eac3to - audio conversion tool

[xkodi]

@madshi

without a doubt TrueHD track found on 'V for Vendetta' HD-DVD is 24bit/48kHz.

why i'm sure about that? what is my test procedure?

TrueHD is based on MLP (aka Packed PCM), so i've used 'SurCode MLP' to create test MLP files from WAV files with different resolution and number of channels e.g. 24bit/44.1kHz/5.1, 16bit/48kHz/2.0, etc. and named them test2444151.mlp, 164820.mlp, etc.

then i've used WinDVD 8.0 B06.109 to open those test MLP files, WinDVD doesn't produce sound while playback MLP files, but every time for every test MLP file, WinDVD shows (from the menu 'Setup->Information') the right resolution and number of channels, exactly the same as the resolution and number of channels of the source WAV file from which the corresponding MLP file was created:

Stream:
Type: Packed PCM
Bitrate: 0 Kbps
Number of main channels: 6
LFE Channel: None
Sampling Frequency: 44.1khz
Bits Per Sample: 24
Output:
Sample Rate: 44 KHz
Bits per Sample: 16

so that means without a doubt, that WinDVD checks the MLP (TrueHD) headers for the real resolution of the encoded audio.

the next step was to demux the 'V for Vendetta' HD-DVD using EVOdemux and to rename the resulted TrueHD file to truehd.mlp and then WinDVD 8.0 B06.109 said that it is:

Stream:
Type: Packed PCM
Bitrate: 0 Kbps
Number of main channels: 6
LFE Channel: None
Sampling Frequency: 48khz
Bits Per Sample: 24
Output:
Sample Rate: 48 KHz
Bits per Sample: 16

so the TrueHD (MLP) track found on 'V for Vendetta' HD-DVD is 24bit/48kHz/5.1

to double check the accuracy of the information that WinDVD outputs, i've also done another test:

this time with the bonus TrueHD track from the 'Nine Inch Nails Live: Beside You in Time' HD-DVD, which is:

633 042 317 bytes in size

and WinDVD says it's 24bit/48kHz/5.1.

i've created, using eac3to, both 16bit and 24bit WAV files from that TrueHD track and then using 'SurCode MLP' compressed them to MLP, the result:

16bit WAV file size: 660 487 104 bytes
compressed to MPL size: 358 807 274 bytes

24bit WAV file size: 990 730 628 bytes
compressed to MPL size: 649 693 162 bytes

the MLP compressed size of the 16bit WAV file is 358 807 274 bytes and MLP compressed size of the 24bit WAV file is 649 693 162 bytes, which is almost the same number as the original TrueHD track - 633 042 317 bytes in size and because TrueHD is optimized MLP compression again can be concluded that the original TrueHD track is 24bit exactly as WinDVD reports.

[Rectal Prolapse]

An AVSForum Insider has mentioned that TrueHD decoders always output 24 bit.

I think you need to do your test again on a disc that is known to have a 16 bit TrueHD audio track. I have a feeling that you will think that all TrueHD releases are 24 bit, when in fact the majority (or even all) of them are 16 bit.

[xkodi]

An AVSForum Insider has mentioned that TrueHD decoders always output 24 bit.

I think you need to do your test again on a disc that is known to have a 16 bit TrueHD audio track. I have a feeling that you will think that all TrueHD releases are 24 bit, when in fact the majority (or even all) of them are 16 bit.

i do not agree, because in the first test i don't use the output of TrueHD decoders, instead i'm using series of homemade MLP tracks with known resolution to prove, that WinDVD can find and shows the real resolution of the MLP/TrueHD tracks correctly and so WinDVD uses the MLP/TrueHD headers to find that information and there is no usage of TrueHD decoders at all in that test.

in the second test, yes, but only to double check the results from the first test, i'm using Nero TrueHD decoder to decompress 600MB TrueHD track (demuxed from real HD-DVD disc) in both 16bit and 24bit WAV files and then compress them again to MLP and that test shows that the initial source for the original TrueHD track is 24bit, because if the original TrueHD track is created from 16bit WAV it will be only 340MB.

so i can't see any flaw in the tests and that is why i'm posting here, if you have something more concrete for the possible flaw, also why you are so sure, that TrueHD tracks are 16bit, WinDVD and the 'Surecode MLP' test shows exactly the opposite and in that test, i'm not using the output of TrueHD decoders at all.

[edit] and here is another prove, even if TrueHD decoders always output 24 bit, WinDVD shows that the stream is 16bit if and only if it is really 16bit (the output is from homemade 16bit MLP track):

Stream:
Type: Packed PCM
Bitrate: 0 Kbps
Number of main channels: 6
LFE Channel: None
Sampling Frequency: 44.1khz
Bits Per Sample: 16
Output:
Sample Rate: 44 KHz
Bits per Sample: 16

[madshi]

without a doubt TrueHD track found on 'V for Vendetta' HD-DVD is 24bit/48kHz.
:eek:

TrueHD is based on MLP (aka Packed PCM), so i've used 'SurCode MLP' to create test MLP files from WAV files with different resolution and number of channels e.g. 24bit/44.1kHz/5.1, 16bit/48kHz/2.0, etc. and named them test2444151.mlp, 164820.mlp, etc.

then i've used WinDVD 8.0 B06.109 to open those test MLP files, WinDVD doesn't produce sound while playback MLP files, but every time for every test MLP file, WinDVD shows (from the menu 'Setup->Information') the right resolution and number of channels, exactly the same as the resolution and number of channels of the source WAV file from which the corresponding MLP file was created:

so that means without a doubt, that WinDVD checks the MLP (TrueHD) headers for the real resolution of the encoded audio.

the next step was to demux the 'V for Vendetta' HD-DVD using EVOdemux and to rename the resulted TrueHD file to truehd.mlp and then WinDVD 8.0 B06.109 said that it is:
Well, this does not fully convince me yet. Why? Because the headers of MLP files differ from the headers of TrueHD files. eac3to is able to analyze MLP files and extract the "bitdepth" information from the MLP header. But eac3to is not able to do the same for TrueHD files. I'm not sure. Maybe WinDVD is able to do that. Maybe WinDVD is cleverer than eac3to, that's possible. But it's also possible that WinDVD is reporting the correct bitdepth for MLP files while always reporting 24bit for TrueHD files.

to double check the accuracy of the information that WinDVD outputs, i've also done another test:

this time with the bonus TrueHD track from the 'Nine Inch Nails Live: Beside You in Time' HD-DVD, which is:

633 042 317 bytes in size

and WinDVD says it's 24bit/48kHz/5.1.

i've created, using eac3to, both 16bit and 24bit WAV files from that TrueHD track and then using 'SurCode MLP' compressed them to MLP, the result:

16bit WAV file size: 660 487 104 bytes
compressed to MPL size: 358 807 274 bytes

24bit WAV file size: 990 730 628 bytes
compressed to MPL size: 649 693 162 bytes

the MLP compressed size of the 16bit WAV file is 358 807 274 bytes and MLP compressed size of the 24bit WAV file is 649 693 162 bytes, which is almost the same number as the original TrueHD track - 633 042 317 bytes in size and because TrueHD is optimized MLP compression again can be concluded that the original TrueHD track is 24bit exactly as WinDVD reports.
Ok, that quite clearly shows that that bonus track is most probably 24bit. But that's still no 100% proof that Vendetta is 24bit, too. Can you do the same test with Vendetta, too? Or could you try to find a TrueHD track which WinDVD reports at being 16bit only?

Now if only we could make eac3to decode demuxed MLP/TrueHD files then we could do another double check and decode your home made MLP files and check whether the Nero decoder produces bit identical output! You don't happen to have an authoring software available which can package your home made MLP files into an EVO container, so that eac3to can decode your MLP files?

[madshi]

when I try to convert a mono track with eac3to I get this

E-AC3, 1.0 channels, 1:58:47, 64kbit/s, 48khz
Mono E-AC3 tracks can only be decoded by Nero's filters, switching to Nero.
Muxing eac3 file to raw. Please wait...
The file size of the raw file doesn't seem to fit.
The expected file size for 16 bit is 5.36 MB.
The expected file size for 24 bit is 8.04 MB.
The real file size is 0 Bytes.
I've just tried it again and it works on my test mono chunk. Would it be possible for you to make that mono track available for me for testing?

[xkodi]

Can you do the same test with Vendetta, too? Or could you try to find a TrueHD track which WinDVD reports at being 16bit only?


in several days i will do the same test with Vendetta and post the results; i have only two HD-DVD discs with TrueHD and they both say 24bit, but i will look for more TrueHD samples.


You don't happen to have an authoring software available which can package your home made MLP files into an EVO container, so that eac3to can decode your MLP files?

actually i can pack them in AOB container and as far as i know EVOdemux can convert AOB to EVOB, but i don't know if 'Nero Audio Decoder 2' is capable to decode pure MLP, but i will try that test in several hours and post back the result.

[madshi]

Thanks, xkodi!

[xkodi]

actually i can pack them in AOB container and as far as i know EVOdemux can convert AOB to EVOB, but i don't know if 'Nero Audio Decoder 2' is capable to decode pure MLP, but i will try that test in several hours and post back the result.

24bit/48kHz/5.1ch WAV file -> 'SurCode MLP' -> 24bit/48kHz/5.1ch MLP file -> 'discWelder CHROME' -> AOB container -> 'EVOdemux 0.626' -> EVOB container

but the resulted EVOB file opened again with EVOdemux says that the stream is LPCM, not MLP (PPCM) and WinDVD also reports LPCM and nothing can playback that EVOB, so seems that is a dead end, because EVOdemux can't convert AOB with MLP to EVOB correctly, maybe MLP can't be muxed in EVOB at all, only the TrueHD modification.

[mp4p2p]

eac3toGUI 0.63 this link:http://www.sendspace.com/file/jwghja

my web can't open,may be is isp question,

please,who can send me?

my email:dhdbase@gmail.com

thanks

[madshi]

maybe MLP can't be muxed in EVOB at all, only the TrueHD modification.
Actually some HD-DVD tracks are MLP and not TrueHD. So it's definitely possible to mux MLP into EVOB. I think EvoDemux just doesn't do it properly, sadly... :(

[Sagittaire]

I have problem ... ???

D:\Mes dossiers\Codec\HDDVD\eac3to>eac3to.exe azerty.eac3 azerty.wav
E-AC3, 5.1 channels, 0:00:30, 1536kbit/s, 48khz
Muxing eac3 file to raw. Please wait...
Getting "Dump" instance failed

[madshi]

I have problem ... ???
You need to register the dump filter which ships with GraphEdit. Use "regsvr32 dump.ax".

[xkodi]

Actually some HD-DVD tracks are MLP and not TrueHD. So it's definitely possible to mux MLP into EVOB. I think EvoDemux just doesn't do it properly, sadly... :(

very conclusive test, at least in my opinion:

1) 'Nine Inch Nails Live: Beside You in Time' bonus track, TrueHD size:

633 042 317 bytes

2) decompressed using eac3to to 16bit WAV, size: 660 487 112 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

358 807 274 bytes

3) decompressed using eac3to to 16bit WAV and then converted to 24bit using Wavewizard v0.54b,
now the least significant bits are zeroes checked with hexeditor, size: 990 730 628 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

362 789 876 bytes

4) decompressed using eac3to to 24bit WAV file, size: 990 730 628 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

649 693 162 bytes

so 2) and 3) shows that if the source WAV file is not real 24bit, but 16bit padded to 24bit with zeroes
for least significant bits MLP compression is enough intelligent and that doesn't significantly increase
the size of the compressed file and that plus 3) and 4) clearly shows that the source WAV file for the
'Nine Inch Nails Live: Beside You in Time' bonus TrueHD track should be real 24bit.

p.s. soon i will do the same test for Vendetta TrueHD track.

[Rectal Prolapse]

xkodi: Well, I'm going by the reported bitdepth from the studios and hidef reviewers - although they aren't always reliable.

Perhaps the only thing we can do is ask the Insiders on AVSForum.com.

Good luck with your tests!

[madshi]

very conclusive test, at least in my opinion
So you got your MLPs successfully packages into EVO now? :)

1) 'Nine Inch Nails Live: Beside You in Time' bonus track, TrueHD size:

633 042 317 bytes

2) decompressed using eac3to to 16bit WAV, size: 660 487 112 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

358 807 274 bytes

3) decompressed using eac3to to 16bit WAV and then converted to 24bit using Wavewizard v0.54b,
now the least significant bits are zeroes checked with hexeditor, size: 990 730 628 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

362 789 876 bytes

4) decompressed using eac3to to 24bit WAV file, size: 990 730 628 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

649 693 162 bytes
Wait a moment! If the Nero decoder works correctly, step 4 should have given you a WAV file with the least significant 8 bits being zero. And if you reencoded this, you should again end up with only a 360MB of MLP. So this clearly shows that the Nero decoder is either not working properly, or that it's doing some additional processing on the audio data after decoding. Do you agree?

2) and 3) shows that if the source WAV file is not real 24bit, but 16bit padded to 24bit with zeroes
for least significant bits MLP compression is enough intelligent and that doesn't significantly increase
the size of the compressed file and that plus 3) and 4) clearly shows that the source WAV file for the
'Nine Inch Nails Live: Beside You in Time' bonus TrueHD track should be real 24bit.
I think the main clue is that the TrueHD file is only barely smaller than the 16bit WAV file. This is a quite clear sign that the TrueHD file has more than 16bit. Otherwise it would have an extremely bad compression ratio, which is very unprobable.

I think we should be able to draw similar conclusions about V for Vendetta: If the TrueHD file is noticably smaller than the 16bit WAV then probably the TrueHD file is only 16bit. If the TrueHD file is however almost as big as the 16bit WAV, then the TrueHD track is probably more than 16bit. Agreed?

[xkodi]

So you got your MLPs successfully packages into EVO now? :)

no, i still can't get MLP stream muxed into EVO, the test file is TrueHD stream taken from real HD-DVD disc and then decompressed with eac3to to 16 and 24 bit WAVs, that are used during the test.


Wait a moment! If the Nero decoder works correctly, step 4 should have given you a WAV file with the least significant 8 bits being zero. And if you reencoded this, you should again end up with only a 360MB of MLP. So this clearly shows that the Nero decoder is either not working properly, or that it's doing some additional processing on the audio data after decoding. Do you agree?

if the PCM source is 16bit and because MLP is lossless, then least significant byte should be zero in the decompressed 24bit WAV, but at the moment i can't test that, because i can't find a way how to mux homemade MLP with known PCM source into EVO.

WinDVD always shows that TrueHD tracks are 24bit, maybe all TrueHD tracks are actually 24bit, no matter from the resolution of the PCM source from which are created, maybe when the PCM source is 16bit, before the MLP compression step, they do some processing to upsample it to 24bit and if that is the case Nero decoder works correctly, because that info is added before the compression.


I think the main clue is that the TrueHD file is only barely smaller than the 16bit WAV file. This is a quite clear sign that the TrueHD file has more than 16bit. Otherwise it would have an extremely bad compression ratio, which is very unprobable.

I think we should be able to draw similar conclusions about V for Vendetta: If the TrueHD file is noticably smaller than the 16bit WAV then probably the TrueHD file is only 16bit. If the TrueHD file is however almost as big as the 16bit WAV, then the TrueHD track is probably more than 16bit. Agreed?

yes, i completely agree with that, my Vendetta test clearly also supports that conclusion especially in combination with the previous 'Nine Inch Nails' test:

1) 'V for Vendetta' TrueHD track, file size:

1 368 231 154 bytes

2) decompressed using eac3to to 16bit WAV, size: 4 580 550 320 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

1 292 730 780 bytes

3) decompressed using eac3to to 24bit WAV file, size: 6 871 150 160 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

3 315 710 672 bytes

so i want to summarize:

- WinDVD always shows that TrueHD tracks are 24bit, but that doesn't mean that the source PCM from, which the TrueHD track is created, is really 24bit

- some TrueHD tracks seems to be created from real 24bit PCM source like the 'Nine Inch Nails Live: Beside You in Time' bonus TrueHD track, so downsampling them to 16bit is probably not so good idea

- other TrueHD tracks seems to be created from 16bit PCM source like 'V for Vendetta' TrueHD track and so downsampling them to 16bit is probably completely save

so maybe eac3to default behavior when converting to WAV/FLAC should be the following: to check the size of the TrueHD track and according to that automatically to decide if the PCM source is 16bit or 24bit and then choose the correct resolution for the output and of course to be possible manually to select 16bit or 24bit. what do you think about that?

[ABOOABOO]

when i used the gui i get

C:\eac3to>eac3to.exe "C:\eac3to\input.eac3" "C:\Documents and Settings\bLaCk MaM
bA\Desktop\TEST.ac3" -640
E-AC3, 5.1 channels, 1:48:25, 1536kbit/s, 48khz
Muxing eac3 file to raw. Please wait...
The audio decoder didn't accept the input stream.

anybody else got this

[The_Keymaker]

Here is the link to the latest version of Eac3toGUI. It is version 0.82 and features all of Madshi's latest options.

http://www.sendspace.com/file/9kjvud

Please let me know if you find any problems.

Regards,
The_Keymaker

EDIT: Version 0.80 had a bug which deleted the last two characters on the command line.

[madshi]

WinDVD always shows that TrueHD tracks are 24bit, maybe all TrueHD tracks are actually 24bit, no matter from the resolution of the PCM source from which are created, maybe when the PCM source is 16bit, before the MLP compression step, they do some processing to upsample it to 24bit and if that is the case Nero decoder works correctly, because that info is added before the compression.
No, there's nothing added before compression. I'm sure of that because of 2 reasons:

(1) Roger Dressler (Dolby) said that if the source was 16bit, the output of a TrueHD decoder should have the least significant 8 bits set to zero.

(2) If there was "real" information added to the source to make it 24bit, the "V for Vendetta" TrueHD file size would have to be much bigger than it really is (see your test 3).

yes, i completely agree with that, my Vendetta test clearly also supports that conclusion especially in combination with the previous 'Nine Inch Nails' test:

1) 'V for Vendetta' TrueHD track, file size:

1 368 231 154 bytes

2) decompressed using eac3to to 16bit WAV, size: 4 580 550 320 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

1 292 730 780 bytes

3) decompressed using eac3to to 24bit WAV file, size: 6 871 150 160 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

3 315 710 672 bytes
Thanks for the test. Good information!

so i want to summarize:

- WinDVD always shows that TrueHD tracks are 24bit, but that doesn't mean that the source PCM from, which the TrueHD track is created, is really 24bit
Agreed.

- some TrueHD tracks seems to be created from real 24bit PCM source like the 'Nine Inch Nails Live: Beside You in Time' bonus TrueHD track, so downsampling them to 16bit is probably not so good idea
Agreed.

- other TrueHD tracks seems to be created from 16bit PCM source like 'V for Vendetta' TrueHD track and so downsampling them to 16bit is probably completely save
I'd agree only if the decoder behaved correctly and set the 8 least significant bits to zero. But the Nero decoder doesn't do that. I've no idea what exactly the Nero decoder is doing. Maybe it's doing range compression? Or changing volume slightly? Anyway, the output of the Nero decoder is clearly "finer" than 16bit, either due to a bug or due to additional processing on the audio data. So downsampling the decoder output to 16bit sounds like losing information to me.

------

Here comes a polite request to all people who have bought the HD-DVD/Blu-Ray plugin: Could you please contact Nero, report about the incorrectly working TrueHD decoder and ask them to fix this? I've already done that, but got no reply at all. Maybe we'll have more success if multiple people complain... Thanks!

[madshi]

Here is the link to the latest version of Eac3toGUI. It is version 0.82 and features all of Madshi's latest options.
Thanks! Just had a look at the GUI. Looks nice to me. Some comments:

(1) I know the help doesn't say it, but theoretically the AC3 bitrate can be anything between 32 and 640. So if you want you could add further bitrates, or even a slider or up/down buttons or whatever you like. Smaller bitrates than 192 may make sense e.g. for mono tracks.

(2) I'd suggest to rename the "Select Filter" text to "Force Filter" cause there are also Sonic filters that eac3to can use, if you don't activate any of the "force" switches. Alternatively you could replace the two checkboxes with 3 radios like "use whatever filter eac3to prefers", "use Nero filters" and "use orbitLee's filter".

(3) 96khz is also a valid option for PCM. 8 channels is also a valid option for TrueHD (although currently it doesn't make much sense, since the Nero decoder currently only supports 5.1 decoding).

(4) eac3to supports any channel order you like, not just 0,1,2,3,4,5 and the Blu-Ray one. E.g. you could do "5,4,0,1,2,3" or "2,1,4,0,3,5" or for 8 channels you could do "5,7,1,2,6,4,3,0" or anything else. I just don't know if there's any use for any other order than the default one (0,1,2,3,4,5[,6,7]) and the Blu-Ray one (0,1,2,5,3,4[,6,7]). But you can tell eac3to to reorder the channels to any strange order you like. Don't know if it's worth it adding this capability to the GUI, though.

(5) The TrueHD options can all be selected at the same time, but the GUI always allows only one option at the same time.

(6) Most PCM options can also be selected at the same time. Of course you can only select either 16bit or 24bit. And you can only select "big" or "little". But you could select e.g. "24bit", "8 channels" and "big" at the same time. Btw, you may want to rename "big" and "little" to "big endian" and "little endian", so people know what the options mean.

(7) Supported input file extensions are: "pcm", "eac3", "dd+", "ddp", "ec3", "wav" and "evo".

(8) It would be nice (but not necessary) if you disabled all options which don't make sense for the input file type. E.g. the TrueHD options could be disabled if the user selects an E-AC3 input file.

These are just some comments and suggestions. You don't need to implement any of this, of course!

[otaku975]

hello
i was trying to do some conversion
but i was wondering one thing
its DTSHD to DTS
wich extension should the dtshd file be set to , and same for dts output
thanks!

[madshi]

i was trying to do some conversion
its DTSHD to DTS
eac3to currently doesn't do this conversion. I think there are other tools that do that, but you won't find them in this thread. Sorry, don't have a link at hand.

[xkodi]

eac3to currently doesn't do this conversion. I think there are other tools that do that, but you won't find them in this thread. Sorry, don't have a link at hand.

http://forum.doom9.org/showpost.php?p=967297&postcount=900

[The_Keymaker]

Thanks for the feedback Madshi.

Here is the link for the latest version of EAC3toGUI (version 0.90):

http://www.sendspace.com/file/vech6e

EAC3toGUI Version 0.90 changelog:

- Changed filter section title to "Force Filter"
- Added "endian" to big and little option labels
- Added 128kbit and 64kbit AC-3 bit rate options
- Now allows simultaneous selection of TureHD options
- Now allows simultaneous selection of PCM options
- Added logic to allow selection of common TrueHD and
PCM options (96KHz, 8 Ch) but not BOTH!

This version does not yet feature allowing specific channel orders or file type checking (i.e., PCM or TrueHD). My program can only determine filetype by extension. If this is sufficient, I will try to implement that in the next version.

Please let me if you discover any problems.

Regards,
The_Keymaker

[madshi]

Here is the link for the latest version of EAC3toGUI (version 0.90)
Looks good - thanks!!

This version does not yet feature allowing specific channel orders or file type checking (i.e., PCM or TrueHD). My program can only determine filetype by extension. If this is sufficient, I will try to implement that in the next version.
If you ask me, filetype by extension is just fine.

[Blackburst67]

I have some probleme to encode the sound of "Nine Inch Nails BESIDE YOU IN TIME".

I have the entiere HD-DVD and encode the video part, but I can't encode the sound.
The TrueHD gives me an error with eac3to

C:\Documents and Settings\Garf\Bureau\eac3>eac3to.exe "D:\55.wav" "D:\nin.ac3"
-448 -keep24
Converting the wav file to ac3. Please wait...

Aften: A/52 audio encoder
Version SVN
(c) 2006-2007 Justin Ruggles, Prakash Punnoor, et al.

input format: Signed 24-bit 48000 Hz 5.1-channel
output format: 48000 Hz 3/2 + LFE

"aften" reported error code "3221225477".

C:\Documents and Settings\Garf\Bureau\eac3>

eac3to already convert my TrueHD in EVO files and gives me an 6ch wav.

And when I demux the other track in E-AC3 and convert with eac3to, I have an eroor with Dolby who can't parsing the header of the file...

[madshi]

I have some probleme to encode the sound of "Nine Inch Nails BESIDE YOU IN TIME".
That error code in hexa is "C0000005" which means that "aften" crashed in a bad way while trying to encode the wav. Can you play that wav? Does it sound ok if you play that file? Which aften version are you using? Try a different aften version. Maybe that one you're using is buggy.

And when I demux the other track in E-AC3 and convert with eac3to, I have an eroor with Dolby who can't parsing the header of the file...
Can't say anything about that without seeing the exact error message.

[Binnugsai]

I have so some HD-DVD in my hard disk and use evodemux.

So i want to know if i have to use for the sound format when demux to "sound.evo" or sound.ddp.... which on do i use exactly !

And when i try for example to use the English sound from King Kong about 1500 k, i have some prob and a error message with Aften revision.......

So please, can someone can upload with eac3togui with inside the good tools, aften, sox etc.....

Thank you for all everybody.

Note : i try to demux one EVO, the full movie from King Kong, Movie and sound, i use VC1 and ddp for the sound and i have a file kingkong.evo, i try to use eac3togui 0.90 and failed. Nothing, no reaction about the software.

[madshi]

@Binnugsai, it's a bit difficult for me to help you without having specific erorr messages. For TrueHD tracks you should use EvoDemux to mux the TrueHD track (and only this track) into a separate EVO file. For E-AC3 tracks you should simply demux the E-AC3 tracks. The TrueHD EVO files and demuxed E-AC3 files can then be converted to AC3 or FLAC by using eac3to (or the GUI).

[Binnugsai]

Thank you Mdshi, nice to, you.

So you said me to use for TrueHD in a separate EVO in output,
feature1.evo and feature2.evo with only one sound, that's it !. and for the dd+( eac3) just demux but fir the output format, dd+ or ddp or another one.....

[foxyshadis]

very conclusive test, at least in my opinion:

1) 'Nine Inch Nails Live: Beside You in Time' bonus track, TrueHD size:

633 042 317 bytes

2) decompressed using eac3to to 16bit WAV, size: 660 487 112 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

358 807 274 bytes

3) decompressed using eac3to to 16bit WAV and then converted to 24bit using Wavewizard v0.54b,
now the least significant bits are zeroes checked with hexeditor, size: 990 730 628 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

362 789 876 bytes

4) decompressed using eac3to to 24bit WAV file, size: 990 730 628 bytes
compressed back to MPL using 'SurCode MLP', MLP file size:

649 693 162 bytes

so 2) and 3) shows that if the source WAV file is not real 24bit, but 16bit padded to 24bit with zeroes
for least significant bits MLP compression is enough intelligent and that doesn't significantly increase
the size of the compressed file and that plus 3) and 4) clearly shows that the source WAV file for the
'Nine Inch Nails Live: Beside You in Time' bonus TrueHD track should be real 24bit.

p.s. soon i will do the same test for Vendetta TrueHD track.

I can't help with identification, but I do want to point out that this method is flawed: The output could be dither rather than true 24 bit, both are about equally hard to compress. (Not that I could ever tell the difference, even with my nice old amp.) Of course, it could just as easily be encoded dither as outputby the decoder, you just have to assume otherwise.

[Beastie Boy]

@Binnugsai, use EVOdemux on Feature_1.EVO with the option selected to automatically process the second EVO. This will give you a seperate file for each of the streams (video, audio, subtitles). Converting a DD+ file is the easiest, this is all I have done so far.

Using eAC3to (and the GUI if you prefer), load the relevent dd+ file as input, select and output directory and file name and click convert.
Ensure that the SOX and Aften files are all in the same directory as eAC3to.
Just a couple of things to be aware of, though they may now be fixed: eAC3toGUI loads ec3, wav & evo files, so you may want to change the file extension of your audio file. Also, I had an error message saying that my audio file couldn't be found, even though I loaded it from the GUI. I moved the file to the root of the drive and it worked OK, so I guess there is a problem with path names.

I have never done a TrueHD track, but from what I have read, you need to remux using EVOdemux and create an EVO with ONLY the audio in it. Also, you must have the Nero HD-DVD/Blu-Ray filter installed and registered.

Hope this helps. Cheers, Beastie.

[Binnugsai]

Thank you Beastie, nice. So i choose in output for the audio dd+, example kingkong.dd+ and change the extension after kingkong.eac3 or another one !.

Thanks

[Binnugsai]

In evodemux we have to choice the extension for the video, audio and subtitle. by default we have mpa for audio. That's why i ask you if i can use between .eac3 or another one to convert it after.

Thanks Bonnugsai

[Beastie Boy]

Demux, then right click the dd+ file and select Rename. Change extension to .ec3 and hit enter. Now load the file in eAC3toGUI. Type output file name (eg King_Kong.ac3), select output directory and click Convert.

Cheers, Beastie.

[Binnugsai]

Thank you so much Beastie Boy. I try and see that tonight.

See you soon. Binnugsai

[Binnugsai]

Hello, so i try to encode a dd+, Pitch.Black.HD-DVD.1080pEnglish 5.1.ec3 to test.ac3 with eac3toGui and i have that :

C:\DVD Outils HD\EAC3toGUI 0.90>eac3to.exe "C:\DVD Outils HD\EAC3toGUI 0.90\Pitc
h.Black.HD-DVD.1080pEnglish 5.1.ec3" "G:\test.ac3" -640 -nero
E-AC3, 5.1 channels, 1:51:54, 1536kbit/s, 48khz
Muxing eac3 file to raw. Please wait...
The file size of the raw file doesn't seem to fit.
The expected file size for 16 bit is 1,66 MB.
The expected file size for 24 bit is 2,49 MB.
The real file size is 0 Bytes.

C:\DVD Outils HD\EAC3toGUI 0.90>

Did i miss something, i force nero and i have Nero burning Premium in my hard disk. Did i install the bluray/Hddvd plugin in plus or not necessary !!! is there something wrong.

Thank you for your help
Binnugsai

[honai]

Does anyone have a neat command-line for converting a TruHD track (inside an EVO container) to 6 discrete mono WAVs (which will be fed into SurCode DTS) using eac3to?

[madshi]

Hello, so i try to encode a dd+, Pitch.Black.HD-DVD.1080pEnglish 5.1.ec3 to test.ac3 with eac3toGui and i have that :

C:\DVD Outils HD\EAC3toGUI 0.90>eac3to.exe "C:\DVD Outils HD\EAC3toGUI 0.90\Pitc
h.Black.HD-DVD.1080pEnglish 5.1.ec3" "G:\test.ac3" -640 -nero
E-AC3, 5.1 channels, 1:51:54, 1536kbit/s, 48khz
Muxing eac3 file to raw. Please wait...
The file size of the raw file doesn't seem to fit.
The expected file size for 16 bit is 1,66 MB.
The expected file size for 24 bit is 2,49 MB.
The real file size is 0 Bytes.

C:\DVD Outils HD\EAC3toGUI 0.90>

Did i miss something, i force nero and i have Nero burning Premium in my hard disk. Did i install the bluray/Hddvd plugin in plus or not necessary !!! is there something wrong.

Thank you for your help
Binnugsai
That seems to be the same problem other people are having with mono E-AC3 files. That's very strange. It seems that the Nero Audio Decoder sometimes doesn't output any data. I wish I could reproduce that on my PC! It works all fine for me...

[madshi]

Does anyone have a neat command-line for converting a TruHD track (inside an EVO container) to 6 discrete mono WAVs (which will be fed into SurCode DTS) using eac3to?
I guess I can add support for that to eac3to. May take a few days, though... BTW, is there a way to automate the SurCode DTS encoder?

[Beastie Boy]

@Binnugsai : I think I will have to leave it to the real experts to help with your problem. I was fortunate, it just worked.
Here's a couple of things to try:
1) Have you tried without forcing the Nero filter? Untick the Nero box.
2) Do you have the correct filters installed as per the first post in this thread?
3) Can you play the mpa file in graphEdit?

Cheers, Beastie.

[Binnugsai]

Beasty : Thanks a lot. I think if you install Nero Burning Premium you have inside Bluray and HDDVD plugin for DD+ and TruHd.
1 ) So about Untick the nero box ! where.....What do you mean.... I try yes without forcng the nero filter.
2 ) filters, i think yes. you can say me witch one!
3) I don't try to play the mpa audio with graphedit.
Perhaps that is my problem, i demux with evodemux in audio.dd+ and change as you said me in audio.ec3.

Thanks

[Beastie Boy]

1) On eAC3toGUI is a panel with check boxes labeled Nero and OrbitLee. Make sure neither are ticked.
2) The required filters are in the original post by madshi.
Nero Premium does not include the HD-DVD/Blu-ray plugin, as far as I know.

[Beastie Boy]

The filters are not freeware, so requesting them is a breech of the forum rules. Consider editing your post to avoid a strike, and I would suggest removing your e-mail address.
Sorry.

[Binnugsai]

Sorry, i will see for the filter. thank you for all.

[homerpez]

Oh, boy.. a new adventure...

I've figured out DD+ from HD-DVD (Orbitlee's filter + Sonic Audio 4.2 and it works), and I've figured out 5.1 losless PCM from Blu-Ray (Tranzcode + Wavwizard + Behappy).

Now I've finally come across a Dolby "True HD" track on Ghost Rider (Blu-Ray).

Orbitlee's filter chokes on it, and Tranzcode GUI says it's not a 6-channel track. So... I'm assuming that Eac3to GUI is the tool I might need. It looks like this movie actually uses the 7.1 channels.

I seem to be a bit confused to start with... can Eac3to GUI handle this one? I'm experimenting via trial-and-error as we speak, but hopefully someone can explain the right way (if this has been figured out).

So far, my experiments have resulted in a horrible-sounding ac3, which sounds like garbled digital static.

Aiming for the backup result to have simple 5.1 Channel AC3 audio... thanks in advance if anyone can give me a crash course in how to do this properly.

[madshi]

@homerpez, currently eac3to only supports TrueHD inside of EVO containers. But if you can send me a little chunk of your Blu-Ray M2TS file (20MB should be enough), I may be able to add support for Blu-Ray TrueHD, too. Much depends on how the Nero filters behave. For sure you'll need the Nero HD-DVD/Blu-Ray plugin. You can try right now if things work by renaming GraphEdit.exe to "Recode.exe" and then using "File Source Async -> Nero Splitter -> Nero Audio Decoder -> Dump". Does that work?

[homerpez]

@homerpez, currently eac3to only supports TrueHD inside of EVO containers. But if you can send me a little chunk of your Blu-Ray M2TS file (20MB should be enough), I may be able to add support for Blu-Ray TrueHD, too. Much depends on how the Nero filters behave. For sure you'll need the Nero HD-DVD/Blu-Ray plugin. You can try right now if things work by renaming GraphEdit.exe to "Recode.exe" and then using "File Source Async -> Nero Splitter -> Nero Audio Decoder -> Dump". Does that work?

Whoops... I'm sorry, I failed to read that part, or for that matter, realize there was any difference between HD-DVD and Blu-Ray versions of the same audio type...

I'd be happy to send you a sample, the question is, how to split up an m2ts file? (I'm quite the newbie)

I think you can somehow do this with a hex editor or something, but I might need a tutorial on how to get 20MB out of it, and then to you.

I guess I should put this backup on the back burner until this support is developed... and keep the original disc away from the kids until then :p

BTW... the method you suggested to test this from the demuxed True HD file (Filesource -> Nero Splitter) did not work... created some wild error and could not connect at step #1, to the splitter. I probably should just have you experiement with the .m2ts file, since you know what you're doing, and I don't. :thanks:

[honai]

I guess I can add support for that to eac3to. May take a few days, though... BTW, is there a way to automate the SurCode DTS encoder?

That would be great, thanks!

I don't know of any way to automate the encoder other than using generic third-party tools that script mouse events.

[ACrowley]

yeah Madshi , a Option to output 6 mono waves with the Output Order L R C LFE SL SR would be very nice

Usually is use Wavewizard to get mono channels form the wave decoded by eac3to (Nero)

@otaku975

Why do you want to reencode a DTSHD to DTS ? Every DTS HD Tracks has encoded a DTS "core" inside ,which is a "standard" DTS 1536kbps 5.1. playable withg all DTS Decoders
You can simply extract this core with dtscore.exe

No reencode