Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion.

Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules.

 

Go Back   Doom9's Forum > General > Audio encoding

Closed Thread
 
Thread Tools Search this Thread Display Modes
Old 27th December 2015, 13:56   #13761  |  Link
Yoshi
Registered User
 
Join Date: Apr 2014
Posts: 32
Dolby TrueHD Atmos issue

I might have found another bug - this time in the decoder used for Dolby TrueHD tracks in conjunction with Dolby Atmos encodings (eac3to 3.31).

In the case of Léon (US Supreme Cinema Series), the decoding of the Atmos track leads to weird cracklings at certain positions whereas the decoding of the embedded AC3 track is flawless.

Yoshi is offline  
Old 27th December 2015, 15:49   #13762  |  Link
Boulder
Pig on the wing
 
Boulder's Avatar
 
Join Date: Mar 2002
Location: Finland
Posts: 5,718
Have you tested decoding with a recent ffmpeg build? If it also causes the issue, you need to report the bug to the ffmpeg devs.
__________________
And if the band you're in starts playing different tunes
I'll see you on the dark side of the Moon...
Boulder is offline  
Old 27th December 2015, 23:40   #13763  |  Link
Yoshi
Registered User
 
Join Date: Apr 2014
Posts: 32
Have now. Using ffmpeg-20151227-git-baf4c48-win64-static to decode the thd file demuxed by eac3to, it's the same result.



Guess I better let them know.

Still confused - wasn't the TrueHD decoder to be bug-free so far at least?
Yoshi is offline  
Old 28th December 2015, 10:39   #13764  |  Link
nevcairiel
Registered Developer
 
Join Date: Mar 2010
Location: Hamburg/Germany
Posts: 10,344
There is no guarantee your TrueHD stream isn't just broken.
__________________
LAV Filters - open source ffmpeg based media splitter and decoders
nevcairiel is offline  
Old 28th December 2015, 21:28   #13765  |  Link
Yoshi
Registered User
 
Join Date: Apr 2014
Posts: 32
@nevcairiel

Which is why I wanted to take some precaution and used the word "might" wisely. Guilty as charged that the source is not really an official one but to double-check this, I used a remuxed one and a nominal 1:1 copy which has to be put together by eac3to.

Just for my defense - NO, I won't buy that movie for the fourth time (two DVDs and the LaserDisc is enough now), just because they finally get a half-way decent video master they could have come up with in the first place (so annoying).

Besides that, I don't consider this Atmos remix to be that great due to its "tinny" acoustics and of course, the 4K remastered Blu-ray doesn't contain the 5.1 mix.

FFmpeg states something about "lossless" check failed, but only with one of the two sources as far as I remember - the crackled PCM result is similar though.

Maybe someone can grab the original and double- (or rather: triple-) check.
Yoshi is offline  
Old 30th December 2015, 02:48   #13766  |  Link
Yoshi
Registered User
 
Join Date: Apr 2014
Posts: 32
Another oddity:

When decoding this DTS-HD MA track, there is some kind of contradiction in the log output of eac3to: on one hand, libDcaDec is allegedly outputting 16 bit data, however the resulting PCM file uses all the 24 bits according to eac3to at the same time.

In any case, the output at least matches the ArcSoft decoder's. Yet again, I don't know if it's really lossless or not.

Yoshi is offline  
Old 30th December 2015, 12:33   #13767  |  Link
ACrowley
Registered User
 
Join Date: Apr 2006
Posts: 1,008
Have anbody here succes with the Sonic Audio Decoder on W7/8.1 x64 ?

I tried v4.2 and 4.3 from the Decoder Pack, also v5 from, Sonic Cinevision.
The Sonic Cinemaster® Audio Decoder 4.3 appears in Directshow registered Filters and i can use it MPC HC etc

But eac3to gives me :

The Sonic Audio Decoder (3.31.0.0) doesn't seem to be installed


I placed the Files in Systems Folder etc etc, nothing helps.
Its a clean System without any Decoder Packages etc.

Is there trick on W8 x64 ?

Thanks
ACrowley is offline  
Old 5th January 2016, 21:34   #13768  |  Link
dts350z
Registered User
 
Join Date: May 2013
Posts: 6
resample to 32bit float

I would like to resample to 32 bit float, and thus avoid any 2nd pass. Can eac3to do that?

ideally output would be in a format that also supports files larger than 4GB, such as w64 or rf64.
dts350z is offline  
Old 6th January 2016, 07:05   #13769  |  Link
dts350z
Registered User
 
Join Date: May 2013
Posts: 6
-r8brain

Anybody notice that the actual usage output of the current version doesn't show the:

-r8brain use r8brain resampler instead of SSRC

option, unlike the output shown in the post at the top of this thread?

Version number match
dts350z is offline  
Old 6th January 2016, 13:06   #13770  |  Link
Nico8583
Registered User
 
Join Date: Jan 2010
Location: France
Posts: 851
Hi I would like to know : what is the best way to convert DTS 5.1 to AC3 5.1 ? Only "eac3to.exe source.dts destination.ac3 -640" ? Or "-normalize" or "-dontPatchDts" are need ? Thank you !
Nico8583 is offline  
Old 6th January 2016, 13:23   #13771  |  Link
Overdrive80
Anime addict
 
Overdrive80's Avatar
 
Join Date: Feb 2009
Location: Spain
Posts: 673
@madshi

Instead of using r8brain, would not it be better use ffmpeg (soxr)?

http://src.infinitewave.ca/
__________________
Intel i7-6700K + Noctua NH-D15 + Z170A XPower G. Titanium + Kingston HyperX Savage DDR4 2x8GB + Radeon RX580 8GB DDR5 + ADATA SX8200 Pro 1 TB + Antec EDG750 80 Plus Gold Mod + Corsair 780T Graphite
Overdrive80 is offline  
Old 6th January 2016, 16:26   #13772  |  Link
dts350z
Registered User
 
Join Date: May 2013
Posts: 6
Quote:
Originally Posted by Nico8583 View Post
Hi I would like to know : what is the best way to convert DTS 5.1 to AC3 5.1 ? Only "eac3to.exe source.dts destination.ac3 -640" ? Or "-normalize" or "-dontPatchDts" are need ? Thank you !

That's a lossy to lossy conversion. Why would you want to do that?

IMHO dts is a better sounding format anyway.
dts350z is offline  
Old 6th January 2016, 16:52   #13773  |  Link
mariner
Registered User
 
Join Date: Nov 2005
Posts: 583
Quote:
Originally Posted by tebasuna51 View Post
@kukushka
Maybe we can open a new thread to speak about AAC and muxer/demuxer's (Mp4Muxer/Mp4Box/MkvToolnix/tsMuxeR), but the relevant questions for this thread are clear:

- The directshow Nero 7 decoder, used by eac3to to decode .aac, cut the first 1024 samples (21,333 ms in 48 KHz) in 2.0 and is broken for 5.1.
Take in mind the problem.

- The encoder NeroAacEnc.exe, user by eac3to to encode .m4a, put the correct delay to compensate and can be decoded without problems with NeroAacDec, Qaac or ffmpeg. No problem with it.

- eac3to works fine extracting AAC from MKV/TS/M2TS containers, I obtain the same aac than I muxed previously.
Then, to avoid problems, you can use eac3to to extract and Qaac or ffmpeg to decode (or LWLibavAudioSource inside AviSynth).
Greetings tebasuna51.

Thanks for the post. I'm using eac3to to convert aac to ac3, and struggling to understanding the delay issue. Appreciate if you could assist.

1. If direct from aac to ac3, 5ms silence seems to be inserted. This appears to agree with ffmpeg. Audacity is used for comparing the two.

2. If going from ts/mkv to ac3, Nero Audio Decoder would be used. This would seem to remove 42ms if the audio begins with silence. So the resulting ac3 would be either -37ms shorter or +5ms longer, depending on the initial content.

Does this behavior look correct to you? Is there a way not to use Nero for step 2?

Many thanks and best regards.
mariner is offline  
Old 6th January 2016, 17:32   #13774  |  Link
tebasuna51
Moderator
 
tebasuna51's Avatar
 
Join Date: Feb 2005
Location: Spain
Posts: 6,890
Quote:
Originally Posted by Nico8583 View Post
Hi I would like to know : what is the best way to convert DTS 5.1 to AC3 5.1 ? Only "eac3to.exe source.dts destination.ac3 -640" ?
Is not the best.
Is correct using eac3to (dcadec) and Aften encoder.

But, talking about free soft, now is better ffmpeg:
ffmpeg -acodec libdcadec -i source.dts -acodec ac3 -center_mixlev 0.707 -surround_mixlev 0.707 -ab 640k destination.ac3

Quote:
Or "-normalize" or "-dontPatchDts" are need?
Nope.

Quote:
Originally Posted by dts350z View Post
That's a lossy to lossy conversion. Why would you want to do that?
To save space and/or make compatible with some players, I supose.

Quote:
IMHO dts is a better sounding format anyway.
Better format? I don't think so.

Talking about size/quality AC3 is better and AAC much better.
Talking about compatibility AC3 is better.
__________________
BeHappy, AviSynth audio transcoder.
tebasuna51 is offline  
Old 6th January 2016, 18:01   #13775  |  Link
LigH
German doom9/Gleitz SuMo
 
LigH's Avatar
 
Join Date: Oct 2001
Location: Germany, rural Altmark
Posts: 6,753
The "core" dts audio format on DVD Video may be a low-loss format when using the bitrate close to LPCM 16-bit stereo (1536 kbps); but it is not lossless. And the bitrate close to LPCM 16-bit mono (768 kbps) is even audibly lossy. It may use less psycho-acoustic filtering than AC3. But that doesn't make it "better": dts tries to retain even probably inaudible frequencies only "audiophiles" believe to recognize (but couldn't prove), so it possibly lacks in accuracy for audible frequencies instead.

Its purpose was to compress multi-channel audio to bitrates close to comparable usual bitrates of 16-bit LPCM in mono or stereo, and it is still a format based on 16-bit integer samples. Dolby Digital (AC3) instead works with floating point parameters, therefore it may have a better dynamic range if it compresses original 24-bit integer or even floating point samples (important especially for almost slient scenes, like in classical music), keeping the audio audible despite a loss of frequency parts excluded by psycho-acoustic filters, because you will probably not recognize them anyway (except on a rather psychological level).
__________________

New German Gleitz board
MediaFire: x264 | x265 | VPx | AOM | Xvid
LigH is online now  
Old 6th January 2016, 18:19   #13776  |  Link
tebasuna51
Moderator
 
tebasuna51's Avatar
 
Join Date: Feb 2005
Location: Spain
Posts: 6,890
Quote:
Originally Posted by mariner View Post
1. If direct from aac to ac3, 5ms silence seems to be inserted. This appears to agree with ffmpeg. Audacity is used for comparing the two.
This is the default behaviour for ALL ac3 encoders.
BTW, with Aften, you can avoid the insertion of 5 ms of silence with, for instance:

eac3to INPUT stdout.wav | Aften -b 192 -pad 0 -readtoeof 1 - OUTPUT.ac3

Quote:
2. If going from ts/mkv to ac3, Nero Audio Decoder would be used. This would seem to remove 42ms if the audio begins with silence. So the resulting ac3 would be either -37ms shorter or +5ms longer, depending on the initial content.
Seems you have misunderstand my post. I say:
"The directshow Nero 7 decoder, used by eac3to to decode .aac, cut the first 1024 samples (21,333 ms in 48 KHz) in 2.0"

Then, always cut 21 ms, and (without -pad 0) finish with 16 ms shorter.

The Nero Audio Decoder (NeroAacDec.exe) is not used by eac3to at all.

Quote:
Is there a way not to use Nero for step 2?
eac3to only can decode aac with the directshow Nero 7 decoder.
__________________
BeHappy, AviSynth audio transcoder.
tebasuna51 is offline  
Old 6th January 2016, 21:09   #13777  |  Link
Nico8583
Registered User
 
Join Date: Jan 2010
Location: France
Posts: 851
Quote:
Originally Posted by dts350z View Post
That's a lossy to lossy conversion. Why would you want to do that?

IMHO dts is a better sounding format anyway.
I don't know if I'll convert DTS to AC3 but I feel sound variation are more important on a DTS track than AC3 track. When actors are speaking, the sound is low but when there is action the sound is high so I'm playing with my remote several times but perhaps it's only a feeling or it's not related to DTS.
And also for a compatibility a little bit

Quote:
Originally Posted by tebasuna51 View Post
Is not the best.
Is correct using eac3to (dcadec) and Aften encoder.

But, talking about free soft, now is better ffmpeg:
ffmpeg -acodec libdcadec -i source.dts -acodec ac3 -center_mixlev 0.707 -surround_mixlev 0.707 -ab 640k destination.ac3
So ArcSoft is useless now to decode DTS ? Do you have a conversion commandline sample with dcadec and Aften ?

Quote:
Originally Posted by tebasuna51
Nope.
Ok

Quote:
Originally Posted by tebasuna51
To save space and/or make compatible with some players, I supose.
Yes for compatibility and also for the other reason (see before)


Quote:
Originally Posted by tebasuna51
Better format? I don't think so.

Talking about size/quality AC3 is better and AAC much better.
Talking about compatibility AC3 is better.
I don't have test AAC so I don't know about the compatibility but I could test it.

Last question, what about remapping channel ? Is there a risk to have an issue with wrong channel remapping if I convert DTS to AC3 ?

Thank you !
Nico8583 is offline  
Old 6th January 2016, 21:30   #13778  |  Link
AlexKane
Registered User
 
AlexKane's Avatar
 
Join Date: Jul 2014
Posts: 20
Quote:
Originally Posted by Nico8583 View Post
I don't know if I'll convert DTS to AC3 but I feel sound variation are more important on a DTS track than AC3 track. When actors are speaking, the sound is low but when there is action the sound is high so I'm playing with my remote several times but perhaps it's only a feeling or it's not related to DTS.
And also for a compatibility a little bit
The dynamic range of the source material (the sound variation you describe above) is intentional. Films are mixed for theaters, not small apartments and tiny TV speakers. Also, i believe DTS, AAC, Vorbis, Opus, etc maintain the dynamic range of the source material, since they don't apply dynamic range compression (DRC).

In the case of AC3, you can use ffmpeg to generate level scaling metadata, but the results are not ideal since obvious pumping artifacts can be introduced during playback.

Last edited by AlexKane; 6th January 2016 at 21:41.
AlexKane is offline  
Old 6th January 2016, 23:00   #13779  |  Link
tebasuna51
Moderator
 
tebasuna51's Avatar
 
Join Date: Feb 2005
Location: Spain
Posts: 6,890
Quote:
Originally Posted by Nico8583 View Post
...So ArcSoft is useless now to decode DTS ? Do you have a conversion commandline sample with dcadec and Aften ?
Now dcadec is the default decoder for DTS, your sintax is enough:
"eac3to.exe source.dts destination.ac3" (640 Kb/s is the default also)

Quote:
Last question, what about remapping channel ? Is there a risk to have an issue with wrong channel remapping if I convert DTS to AC3 ?
No problem.
__________________
BeHappy, AviSynth audio transcoder.
tebasuna51 is offline  
Old 6th January 2016, 23:27   #13780  |  Link
Nico8583
Registered User
 
Join Date: Jan 2010
Location: France
Posts: 851
Quote:
Originally Posted by AlexKane View Post
The dynamic range of the source material (the sound variation you describe above) is intentional. Films are mixed for theaters, not small apartments and tiny TV speakers. Also, i believe DTS, AAC, Vorbis, Opus, etc maintain the dynamic range of the source material, since they don't apply dynamic range compression (DRC).

In the case of AC3, you can use ffmpeg to generate level scaling metadata, but the results are not ideal since obvious pumping artifacts can be introduced during playback.
Ok thanks, I believed DTS was the only one to use dynamic range...

Quote:
Originally Posted by tebasuna51 View Post
Now dcadec is the default decoder for DTS, your sintax is enough:
"eac3to.exe source.dts destination.ac3" (640 Kb/s is the default also
Ok thanks but I don't understand why do you say "talking about free soft" for ffmpeg, dcadec and Aften are commercial softwares ?
Nico8583 is offline  
Closed Thread

Tags
eac3to

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 20:24.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.