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 > Capturing and Editing Video > New and alternative a/v containers

Closed Thread
 
Thread Tools Search this Thread Display Modes
Old 26th March 2015, 18:24   #3081  |  Link
filler56789
SuperVirus
 
filler56789's Avatar
 
Join Date: Jun 2012
Location: Antarctic Japan
Posts: 1,351
The next step is, the Linux folks convince foo86 that LBR-decoding *is* important for libdcadec ;-)

https://github.com/foo86/dcadec/issu...mment-85062370

Last edited by filler56789; 26th March 2015 at 23:07.
filler56789 is offline  
Old 26th March 2015, 19:25   #3082  |  Link
hubblec4
Matroska find' ich toll
 
Join Date: Apr 2008
Posts: 1,370
Quote:
Originally Posted by Mosu View Post
In MPEG transport streams?
yes, m2ts.

Thanks a lot for your continuous work.
hubblec4 is offline  
Old 27th March 2015, 21:32   #3083  |  Link
Mosu
MKVToolNix author
 
Mosu's Avatar
 
Join Date: Sep 2002
Location: Braunschweig, Germany
Posts: 4,278
MKVToolNix v7.8.0 released

Hey,

I've released MKVToolNix v7.8.0. Tons of bug fixes, among them an important one that fixes a corruption when dealing with HEVC tracks without aspect ratio information. New features include support for DTS Exrepss and a Swedish translation (work in progress).

Changes for package maintainers: Boost's Math library is now required.

Here are the usual links: the MKVToolNix home page, the Windows installer and portable version and the source code.

All binaries that I provide have been uploaded already.

Here's the full ChangeLog since the previous release:

Quote:
2015-03-27 Moritz Bunkus <moritz@bunkus.org>
* Released v7.8.0.
* build system: Boost's »Math« library is now required.
* mkvmerge: bug fix: HEVC tracks which did not have an aspect ratio present in their sequence parameter set were copied incorrectly; the resulting sequence parameter set was invalid. Fixes #1081.
* mkvextract: bug fix: When extracting HEVC tracks mkvextract will use the same start code lengths that x265 uses (four bytes 0x00000001 for the first and video/picture/sequence parameter set NALs and three bytes 0x000001 for all others).

2015-03-26 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: enhancement: File type identification will output a more detailed description of the DTS type for DTS audio tracks (DTS-HD Master Audio, DTS-HD High Resolution, DTS Express or just plain DTS). Implements #1109.
* mkvmerge: new feature: Implemented support for DTS Express.

2015-03-25 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: bug fix: The number of channels in DTS tracks with more than six channels is now recognized correctly by parsing the DTS HD extensions, too. Fixes #1139.

2015-03-24 Moritz Bunkus <moritz@bunkus.org>
* all: new feature: added a Swedish translation of the programs by Kristoffer Grundström (see AUTHORS).

2015-03-24 Mats Peterson <matsp888@yahoo.com>
* mkvmerge: bug fix: Fixed handling of the BITIMAPINFOHEADER extra data size handling during merging and extraction for codecs like HuffYUV.

2015-03-23 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: bug fix: When appending unframed HEVC/h.265 tracks and setting the default duration the second and all following source parts will use the same default duration as set for the first part. Fixes #1147.

2015-03-17 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: bug fix: enabled the use of tags in WebM files. Tagging elements not supported by the WebM specs are removed. Fixes #1143.

2015-03-13 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: bug fix: fixed detection of audio tracks in QuickTime files whose FourCC code is unknown to mkvmerge.

2015-03-11 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: bug fix: fixed detection of video tracks in QuickTime files whose FourCC code is unknown to mkvmerge.

2015-03-10 Moritz Bunkus <moritz@bunkus.org>
* mkvextract: bug fix: Fixed VobSub file naming when mkvextract is built against Boost::Filesystem older than 1.50.0. Fixes #1140.

2015-03-09 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: bug fix: fixed detection of Cinepak video tracks in QuickTime files.
* mkvmerge: bug fix: fixed detection of PCM audio tracks in QuickTime files using the »raw « FourCC.
* mkvmerge: bug fix: fixed detection of tracks in Flash Video files for which the headers do not signal a track.
* mkvmerge: bug fix: fixed a segfault in the Flash Video file format reader.

2015-03-03 Moritz Bunkus <moritz@bunkus.org>
* mmg: The change making the window wider by default has been reverted.
* mkvmerge: bug fix: Fixed file type detection for MP3 files with big ID3 tags at the start of the file (e.g. if they contain cover images).
* mkvinfo (Qt interface): enhancement: implemented support for opening files via drag & drop.

2015-03-01 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: enhancement: added an option (--engage no_delay_for_garbage_in_avi) for disabling deriving a delay from garbage in audio tracks in AVI files. Requested in #1137.
* mkvmerge: bug fix: The formula used for calculating the audio delay for garbage data at the start of tracks in AVI files has been fixed. Fixes #1137.
Have fun.
__________________
Latest MKVToolNix is v83.0

If I ever ask you to upload something, please use my file server.
Mosu is offline  
Old 27th March 2015, 21:57   #3084  |  Link
filler56789
SuperVirus
 
filler56789's Avatar
 
Join Date: Jun 2012
Location: Antarctic Japan
Posts: 1,351
^ Many for the new version.
filler56789 is offline  
Old 28th March 2015, 13:49   #3085  |  Link
filler56789
SuperVirus
 
filler56789's Avatar
 
Join Date: Jun 2012
Location: Antarctic Japan
Posts: 1,351
Testing the latest MKVmerge and the latest LAV Filters...

Re: DTS Express — muxing went fine, and playback went fine

Re: "Pure Lossless" DTS —
mmg says "everything went fine", BUT the original 35.9MB .dts(hd) file became a 5KB .mka file

LAV Splitter and LAV Audio do play the "pure lossless" source, BUT at the wrong speed

Last edited by filler56789; 28th March 2015 at 13:50. Reason: grammar
filler56789 is offline  
Old 28th March 2015, 14:00   #3086  |  Link
Mosu
MKVToolNix author
 
Mosu's Avatar
 
Join Date: Sep 2002
Location: Braunschweig, Germany
Posts: 4,278
I'll need that file you're trying to mux in order to debug it. Please upload it to my FTP server. Thanks.
__________________
Latest MKVToolNix is v83.0

If I ever ask you to upload something, please use my file server.
Mosu is offline  
Old 28th March 2015, 14:53   #3087  |  Link
filler56789
SuperVirus
 
filler56789's Avatar
 
Join Date: Jun 2012
Location: Antarctic Japan
Posts: 1,351
^ Done
(archive's name = "MARINA_LIMA~Criancza.zip").
filler56789 is offline  
Old 28th March 2015, 14:59   #3088  |  Link
Mosu
MKVToolNix author
 
Mosu's Avatar
 
Join Date: Sep 2002
Location: Braunschweig, Germany
Posts: 4,278
That's an interesting file, haven't seen such a one yet. ffmpeg generally knows about the format (see here), but cannot decode it either.

Meaning: this kind of DTS-HD container is currently not supported by mkvmerge. Maybe I'll add support for it, but don't count on it.
__________________
Latest MKVToolNix is v83.0

If I ever ask you to upload something, please use my file server.
Mosu is offline  
Old 28th March 2015, 15:09   #3089  |  Link
SeeMoreDigital
Life's clearer in 4K UHD
 
SeeMoreDigital's Avatar
 
Join Date: Jun 2003
Location: Notts, UK
Posts: 12,219
Quote:
Originally Posted by Mosu View Post
I'll need that file you're trying to mux in order to debug it. Please upload it to my FTP server. Thanks.
Quote:
Originally Posted by filler56789 View Post
^ Done
(archive's name = "MARINA_LIMA~Criancza.zip").
May I have that sample too please for my collection?
__________________
| I've been testing hardware media playback devices and software A/V encoders and decoders since 2001 | My Network Layout & A/V Gear |
SeeMoreDigital is online now  
Old 28th March 2015, 15:34   #3090  |  Link
filler56789
SuperVirus
 
filler56789's Avatar
 
Join Date: Jun 2012
Location: Antarctic Japan
Posts: 1,351
@SMD:

Code:
http://www.mediafire.com/download/g7dzip5a350e75b/MARINA_LIMA~Criancza.zip
filler56789 is offline  
Old 28th March 2015, 15:42   #3091  |  Link
sneaker_ger
Registered User
 
Join Date: Dec 2002
Posts: 5,565
Quote:
Originally Posted by Mosu View Post
That's an interesting file, haven't seen such a one yet. ffmpeg generally knows about the format (see here), but cannot decode it either.
The native ffmpeg decoder can't, but ffmpeg can decode it when compiled with (lib)dcadec.
sneaker_ger is offline  
Old 28th March 2015, 15:54   #3092  |  Link
SeeMoreDigital
Life's clearer in 4K UHD
 
SeeMoreDigital's Avatar
 
Join Date: Jun 2003
Location: Notts, UK
Posts: 12,219
Quote:
Originally Posted by filler56789 View Post
@SMD:

Code:
http://www.mediafire.com/download/g7dzip5a350e75b/MARINA_LIMA~Criancza.zip
Many thanks...

Well I managed to mux the 'core-less' DTS-HD MA stream into the .m2ts container, in the hope that it might trick MKVmerge into muxing it into .mka, but it did not work.

I also tried playing the 'core-less' DTS-HD MA stream using my Oppo BDP-103 player, but it did not work either...
__________________
| I've been testing hardware media playback devices and software A/V encoders and decoders since 2001 | My Network Layout & A/V Gear |
SeeMoreDigital is online now  
Old 28th March 2015, 15:58   #3093  |  Link
Mosu
MKVToolNix author
 
Mosu's Avatar
 
Join Date: Sep 2002
Location: Braunschweig, Germany
Posts: 4,278
mkvmerge has two problems with that file The first is that it uses that strange DTSHDHDR file format. This is not really a big problem as mkvmerge can skip the first 112 bytes and treat the rest as a pure DTS elementary stream. However, even if there weren't a second problem one result of this file format is that mkvmerge would emit a warning about garbage at the beginning of the file. So this has to be fixed, too.

More important than that is that mkvmerge currently only supports core-less DTS files which use LBR extension substreams (that's what DTS Express is: a core-less extension substream consisting solely of LBR extensions). The file in question uses XLL extensions. These are currently not supported by mkvmerge if they're not accompanied by cores.
__________________
Latest MKVToolNix is v83.0

If I ever ask you to upload something, please use my file server.
Mosu is offline  
Old 28th March 2015, 17:57   #3094  |  Link
kasper93
MPC-HC Developer
 
Join Date: May 2010
Location: Poland
Posts: 586
Quote:
Originally Posted by Mosu View Post
That's an interesting file, haven't seen such a one yet. ffmpeg generally knows about the format (see here), but cannot decode it either.

Meaning: this kind of DTS-HD container is currently not supported by mkvmerge. Maybe I'll add support for it, but don't count on it.
You need to build ffmpeg with libdcadec which is superior to ffmpeg's dca decoder in every way. Simply do ./configure --disable-decoder=dca --enable-libdcadec

Last edited by kasper93; 28th March 2015 at 18:50.
kasper93 is offline  
Old 28th March 2015, 20:26   #3095  |  Link
Mosu
MKVToolNix author
 
Mosu's Avatar
 
Join Date: Sep 2002
Location: Braunschweig, Germany
Posts: 4,278
Hmm, I've looked through the DTS spec and am somewhat unclear how to retrieve the sampling frequency from a core-less XLL extension sub-stream… The general EXSS headers do have a field called »nuMaxSampleRate« in the asset descriptor, but is that the actual sampling frequency? LBR extension sub-streams have an additional field »nLBRSampleRateCode« in their own headers (LBR audio asset), but there's no such field in the XLL sub-stream headers.

The libdcadec sources aren't helping either.
__________________
Latest MKVToolNix is v83.0

If I ever ask you to upload something, please use my file server.
Mosu is offline  
Old 28th March 2015, 21:04   #3096  |  Link
filler56789
SuperVirus
 
filler56789's Avatar
 
Join Date: Jun 2012
Location: Antarctic Japan
Posts: 1,351
Assuming I have not misunderstood everything...

Quote:
Originally Posted by ETSI TS 102 114 V1.3.1 (2011-08), page 119

sFreqIndex (original sampling frequency)

This is the sample rate of audio in a channel set. It corresponds to the resulting sampling frequency after interpolation by the nFsInterpolate factor, on the encode side. The 4-bit sFreqIndex field is interpreted as described in Table 8-5. Remember that certain extra fields are encountered in the data stream once the sample rate is greater than 96 kHz.
filler56789 is offline  
Old 28th March 2015, 21:36   #3097  |  Link
Mosu
MKVToolNix author
 
Mosu's Avatar
 
Join Date: Sep 2002
Location: Braunschweig, Germany
Posts: 4,278
That's the sampling frequency for one channel set, and I'm not so sure how to interpret that value.

Your FLAC file has a sampling frequency of 48000 Hz. Your XLL DTS file has sFreqIndex == 5 (meaning 22050 Hz) with a sFsInterpolate == 4. These just don't add up.

For some packets sFreqIndex is even == 8 (indicating 176400 Hz). Yikes?

At least nuMaxSampleRate from the EXSS header audio asset stays constant at 48000 Hz…
__________________
Latest MKVToolNix is v83.0

If I ever ask you to upload something, please use my file server.
Mosu is offline  
Old 28th March 2015, 22:34   #3098  |  Link
filler56789
SuperVirus
 
filler56789's Avatar
 
Join Date: Jun 2012
Location: Antarctic Japan
Posts: 1,351
My bad, it seems you are right.
filler56789 is offline  
Old 29th March 2015, 08:16   #3099  |  Link
Mosu
MKVToolNix author
 
Mosu's Avatar
 
Join Date: Sep 2002
Location: Braunschweig, Germany
Posts: 4,278
I've created pre-build 682 which can read your file just fine. It still relies on that »maximum sampling frequency« field, though, and like I've said I'm not certain at all that this is the right thing to do.
__________________
Latest MKVToolNix is v83.0

If I ever ask you to upload something, please use my file server.
Mosu is offline  
Old 29th March 2015, 12:13   #3100  |  Link
filler56789
SuperVirus
 
filler56789's Avatar
 
Join Date: Jun 2012
Location: Antarctic Japan
Posts: 1,351
^ Now I understand you; I had forgotten that almost two years ago, Kostya said to me:

Quote:
Originally Posted by codecs.multimedia.cx/?p=313
...that specification is known to be incomplete and sometimes wrong.
filler56789 is offline  
Closed Thread

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 09:16.


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