are grouped and compressed. The packet lenght and duration is fixed and depends on the samplerate (44.1 KHz, 48 KHz...). If you want to do a lossless operation with a AC3 stream you need to do at packet level. Supose (I don't remember the real number) a packet is 16 ms, you want to delay the stream 20 ms, eac3to only can add a complete ac3 silence packet (16 ms) or two (32 ms). In the first case the error is -4ms and in the second +12ms.
Thank you for your reply, but my question is not specifically about delays in AC3 tracks, but about how to interpret a particular type of messages in the eac3to log.

[tebasuna51]

Or does it mean that eac3to managed to apply only (X-1)ms delay, where -X is the full audio delay determined by eac3to, and I still need to apply +1 ms (hypothetically) delay to the audio track to fix it completely
That is correct.

[Boulder]

Regarding audio delay: if DGIndex tells me that the audio track has +144ms delay, do I need to use +144ms or -144ms in eac3to's command line? It's a silly question but there is no obvious answer ;)

[Marin85]

Regarding audio delay: if DGIndex tells me that the audio track has +144ms delay, do I need to use +144ms or -144ms in eac3to's command line? It's a silly question but there is no obvious answer ;)
I feel like I am being mocked here :) What tebasuna51 (btw, thank you for the clear answer!) explained is - as I suspected - sort of the opposite of what eac3to is actually suggesting by the above message IMHO. It is confusing because eac3to message refers to the delay it tries to apply to the audio track, and not to the (resulting) delay the audio track already has.

[Boulder]

I feel like I am being mocked here :)Not at all :) It's just very confusing at times. I think I've always used the opposite but since the delays are usually rather small, it's not easy to determine which way is the correct one.

[tebasuna51]

... if DGIndex tells me that the audio track has +144ms delay, do I need to use +144ms...
Yes, you must always respect the sign.
The info (also MediaInfo) always show the delay than you need apply.

[tormento]

There is a stream that eac3to really doesn't like:

9: DTS Express, English, 1.0 channels, 24 bits, 96kbps, 48kHz

I can demux it but nothing more. Tried with TotalMedia libraries, LibAV ones but whenever I am going to export to wavs, it tells me some kind of errors. Any idea about how to convert it?

[Brazil2]

There is a stream that eac3to really doesn't like:

9: DTS Express, English, 1.0 channels, 24 bits, 96kbps, 48kHz

I can demux it but nothing more. Tried with TotalMedia libraries, LibAV ones but whenever I am going to export to wavs, it tells me some kind of errors. Any idea about how to convert it?
A sample would be nice for testing purposes :)

[tormento]

A sample would be nice for testing purposes :)
If only could be possible to demux it somehow..

[Brazil2]

If only could be possible to demux it somehow..
Well, you said you did:
I can demux it but nothing more.


Anyway, post a muxed sample so we can check it ?

[Bryce2]

Hi to all!
Trying to convert a DTS track to AC3 using "eac3to" leads me to an unusual problem.
DTS (ES):2.77GiB, 4h 23mn = AC3: 1.34GiB, 5h 0mn.
I want to mention that this is only the 2nd time I'm facing this problem after more than 200 conversions. The first time, I found a solution in this forum with adding a parameter to eac3to, but now I'm searching hours & hours with no luck.

[Sparktank]

Hi to all!
Trying to convert a DTS track to AC3 using "eac3to" leads me to an unusual problem.
DTS (ES):2.77GiB, 4h 23mn = AC3: 1.34GiB, 5h 0mn.
I want to mention that this is only the 2nd time I'm facing this problem after more than 200 conversions. The first time, I found a solution in this forum with adding a parameter to eac3to, but now I'm searching hours & hours with no luck.

Can you post the full log?

[Bryce2]

Hi to all!
Trying to convert a DTS track to AC3 using "eac3to" leads me to an unusual problem.
DTS (ES):2.77GiB, 4h 23mn = AC3: 1.34GiB, 5h 0mn.
I want to mention that this is only the 2nd time I'm facing this problem after more than 200 conversions. The first time, I found a solution in this forum with adding a parameter to eac3to, but now I'm searching hours & hours with no luck.

Solved:
Using as usual the default command "eac3to.exe input.dts output.ac3" was the reason of this weird effect (for this & only dts stream).
After trial & error and changing the command to "eac3to.exe input.dts output.ac3 -libav" ..did the thing.
Anyway ..:thanks: "Sparktank" for your interest to give a help.

[DJ-1]

Can I convert maintain DTS-MA track using eac32 ?
Sent from my GT-I9100 using Tapatalk 2

[DJ-1]

Oops, double post

[sneaker_ger]

eac3to can do that with the help of external libraries like Arcsoft. Read the start post.

[DJ-1]

eac3to can do that with the help of external libraries like Arcsoft. Read the start post.
OK, will read it, I was trying out Multi avchd Wichita uses eac32.... tried converting dts-ma to LPCM, (as my media box does play dts, but not .dts-ma..... ) so I was trying g to passtgrough the pre-decoded LPCM stream.. got lots of hissing...nothing more.
I have an Onkyo Tx nr609, (does support dts-ma )
Sent from my GT-I9100 using Tapatalk 2

[frumble]

Hello,
I have a problem with eac3to that drives me crazy. I want to decode the DTS-HD MA streams of my German release of the LotR Ext. BDs and encode them to FLAC so that my Linux audio players can play the full sound, not just the DTS core.
I have downloaded eac3to and the ArcSoft DTS Decoder 1.1.0.0 plus the HdBrStreamExtractor GUI.
The output WAVE files (or FLACs) have all 7 channels and the sound is nice but they appear to have the wrong channel mapping: Voices are very low and often come from only one side (I have only stereo speakers) and once in a while come with strange echo. But when I import them into Audacity and play them with the editor the output is correct. This however has no effect on the Audacity export: 6.1 channel export has the same mapping problems as the original file. My players are not the problem, they can play every DTS, AC3 and TrueHD stream correctly.
I read something about "strange setup" but I can't understand if this might be the problem. I have absolutely no knowledge about audio mastering. The issue affects both English and German streams. The Audacity export window offers the option to remap the channels but I have no clue what could be the right choice. So I really hope you can help me: I uploaded a piece of 6.1 export from Audacity in Ogg Vorbis and it would be really nice if someone could tell me the right mapping. Thank you very much in advance!
http://www.2shared.com/audio/boFrWT5B/Export-example-eac3to-wrong-ch.html - I hope not to violate against forum rules with this.

Off this topic I want to say something: It may sound ridiculous but I miss a "tutorial". I lost hours in trying to understand what these freeware tools do and how to set them up with decoder and codecs. The learning curve is very steep but not because I am a idiot but 'cause most of your tools lack a proper documentation. I am a Linux user and I am used to find answers in reading docs.

It's really great that there are freeware tools like eac3to but I can't understand why the authors of such media helpers don't make the source code available. The last version of eac3to is from 2010? I read something about a GPU video encoder the author is writing since then and the promise to get back to work on eac3to when this encoder is final. But since 2010 no progress with eac3to. In this time the program could have been matured and bugs could have been fixed from others but they can't do it because they don't have the source code. It is his right to hold the source for himself but I truly can't understand the reason. It doesn't seams to be his plan to make commercial profit out of it. But no hard feelings, this are just my thoughts.

[NanoBot]

Hi frumble,

this workaround might help: http://forum.doom9.org/showthread.php?p=1524301#post1524301

[sneaker_ger]

Unfortunately 6.1 is not defined in FLAC. So players might decode it not as you expect, unless you do a downmix.

In addition to NanoBot's link:
http://forum.doom9.org/showthread.php?p=1538930#post1538930

[Sparktank]

Unfortunately 6.1 is not defined in FLAC.

This has actually been a bit of a constant problem for FLAC.
One user here goes on about it...
Dear (people), please stop using 6.1 FLAC (http://anonym.to/?http://mod16.org/hurfdurf/?p=184)

I remember reading about 6.1 FLAC in other places.

Personally, I'd stick to the CORE or even a conversion to AC3@640, as 640 is perceived to have transparent quality to the master file.

With some BD rips I do, I'm quite happy using only the core for maximum compatibility with my BD player that can playback MKV (with original h264/VC1/MPEG).

It's not a significant compromise between lossless and core, I'd stick to CORE audio. The human ear wouldn't really be able to pick pu the difference too much.

[Midzuki]

Another possible approach: drop FLAC :p and "move house" to WavPack :)

[frumble]

Thank you for your help! So I made WAV files with eac3to and want to convert them to WavPack, but the Encoder says: "can't handle .WAV files larger than 4 GB (non-standard)!" - sadly booth streams are over 4 GB. Also Monkey's audio can't encode them: "Error: 1002".
Export to AC3 with ea3to works fine but I absolutely want a lossless codec (it feels better ;) ). In my desperation I even tried to merge the movie MKV with the WAVE files but MKVmerge GUI gives the error "87".
Both WAVE streams seams to be a little bit strange: Non of my audio players can play more than a few minutes (even they display only a few minutes of playtime) and also Audacity can only show me a few seconds!
The link from @Sparktank is very confusing for me. Maybe I try it later.
Downmixing the stream would be the last option but I would be very pleased if it's possible to preserve the original channels...

[Midzuki]

^ @ frumble:

1) forget Monkey Audio, it doesn't support multichannel ;

2) try "wavpack --help" in the command prompt ;)

[Midzuki]

Off this topic...

It's really great that there are freeware tools like eac3to but I can't understand why the authors of such media helpers don't make the source code available...
But since 2010 no progress with eac3to. In this time the program could have been matured and bugs could have been fixed from others but they can't do it because they don't have the source code.

Sadly you're not entirely correct. Open source-code is NO guarantee that there will ever be other capable programmers interested in fixing the bugs of the software. Take a look at MaestroSBT, K-Meleon, mpeg2enc, mkisofs, etc.

[frumble]

@Midzuki Sorry, the WaPack help page hasn't a option close to "allow input files bigger than 4 GB", or what do you mean?
You are right, the fact that a software is OSS doesn't mean that there are other developers capable with the task. But look at ffmpeg/libav, x264, VLC and mplayer in the media field. It's not soo bad. And consider the chance @madshi, eac3to's author can't work on freeware projects anymore or dies. Then the community would be forced to rewrite the tool from scratch. OSS is freedom and opportunity, not constraint.

[Midzuki]

@Midzuki Sorry, the WaPack help page hasn't a option close to "allow input files bigger than 4 GB", or what do you mean?
...

-i ignore length in wav header

[MuteyM]

Hi, I am decoding a DTS-ES 6.1 soundtrack to wav format. If I use the Sonic decoder then everything works correctly. However if I use libav, then I end up with a wav file with a corrupt channel mask. It looks like eac3to is not taking into account libav's lack of back channel decoding:

eac3to j:\backup 1) 4: c:\temp\orig.wav -libav
M2TS, 1 video track, 2 audio tracks, 4 subtitle tracks, 1:38:52, 24p /1.001
1: Chapters, 20 chapters
2: MPEG2, 1080p24 /1.001 (16:9)
3: AC3, English, 5.1 channels, 640kbps, 48kHz, dialnorm: -27dB
4: DTS-ES, English, 6.1 channels, 24 bits, 1509kbps, 48kHz, dialnorm: -4dB
5: Subtitle (PGS), English
6: Subtitle (PGS), French
7: Subtitle (PGS), Spanish
8: Subtitle (PGS), English
a04 The libav DTS decoder doesn't decode the back channels.
a04 Extracting audio track number 4...
a04 Removing DTS dialog normalization...
a04 Removing XCh extension...
a04 Decoding with libav/ffmpeg...
a04 Reducing depth from 64 to 24 bits...
a04 Writing WAV...
a04 Creating file "c:\temp\orig.wav"...

I end up with a wav file containing 5.1 channels of data, but with a channel mask of 0x70f (6.1) instead of the expected 0x60f (5.1). This confuses the heck out of any software that attempts to play it :)

I thought using -down6 might force eac3to to work properly, but this ended up with the same problem. So I think the solution is for madshi to update eac3to to write the correct channel mask when libav decodes 6.1 (and probably 7.1) DTS files.

[tebasuna51]

I end up with a wav file containing 5.1 channels of data, but with a channel mask of 0x70f (6.1) instead of the expected 0x60f (5.1).
Yes, is a know bug.

[frumble]

@Midzuki Thank you very much! The conversion worked with all streams and the end MKVs play nice with the WavPack codec! :)

[Midzuki]

^ Glad to see it worked :)

Now let's spread the word,
WavPack is the future! :sly: :D

[Bigmango]

Is it possible to convert TrueHD to DTS-HDMA (TrueHD > WAVS > DTS HD Encoder) and keep the same sound for the TrueHD tracks that apply DRC to improve the sound? (at least I think it's the DRC that's doing this ?).

This problem arises with the Transformers 3 TrueHD 7.1 track (and people tell me Iron Man is probably the only other movie doing the same). The converted DTS-HDMA track does not sound the same as the original TrueHD. Some sounds at different moments in the movie have more presence with the original TrueHD track. It makes the sound better, more dynamic.

I think this is caused by the DRC (?) that's applied differently on the different channels at different moments.

The TrueHD track sounds better with PowerDVD 12 compared to the converted DTS-HDMA, but with MPC I don't seem to hear a difference (this leads me to think that MPC doesn't apply the effects and so it doesn't handle the TrueHD metadata correctly (?).

PCM and DTS-HDMA can't modify the sound at playback time in the way TrueHD does. How can I save the wavs with the full TrueHD sound experience?

To summarize, I would like to convert TrueHD to DTS-HDMA and have it sound as good on hardware receivers and PowerDVD (now with eac3to the audio source quality is the same, but many sounds are lacking presence in different parts of the movie).


(This is the first movie I am having this problem with. It seems the Transformers 3 sound engineers wanted to TrueHD track to be reproduced with improvements applied to the Lossless track it contains - and the whole experience is indeed more enjoyable to my ears with these improvements).



PS: sorry mods, I have opened this thread (http://forum.doom9.org/showthread.php?p=1570771#post1570771) before, but I think here is the right place as it concerns eac3to

[tebasuna51]

Bimango, please continue this discussion in your original thread if you want.
The discussion about this topic in this thread was closed in first month.

eac3to always ignore DRC because is a effect to apply at play time (with 'Night mode' for instance), never when recoding, and don't improve the audio just make it less dynamic.

[Bigmango]

Bimango, please continue this discussion in your original thread if you want.
The discussion about this topic in this thread was closed in first month.

eac3to always ignore DRC because is a effect to apply at play time (with 'Night mode' for instance), never when recoding, and don't improve the audio just make it less dynamic.

Ok so if it's not DRC it's another effect that's applied to the sound. (continued in my other thread (http://forum.doom9.org/showthread.php?t=164724)).

Regarding eac3to: Can eac3to be used with another decoder that can save the "complete" TrueHD to wavs? It seems the open source decoders don't fully support TrueHD as they only extract the lossless track without the effects that enhance it by giving more presence to specific sounds at specific times (most movies only play the lossless track as is, so only a handful of movies seem to be concerned by this issue).

Thanks.

[tebasuna51]

eac3to save the complete TrueHD to wav's. Don't exist "effects that enhance it by giving more presence to specific sounds".

[Joniii]

I'm converting 5.1 AC3 from DVD (25.000 -> 23.976), Is it normal that eac3to remaps channels?

eac3to f:\al.ac3 g:\al.ac3 -slowdown
AC3, 5.1 channels, 1:26:43, 384kbps, 48kHz, dialnorm: -27dB
The Nero decoder doesn't seem to work, will use libav instead.
Removing AC3 dialog normalization...
Decoding with libav/ffmpeg...
Remapping channels...
Changing FPS from 25.000 to 23.976...
Encoding AC3 <640kbps> with libAften...
Creating file "g:\al.ac3"...
eac3to processing took 3 minutes, 42 seconds.
Done.

[tebasuna51]

Yes, internal AC3 channel order is different than standard channel order used to resample and send to encoder.
Don't worry.

[ilomambo]

Hello, I am having the following problem:

eac3to v3.24
command line: "c:\Program Files\Utils\eac3to\eac3to.exe" "track01.dts" "track01.wav" -libav -simple
------------------------------------------------------------------------------
VOB, 1 audio track, 0:08:57
1: DTS-96/24, 5.1 channels, 24 bits, 1510kbps, 96kHz
Track 1 is used for destination file "track01.wav".
[a01] Extracting audio track number 1...
[a01] Decoding with libav/ffmpeg...
[a01] Reducing depth from 64 to 24 bits...
[a01] Writing WAV...
[a01] Creating file "track01.wav"...
[a01] Clipping detected, a 2nd pass will be necessary. <WARNING>
[a01] Starting 2nd pass...
[a01] Extracting audio track number 1...
[a01] Decoding with libav/ffmpeg...
[a01] Reducing depth from 64 to 24 bits...
[a01] Writing WAV...
[a01] Applying -0.73dB gain...
[a01] Creating file "track01.wav"...
eac3to processing took 1 minute, 27 seconds.
Done.


The wav file seems fine, I can see it plays all 5.1 channels
But look at the file sizes:

>>dir track01*.*
Volume in drive K is Music
Volume Serial Number is 884C-4DCD

Directory of K:\DTS

29/04/2012 14:09 105,261,056 track01.dts
05/05/2012 23:54 987 track01 - Log.txt
05/05/2012 23:54 463,804,460 track01.wav
3 File(s) 569,066,503 bytes
0 Dir(s) 88,401,051,648 bytes free


Is this correct, WAV has a factor of 4.5 times the DTS size?
I don't know, other WAV DTS files I have are much smaller (given the average MB/min of audio, this one is 8:56 min)
Is there any command line switch to keep the WAV size smaller?

[tebasuna51]

...
Is this correct, WAV has a factor of 4.5 times the DTS size?

eac3to/libav decode your DTS-96/24, 5.1 channels, 24 bits, 1510kbps, 96kHz to a WAV 5.1 channels, 24 bits, 48kHz with a bitrate of:
6 channels x 24 bits x 48KHz = 6912 kb/s

6912/1510 = 4,57

Then, yes, is correct.

Is there any command line switch to keep the WAV size smaller?
You can use -down16 to obtain a WAV 5.1 channels, 16 bits, 48kHz with a bitrate of:
6 channels x 16 bits x 48KHz = 4608 kb/s

4608/1510 = 3,05

[ilomambo]

You can use -down16 to obtain a WAV 5.1 channels, 16 bits, 48kHz with a bitrate of:
6 channels x 16 bits x 48KHz = 4608 kb/s

4608/1510 = 3,05

Thanks.

I am dumb regarding the technical details. I think the DTS file has the same information as the final WAV, that's why the increase in size seemed too much.
But, I assume from your explanation, that much of the DTS info is replicated to create the 6 CH WAV, and that's why the file is so much bigger, isn't it?

EDIT:
I just looked in another song I have in DTS 5.1 WAV format (in the AC3 filter properties while it was playing) and it showed this:


Decoder:
Stream format: DTS 3/2.1 (5.1) 44100Hz
Bitstream type: 14bit low endian
Frame size: free format
Samples: 1024
Bitrate: unknown
SPDIF stream type: 0xc
Frame interval: 4096
Actual bitrate: 1411kbps
DTS
speakers: 3/2.1 (5.1)
sample rate: 44100Hz
bitrate: 1411kbps
stream: 14bit LE
frame size: 3584 bytes
nsamples: 1024
amode: 9
No CRC

Tebasuna51, If I follow your math 6ch x 14bit x 44KHz = 3696 kbps != 1411 kbps reported by AC3
It is 13:40 min song and the file only takes 141MB
Something is not fitting, according to my understanding

On the other hand when I play the file created by eac3to, AC3 filter shows this:


Input format: PCM24 3/2.1 (5.1) 48000
User format: PCM16 - 0
Output format: PCM16 3/2.1 (5.1) 48000

That's why I think I am using the wrong tool. I just wanted to wrap the DTS file in WAV format, not to convert it. It seems the file I got is pure PCM.

[MrVideo]

I just wanted to wrap the DTS file in WAV format, not to convert it. It seems the file I got is pure PCM.

WAV is not a wrapper, it is a format, just like MPEG-2 and MPEG-4 are video formats, that get wrapped into various containers. Neither DTS or WAV are containers.

DTS is a compressed audio format, while WAV is not. Hence the reason that the WAV file is larger after you uncompressed the DTS file.

[Midzuki]

...

DTS is a compressed audio format, while WAV is not.

But yes, .WAV is a RIFF-based container, and may contain compressed audio. Regarding DTS-in-WAV especifically, there are two types, 1) normal, without SPDIF-padding, and with a .dca TwoCC, and 2) hacky, with SPDIF-padding, disguised as stereo PCM @ 32 / 44.1 / 48 kHz.

[MrVideo]

Interesting. I've never seen anything but WAV uncompressed PCM audio.

Then, that leads me to this question... why put DTS compressed data into a WAV file? I'd just leave it as a DTS file.

[tebasuna51]

But, I assume from your explanation, that much of the DTS info is replicated to create the 6 CH WAV, and that's why the file is so much bigger, isn't it?
The audio data in DTS is encoded (compressed like a zip) the audio data in WAV file is decoded (uncompressed PCM data)

I just looked in another song I have in DTS 5.1 WAV format
A dts_5.1_wav is not a correct WAV file because the header don't show the content. The fake header show a PCM 2.0 44.1KHz just to lies a burner and burn a CD-Audio.

I just wanted to wrap the DTS file in WAV format, not to convert it.
Then you want burn CD-Audios 5.1
Try spdifer (http://ac3filter.net/wiki/AC3Filter_tools), a AC3Filter tool.

Then, that leads me to this question... why put DTS compressed data into a WAV file?

Like you can read, only to burn a CD Audio.
Some CD players can send the DTS 5.1 and play surround.

[ilomambo]

Then you want burn CD-Audios 5.1
Try spdifer (http://ac3filter.net/wiki/AC3Filter_tools), a AC3Filter tool.


tebasuna51, you are the man! :goodpost:

:thanks:

spdifer worked like a charm! I got my WAV file 5.1ch, at the same size as the DTS original file !!

... and I learned something about DTS and WAV files.

[RazvanuZu]

I get an error when I try to use eac3to with SurCode DVD–DTS v1029. Is this a problem with eac3to or SurCode DVD–DTS v1029? I mean, is eac3to incompatible with SurCode DVD–DTS v1029?
When I use SurCode DVD–DTS v1021 I get no error.

The error sounds like: "Pressing the Surcode "Encode" button didn't seem to work", but when I directly use SurCode DVD–DTS v1029 I don't get any errors at all.

What is the latest compatible version of SurCode DVD–DTS?

Thanks.

[ramicio]

I wish there was a way to have eac3to in Linux without wine. I don't have a GUI installed, so I can't do wine. I run eac3to on my Windows machine to rip discs, but sometimes I forget something and need to fiddle with large files after I've written them, and my disc is across a network. Being able to have it on my Linux would speed things up quite a bit. I would only need basic libavcodec functionality.

[Furiousflea]

No matter the source/decoder or encoder, when you downmix 6.1 to 5.1 with eac3to need use:

-0,1,2,3,5,6,4 -down6

Is this still needed, may I ask. Changelog mentions a bug fixed in version 3.21...

fixed: 6.1 DTS decoding with ArcSoft resulted in wrong channel order

Many thanks.

[tebasuna51]

The decode is OK, only need a remap when downmix 6.1 to 5.1

[Furiousflea]

The decode is OK, only need a remap when downmix 6.1 to 5.1

Thanks for confirmation :)

[Joniii]

I have previously converted 25.000 fps audio to 23.976 with -slowdown switch, how do I convert 25.000 to 24.000?

I have import Blu-ray with strange 24p (not 24.000/1.001) h.264 and DTS. I'm trying to add AC3 audio into it from my older PAL DVD.

[ramicio]

Are you even sure the audio is going to line up? What is the fixation with the audio from the DVD?

Anyway, you would first add the "-25.000" switch, to tell it to assume the material is 25 FPS, and then "-changeto24.000".

[Joniii]

Are you even sure the audio is going to line up? What is the fixation with the audio from the DVD?

Anyway, you would first add the "-25.000" switch, to tell it to assume the material is 25 FPS, and then "-changeto24.000".

I've always adjusted the delay after conversion. I'm adding dubbed audio from DVD to import blu-ray for my 3 year old.

I've thought you need to convert audio with -slowdown not -changeto switch?

[ramicio]

I still don't get it. Slowdown is deprecated, and only for 25 to 23.976 conversion.