From kurtnoise at free.fr Tue Oct 4 20:33:42 2005 From: kurtnoise at free.fr (Kurtnoise) Date: Tue, 04 Oct 2005 20:33:42 +0200 Subject: [Matroska-devel] MKVE Wizard: a new tool to update mkvextractGUI Message-ID: <4342CB06.2050001@free.fr> Hi guys, I'm rewriting a tool to extract tracks/chapters/attachments/cuesheets/tags...from mkv & mka files. Could you test this please ? It's just a Wizard. No complicated to use it. There are only 3 steps. (Check out this post to see some snapshoots.) mka support is not finished yet. So, test it only with mkv files please. Suggestions are welcome of course... All credits for Alexnoe & Mosu. Many thanks guys for your work & tools. ++ Kurtnoise From ivanburnin at hushmail.com Wed Oct 5 13:19:26 2005 From: ivanburnin at hushmail.com (Ivan Burnin) Date: Wed, 5 Oct 2005 04:19:26 -0700 Subject: [Matroska-devel] Licensing question Message-ID: I've got a licensing question, I know this horse has been beaten to death, but it had a foal. Here's the situation. I have an application (pending funding and negotiations) that will be legally distributing video. Because of the nature of the video I have pseudonymed myself because I don't know if the Matroska developers would want to be associated with it (I will gladly denym myself if the developers agree under private disclosure). Additionally, because of the DRM necessary it may be necessary to make changes to the root code (obviously donated back). The initial target viewer is Windows Media Player, so the changes should be fairly minor. I'm not concerned about the legal aspects, any half-witted lawyer could guide me through that process, in fact I wouldn't even need a lawyer. Instead I'm asking about the social contract. My view is quite simple; this was developed by people that deserve recognition of their preferred form. So what I'm asking is what form do the developers want this recognition in? I had considered revisiting the earlier suggestion of a licensing authority which would pay some amount to the core developers, or alternately attempting to keep a position available should one of the developers need it. I'm willing to listen to opinions and would actually _like_ to pay some form of licensing fee. On the licensing fee, in whatever form. I want to see Matroska become better, and I realize that having licensing fees (in whatever form) involved greatly increases the odds of that happening. Obviously this would be purely a social contract proposition, but you all deserve recognition for your work and I do my best to give recognition where recognition is due. From chris at matroska.org Thu Oct 6 07:44:47 2005 From: chris at matroska.org (Christian HJ Wiesner) Date: Thu, 06 Oct 2005 07:44:47 +0200 Subject: [Matroska-devel] Licensing question In-Reply-To: References: Message-ID: <4344B9CF.6090907@matroska.org> Hi Ivan, i am in a hurry writing this, so i bring it briefly to the point : matroska, the format, is license free. However, some of the code is not. If you say you are using matroska for your app, exactly what parts of the code have you been using, and what kind of changes did you add ? About your will to give some form of donations to the project, thanks a lot. Regards Christian matroska project admin Ivan Burnin schrieb: >I've got a licensing question, I know this horse has been beaten to death, >but it had a foal. > >Here's the situation. I have an application (pending funding and >negotiations) that will be legally distributing video. Because of the nature >of the video I have pseudonymed myself because I don't know if the Matroska >developers would want to be associated with it (I will gladly denym myself >if the developers agree under private disclosure). Additionally, because of >the DRM necessary it may be necessary to make changes to the root code >(obviously donated back). The initial target viewer is Windows Media Player, >so the changes should be fairly minor. > >I'm not concerned about the legal aspects, any half-witted lawyer could >guide me through that process, in fact I wouldn't even need a lawyer. >Instead I'm asking about the social contract. My view is quite simple; this >was developed by people that deserve recognition of their preferred form. So >what I'm asking is what form do the developers want this recognition in? I >had considered revisiting the earlier suggestion of a licensing authority >which would pay some amount to the core developers, or alternately >attempting to keep a position available should one of the developers need >it. I'm willing to listen to opinions and would actually _like_ to pay some >form of licensing fee. > >On the licensing fee, in whatever form. I want to see Matroska become >better, and I realize that having licensing fees (in whatever form) involved >greatly increases the odds of that happening. Obviously this would be purely >a social contract proposition, but you all deserve recognition for your work >and I do my best to give recognition where recognition is due. > > From paul at msn.com Thu Oct 6 20:45:33 2005 From: paul at msn.com (Paul Bryson) Date: Thu, 6 Oct 2005 13:45:33 -0500 Subject: [Matroska-devel] Re: Licensing question References: Message-ID: "Ivan Burnin" wrote in message news:di0ct4$dmj$1 at sea.gmane.org... > Additionally, because of the DRM necessary it may be necessary to make > changes to the root code (obviously donated back). The initial target > viewer is Windows Media Player, so the changes should be fairly minor. Note that there already exists a framework in the specifications for handling DRM. Use a ContentEncodingType of "encryption". http://www.matroska.org/technical/specs/index.html#ContentEncodingType I don't think the code exists to handle this yet, but it would be nice if you made it to spec. Atamido (Always reply to the list) From paul at msn.com Thu Oct 6 20:51:39 2005 From: paul at msn.com (Paul Bryson) Date: Thu, 6 Oct 2005 13:51:39 -0500 Subject: [Matroska-devel] Re: Licensing question References: Message-ID: "Ivan Burnin" wrote in message news:di0ct4$dmj$1 at sea.gmane.org... > Additionally, because of the DRM necessary it may be necessary to make > changes to the root code (obviously donated back). The initial target > viewer is Windows Media Player, so the changes should be fairly minor. There already exists a method to do DRM in the specs. Set ContentEncodingType to encryption. http://www.matroska.org/technical/specs/index.html#ContentEncodingType I don't believe the code exists to handle this, but it would be nice if you used the existing structures. Atamido From ivanburnin at hushmail.com Fri Oct 7 01:06:07 2005 From: ivanburnin at hushmail.com (ivanburnin at hushmail.com) Date: Thu, 6 Oct 2005 16:06:07 -0700 Subject: [Matroska-devel] Licensing question Message-ID: <200510062306.j96N6AG3093482@mailserver2.hushmail.com> On Wed, 05 Oct 2005 22:44:47 -0700 Christian HJ Wiesner wrote: >matroska, the format, is license free. However, some of the code >is not. >If you say you are using matroska for your app, exactly what parts >of >the code have you been using, and what kind of changes did you add >? So far there isn't any code, I'm in early negotiations with the copyright holders. Eventually though the preference would be to use the DirectShow plugin architecture. The modifications would revolve primarily around establishing the ability to embed DRM into all the components, obviously the video and audio codecs support it (we'll be writing our own wrapper codecs) but I haven't dug deep enough into the subtitle and menu architectures to see if they already support it. I know how much we all enjoy DRM, but copyright holders generally really like it. I was thinking about creating an architecture for subtitles that creates an alpha-based overlay, so the subtitles would be seperately rendered, and overlayed via alpha- channels. Assuming that the subtitle subsystem was then built on a plugin architecture this would eliminate all of our problems (we could create an encrypted one), and the rendering would provide a more powerful architecture for the rest of Matroska. > >About your will to give some form of donations to the project, >thanks a lot. Of course. Concerned about your privacy? Follow this link to get secure FREE email: http://www.hushmail.com/?l=2 Free, ultra-private instant messaging with Hush Messenger http://www.hushmail.com/services-messenger?l=434 Promote security and make money with the Hushmail Affiliate Program: http://www.hushmail.com/about-affiliate?l=427 From andrew.abbass at gmail.com Fri Oct 7 06:04:38 2005 From: andrew.abbass at gmail.com (Andrew Abbass) Date: Fri, 7 Oct 2005 01:04:38 -0300 Subject: [Matroska-devel] intellectual property rights management Message-ID: Hello, Do you plan on eventually incorporating any DRM systems into your media container? -Andrew Abbass Newdor.com Media Inc. From paul at msn.com Fri Oct 7 10:07:43 2005 From: paul at msn.com (Paul Bryson) Date: Fri, 7 Oct 2005 03:07:43 -0500 Subject: [Matroska-devel] Re: intellectual property rights management References: Message-ID: The specifications include a method to encrypt streams, which is what you do with DRM. However, there isn't any existing code to use these methods. At this time none of the team members have plans to write a full fledged DRM system, but anyone else could certainly do it. Atamido (Always reply to the list) "Andrew Abbass" wrote... Hello, Do you plan on eventually incorporating any DRM systems into your media container? -Andrew Abbass Newdor.com Media Inc. _______________________________________________ Matroska-devel mailing list Matroska-devel at lists.matroska.org http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-devel Read Matroska-Devel on GMane: http://dir.gmane.org/gmane.comp.multimedia.matroska.devel From paul at msn.com Fri Oct 7 15:17:06 2005 From: paul at msn.com (Paul Bryson) Date: Fri, 7 Oct 2005 08:17:06 -0500 Subject: [Matroska-devel] Re: intellectual property rights management References: Message-ID: ----- Original Message ----- From: "Andrew Abbass" To: "Paul Bryson" Sent: Friday, October 07, 2005 7:58 AM Subject: Re: intellectual property rights management Have you ever heard of Shared Media Licensing? I believe they're currently gunning for the patent to place the purchasing system for DRM protected files inside the file itself. This makes burning the data to CD or DVD for bulk distribution a cheaper alternative to direct download. Will this container eventually cover text and graphic files as well? -Andrew From steve.lhomme at free.fr Fri Oct 7 21:02:49 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Fri, 07 Oct 2005 21:02:49 +0200 Subject: [Matroska-devel] New SimpleBlock Message-ID: <4346C659.2090305@free.fr> Hi, We are in the process of adding a new Block element to the specs. The idea is to make it as simple as possible to write and to read. That is no complex referencing system and such. So the new SimpleBlock is now at the same level as BlockGroup (which then lose his mandatory feature in the block). It has the same format as our good old Block, with the addition of 2 flags : - keyframe (set when the frames of the Block are stand alone) - discardable (set when the frame can be safely discarded on playback) This change was motivated by Haali and Mosu to avoid parsing the codec to find references and the required timecode. There should be no problem on playback as it's the way AVI and MP4 work. Files with that new block should be available soon from mkvmerge and Haali's muxer (and hopefully AVI-Mux GUI), so you'll have to update your player for such files. Haali already supports it in his experimental build and I'll it to VLC this week-end when mkvmerge can create such files. -- robUx4 on blog From steve.lhomme at free.fr Fri Oct 7 21:10:36 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Fri, 07 Oct 2005 21:10:36 +0200 Subject: [Matroska-devel] Compression track 3 Message-ID: <4346C82C.5060303@free.fr> Hi (again), Another addition for the coming days in Matroska is the new compression feature of each track. The idea is to strip the header of each frame in the track, because that header is always the same. It's then readded on playback. This feature could seem like ridiculous if you only gain 1 byte per Block. But for a movie of 1h30 you could save 150 KB on an MP3 track. And for MPEG4 part 2 (25 fps, 4 bytes saved) it's 540 KB. This feature is already working in Haali's experimental. We now have to implement it in mkvmerge and VLC. -- robUx4 on blog From paul at msn.com Fri Oct 7 22:03:07 2005 From: paul at msn.com (Paul Bryson) Date: Fri, 7 Oct 2005 15:03:07 -0500 Subject: [Matroska-devel] Re: Compression track 3 References: <4346C82C.5060303@free.fr> Message-ID: "Steve Lhomme" wrote > Hi (again), > > Another addition for the coming days in Matroska is the new compression > feature of each track. The idea is to strip the header of each frame in > the track, because that header is always the same. It's then readded on > playback. As we already have a "compression" feature, perhaps a different term should be used for this feature to avoid confusion. Maybe BlockAppend or BlockStrip? BlockRepeater? I'm sure someone could come up with a really good term for this. I don't know if they are common, but what about stripping the footer also? Do many codecs use a byte sequence to close their frame? Paul Bryson From steve.lhomme at free.fr Fri Oct 7 22:08:17 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Fri, 07 Oct 2005 22:08:17 +0200 Subject: [Matroska-devel] Re: Compression track 3 In-Reply-To: References: <4346C82C.5060303@free.fr> Message-ID: <4346D5B1.7000908@free.fr> Paul Bryson wrote: > "Steve Lhomme" wrote > >>Hi (again), >> >>Another addition for the coming days in Matroska is the new compression >>feature of each track. The idea is to strip the header of each frame in >>the track, because that header is always the same. It's then readded on >>playback. > > > As we already have a "compression" feature, perhaps a different term should > be used for this feature to avoid confusion. Maybe BlockAppend or > BlockStrip? BlockRepeater? I'm sure someone could come up with a really > good term for this. No, it's a lossless compression handled at the container level for the specified track. It's perfectly fine to call it compression. There is a "compression" and "decompression" involved, even though it's the most basic possible (apart from null transformation). The name of the codec is "Header Stripping". > I don't know if they are common, but what about stripping the footer also? > Do many codecs use a byte sequence to close their frame? No, I don't think so. Not worth the effort. From paul at msn.com Fri Oct 7 23:13:59 2005 From: paul at msn.com (Paul Bryson) Date: Fri, 7 Oct 2005 16:13:59 -0500 Subject: [Matroska-devel] Re: Re: Compression track 3 References: <4346C82C.5060303@free.fr> <4346D5B1.7000908@free.fr> Message-ID: "Steve Lhomme" wrote... > No, it's a lossless compression handled at the container level for the > specified track. It's perfectly fine to call it compression. There is a > "compression" and "decompression" involved, even though it's the most > basic possible (apart from null transformation). The name of the codec is > "Header Stripping". You are using the ContentCompression and ContentCompAlgo elements then? Atamido From steve.lhomme at free.fr Sat Oct 8 07:24:42 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Sat, 08 Oct 2005 07:24:42 +0200 Subject: [Matroska-devel] Re: Re: Compression track 3 In-Reply-To: References: <4346C82C.5060303@free.fr> <4346D5B1.7000908@free.fr> Message-ID: <4347581A.30508@free.fr> Paul Bryson wrote: > "Steve Lhomme" wrote... > >>No, it's a lossless compression handled at the container level for the >>specified track. It's perfectly fine to call it compression. There is a >>"compression" and "decompression" involved, even though it's the most >>basic possible (apart from null transformation). The name of the codec is >>"Header Stripping". > > > You are using the ContentCompression and ContentCompAlgo elements then? Yes. I forgot to mention that. You can see a new value in the specs there. From steve.lhomme at free.fr Sat Oct 8 10:19:42 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Sat, 08 Oct 2005 10:19:42 +0200 Subject: [Matroska-devel] New SimpleBlock In-Reply-To: <4346C659.2090305@free.fr> References: <4346C659.2090305@free.fr> Message-ID: <4347811E.6040608@free.fr> OK, some comments on this: - It will break playback of some files in all older players as there is a track they recognize but they never find a BlockGroup for. In that case we should bump the Matroska min version number (DocTypeReadVersion in the EBML header) to 2 as well as the Matroska version number (DocTypeVersion). But I'm not sure any current player verify that they match the required version (in that case they are not spec compliant...). So I suggest we go without any change (and implement that check where we can). It won't be a big problem as this new SimpleBlock won't be used by default. - The Frames in the SimpleBlock have no duration. That means SimpleBlock is not suitable for VFR content (VFR video or Vorbis). Or we could fall back in using BlockGroup when the duration of the frames are not the default ones... Steve Lhomme wrote: > Hi, > > We are in the process of adding a new Block element to the specs. The > idea is to make it as simple as possible to write and to read. That is > no complex referencing system and such. So the new SimpleBlock is now at > the same level as BlockGroup (which then lose his mandatory feature in > the block). It has the same format as our good old Block, with the > addition of 2 flags : > - keyframe (set when the frames of the Block are stand alone) > - discardable (set when the frame can be safely discarded on playback) > > This change was motivated by Haali and Mosu to avoid parsing the codec > to find references and the required timecode. There should be no problem > on playback as it's the way AVI and MP4 work. > > Files with that new block should be available soon from mkvmerge and > Haali's muxer (and hopefully AVI-Mux GUI), so you'll have to update your > player for such files. Haali already supports it in his experimental > build and I'll it to VLC this week-end when mkvmerge can create such files. > -- robUx4 on blog From steve.lhomme at free.fr Sat Oct 8 11:46:00 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Sat, 08 Oct 2005 11:46:00 +0200 Subject: [Matroska-devel] New Pack v1.1.2 Message-ID: <43479558.6040802@free.fr> Hi, We are about to release a new official pack. The main motivation is that the H264 decoder we are using is too old and it cannot decode some of the new files floating around. And the pack is supposed to support such basic files with codecs we support. I waited a little bit longer after rc2 because I wanted to integrate the new SimpleBlock and the new header compression method. So that we don't make files that can't be played by our pack. But it turns out creating such files in a stable way in mkvmerge may take long and won't be enabled by default in mkvmerge. So I think it's OK to release the pack ASAP, as long as Haali is confident with his support for the 2 new additions. Apart from that, the pack is ready to go. I'll release a rc3 before making it public to make sure Haali's new pack has no problem when installed from the pack. -- robUx4 on blog From steve.lhomme at free.fr Sat Oct 8 14:31:47 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Sat, 08 Oct 2005 14:31:47 +0200 Subject: [Matroska-devel] Re: [Matroska-cvs] [matroska] r1211 - trunk/MatroskaPack/Satsuki In-Reply-To: <200510081206.j98C6FTs052026@matroska.org> References: <200510081206.j98C6FTs052026@matroska.org> Message-ID: <4347BC33.6070306@free.fr> Here is the patch to ffmpeg in ffdshow, in case someday says we're violating the license by releasing a binary without the code... robux4 at matroska.org wrote: > Author: robux4 > Date: 2005-10-08 16:06:12 +0400 (Sat, 08 Oct 2005) > New Revision: 1211 > > Modified: > trunk/MatroskaPack/Satsuki/Matroska_Pack_var.nsh > trunk/MatroskaPack/Satsuki/ReadMe_Full.html > trunk/MatroskaPack/Satsuki/ReadMe_Lite.html > Log: > pack: new rc3 version > > Modified: trunk/MatroskaPack/Satsuki/Matroska_Pack_var.nsh > =================================================================== > --- trunk/MatroskaPack/Satsuki/Matroska_Pack_var.nsh 2005-10-06 21:30:13 UTC (rev 1210) > +++ trunk/MatroskaPack/Satsuki/Matroska_Pack_var.nsh 2005-10-08 12:06:12 UTC (rev 1211) > @@ -1,6 +1,6 @@ > !define VERSION_MAJOR "1.1.2" > !define VERSION_MINOR "1208" ; should correspond to the latest SVN version > -!define VERSION_RC "2" > +!define VERSION_RC "3" > !define VERSION_BETA "2" > > ; define one of these for release candidate and beta versions > @@ -8,7 +8,7 @@ > ;!define BETA "yep" > > > -!define VerHaali "v21-05-2005" > +!define VerHaali "v05-10-2005" > !define VerFFDShow "v30-09-2005 MSVC" > !define VerVSFilter "v2.36+border fix" > !define VerGabestMPEG2 "v0.1.2.0" > > Modified: trunk/MatroskaPack/Satsuki/ReadMe_Full.html > =================================================================== > --- trunk/MatroskaPack/Satsuki/ReadMe_Full.html 2005-10-06 21:30:13 UTC (rev 1210) > +++ trunk/MatroskaPack/Satsuki/ReadMe_Full.html 2005-10-08 12:06:12 UTC (rev 1211) > @@ -37,8 +37,8 @@ > > > Haali Media Splitter > -v18-09-2005 > -2005-09-18 > +v05-10-2005 > +2005-10-05 > Can also be used to demux AVI, DivX and MP4 files > > > @@ -86,8 +86,8 @@ > > > ffdshow > -v2005-09-30 (decoding only) > -2005-08-30 > +v2005-09-30+bugfix (decoding only) > +2005-08-30 + 2005-10-08 > Audio and video codecs > > > > Modified: trunk/MatroskaPack/Satsuki/ReadMe_Lite.html > =================================================================== > --- trunk/MatroskaPack/Satsuki/ReadMe_Lite.html 2005-10-06 21:30:13 UTC (rev 1210) > +++ trunk/MatroskaPack/Satsuki/ReadMe_Lite.html 2005-10-08 12:06:12 UTC (rev 1211) > @@ -37,8 +37,8 @@ > > > Haali Media Splitter > -v18-09-2005 > -2005-09-18 > +v05-10-2005 > +2005-10-05 > Can also be used to demux AVI, DivX and MP4 files > > > > _______________________________________________ > Matroska-cvs mailing list > Matroska-cvs at lists.matroska.org > http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-cvs -- robUx4 on blog -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ffmpeg.patch URL: From steve.lhomme at free.fr Sat Oct 8 15:51:54 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Sat, 08 Oct 2005 15:51:54 +0200 Subject: [Matroska-devel] Licensing question In-Reply-To: <200510062306.j96N6AG3093482@mailserver2.hushmail.com> References: <200510062306.j96N6AG3093482@mailserver2.hushmail.com> Message-ID: <4347CEFA.4010603@free.fr> Hi, Sorry for not replying earlier, but I was kinda busy... ivanburnin at hushmail.com wrote: > On Wed, 05 Oct 2005 22:44:47 -0700 Christian HJ Wiesner > wrote: > >>matroska, the format, is license free. However, some of the code >>is not. >>If you say you are using matroska for your app, exactly what parts > > >>of >>the code have you been using, and what kind of changes did you add > > >>? > > > So far there isn't any code, I'm in early negotiations with the > copyright holders. Eventually though the preference would be to use > the DirectShow plugin architecture. The modifications would revolve > primarily around establishing the ability to embed DRM into all the > components, obviously the video and audio codecs support it (we'll > be writing our own wrapper codecs) but I haven't dug deep enough > into the subtitle and menu architectures to see if they already > support it. I know how much we all enjoy DRM, but copyright holders > generally really like it. I was thinking about creating an > architecture for subtitles that creates an alpha-based overlay, so > the subtitles would be seperately rendered, and overlayed via alpha- > channels. Assuming that the subtitle subsystem was then built on a > plugin architecture this would eliminate all of our problems (we > could create an encrypted one), and the rendering would provide a > more powerful architecture for the rest of Matroska. IMO there is a big problem with DRM mixed with open-source. I still haven't found a way to make it work. Because unlike most security features existing in this world, it's made to prevent legitimate users (including coders) to use the content they decode a certain way. That means once the content is decoded, it is still in open-source code, and therefore not "protected" (or Managed as in DRM) at all. So the only real solution I see is some form of closed source between the decoder and the renderer. So, that means whatever codec you codec you want to use, you have to have a closed version of the decoder. And obviously a closed version of the renderer... Haali can probably help here as he wrote a DirectShow renderer for "fun". For the codec, there must be plenty, but you have to contact the people directly. Unfortunately we don't have any good video codec on CoreCodec for that (CorePNG wouldn't qualify for internet distribution). The other option is to contact people who make XviD or ffmpeg to see if they are willing to license the code for use in a closed source app. The people also deserve help :D Of course once this is done, matroska is a great format and the code could be written to support this special edition codec. Maybe all (splitter, codec, render) could compiled in one single file... For such things, contact CoreCodec Inc which will deal with the matroska licenses and closed development. From paul at msn.com Mon Oct 10 05:41:57 2005 From: paul at msn.com (Paul Bryson) Date: Sun, 9 Oct 2005 22:41:57 -0500 Subject: [Matroska-devel] Re: Re: Re: Compression track 3 References: <4346C82C.5060303@free.fr> <4346D5B1.7000908@free.fr> <4347581A.30508@free.fr> Message-ID: "Steve Lhomme" wrote... > Paul Bryson wrote: >> You are using the ContentCompression and ContentCompAlgo elements then? > > Yes. I forgot to mention that. You can see a new value in the specs there. Ah, good news then. That is most definitely a good and proper way to do it. Atamido From notorious_namer at hotmail.com Mon Oct 10 09:58:36 2005 From: notorious_namer at hotmail.com (Erik Nguyen) Date: Mon, 10 Oct 2005 00:58:36 -0700 Subject: [Matroska-devel] (no subject) Message-ID: My Name is Erik Nguyen I am wondering if Your making a program that convert MKV to a AVI format and also splits up the languages as well. When i was converting a MKV filr i realize that it had no sound and it decompress to a 3GB file. I want a program thats split them up separately. like one english avi one japanese avi etc. i realize that the converted file was not just one avi but more than one. And becasue of it i can't seem to make it play one langauge. this lead me to think that the sound was not able to work because of it. if u have a program that convert it properply can u send me a link thank you PS If you don't know any or have a program like that can you develop one as a project. _________________________________________________________________ Take advantage of powerful junk e-mail filters built on patented Microsoft? SmartScreen Technology. http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=http://hotmail.com/enca&HL=Market_MSNIS_Taglines Start enjoying all the benefits of MSN? Premium right now and get the first two months FREE*. From chris at matroska.org Mon Oct 10 20:40:57 2005 From: chris at matroska.org (Christian HJ Wiesner) Date: Mon, 10 Oct 2005 20:40:57 +0200 Subject: [Matroska-devel] (no subject) In-Reply-To: References: Message-ID: <434AB5B9.7030009@matroska.org> Hi Erik, thanks for your interest in our project, especially about converting MKV files into other, obviously preferred, formats. However, i am having a number of problems with your email : 1. Could you please tell us what program you were using to 'convert' your MKV into a 3 GB AVI file ? Do you really think that we have the time to guess that ? 2. May we know the purpose of your planned conversion into AVI ? 3. AVI can hold up to 99 tracks in one file IIRC, makes 98 audio tracks if one video track is present. So why would you insist in getting separate AVIs for every language then ( see 2.) ? 4. MKV is developed to overcome the numerous existing limitations of AVI. For this very reason, most MKV files are simply speaking not convertable into AVI, depending on the used codecs and if the file is VFR or not. What codecs are being used in your MKV file ? Is it VFR (Variable Framerate) ? and where did you get it from ? Thanks for answering our questions, enabling us to help you with what you are trying to do (whatever it is) Christian matroska project admin http://www.matroska.org OpenDML AVI can Erik Nguyen schrieb: > My Name is Erik Nguyen > > I am wondering if Your making a program that convert MKV to a > AVI format and > also splits up the languages as well. When i was converting a MKV > filr i realize that it > had no sound and it decompress to a 3GB file. I want a program thats > split them up > separately. > like > one english avi > one japanese avi > etc. > i realize that the converted file was not just one avi but more than > one. And becasue of it i can't seem to make it play one langauge. > this lead me to think that the sound was not able to work because of > it. if u have a program that convert it properply can u send me a link > thank you > PS If you don't know any or have a program like that can you develop > one as a project. From steve.lhomme at free.fr Tue Oct 18 09:35:44 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Tue, 18 Oct 2005 09:35:44 +0200 Subject: [Matroska-devel] Does Matroska Suck ? Message-ID: <4354A5D0.1070804@free.fr> http://www.scottberkun.com/essays/essay46.htm The asnwer is not there. But it's a good article to understanding what it takes to do something good. -- robUx4 on blog From chris at matroska.org Tue Oct 18 11:03:36 2005 From: chris at matroska.org (Christian HJ Wiesner) Date: Tue, 18 Oct 2005 11:03:36 +0200 Subject: [Matroska-devel] Does Matroska Suck ? In-Reply-To: <4354A5D0.1070804@free.fr> References: <4354A5D0.1070804@free.fr> Message-ID: <4354BA68.7020304@matroska.org> Steve Lhomme schrieb: > http://www.scottberkun.com/essays/essay46.htm > The answer is not there. But it's a good article to understanding what > it takes to do something good. Great article ! As i am basically presenting the user fraction in the team, forming a 'counterforce' to you guys being developers, i can't agree more on what he writes for example about the 'beauty of code'. As this may be very relevant to you, we users couldn't care less about it. The biggest downsides of matroska for us users, compared to AVI, is still the missing convertability of MKV files into other formats, DVD having the highest priority. For this reason i was trying to find a cooperation partner here for us, but you see what is happening to my emails, for example the ones i sent the people from DVD Santa. These people, having decided to read files into their app via DShow, are so close to a working solution, but still they don't care nor respond (i might ask Dan Marlin to write them an email, signed with 'CEO Corecodec' ... maybe this will help ;-) ). Another thing is STILL, despite all the hard work from Haali, the serious playback problems we are facing on some machines. Now, its easy to stand at the position '...well, this is their fault, they shouldnt have installed so many codec packs and other shit....' but i believe we are making things to easy for us here. Did you know that a CD/DVD burning app is installing about a dozen different DirectShow filters these days, for their multimedia support ? Now lets think for a moment you knew that, but do you think a normal user has the faintest clue of what Nero is doing to his machine, when he decides to install even the Demo Version only, for testing ? In other words, its too easy to say its their fault. Its not, because many apps and 'codec packs' are raping their machines, and they won't know. All they experience, when trying to play this fancy MKV files they jut got, that - they dont play at all - no video or audio files will play anymore - the PC is crashing when trying to play MKV files Being a normal user, who would you believe is the guilty, if either one will happen after installing our DShow playback pack ? Nero ? A codec pack you installed more than a year ago, and about which you forgot about in the mean time ? Now, we could show a state of mind like '.... well, we cant care about anybody ....' , but then we have to be aware that one sole negative voice is often louder than 100 positive voices. If people are frustrated, they make themselves heard, and the internet will help them because it will save your anonymity. For this very reason i am more and more inclined to believe, that we should be able to offer an alternative (NOT a replacement) to DShow, and soon. This can be VLC or Gstreamer or anything else built on top of FFMPEG, but it needs to have - a nice GUI, similar to what people are used to from DShow players (VLC ?) - a DShow reader module, to be able to use DShow decoders like Gabest's Realmedia splitter, or the ON2 decoder - another significant advantage to DShow players, like lower CPU consumption for h.264, to make at least some Pro's (multiplicators) using it Besides that, the success of our project is based on the user friendliness of the matroska tools compared to similar projects. This should not be reason for us to stand still, there is still a lot of work to do to be the best. On the other, without mmg.exe, VdubMod (especially in the beginning) or the DShow splitters we could offer, matroska would be nowhere. Lets build on this experience, try to progress from where we are today and make sure that only a very small percentage of the matroska users are of different opinion about this. Christian matroska project admin From moritz at bunkus.org Tue Oct 18 21:03:27 2005 From: moritz at bunkus.org (Moritz Bunkus) Date: Tue, 18 Oct 2005 21:03:27 +0200 Subject: [Matroska-devel] libebml 0.7.6 and libmatroska 0.8.0 have been released Message-ID: <200510182106.42462.moritz@bunkus.org> Hey, we've just released new versions of our libraries: 0.7.6 for libebml and 0.8.0 for libmatroska. The former contains a couple of build system fixes and a bug fix regarding elements with an unknown size. The latter contains quite a few new elements, e.g. our new "simple" block. The spec compliance has been improved/changed, too. The official source download links are http://dl.matroska.org/downloads/libebml/libebml-0.7.6.tar.bz2 and http://dl.matroska.org/downloads/libmatroska/libmatroska-0.8.0.tar.bz2 I've already updated all the binary packages that I host. Have a nice day :) Mosu -- If Darl McBride was in charge, he'd probably make marriage unconstitutional too, since clearly it de-emphasizes the commercial nature of normal human interaction, and probably is a major impediment to the commercial growth of prostitution. - Linus Torvalds -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From steve.lhomme at free.fr Wed Oct 19 18:18:27 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Wed, 19 Oct 2005 18:18:27 +0200 Subject: [Matroska-devel] Does Matroska Suck ? In-Reply-To: <4354BA68.7020304@matroska.org> References: <4354A5D0.1070804@free.fr> <4354BA68.7020304@matroska.org> Message-ID: <435671D3.9040708@free.fr> Christian HJ Wiesner wrote: > Steve Lhomme schrieb: > >> http://www.scottberkun.com/essays/essay46.htm >> The answer is not there. But it's a good article to understanding what >> it takes to do something good. > > Another thing is STILL, despite all the hard work from Haali, the > serious playback problems we are facing on some machines. Now, its easy > to stand at the position '...well, this is their fault, they shouldnt > have installed so many codec packs and other shit....' but i believe > we are making things to easy for us here. Did you know that a CD/DVD > burning app is installing about a dozen different DirectShow filters > these days, for their multimedia support ? Now lets think for a moment > you knew that, but do you think a normal user has the faintest clue of > what Nero is doing to his machine, when he decides to install even the > Demo Version only, for testing ? Yeah, DShow is a big problem. VLC is one good solution. And hopefully TC(P)MP will be another one. Anyway, for the pack I try to override as few default FFDShow settings as possible to avoid weird behaviours. But it seems having old ffdshow creates big problems. Uninstalling and reinstalling seem to fix the pb... /note to self, uninstall the previous packs & ffdshow before installing... From steve.lhomme at free.fr Wed Oct 19 18:22:07 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Wed, 19 Oct 2005 18:22:07 +0200 Subject: [Matroska-devel] Re: Matroska video blocking problem In-Reply-To: References: Message-ID: <435672AF.6020607@free.fr> Hi, I forward this email to our ML where Haali (who makes our splitter) will be able to read it. It sounds like your file only "indexes" the first video track and not the other one. Matroska (the format) can support index on all tracks, even partial index (called Cue). Maybe you're missing a command-line to do that or it's a bug in the muxing program (mkvmerge ?) or Haali's splitter that uses the wrong index for this track. Andrew Bromwich wrote: > Hi Steve, > > I have been having some problems with what i believe is with the > Matroska splitter. I wasnt too sure if posting a bug on the corecodec > wesite was the best option or emailing you (so i just emailed you, sorry > if this is incorrect). > > The problem I am having is that I have some video files, with a number > of video streams (normally 4 streams) and an audio stream, but when the > video streams are seeked through i get a large amount of blocking (or > periods of video freezing depending on video decompressor) on the 2nd, > 3rd and 4th stream. I did quite a lot of testing with a number of > different compressors and decompressors but with the same results. The > first screen always seems to work great which is really what is throwing > me off! I have not personally really got into the insides of the > Matroska splitter or muxer but am guessing that it might be some sort of > buffer that has been optimised for single video stream playback?? This > problem wouldnt be so much of an issue but the software I am developing > requires a fair bit of skipping through the video file, then every time > i seek it either blocks up or freezes altogether until the next > key-frame comes along. > > I have also tried a fair amount of testing using a number of different > settings with the video compressors i have been using (w.r.t keyframe > rates) but this really doesnt seem to make much difference to the > outcome either. > > I do appologise if i have emailed the wrong person (if this is the case > can you please point me in the direction of someone who might be able to > help). > > Thankyou, > > Andrew -- robUx4 on blog From aaa at aaaa.com Wed Oct 19 23:22:45 2005 From: aaa at aaaa.com (Zen) Date: Wed, 19 Oct 2005 23:22:45 +0200 Subject: [Matroska-devel] Difficulties to link libebml with DevCpp Message-ID: <4356B925.5040505@aaaa.com> I try (and succeed, but with difficulties) to compile and link libebml with DevCpp 4.9.9.2 (MinGW32 3.4) The project files are out of date (if you are interested, I can send you updated files), but I have difficulties to link if I create a library with all ebml files : - with all files in one library, I have link errors I don't understand (linker search for EbmlSInteger class when linking Matroska library, but the EbmlSInteger.cpp is included in the project of Ebml library!) - with all files in one library, AND one more library with only EbmlSInteger.cpp, when I link with the BOTH libraries, this is OK. Did you already have this kind of problem? Note : libebml works fine with Borland C++ Builder 6 and MS Visual Studio 2005 beta, libmatroska works fine with Borland C++ Builder 6, MS Visual Studio 2005 beta, and DevCpp 4.9.9.2. Zen http://mediainfo.sourceforge.net Below my linking problems : Compiler: Default compiler Building Makefile: "E:\Personnel\Programmation\MediaInfo\Project\DevCpp\GUI\Makefile.win" Executing make... make.exe -f "E:\Personnel\Programmation\MediaInfo\Project\DevCpp\GUI\Makefile.win" all g++.exe ../../../Source/GUI/WxWidgets/GUI_Main.o ../../../Source/GUI/WxWidgets/App.o MediaInfo_private.res -o "..\..\..\Release\GCC_MinGW32\GUI\MediaInfo.exe" -L"C:/Dev-Cpp/lib" -L"C:/Dev-Cpp/lib/msw-unicode-2.6.2/Static" -mwindows -mwindows -lwx_mswu_fl-2.6 -lwx_mswu_mmedia-2.6 -lwx_mswu_plot-2.6 -lwx_mswu_stc-2.6 -lwx_mswu_svg-2.6 -lwx_mswu_gizmos_xrc-2.6 -lwx_mswu_gizmos-2.6 -lwx_mswu_ogl-2.6 -lwx_mswu_deprecated-2.6 -lwx_mswu_animate-2.6 -lwx_mswu_gl-2.6 -lopengl32 -lglu32 -lwx_mswu_xrc-2.6 -lwx_mswu_html-2.6 -lwx_mswu_dbgrid-2.6 -lwx_baseu_odbc-2.6 -lwx_mswu_adv-2.6 -lwx_mswu_qa-2.6 -lwx_mswu_core-2.6 -lwx_baseu_xml-2.6 -lwx_baseu_net-2.6 -lwx_baseu-2.6 -lwx_mswu_media-2.6 -lwxtiff-2.6 -lwxjpeg-2.6 -lwxpng-2.6 -lwxzlib-2.6 -lwxregexu-2.6 -lwxexpat-2.6 -lrpcrt4 -loleaut32 -lole32 -luuid -lodbc32 -lole32 -loleaut32 -lwinspool -lwinmm -lshell32 -lcomctl32 -lcomdlg32 -lctl3d32 -ladvapi32 -lwsock32 -lgdi32 ../../../../MediaInfoLib/Release/GCC_MinGW32/Library/MediaInfoLib.a ../../../../Shared/Library/GCC_MinGW32/Unicows.a ../../../../Shared/Library/GCC_MinGW32/ZenLib.a ../../../../Shared/Library/GCC_MinGW32/ZLib.a ../../../../Shared/Library/GCC_MinGW32/WxBase.a ../../../../Shared/Library/GCC_MinGW32/WxWidgets_Core.a ../../../../Shared/Library/GCC_MinGW32/Ebml.a ../../../../Shared/Library/GCC_MinGW32/Matroska.a ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxTag.o)(.rdata$_ZTVN 11libmatroska19KaxTagPopularimeterE[vtable for libmatroska::KaxTagPopularimeter]+0x20):KaxTag.cpp: undefined reference to `libebml::EbmlSInteger::UpdateSize(bool, bool)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxTag.o)(.rdata$_ZTVN 11libmatroska19KaxTagPopularimeterE[vtable for libmatroska::KaxTagPopularimeter]+0x28):KaxTag.cpp: undefined reference to `libebml::EbmlSInteger::ReadData(libebml::IOCallback&, libebml::ScopeMode)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxTag.o)(.rdata$_ZTVN 11libmatroska19KaxTagPopularimeterE[vtable for libmatroska::KaxTagPopularimeter]+0x48):KaxTag.cpp: undefined reference to `libebml::EbmlSInteger::RenderData(libebml::IOCallback&, bool, bool)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxTag.o)(.text$_ZN11l ibmatroska19KaxTagPopularimeter6CreateEv[libmatroska::KaxTagPopularimete r::Create()]+0x5c):KaxTag.cpp: undefined reference to `libebml::EbmlSInteger::EbmlSInteger()' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxTag.o)(.text$_ZNK11 libmatroska19KaxTagPopularimeter5CloneEv[libmatroska::KaxTagPopularimete r::Clone() const]+0x66):KaxTag.cpp: undefined reference to `libebml::EbmlSInteger::EbmlSInteger(libebml::EbmlSInteger const&)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxBlockData.o)(.text+ 0x1af):KaxBlockData.cpp: undefined reference to `libebml::EbmlSInteger::UpdateSize(bool, bool)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxBlockData.o)(.rdata $_ZTVN11libmatroska17KaxReferenceBlockE[vtable for libmatroska::KaxReferenceBlock]+0x28):KaxBlockData.cpp: undefined reference to `libebml::EbmlSInteger::ReadData(libebml::IOCallback&, libebml::ScopeMode)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxBlockData.o)(.rdata $_ZTVN11libmatroska17KaxReferenceBlockE[vtable for libmatroska::KaxReferenceBlock]+0x48):KaxBlockData.cpp: undefined reference to `libebml::EbmlSInteger::RenderData(libebml::IOCallback&, bool, bool)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxBlockData.o)(.rdata $_ZTVN11libmatroska19KaxReferenceVirtualE[vtable for libmatroska::KaxReferenceVirtual]+0x20):KaxBlockData.cpp: undefined reference to `libebml::EbmlSInteger::UpdateSize(bool, bool)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxBlockData.o)(.rdata $_ZTVN11libmatroska19KaxReferenceVirtualE[vtable for libmatroska::KaxReferenceVirtual]+0x28):KaxBlockData.cpp: undefined reference to `libebml::EbmlSInteger::ReadData(libebml::IOCallback&, libebml::ScopeMode)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxBlockData.o)(.rdata $_ZTVN11libmatroska19KaxReferenceVirtualE[vtable for libmatroska::KaxReferenceVirtual]+0x48):KaxBlockData.cpp: undefined reference to `libebml::EbmlSInteger::RenderData(libebml::IOCallback&, bool, bool)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxBlockData.o)(.text$ _ZN11libmatroska17KaxReferenceBlock6CreateEv[libmatroska::KaxReferenceBl ock::Create()]+0x5c):KaxBlockData.cpp: undefined reference to `libebml::EbmlSInteger::EbmlSInteger()' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxBlockData.o)(.text$ _ZNK11libmatroska17KaxReferenceBlock5CloneEv[libmatroska::KaxReferenceBl ock::Clone() const]+0x66):KaxBlockData.cpp: undefined reference to `libebml::EbmlSInteger::EbmlSInteger(libebml::EbmlSInteger const&)' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxBlockData.o)(.text$ _ZN11libmatroska19KaxReferenceVirtual6CreateEv[libmatroska::KaxReference Virtual::Create()]+0x5c):KaxBlockData.cpp: undefined reference to `libebml::EbmlSInteger::EbmlSInteger()' ../../../../Shared/Library/GCC_MinGW32/Matroska.a(KaxBlockData.o)(.text$ _ZNK11libmatroska19KaxReferenceVirtual5CloneEv[libmatroska::KaxReference Virtual::Clone() const]+0x66):KaxBlockData.cpp: undefined reference to `libebml::EbmlSInteger::EbmlSInteger(libebml::EbmlSInteger const&)' collect2: ld returned 1 exit status make.exe: *** [../../../Release/GCC_MinGW32/GUI/MediaInfo.exe] Error 1 Execution terminated From moritz at bunkus.org Thu Oct 20 08:55:30 2005 From: moritz at bunkus.org (Moritz Bunkus) Date: Thu, 20 Oct 2005 08:55:30 +0200 Subject: [Matroska-devel] Difficulties to link libebml with DevCpp In-Reply-To: <4356B925.5040505@aaaa.com> References: <4356B925.5040505@aaaa.com> Message-ID: <200510200856.26619.moritz@bunkus.org> Hey, On Wednesday 19 October 2005 23:22, Zen wrote: > The project files are out of date (if you are interested, I can send you > updated files), That would be nice. > g++.exe ../../../Source/GUI/WxWidgets/GUI_Main.o > ../../../Source/GUI/WxWidgets/App.o MediaInfo_private.res -o > "..\..\..\Release\GCC_MinGW32\GUI\MediaInfo.exe" -L"C:/Dev-Cpp/lib" > -L"C:/Dev-Cpp/lib/msw-unicode-2.6.2/Static" -mwindows -mwindows > -lwx_mswu_fl-2.6 -lwx_mswu_mmedia-2.6 -lwx_mswu_plot-2.6 > -lwx_mswu_stc-2.6 -lwx_mswu_svg-2.6 -lwx_mswu_gizmos_xrc-2.6 > -lwx_mswu_gizmos-2.6 -lwx_mswu_ogl-2.6 -lwx_mswu_deprecated-2.6 > -lwx_mswu_animate-2.6 -lwx_mswu_gl-2.6 -lopengl32 -lglu32 > -lwx_mswu_xrc-2.6 -lwx_mswu_html-2.6 -lwx_mswu_dbgrid-2.6 > -lwx_baseu_odbc-2.6 -lwx_mswu_adv-2.6 -lwx_mswu_qa-2.6 > -lwx_mswu_core-2.6 -lwx_baseu_xml-2.6 -lwx_baseu_net-2.6 -lwx_baseu-2.6 > -lwx_mswu_media-2.6 -lwxtiff-2.6 -lwxjpeg-2.6 -lwxpng-2.6 -lwxzlib-2.6 > -lwxregexu-2.6 -lwxexpat-2.6 -lrpcrt4 -loleaut32 -lole32 -luuid -lodbc32 > -lole32 -loleaut32 -lwinspool -lwinmm -lshell32 -lcomctl32 -lcomdlg32 > -lctl3d32 -ladvapi32 -lwsock32 -lgdi32 > ../../../../MediaInfoLib/Release/GCC_MinGW32/Library/MediaInfoLib.a > ../../../../Shared/Library/GCC_MinGW32/Unicows.a > ../../../../Shared/Library/GCC_MinGW32/ZenLib.a > ../../../../Shared/Library/GCC_MinGW32/ZLib.a > ../../../../Shared/Library/GCC_MinGW32/WxBase.a > ../../../../Shared/Library/GCC_MinGW32/WxWidgets_Core.a > ../../../../Shared/Library/GCC_MinGW32/Ebml.a > ../../../../Shared/Library/GCC_MinGW32/Matroska.a Usually linkers care about the order in which you specify the libraries. It tries to resolve all unknown symbols from all the following libraries and discards all unneeded symbols from the current library. So when it reaches Ebml.a the stuff from Matroska.a hasn't been found yet, so the linker discards everything from Ebml.a. Next if examines Matroska.a and needs all those symbols from Ebml.a, but those have already been discarded. The solution should be to swap Ebml.a and Matroska.a. Mosu -- If Darl McBride was in charge, he'd probably make marriage unconstitutional too, since clearly it de-emphasizes the commercial nature of normal human interaction, and probably is a major impediment to the commercial growth of prostitution. - Linus Torvalds -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From steve.lhomme at free.fr Thu Oct 20 09:52:39 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Thu, 20 Oct 2005 09:52:39 +0200 Subject: [Matroska-devel] libebml bug on BeOS Message-ID: <43574CC7.2060800@free.fr> http://buildbot.videolan.org/I386-BeOS-Trunk/builds/195/Build%20Contribs/0 I'll have a look at the problem. -- robUx4 on blog From steve.lhomme at free.fr Thu Oct 20 11:16:24 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Thu, 20 Oct 2005 11:16:24 +0200 Subject: [Matroska-devel] Free VMWare Player Message-ID: <43576068.8020505@free.fr> http://www.vmware.com/products/player/ That could be interresting to test softwares on Win98, W2K, various Linux distro and BeOS. We'd just have to share our OS image. By testing I mean compiling and basic support (as it's a bit slower than the regular OS and not as multimedia friendly). -- robUx4 on blog From paul at msn.com Thu Oct 20 15:44:40 2005 From: paul at msn.com (Paul Bryson) Date: Thu, 20 Oct 2005 08:44:40 -0500 Subject: [Matroska-devel] Re: Free VMWare Player References: <43576068.8020505@free.fr> Message-ID: Okay, that is really cool. "Steve Lhomme" wrote in message news:43576068.8020505 at free.fr... > http://www.vmware.com/products/player/ > > That could be interresting to test softwares on Win98, W2K, various Linux > distro and BeOS. We'd just have to share our OS image. > > By testing I mean compiling and basic support (as it's a bit slower than > the regular OS and not as multimedia friendly). > > -- > robUx4 on blog > _______________________________________________ > Matroska-devel mailing list > Matroska-devel at lists.matroska.org > http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-devel > Read Matroska-Devel on GMane: > http://dir.gmane.org/gmane.comp.multimedia.matroska.devel > From aaa at aaaa.com Fri Oct 21 10:56:42 2005 From: aaa at aaaa.com (Zen) Date: Fri, 21 Oct 2005 10:56:42 +0200 Subject: [Matroska-devel] Re: Difficulties to link libebml with DevCpp In-Reply-To: <200510200856.26619.moritz@bunkus.org> References: <4356B925.5040505@aaaa.com> <200510200856.26619.moritz@bunkus.org> Message-ID: <4358AD4A.4070004@aaaa.com> Moritz Bunkus wrote: > Hey, > > On Wednesday 19 October 2005 23:22, Zen wrote: > >> The project files are out of date (if you are interested, I can send you >> updated files), > > That would be nice. I sent to your private adress the files... But please : is it possible for libebml to stop to have an exception for WinIOCallback? At least for the .h file, because all .h are in the same directory, except ONE, it is always forgot... For .cpp, maybe a #ifdef _WIN32 for all file would be better than to have a specific directory... > Usually linkers care about the order in which you specify the > libraries. It tries to resolve all unknown symbols from all the > following libraries and discards all unneeded symbols from the current > library. So when it reaches Ebml.a the stuff from Matroska.a hasn't been > found yet, so the linker discards everything from Ebml.a. Next if > examines Matroska.a and needs all those symbols from Ebml.a, but those > have already been discarded. The solution should be to swap Ebml.a and > Matroska.a. And the result is... Exact! GCC (MinGW32) is surprising, only EbmlSInteger, not others... I don't want to know why ;-) From steve.lhomme at free.fr Sat Oct 22 15:22:26 2005 From: steve.lhomme at free.fr (Steve Lhomme) Date: Sat, 22 Oct 2005 15:22:26 +0200 Subject: [Matroska-devel] Re: Difficulties to link libebml with DevCpp In-Reply-To: <4358AD4A.4070004@aaaa.com> References: <4356B925.5040505@aaaa.com> <200510200856.26619.moritz@bunkus.org> <4358AD4A.4070004@aaaa.com> Message-ID: <435A3D12.2040900@free.fr> Zen wrote: > Moritz Bunkus wrote: > >> Hey, >> >> On Wednesday 19 October 2005 23:22, Zen wrote: >> >>> The project files are out of date (if you are interested, I can send you >>> updated files), >> >> >> That would be nice. > > > I sent to your private adress the files... > But please : is it possible for libebml to stop to have an exception for > WinIOCallback? > At least for the .h file, because all .h are in the same directory, > except ONE, it is always forgot... > For .cpp, maybe a #ifdef _WIN32 for all file would be better than to > have a specific directory... Well, the IO classes are just helpers. Usually each app create it own IO class as a child of our virtual class. >> Usually linkers care about the order in which you specify the >> libraries. It tries to resolve all unknown symbols from all the >> following libraries and discards all unneeded symbols from the current >> library. So when it reaches Ebml.a the stuff from Matroska.a hasn't been >> found yet, so the linker discards everything from Ebml.a. Next if >> examines Matroska.a and needs all those symbols from Ebml.a, but those >> have already been discarded. The solution should be to swap Ebml.a and >> Matroska.a. > > > And the result is... Exact! > GCC (MinGW32) is surprising, only EbmlSInteger, not others... I don't > want to know why ;-) Yeah, too many people think gcc is a really good compiler ;) From ivanburnin at hushmail.com Sun Oct 23 07:28:14 2005 From: ivanburnin at hushmail.com (ivanburnin at hushmail.com) Date: Sat, 22 Oct 2005 22:28:14 -0700 Subject: [Matroska-devel] Question Message-ID: <20051023052820.1CD3933C23@mailserver5.hushmail.com> I haven't seen these in anything yet, but I admit I haven't looked all that deeply. I know Matroska is capable of having multiple audio channels, and multiple video channels. Is there currently any way to create a scripting mixing board for them (this allows overriding, and can handle multiple scripts for multiple purposes), as well as hopefully a 3D sound positioning mechanism (again scriptable). I'm thinking about something like a Karaoke situation, where storing both the background only and full audio versions would require nearly double the space of background only & speex compressed voice. Alternately a situational performance where for example in a Jazz combo at may be desirable to bring out the trumpet, but not lose the original data, allowing a great deal of flexibility in playback. With a suitable amplification/speaker environment it would be possible to do a great many very interesting things with this, and with live recordings the flexibility could open many more possibilities. So my question is: Is there a positioning and mixing script facility available? And assuming there isn't, is there any other interest in seeing one? Concerned about your privacy? Instantly send FREE secure email, no account required http://www.hushmail.com/send?l=480 Secure your website with a Hush SSL certificate from https://www.hushssl.com?l=485 From christian at matroska.org Mon Oct 24 03:54:28 2005 From: christian at matroska.org (ChristianHJW) Date: Mon, 24 Oct 2005 03:54:28 +0200 Subject: [Matroska-devel] Re: Mastroska muxer 0.9 port In-Reply-To: <1129283936.13509.2.camel@localhost.localdomain> References: <1129283936.13509.2.camel@localhost.localdomain> Message-ID: <435C3ED4.6020605@matroska.org> Hi Michal, please note that a new type of matroska file has been introduced recently, called 'matroska v2' . It will ad a new block element called 'simple block', allowing to create MKV files very easily and with even lower overhead. Specs are here : http://www.matroska.org/technical/specs/index.html#SimpleBlock . Could you please care about adding support for this in the gstreamer matroska demuxer, by time ? Good news is, the amount of extra code needed for that is almost neglectible, as changes are minimal . You will have about 4 weeks to do that still, because then Mosu will release a new mkvtoolnix version with simple block writing capabilities :) .... Thanks Christian matroska project admin http://www.matroska.org Michal Benes schrieb: > Hello, > > I have ported Matroska muxer to 0.9. The patch is available in bugzilla > as bug #318847 > http://bugzilla.gnome.org/show_bug.cgi?id=318847 > > I have tried to follow all the guidelines (added many missing docs, > tested by valgrind, used the version from 0.8 branch, ...) > > Cheers, > > Michal > > > > ------------------------------------------------------- > This SF.Net email is sponsored by: > Power Architecture Resource Center: Free content, downloads, discussions, > and more. http://solutions.newsforge.com/ibmarch.tmpl From christian at matroska.org Mon Oct 24 04:05:14 2005 From: christian at matroska.org (ChristianHJW) Date: Mon, 24 Oct 2005 04:05:14 +0200 Subject: [Matroska-devel] Re: Containers overhead In-Reply-To: <200509241555.39984.davidecapod@inwind.it> References: <200509241555.39984.davidecapod@inwind.it> Message-ID: <435C415A.7000402@matroska.org> Hi Davide, i assume these are bytes ? Pls. note that your comparison is not correct for such very small file sizes, because mkvmerge will by default reserve a few KB in the header for tags, so that these can be written into the file afterwards with our tag editor, without having to remux the complete file. In your case, this is totally wasted space. If you create such small files, you should deactivate that with the right switch in mkvmerge ( ??no?tags ) . Also, it may happen that our file header structure will not allow very small overheads on such untypically small files of just 2 MB. Retry with real video file sizes, like 50 - 1000 MB, and you'll see the difference. Christian matroska project admin http://www.matroska.org Davide Capodaglio schrieb: > Sorry if maybe a little OT... > I read an article about Matroska container having the less overhead for > muxing, so I tried some experiments using a small video track containing a > h264 video and an mp3 audio, but I found out that mp4 has an even less > overhead. > > > 2333916 set 20 23:39 264.avi > > avi2raw --video 264.avi 264.264 > avi2raw --audio 264.avi 264.mp3 > > 1847349 set 24 15:44 264.264 > 473710 set 24 15:44 264.mp3 > > > MP4Box -add 264.264 -fps 10 -add 264.mp3 264.mp4 > > 2325001 set 24 15:46 264.mp4 > > > MP4Box -add 264.264 -fps 10 264.mp4 > mkvmerge -o 264.mkv 264.mp4 264.mp3 > > 2331111 set 24 15:47 264.mkv > > > mkvmerge -o 264_avi.mkv --engage allow_avc_in_vfw_mode 264.avi > > 2331484 set 24 15:52 264_avi.mkv > > > 264+mp3 = 2321059 > avi: 2333916-2321059 = 12857 > mp4: 2325001-2321059 = 3942 > mkv: 2331111-2321059 = 10052 > mkv (avi): 2331484-2321059 = 10425 > > > So the .mp4 seem to be the one with MUCH LESS overhead than the others, even > that Matroska... > Is my test right?? > > Thanks & bye > Davide > > From michal.benes at xeris.cz Mon Oct 24 16:40:51 2005 From: michal.benes at xeris.cz (Michal Benes) Date: Mon, 24 Oct 2005 16:40:51 +0200 Subject: [Matroska-devel] Re: [gst-devel] Re: Mastroska muxer 0.9 port In-Reply-To: <435C3ED4.6020605@matroska.org> References: <1129283936.13509.2.camel@localhost.localdomain> <435C3ED4.6020605@matroska.org> Message-ID: <1130164851.7417.6.camel@localhost.localdomain> Hi Christian, thank you for the information. These are good news. I will implement this feature this week to both muxer and demuxer. Is there a list of the changes in matroska v2 somewhere or is the 'simple block' the only one? Cheers, Michal ChristianHJW p??e v Po 24. 10. 2005 v 03:54 +0200: > > Hi Michal, > > please note that a new type of matroska file has been introduced > recently, called 'matroska v2' . It will ad a new block element called > 'simple block', allowing to create MKV files very easily and with even > lower overhead. > > Specs are here : > http://www.matroska.org/technical/specs/index.html#SimpleBlock . > > Could you please care about adding support for this in the gstreamer > matroska demuxer, by time ? Good news is, the amount of extra code > needed for that is almost neglectible, as changes are minimal . > > You will have about 4 weeks to do that still, because then Mosu will > release a new mkvtoolnix version with simple block writing capabilities > :) .... > > Thanks > > Christian > matroska project admin > http://www.matroska.org > > > Michal Benes schrieb: > > > Hello, > > > > I have ported Matroska muxer to 0.9. The patch is available in bugzilla > > as bug #318847 > > http://bugzilla.gnome.org/show_bug.cgi?id=318847 > > > > I have tried to follow all the guidelines (added many missing docs, > > tested by valgrind, used the version from 0.8 branch, ...) > > > > Cheers, > > > > Michal > > > > > > > > ------------------------------------------------------- > > This SF.Net email is sponsored by: > > Power Architecture Resource Center: Free content, downloads, discussions, > > and more. http://solutions.newsforge.com/ibmarch.tmpl > > > > ------------------------------------------------------- > This SF.Net email is sponsored by the JBoss Inc. > Get Certified Today * Register for a JBoss Training Course > Free Certification Exam for All Training Attendees Through End of 2005 > Visit http://www.jboss.com/services/certification for more information > _______________________________________________ > gstreamer-devel mailing list > gstreamer-devel at lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/gstreamer-devel From a.bromwich at gmail.com Tue Oct 25 06:04:25 2005 From: a.bromwich at gmail.com (Andrew Bromwich) Date: Tue, 25 Oct 2005 14:04:25 +1000 Subject: [Matroska-devel] Re: Matroska video blocking problem In-Reply-To: <435672AF.6020607@free.fr> References: <435672AF.6020607@free.fr> Message-ID: Hi All, > It sounds like your file only "indexes" the first video track and not the other one. I do not "index" any of the streams with my code explicitly so does Matroska just index the first video stream if this is the case?? I have tried to find some examples of how to setup indexing properly using Matroska but i haven't had any luck. Would someone be able to post an example or point me in the direction of one? > Maybe you're missing a command-line to do that or it's a bug in the muxing program (mkvmerge ?) I have created a "recording software" that takes multiple video and audio streams and puts them into a matroska A/V file. Is this where is indexing must be done?? Also i have noticed that when using GraphEdit to render a file using Haali's splitter (using the "Render Media File" option in the File menu) only pins for the first video and audio streams are exposed? I havent tried creating this graph in code but is this a limitation of the filter?? Thanks, Andrew On 10/20/05, Steve Lhomme wrote: > > Hi, > > I forward this email to our ML where Haali (who makes our splitter) will > be able to read it. It sounds like your file only "indexes" the first > video track and not the other one. Matroska (the format) can support > index on all tracks, even partial index (called Cue). Maybe you're > missing a command-line to do that or it's a bug in the muxing program > (mkvmerge ?) or Haali's splitter that uses the wrong index for this track. > > Andrew Bromwich wrote: > > Hi Steve, > > > > I have been having some problems with what i believe is with the > > Matroska splitter. I wasnt too sure if posting a bug on the corecodec > > wesite was the best option or emailing you (so i just emailed you, sorry > > if this is incorrect). > > > > The problem I am having is that I have some video files, with a number > > of video streams (normally 4 streams) and an audio stream, but when the > > video streams are seeked through i get a large amount of blocking (or > > periods of video freezing depending on video decompressor) on the 2nd, > > 3rd and 4th stream. I did quite a lot of testing with a number of > > different compressors and decompressors but with the same results. The > > first screen always seems to work great which is really what is throwing > > me off! I have not personally really got into the insides of the > > Matroska splitter or muxer but am guessing that it might be some sort of > > buffer that has been optimised for single video stream playback?? This > > problem wouldnt be so much of an issue but the software I am developing > > requires a fair bit of skipping through the video file, then every time > > i seek it either blocks up or freezes altogether until the next > > key-frame comes along. > > > > I have also tried a fair amount of testing using a number of different > > settings with the video compressors i have been using (w.r.t keyframe > > rates) but this really doesnt seem to make much difference to the > > outcome either. > > > > I do appologise if i have emailed the wrong person (if this is the case > > can you please point me in the direction of someone who might be able to > > help). > > > > Thankyou, > > > > Andrew > > -- > robUx4 on blog > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mike at po.cs.msu.su Tue Oct 25 11:45:01 2005 From: mike at po.cs.msu.su (Mike Matsnev) Date: Tue, 25 Oct 2005 13:45:01 +0400 Subject: [Matroska-devel] Re: Matroska video blocking problem In-Reply-To: References: <435672AF.6020607@free.fr> Message-ID: <20051025094501.GA69074@azog.po.cs.msu.su> On Tue, Oct 25, 2005 at 02:04:25PM +1000, Andrew Bromwich wrote: > > It sounds like your file only "indexes" the first video track and not the > other one. > > I do not "index" any of the streams with my code explicitly so does Matroska > just index the first video stream if this is the case?? I have tried to find > some examples of how to setup indexing properly using Matroska but i haven't > had any luck. Would someone be able to post an example or point me in the > direction of one? > > > Maybe you're missing a command-line to do that or it's a bug in the muxing > program (mkvmerge ?) > > I have created a "recording software" that takes multiple video and audio > streams and puts them into a matroska A/V file. Is this where is indexing > must be done?? Yes. But my filter does not require an index. I did some testing and could not reproduce your problem. My guess is you are not writing keyframe flags properly (references in matroska). > Also i have noticed that when using GraphEdit to render a file using Haali's > splitter (using the "Render Media File" option in the File menu) only pins > for the first video and audio streams are exposed? I havent tried creating > this graph in code but is this a limitation of the filter?? Yes. Other streams can be selected via tray icon context menu or programmatically with IAMStreamSelect interface. /Mike