mkvtoolnix (Matroska toolkit): new release

[Mosu]

(current version: 0.9.7, direct link to Windows binaries: http://www.bunkus.org/videotools/mkv...lnix-0.9.7.rar, if you're upgrading from prior 0.8.2 you need a new runtime package

ChangeLog from 0.9.6 to 0.9.7:
2004-10-24 Moritz Bunkus <moritz@bunkus.org>
* Released v0.9.7.
* mkvmerge: bug fix: The handling of external timecode files was still not correct but should be OK now.

2004-10-17 Moritz Bunkus <moritz@bunkus.org>
* mmg: Added an error message if the user selects 'mmg' as the 'mkvmerge executable' because that would lead to an infinite number of 'mmg's being spawned.

2004-10-16 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: bug fix: If LFE is on for DTS then the number of channels is one more than what the DTS frame header says.
* mkvmerge: bug fix: Timecodes for Vorbis were wrong on rare occasions (when reading laced Vorbis from a Matroska file and changing the lacing, e.g. when splitting for the second and all following files).

2004-10-10 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge/mkvinfo/mkvextract: bug fix: The chapter and tag element tables were not always intialized correctly depending on the compiler and the optimization flags used.
* mkvmerge: bug fix: The OGM reader was broken if at least one track was not to be copied from the file (happened between 0.9.5 and 0.9.6).
* mmg: bug fix: After loading saved mmg settings the track input box listed the tracks always coming from the last input file and not from the one they really came from.

2004-10-09 Moritz Bunkus <moritz@bunkus.org>
* mmg: enhancement: Made mmg's main window properly resizable.
* mkvmerge: Rewrote the code for the external timecode files. This also fixes bug 99: The durations for the individual tracks were not correct for those tracks for which --timecodes was used.

2004-10-08 Moritz Bunkus <moritz@bunkus.org>
* mmg: bug fix: Crash when saving chapters from the chapter editor. Same as the mkvinfo issue below but on all OS.
* mkvinfo: bug fix: The chapter and tag element tables were not initialized on Windows resulting in a crash when one of those elements was encountered.)

Heya.

(Some might call it advertizing as I already have such a thread, but it is rather old now and began with version 0.4.4. I think that more users might profit from this new thread.)

Here it is, the newest Incarnation of Evil, uh, I meant mktoolnix release. A lot of bug fixes, a couple of new features. The most prominent new feature is probably... *drummroll* support for zlib compressed VobSubs. Yes, Gabest is working hard on supporting them in his demuxer, so they should become a reality for all of you very soon.

The URL, as always: http://www.bunkus.org/videotools/mkvtoolnix/ and the Windows binaries:
http://www.bunkus.org/videotools/mkv...lnix-0.7.5.rar

Note: Current versions of libebml and libmatroska introduced new lacing schemes. This means that files will be smaller, but you also need a current version of Gabest's Matroska splitter (at least version 1.0.1.8). You can get it on http://sourceforge.net/projects/guliverkli/. You also MUST disable the internal Matroska splitter if you use MPC (Media Player Classic).

Here's the ChangeLog:
--------------------------------
2003-11-05 Moritz Bunkus <moritz@bunkus.org>
* Released v0.7.5.

* mkvmerge: bug fix: The VobSub .idx parser was dividing by 0 if a track only contained one entry.

* mkvmerge: Fixed the timecode reader code and made it a bit more flexible. Added more documentation for this feature along with an example file (examples/example-timecodes.txt).

* mmg: bug fix: When 'default track' is selected then all other tracks of the same type will have their 'default track' flag cleared.

* mkvmerge: SPU packets belonging to the same timecode are grouped together, and the duration is extracted directly from the SPU stream.

2003-10-29 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: The VobSubs are now stripped of the MPEG program stream, and only the SPU packets are kept.

2003-10-28 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge, mkvextract: The Matroska reader and the OGM reader (mkvmerge) as well as mkvextract will discard empty or 'cleaning only' subtitle packets as they are appear in OGMs in order to mark the end of an entry.

2003-10-27 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: new feature: Added the ability to read timecodes from text files which override the timecodes mkvmerge calculates normally.

2003-10-26 Moritz Bunkus <moritz@bunkus.org>
* mkvextract: bug fix: Add all the mandatory elements when extracting chapters so that the resulting XML can always be used directly with mkvmerge again without having to manually add e.g. ChapterLanguage.

* mkvmerge: bug fix; Handle audio tracks from Matroska files with the CodecID A_MS/ACM correctly.

2003-10-25 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: Changes to use libmatroska's new lacing code.

2003-10-23 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: bug fix: The VobSub .idx parser was mis-calculating the subtitle entry frame sizes.

* mkvmerge: bug fix: The Vorbis packetizer was miscalculating the number of samples to add/remove when using audio sync.

2003-10-22 Moritz Bunkus <moritz@bunkus.org>
* mmg: new feature: Added a new menu entry "File -> new" which will clear all the current muxing settings.

* mmg: Added support for VobSub subtitles including their compression options. Added the .m4a extension to the 'add file' dialog.

* mmg: bug fix: Made the input boxes for file names (tags and chapters) editable so that their contents can be deleted.

* mkvmerge: Adjusted the compression handling to the final content encoding specs.

2003-10-19 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: new feature: Implemented generic support for frame compression (mostly useful for VobSub subtitles but could also be used for others) and the complete framework for handling content encodings in the Matroska reader.

2003-10-18 Moritz Bunkus <moritz@bunkus.org>
* mkvinfo: new feature: Dump unknown elements recursively.

2003-10-15 Moritz Bunkus <moritz@bunkus.org>
* mkvmerge: bug fix: Made the SRT reader more tolerant regarding empty lines.
----------------------

Have fun!

[bilu]

Originally posted by Mosu
The most prominent new feature is probably... *drummroll* support for zlib compressed VobSubs. Yes, Gabest is working hard on supporting them in his demuxer, so they should become a reality for all of you very soon.

Wow... i'd love to see this stuff working

Bilu

[Liisachan]

xvid+vorbis=mkv with 0.7.2 plays with MPC6.4.6.7 (&6.4.6.6)
while the same xvid+vorbis=mkv with 0.7.5 doesn't play ;_;
("This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.")

I tested more than one patterns, and always MPC doesnt like MKV files muxed by 0.7.5. Is there anything that breaks backward compatibility?
sample files


PS
Only good combination so far is vid+ac3.
vid+aac = MPC crashes; vid+mp3 = wrong audio speed

Doesn't work with MatroskaSplitter 1.0.1.8 either.
However, the resulted files play if you "Disable lacing" in MKV 0.7.5

[Suiryc]

btw forget VirtualDubMod if you want to edit files using the new lacing method or compression/encryption methods.

The new lacing method is incompatible with old libs (and thus current VDM will crash with this new feature). You will have to wait for the next version to support it.

Also VDM doesn't copy all the extra information from the input file (that includes the full tag tree, the full chapter tree, and all elements that it doesn't know - including compression/encryption information about a track).

[gizmotech]

Yes,

This is to be expected. MPC's built in mkv splitter currently is incapable of understanding the new lacing scheme. Please disable it and ensure you have the latest mkv splitter from the http://sourceforge.net/projects/guliverkli/ guliverkli website.

Gizmo

[JagPanzer]

Mosu, tnx for a new release! I love Matroska.

[Liisachan]

ok fixed
i just had to re-register 1.0.1.8 *phew*....

but i m afraid that is not very nice...
hope wont cause much confusion.......

[Human_USB]

I get MV C++ errors in MPC, ZoomPlayer, and CoreCodec. I have the new splitter and it's just a MKV with a Dvix video and one OGG audio.

[gizmotech]

It actually causes quite a bit of confusion, but only w/ files created w/ this new release (or the earlier Timecode release I used in the VFR Proof of Concept).

I'm actually hoping Gabest might read this and push out an updated version of mpc with an updated mkv splitter in it. I figure that would solve quite a bit of confusion. Either that or remove it completely.

Gizmo.

[Human_USB]

But I have the MPC splitter off and I get the MV C++ errors.... and just not in MPC.

[Human_USB]

Originally posted by Liisachan
xvid+vorbis=mkv with 0.7.2 plays with MPC6.4.6.7 (&6.4.6.6)
while the same xvid+vorbis=mkv with 0.7.5 doesn't play ;_;
("This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.")

I tested more than one patterns, and always MPC doesnt like MKV files muxed by 0.7.5. Is there anything that breaks backward compatibility?
sample files


PS
Only good combination so far is vid+ac3.
vid+aac = MPC crashes; vid+mp3 = wrong audio speed

Doesn't work with MatroskaSplitter 1.0.1.8 either.
However, the resulted files play if you "Disable lacing" in MKV 0.7.5

I get this too.... but I have splitter 1.0.1.8 and I registered it. I think the new splitter needs an installer or something.

-Jason

[Atamido]

You MUST register the latest splitter. (1.0.1.8)
You MUST register the latest VSFilter. (2.29)
If you are using MPC, you MUST disable the built in Matroska splitter.

If you fail to do any of these things, the fill will NOT play.

[Human_USB]

I did not know I had to get the new VSplitter. I'm trying that now.


Thanks
Jason


*EDIT*

Using Matroska Splitter 20031015 (1.0.1.8), regsvr32 MatroskaSplitter.ax with it placed into System32 and using VSFilter 20030829 (2.29), regsvr32 VSFilter.dll with it placed into System32 and I turned off the MPC MKV splitter and I still get the error.



Help

[Liisachan]

@Human_USB
like i had posted up above, i re-registered 1.0.1.8 and my problem had already been fixed. I once installed 1.0.1.8 a while ago, which had also fixed my problem with MP2 audio, but an older filter was re-registered by somone while I didn't know (possibly by MatroskaMP4Pack.exe). SO I was wrongly believing that I was using 1.0.1.8 when I wrote "1.0.1.8 didn't work!" Actually I was using an older filter when I first disabled MPC's inner filter.

On top ot that, I didn't notice that critical Note ("Current versions of libebml...") in the first post by Mosu; I first read about the new version not here but in the list, and the mail posted in the list didn't have that note.

Anyway, IDX+SUB support is a great step forward. Probebly that's the very reason I didn't read the Note carefully--I was too excited and impatiently wanted to try it. I guess the install pack 0.7, which will install the newer splitter, will be out after Gabest releases a new VSFilter with IDX+SUB support.

I should soon love the new lacing too if it makes files more compact and functional, and I always admir their (dev ppl's) creativity.

That said, I really do hope that coming newer filters (1.0.1.9+) and all the related components can parse the old files too forever, because I already have some MKV files that have "the old structure" and I already , ah, presented some of my MKV files to...somone else, so I can't remux all the files to let them have the new lacing sysytem...I mean, what I really fear is hearing someday "the system in libmatroska 0.5.x is obsolete and playing back such old MKV files will be no more supported." But maybe only a few ppl think that way.

[Human_USB]

@Liisachan

I re-registered using the newest stuff off SF.net and I still can't play. I'm think it's just my computer now acting dumb. I tryed everything in this topic and still no luck. Unless I missed something I will wait for 1.0.1.9 or a installer for all this crap.



I've been waiting for VOBSUB support and now that is somewhat work, I'm waiting for menus.

[Atamido]

Originally posted by Liisachan
I should soon love the new lacing too if it makes files more compact and functional, and I always admir their (dev ppl's) creativity.

That said, I really do hope that coming newer filters (1.0.1.9+) and all the related components can parse the old files too forever, because I already have some MKV files that have "the old structure" and I already , ah, presented some of my MKV files to...somone else, so I can't remux all the files to let them have the new lacing sysytem...I mean, what I really fear is hearing someday "the system in libmatroska 0.5.x is obsolete and playing back such old MKV files will be no more supported." But maybe only a few ppl think that way.

A quick explanation of the way that lacing worked will show that old files are fine. Originally the lacing system used by Matroska was the lacing system designed by Xiph for Vorbis. You have a normal Block with a timecode. There is a bit set in the Block that indicates lacing is being used. Next you have the number of laced packets. (The spec allows for up to 255, but libmatroska only writes up to 8.) Next you have the size of each laced packet, and then the packets themselves to the end of the block. It actually turned out to be only slightly more efficient for a lot of codecs and not really worth the hassle. Recently a few ideas were tossed around such as storing the the size of the packets as an EBML number as it would be more efficient for larger packet sizes.

Also, there was an idea to set a flag if all of the packets were the same size. This would mean a single bit and the number of packets in the lace. Then all of the sizes are known because you know the total size and the number of equal size packets.

Some tests were performed and it was discovered that codecs with larger packets were more efficient with the EBML numbers. And codecs with fixed packet sizes, like AC3, were super efficient with the fixed size flag.

So, three settings were made for lacing. The original setting is still the Xiph lacing. This means that all of the old files that were made with Xiph lacing are still valid and this lacing is still being used. The next is for EBML lacing, and the last is for fixed size lacing.

When muxing, the muxer automatically determines for each block which lacing scheme would be most efficient. So most streams will likely have a mix of all three. See here for a list of different overheads and see how much of a difference the new system made from the Xiph-only system that was used previously.

[Mosu]

Pamel has effectively said all that I would have said Yes, we knew there would be some confusion regarding the new lacing, but we hoped that with Gabest's 1.0.1.8 it would work all right. I hope he'll update MPC soon so that at least the need for deactivating its internal Matroska handling code disappears.

A workaround is to disable lacing on mmg's 'global' tab. Of course this will make files larger.

[Lobuz]

Small question. What is the status of adjusting mkv tools to handle audio targeted applications? Using CUE sheets from CD with audio image, or many files collected in album. With additional stuff (covers,lirycs etc.). What about muxing APE or other audio codecs? Would be nice to have support for mka in foobar player too.

Regards
Lobuz

[Liisachan]

Originally posted by Pamel
When muxing, the muxer automatically determines for each block which lacing scheme would be most efficient. So most streams will likely have a mix of all three.

Thank you for clarification. So the "old lacing" is actually a special case of the new extended lacing system, using always only one of the three types for all the blocks, and so automatically will be supported by new persers who know all the three types, right? Good to know. Thanx so much. So the lacing system has actually not totally changed but rather "generalized"...And that overhead comparison you pointed out is really convincing

[ivanova]

Anyone tested the idx/sub in matroska yet? It muxes fine, but I haven't been able to play it. OR do we have to wait for a new matroska splitter?