From chris at matroska.org  Tue Mar  1 21:05:19 2005
From: chris at matroska.org (Christian HJ Wiesner)
Date: Tue, 01 Mar 2005 21:05:19 +0100
Subject: [Matroska-devel] [Fwd: Second Annual Free Software Contest OPENNED]
Message-ID: <4224CAFF.9040500@matroska.org>

FYI

-------- Original-Nachricht --------
Betreff: 	Second Annual Free Software Contest OPENNED
Datum: 	Tue, 1 Mar 2005 16:45:28 +0100
Von: 	Christophe Uzan <cuzan at le-public-systeme.fr>
An: 	<chris at matroska.org>



 

*Matroska***

Second Annual Free Software Awards

www.tropheesdulibre.org

For the 2nd edition of "Troph?es du Libre", Soissons Informatique Libre 
- European Free Software Transfer and Research Center, invite you in 
Soissons, France, on Thursday, May the 26th 2005.

"Free Software Trophies" goals are to give credit and reward the work of 
individuals, companies and institutions which contribute to development 
and enhancement of Free and Open Source software all around the world.

This year the ? Troph?es du Libre ? will award projects from six categories:

-  Security
-  Applications for public infrastructures and for communities
-  Companies management
-  Education
-  Multimedia / Games
-  Mobility
-  Embedded Systems.

For the first time, the event, partnering with Nexen Services, the Open 
Source hosting specialist, will award a *Special Price for PHP Projects*.

A jury, directed by David Axmark, co-founder of MySQL AB, will evaluate 
the projects. The members of the jury will be famous personalities of 
the open source community, including:

-  Rasmus Lerdorf, creator of PHP,
-  Edouard Gomez, Xvid Project
-  Ben Laurie, creator of Open SSL and representative for the Apache 
Software Foundation,
-  Nat Makarevitch, major player who promotes the Open Source software 
in the French speaking community

The jury will also include some professional users and contributors such 
as Laurent Bloch, information security officer at the INSERM and Patrick 
de Carn? DSI at the Paris 5 University.

The Awards ceremony will take place in Soissons on May 26, 2005. The 
price is an encouragement for the creators, and a quality seal for the 
users. Additionally, the event sponsors and partners will hand over 
other prices.

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.matroska.org/pipermail/matroska-devel/attachments/20050301/d587416f/attachment.html>

From rvmarce at free.fr  Fri Mar  4 10:07:29 2005
From: rvmarce at free.fr (herve marce)
Date: Fri, 4 Mar 2005 10:07:29 +0100
Subject: [Matroska-devel] gestion des drm
Message-ID: <2fb96a6ab4f18baf0e62d6e445126298@free.fr>

Bonjour
D?sol? de m'exprimer en fran?ais sur une mailing list qui est 
certainement en anglais, mais mon anglais n'est pas tr?s bon.
Est il pr?vu dans Matroska une gestion des DRM, j'ai un projet de 
proposer sur le net des vid?os de producteurs ind?pendant mais malgr? 
que cela soit des ind?pendants, il serait souhaitable de pouvoir 
prot?ger un minimum les oeuvres. Je suis donc ? la recherche d'une 
solution alternative que celle propos? par les grands groupes tels que 
Microsoft.
Merci pour vos r?ponses



try i translate

Hello

I'm sorry I write in french, but my english is not very good.
  it is envisaged in matroska a Digital Rights Management, I'm a project 
to propse on net movies of independant producers, but  although that is 
the independent it would be desirable to be able to protect a minimum 
works.  I search a alternaive solution that proposed by the great 
groups such as Microsoft.

Thank's for your answer





From trykos at mail.ru  Wed Mar  9 10:00:00 2005
From: trykos at mail.ru (=?koi8-r?Q?=EB=CF=CE=D3=D4=C1=CE=D4=C9=CE=20?=)
Date: Wed, 09 Mar 2005 12:00:00 +0300
Subject: [Matroska-devel] Matroska WWW SVN down again ?
Message-ID: <E1D8x2u-000Nqb-00.trykos-mail-ru@f37.mail.ru>


Sorry to bring it up again, but i just cant get anything from http://svn.matroska.org/svn/www/
No error, but no reaction (browser just shows "Loading" msg infinite).

Best regards,
 Konstantin Troubine aka Takuto


From steve.lhomme at free.fr  Wed Mar  9 11:07:10 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Wed, 09 Mar 2005 11:07:10 +0100
Subject: [Matroska-devel] Matroska WWW SVN down again ?
In-Reply-To: <E1D8x2u-000Nqb-00.trykos-mail-ru@f37.mail.ru>
References: <E1D8x2u-000Nqb-00.trykos-mail-ru@f37.mail.ru>
Message-ID: <422ECACE.60408@free.fr>

Yes,

As I said, we have problems only with this repository and we don't know 
why. It's fixed now. But next time we are going to have a deeper look at 
what the pb is...

?????????? a ?crit :
> Sorry to bring it up again, but i just cant get anything from http://svn.matroska.org/svn/www/
> No error, but no reaction (browser just shows "Loading" msg infinite).
> 
> Best regards,
>  Konstantin Troubine aka Takuto
> _______________________________________________
> Matroska-devel mailing list
> Matroska-devel at lists.matroska.org
> http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-devel

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From steve.lhomme at free.fr  Fri Mar 11 14:54:09 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Fri, 11 Mar 2005 14:54:09 +0100
Subject: [Matroska-devel] Segment hard/soft/medium linking
Message-ID: <4231A301.7070706@free.fr>

Hi,

As I'm adding support for more complex stuff in VLC, I encountered a pb 
with segment linking and the demos of Haali.

Haali uses tags to define what segments should be played and when. I 
didn't look precisely at the content of his tags, because this is not 
the way to do it anyway. In short, as I said on IRC, "tags are not meant 
to influence the content, especially the way it's played". Matroska is 
modular and tags are not a part that is used for playback.

Nevertheless, we found a grey area in the specs and some use case might 
not be covered by our current system. So let's find out what and clarify 
things for everyone.

Segment linking is a way to virtually link some segment so that when 
playback one, you can easily reference the other. So far we have the 
following :

- hard linking, that's when you split a segment. The resulting segments 
have continuous timecodes. They are linked using PrevUID and NextUID. In 
the end, on playback the user should see them as just one (the original 
unsplit). Chapters only reference content from the segment they are in 
(so segments should be merged on playback).

Hard linking is somehow like .VOB files but in a more liberal way.

Mosu and alexnoe, could you confirm that splitting works this way in 
your software ?

- soft linking, that was recently introduced for the DVD domains. Each 
chapter is related to the other because they share the same family UID. 
The timecodes of each segment are independant. That means each segment 
is standalone, but a chapter codec could reference a segment in the same 
family, to jump there. This reference is done with the ChapterTranslate 
table in the segment Info (translate a codec ID to a segment UID).

Now what Haali is trying to do is somehow different. He wants segments 
to have the following playback scheme :
intro -> ep1 -> outro -> intro -> ep2 -> outro

- I will call that medium linking as it needs features from hard and 
soft linking. The timecode of the outro has to be independant of the 
episode. That means hard linking is not possible. Also the NextUID of 
the intro and the PrevUID of the outro should not be set because they 
can only reference one item. It could be done with soft linking but we 
would miss a key feature : each intro+epX+outro would not be seen as one 
item. That means the total duration would not be specified and the 
tracks might differ.

So we might need an additional system for medium linking.

I'll think about it. But until then, comments/ideas are welcome.
They are always welcome, anyway..

-- 
robUx4 on blog <http://robux4.blogspot.com/>


From moritz at bunkus.org  Fri Mar 11 14:58:48 2005
From: moritz at bunkus.org (Moritz Bunkus)
Date: Fri, 11 Mar 2005 14:58:48 +0100
Subject: [Matroska-devel] Segment hard/soft/medium linking
In-Reply-To: <4231A301.7070706@free.fr>
References: <4231A301.7070706@free.fr>
Message-ID: <200503111458.52353.moritz@bunkus.org>

Hey,

> - hard linking, that's when you split a segment. The resulting segments
> have continuous timecodes. They are linked using PrevUID and NextUID. In
> the end, on playback the user should see them as just one (the original
> unsplit). Chapters only reference content from the segment they are in
> (so segments should be merged on playback).
>
> Hard linking is somehow like .VOB files but in a more liberal way.
>
> Mosu and alexnoe, could you confirm that splitting works this way in
> your software ?

"yes" for mkvmerge.

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: <http://lists.matroska.org/pipermail/matroska-devel/attachments/20050311/2c7f070a/attachment.pgp>

From playful at gmx.at  Fri Mar 11 14:55:13 2005
From: playful at gmx.at (filewalker)
Date: Fri, 11 Mar 2005 13:55:13 +0000 (UTC)
Subject: [Matroska-devel] Haali's parser: feature request for quicktime
	support
Message-ID: <loom.20050311T144349-221@post.gmane.org>

Hi Mike,

thanks for the new features(support for MP4) in your parser!!
(I kicked out now the Nero's one :) ) works great!

Your parser makes it now possible to use only non comercial filters(fffshow, 
your parser) for MP4, too. THANKS!

My feature request is:
Is it possible to support the quicktime formats...because the mp4 format is 
pretty much like the same as quicktime, so I thought that it maybe wouldn't be 
too hard for you to add support for the quicktime formats in your parser?

It would be really  great to have only one good parser, that rocks! :)


Cu filewalker





From mike at po.cs.msu.su  Fri Mar 11 15:19:29 2005
From: mike at po.cs.msu.su (Mike Matsnev)
Date: Fri, 11 Mar 2005 17:19:29 +0300
Subject: [Matroska-devel] Haali's parser: feature request for quicktime
	support
In-Reply-To: <loom.20050311T144349-221@post.gmane.org>
References: <loom.20050311T144349-221@post.gmane.org>
Message-ID: <4231A8F1.3090708@po.cs.msu.su>

filewalker wrote:
> My feature request is:
> Is it possible to support the quicktime formats...because the mp4 format is 
> pretty much like the same as quicktime, so I thought that it maybe wouldn't be 
> too hard for you to add support for the quicktime formats in your parser?
I have no plans to add quicktime support at this time. Maybe later if I get
reasonable docs.

/Mike


From office at digitalrise.biz  Sat Mar 12 14:23:19 2005
From: office at digitalrise.biz (DigitalRise)
Date: Sat, 12 Mar 2005 14:23:19 +0100
Subject: [Matroska-devel] Matroska DVD's
Message-ID: <009d01c52706$a80b0a30$1a00a8c0@marel.local>

Hallo,

Wir planen momentan die Entwicklung eines neuen DVD Players und wollten deshalb fragen, wie es mit "MKV DVD's" aussieht? D.h.

1. Gibt es eine offizielle DVD Filestruktur wie bei den herk?mmlichen DVD's oder auch bei WMV HD DVD's (http://go.microsoft.com/fwlink/?LinkId=41935)?

2. Gibt es ein Zertifikat f?r DVD Player a la "MKV certified"?

3. Wenn es ein Zertifikat gibt, gibt es dann auch Zertifizierungs-DVD's die laufen m?ssen, um das Zertifikat zu erhalten? Wenn ja, wo bekommt man diese?

mfg
Martin Jussel
www.DigitalRise.biz
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.matroska.org/pipermail/matroska-devel/attachments/20050312/da624efd/attachment.html>

From office at digitalrise.biz  Sat Mar 12 15:29:00 2005
From: office at digitalrise.biz (DigitalRise)
Date: Sat, 12 Mar 2005 15:29:00 +0100
Subject: [Matroska-devel] MKV DVD structure & player certification?
Message-ID: <00bc01c5270f$d51d5830$1a00a8c0@marel.local>

Hi,

We are planning a new DVD player and want to know some points about "MKV DVD's". This means:

1. Is there an official MKV DVD structure defined like for Standard-DVD's or WMV HD DVD's (http://go.microsoft.com/fwlink/?LinkId=41935)?

2. A certificate for DVD Players a la "MKV certified" is available?

3. If a certificate is available, how to get it?

Regards,
Martin
www.DigitalRise.biz
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.matroska.org/pipermail/matroska-devel/attachments/20050312/682751d3/attachment.html>

From steve.lhomme at free.fr  Sun Mar 13 11:20:51 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Sun, 13 Mar 2005 11:20:51 +0100
Subject: [Matroska-devel] MKV DVD structure & player certification?
In-Reply-To: <00bc01c5270f$d51d5830$1a00a8c0@marel.local>
References: <00bc01c5270f$d51d5830$1a00a8c0@marel.local>
Message-ID: <42341403.8050708@free.fr>

DigitalRise a ?crit :
> Hi,

Hi,

> We are planning a new DVD player and want to know some points about "MKV 
> DVD's". This means:

Great idea !

> 1. Is there an official MKV DVD structure defined like for 
> Standard-DVD's or WMV HD DVD's 
> (http://go.microsoft.com/fwlink/?LinkId=41935)?

Nop, we don't have such a thing. The Matroska DVDs would probably be 
like DivX CDRs or DVDRs. The user is free to put his files in any 
directory. Non .mkv / .mka files should be ignored (unless they are 
other file types you support).

> 2. A certificate for DVD Players a la "MKV certified" is available?

Not yet. We are going to redefine (or refine) the hardware profiles we 
created a long time ago. They will include the list of codec to support, 
the list of features to support, the resolutions, max bitrate, etc.

> 3. If a certificate is available, how to get it?

We'll create some test files for each profile. But we don't have any 
"official" certification planned. It might come when we define the logo 
to put on hardware devices (like yours).

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From office at digitalrise.biz  Sun Mar 13 12:12:20 2005
From: office at digitalrise.biz (DigitalRise)
Date: Sun, 13 Mar 2005 12:12:20 +0100
Subject: [Matroska-devel] MKV DVD structure & player certification?
References: <00bc01c5270f$d51d5830$1a00a8c0@marel.local>
	<42341403.8050708@free.fr>
Message-ID: <002401c527bd$86b6b430$1a00a8c0@marel.local>

Hi,

No DVD structure? Don't forget the PC users. Most of the PC users are not
freaks. So the structure should contain a directory with the needed playback
filters and an autorun.inf which check/install the needed filters and start
the playback. So minimum requirement for a MKV DVD should be the usage on
"Pure Windows" with no filters installed.

For standalone players a content.xml in the main directory should be also
available which contains information about the files on the disc. These file
can be read from the player to take the right action (autoplay or show the
content for selection).

My suggestion for the content.xml / player requirements:

+ if the content.xml (the DVD) contains more than one file, the player has
to display the content list. Otherwise the single file is played immediate.

+ each file definition contains:
    - short description
    - description
    - length
    - audio / video field (the file contains only audio, only video or both)
    - video resolution (maybe interested for the player?)
    - video aspect ratio (also interested for the player?)
    - audio information (num. of channels?)
    - additional information (interlaced, progressive, etc for the player)
    - etc

The file definitions from content.xml are displayed if the DVD is inserted
and more than one files are on the DVD. So the user can check the
description, see the length of the film, etc and play the stuff he wants.
Also a player should display these information via OSD when user calling the
"MKV container information" while playback.

What do you think? A DVD structure and/or the content.xml is not an useful
thing?

Martin


----- Original Message ----- 
From: "Steve Lhomme" <steve.lhomme at free.fr>
To: "Discussion about the current and future development of Matroska"
<matroska-devel at lists.matroska.org>; <office at digitalrise.biz>
Sent: Sunday, March 13, 2005 11:20 AM
Subject: Re: [Matroska-devel] MKV DVD structure & player certification?


DigitalRise a ?crit :
> Hi,

Hi,

> We are planning a new DVD player and want to know some points about "MKV
> DVD's". This means:

Great idea !

> 1. Is there an official MKV DVD structure defined like for
> Standard-DVD's or WMV HD DVD's
> (http://go.microsoft.com/fwlink/?LinkId=41935)?

Nop, we don't have such a thing. The Matroska DVDs would probably be
like DivX CDRs or DVDRs. The user is free to put his files in any
directory. Non .mkv / .mka files should be ignored (unless they are
other file types you support).

> 2. A certificate for DVD Players a la "MKV certified" is available?

Not yet. We are going to redefine (or refine) the hardware profiles we
created a long time ago. They will include the list of codec to support,
the list of features to support, the resolutions, max bitrate, etc.

> 3. If a certificate is available, how to get it?

We'll create some test files for each profile. But we don't have any
"official" certification planned. It might come when we define the logo
to put on hardware devices (like yours).

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From steve.lhomme at free.fr  Sun Mar 13 14:19:48 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Sun, 13 Mar 2005 14:19:48 +0100
Subject: [Matroska-devel] MKV DVD structure & player certification?
In-Reply-To: <002401c527bd$86b6b430$1a00a8c0@marel.local>
References: <00bc01c5270f$d51d5830$1a00a8c0@marel.local>
	<42341403.8050708@free.fr>
	<002401c527bd$86b6b430$1a00a8c0@marel.local>
Message-ID: <42343DF4.9050101@free.fr>

DigitalRise a ?crit :
> Hi,
> 
> No DVD structure? Don't forget the PC users. Most of the PC users are not
> freaks. So the structure should contain a directory with the needed playback
> filters and an autorun.inf which check/install the needed filters and start
> the playback. So minimum requirement for a MKV DVD should be the usage on
> "Pure Windows" with no filters installed.

We don't intend to playable "just" on Windows. Matroska files can 
already read on Windows, Linux, OS X, BeOS and PocketPC. So we should 
not create a directory for each possible OS in the world. DVDs usually 
don't do that. And if we ever need to do something like this, it will be 
cross-platform.

> For standalone players a content.xml in the main directory should be also
> available which contains information about the files on the disc. These file
> can be read from the player to take the right action (autoplay or show the
> content for selection).

That is a good idea. But then it should not be restricted to Matroska 
(AVI, MP4, MPG, etc).

> My suggestion for the content.xml / player requirements:
> 
> + if the content.xml (the DVD) contains more than one file, the player has
> to display the content list. Otherwise the single file is played immediate.
> 
> + each file definition contains:
>     - short description
>     - description
>     - length
>     - audio / video field (the file contains only audio, only video or both)
>     - video resolution (maybe interested for the player?)
>     - video aspect ratio (also interested for the player?)
>     - audio information (num. of channels?)
>     - additional information (interlaced, progressive, etc for the player)
>     - etc

Yes, that sounds like a good idea. But in the end the player doesn't 
know if it can play the file or not. Only scanning the file with the 
embedded software can make sure it could work (codec & container used). 
So maybe if you add these info (codecs, container, number of segments, 
target profile) it would give a good idea if the player can read it or not.

> The file definitions from content.xml are displayed if the DVD is inserted
> and more than one files are on the DVD. So the user can check the
> description, see the length of the film, etc and play the stuff he wants.
> Also a player should display these information via OSD when user calling the
> "MKV container information" while playback.
> 
> What do you think? A DVD structure and/or the content.xml is not an useful
> thing?

Yes it is !!! But it's almost a project on its own. IMO this format 
should be flexible enough to support all known/main containers... I 
don't have the time to work on this right now. But if you want to start 
it, I'll be glad to contribute ideas and comments.

> Martin
> 
> 
> ----- Original Message ----- 
> From: "Steve Lhomme" <steve.lhomme at free.fr>
> To: "Discussion about the current and future development of Matroska"
> <matroska-devel at lists.matroska.org>; <office at digitalrise.biz>
> Sent: Sunday, March 13, 2005 11:20 AM
> Subject: Re: [Matroska-devel] MKV DVD structure & player certification?
> 
> 
> DigitalRise a ?crit :
> 
>>Hi,
> 
> 
> Hi,
> 
> 
>>We are planning a new DVD player and want to know some points about "MKV
>>DVD's". This means:
> 
> 
> Great idea !
> 
> 
>>1. Is there an official MKV DVD structure defined like for
>>Standard-DVD's or WMV HD DVD's
>>(http://go.microsoft.com/fwlink/?LinkId=41935)?
> 
> 
> Nop, we don't have such a thing. The Matroska DVDs would probably be
> like DivX CDRs or DVDRs. The user is free to put his files in any
> directory. Non .mkv / .mka files should be ignored (unless they are
> other file types you support).
> 
> 
>>2. A certificate for DVD Players a la "MKV certified" is available?
> 
> 
> Not yet. We are going to redefine (or refine) the hardware profiles we
> created a long time ago. They will include the list of codec to support,
> the list of features to support, the resolutions, max bitrate, etc.
> 
> 
>>3. If a certificate is available, how to get it?
> 
> 
> We'll create some test files for each profile. But we don't have any
> "official" certification planned. It might come when we define the logo
> to put on hardware devices (like yours).
> 

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From steve.lhomme at free.fr  Sun Mar 13 21:04:31 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Sun, 13 Mar 2005 21:04:31 +0100
Subject: [Matroska-devel] Segment hard/soft/medium linking
In-Reply-To: <4231A301.7070706@free.fr>
References: <4231A301.7070706@free.fr>
Message-ID: <42349CCF.2050908@free.fr>

utiliser ChapterSegmentUID (permet de lier ? la m?me ?dition ou aucune 
si l'?dition n'est pas trouv?e).

Steve Lhomme a ?crit :
> Hi,
> 
> As I'm adding support for more complex stuff in VLC, I encountered a pb 
> with segment linking and the demos of Haali.
> 
> Haali uses tags to define what segments should be played and when. I 
> didn't look precisely at the content of his tags, because this is not 
> the way to do it anyway. In short, as I said on IRC, "tags are not meant 
> to influence the content, especially the way it's played". Matroska is 
> modular and tags are not a part that is used for playback.
> 
> Nevertheless, we found a grey area in the specs and some use case might 
> not be covered by our current system. So let's find out what and clarify 
> things for everyone.
> 
> Segment linking is a way to virtually link some segment so that when 
> playback one, you can easily reference the other. So far we have the 
> following :
> 
> - hard linking, that's when you split a segment. The resulting segments 
> have continuous timecodes. They are linked using PrevUID and NextUID. In 
> the end, on playback the user should see them as just one (the original 
> unsplit). Chapters only reference content from the segment they are in 
> (so segments should be merged on playback).
> 
> Hard linking is somehow like .VOB files but in a more liberal way.
> 
> Mosu and alexnoe, could you confirm that splitting works this way in 
> your software ?
> 
> - soft linking, that was recently introduced for the DVD domains. Each 
> chapter is related to the other because they share the same family UID. 
> The timecodes of each segment are independant. That means each segment 
> is standalone, but a chapter codec could reference a segment in the same 
> family, to jump there. This reference is done with the ChapterTranslate 
> table in the segment Info (translate a codec ID to a segment UID).
> 
> Now what Haali is trying to do is somehow different. He wants segments 
> to have the following playback scheme :
> intro -> ep1 -> outro -> intro -> ep2 -> outro
> 
> - I will call that medium linking as it needs features from hard and 
> soft linking. The timecode of the outro has to be independant of the 
> episode. That means hard linking is not possible. Also the NextUID of 
> the intro and the PrevUID of the outro should not be set because they 
> can only reference one item. It could be done with soft linking but we 
> would miss a key feature : each intro+epX+outro would not be seen as one 
> item. That means the total duration would not be specified and the 
> tracks might differ.
> 
> So we might need an additional system for medium linking.
> 
> I'll think about it. But until then, comments/ideas are welcome.
> They are always welcome, anyway..
> 

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From playful at gmx.at  Mon Mar 14 10:54:13 2005
From: playful at gmx.at (filewalker)
Date: Mon, 14 Mar 2005 09:54:13 +0000 (UTC)
Subject: [Matroska-devel] Re: Haali's parser: feature request for quicktime
	support
References: <loom.20050311T144349-221@post.gmane.org>
	<4231A8F1.3090708@po.cs.msu.su>
Message-ID: <loom.20050314T104601-511@post.gmane.org>

Mike Matsnev <mike <at> po.cs.msu.su> writes:


> I have no plans to add quicktime support at this time. Maybe later if I get
> reasonable docs.

It's a real pitty because such a filter is really missing. :(
IIRC I read about it that "QuickTime is the same format as MP4, just with 
additional formats (Sorenson and such)".

Cu filewalker
> 
> /Mike
> 






From steve.lhomme at free.fr  Mon Mar 14 11:39:55 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Mon, 14 Mar 2005 11:39:55 +0100
Subject: [Matroska-devel] Segment hard/soft/medium linking
In-Reply-To: <4231A301.7070706@free.fr>
References: <4231A301.7070706@free.fr>
Message-ID: <423569FB.3010900@free.fr>

I have added a new element in the specs to handle medium linking. It's 
"ChapterSegmentUID" which is a segment UID that takes place of a dummy 
chapter. So that the content of that segment is played in place of that 
chapter. There is still a pending problem with that, since start/end 
time shouldn't apply here (to allow changing them independantly of other 
segments). But so far "ChapterTimeStart" is mandatory. This change is 
easy to do in the specs/code. But does it make sense and what is the 
impact ???

KaxChapterSegmentUID was added to libmatroska. An recompilation of 
mkvmerge with the SVN build of libmatroska should allow this from XML 
files directly.

Steve Lhomme a ?crit :
> Hi,
> 
> As I'm adding support for more complex stuff in VLC, I encountered a pb 
> with segment linking and the demos of Haali.
> 
> Haali uses tags to define what segments should be played and when. I 
> didn't look precisely at the content of his tags, because this is not 
> the way to do it anyway. In short, as I said on IRC, "tags are not meant 
> to influence the content, especially the way it's played". Matroska is 
> modular and tags are not a part that is used for playback.
> 
> Nevertheless, we found a grey area in the specs and some use case might 
> not be covered by our current system. So let's find out what and clarify 
> things for everyone.
> 
> Segment linking is a way to virtually link some segment so that when 
> playback one, you can easily reference the other. So far we have the 
> following :
> 
> - hard linking, that's when you split a segment. The resulting segments 
> have continuous timecodes. They are linked using PrevUID and NextUID. In 
> the end, on playback the user should see them as just one (the original 
> unsplit). Chapters only reference content from the segment they are in 
> (so segments should be merged on playback).
> 
> Hard linking is somehow like .VOB files but in a more liberal way.
> 
> Mosu and alexnoe, could you confirm that splitting works this way in 
> your software ?
> 
> - soft linking, that was recently introduced for the DVD domains. Each 
> chapter is related to the other because they share the same family UID. 
> The timecodes of each segment are independant. That means each segment 
> is standalone, but a chapter codec could reference a segment in the same 
> family, to jump there. This reference is done with the ChapterTranslate 
> table in the segment Info (translate a codec ID to a segment UID).
> 
> Now what Haali is trying to do is somehow different. He wants segments 
> to have the following playback scheme :
> intro -> ep1 -> outro -> intro -> ep2 -> outro
> 
> - I will call that medium linking as it needs features from hard and 
> soft linking. The timecode of the outro has to be independant of the 
> episode. That means hard linking is not possible. Also the NextUID of 
> the intro and the PrevUID of the outro should not be set because they 
> can only reference one item. It could be done with soft linking but we 
> would miss a key feature : each intro+epX+outro would not be seen as one 
> item. That means the total duration would not be specified and the 
> tracks might differ.
> 
> So we might need an additional system for medium linking.
> 
> I'll think about it. But until then, comments/ideas are welcome.
> They are always welcome, anyway..
> 

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From mike at po.cs.msu.su  Mon Mar 14 12:08:33 2005
From: mike at po.cs.msu.su (Mike Matsnev)
Date: Mon, 14 Mar 2005 14:08:33 +0300
Subject: [Matroska-devel] Segment hard/soft/medium linking
In-Reply-To: <423569FB.3010900@free.fr>
References: <4231A301.7070706@free.fr> <423569FB.3010900@free.fr>
Message-ID: <423570B1.6070807@po.cs.msu.su>

Steve Lhomme wrote:
> I have added a new element in the specs to handle medium linking. It's 
> "ChapterSegmentUID" which is a segment UID that takes place of a dummy 
> chapter. So that the content of that segment is played in place of that 
> chapter. There is still a pending problem with that, since start/end 
> time shouldn't apply here (to allow changing them independantly of other 
> segments). But so far "ChapterTimeStart" is mandatory. This change is 
> easy to do in the specs/code. But does it make sense and what is the 
> impact ???
Chapter times are essential there, because they allow pulling arbitrary
parts from other segments.

/Mike


From steve.lhomme at free.fr  Mon Mar 14 13:37:23 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Mon, 14 Mar 2005 13:37:23 +0100
Subject: [Matroska-devel] Segment hard/soft/medium linking
In-Reply-To: <423570B1.6070807@po.cs.msu.su>
References: <4231A301.7070706@free.fr> <423569FB.3010900@free.fr>
	<423570B1.6070807@po.cs.msu.su>
Message-ID: <42358583.9000006@free.fr>

Mike Matsnev a ?crit :
> Steve Lhomme wrote:
> 
>> I have added a new element in the specs to handle medium linking. It's 
>> "ChapterSegmentUID" which is a segment UID that takes place of a dummy 
>> chapter. So that the content of that segment is played in place of 
>> that chapter. There is still a pending problem with that, since 
>> start/end time shouldn't apply here (to allow changing them 
>> independantly of other segments). But so far "ChapterTimeStart" is 
>> mandatory. This change is easy to do in the specs/code. But does it 
>> make sense and what is the impact ???
> 
> Chapter times are essential there, because they allow pulling arbitrary
> parts from other segments.

Yes, but it breaks the modularity/object-oriented aspect of matroska. A 
segment shouldn't depend on other segments when playing back. At least 
not when it's supposed to be blind (like an intro/outro). Also you could 
edit the intro/outro without having to remux and update all the episodes 
(in your example).

Now the start time pb is not really an issue, we should just use 0, 
which means we start at the beggining of the other segment, then it's up 
to that segment to play how it prefers (using ordrered chapters for 
example).

> /Mike
> _______________________________________________
> Matroska-devel mailing list
> Matroska-devel at lists.matroska.org
> http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-devel

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From steve.lhomme at free.fr  Mon Mar 14 14:57:51 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Mon, 14 Mar 2005 14:57:51 +0100
Subject: [Matroska-devel] Profiles v2
Message-ID: <4235985F.2080106@free.fr>

Hi,

I just uploaded a draft of hardware profiles that we could use. Any 
comments are welcome, because it's an important part if manufacturers 
consider using Matroska.

http://www.matroska.org/technical/specs/profiles/new_draft.html

RealVideo was left out on purpose. Subs formats could be included too.
You can also see how important it is to have native formats, instead of 
any ACM/VfW codec you could think of.

-- 
robUx4 on blog <http://robux4.blogspot.com/>


From Liisachan at faireal.net  Mon Mar 14 15:59:56 2005
From: Liisachan at faireal.net (Liisachan)
Date: Mon, 14 Mar 2005 23:59:56 +0900
Subject: [Matroska-devel] Profiles v2
In-Reply-To: <4235985F.2080106@free.fr>
References: <4235985F.2080106@free.fr>
Message-ID: <20050314235956%evC7T@faireal.net>

Steve Lhomme <steve.lhomme at free.fr> wrote:

> Hi,
> 
> I just uploaded a draft of hardware profiles that we could use. Any 
> comments are welcome, because it's an important part if manufacturers 
> consider using Matroska.
> 
> http://www.matroska.org/technical/specs/profiles/new_draft.html
> 
> RealVideo was left out on purpose. Subs formats could be included too.

I still remember some of you were trying to re-motivate USF devs 
in 2003, when (iirc) there was an offer for matroska hw support:
http://lists.matroska.org/pipermail/matroska-devel/2003-October/000972.html

USF would be ideal, but I'm afraid it is not going to come in 
time at this rate...or perhaps never.

Realistically, they should support at least SRT, and maybe 
S_VOBSUB too. Another possibility is ttxt from mp4 world?

Liisachan



From steve.lhomme at free.fr  Mon Mar 14 16:30:11 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Mon, 14 Mar 2005 16:30:11 +0100
Subject: [Matroska-devel] Profiles v2
In-Reply-To: <20050314235956%evC7T@faireal.net>
References: <4235985F.2080106@free.fr> <20050314235956%evC7T@faireal.net>
Message-ID: <4235AE03.3030506@free.fr>

Liisachan a ?crit :
> Steve Lhomme <steve.lhomme at free.fr> wrote:
> 
> 
>>Hi,
>>
>>I just uploaded a draft of hardware profiles that we could use. Any 
>>comments are welcome, because it's an important part if manufacturers 
>>consider using Matroska.
>>
>>http://www.matroska.org/technical/specs/profiles/new_draft.html
>>
>>RealVideo was left out on purpose. Subs formats could be included too.
> 
> 
> I still remember some of you were trying to re-motivate USF devs 
> in 2003, when (iirc) there was an offer for matroska hw support:
> http://lists.matroska.org/pipermail/matroska-devel/2003-October/000972.html
> 
> USF would be ideal, but I'm afraid it is not going to come in 
> time at this rate...or perhaps never.
> 
> Realistically, they should support at least SRT, and maybe 
> S_VOBSUB too. Another possibility is ttxt from mp4 world?

Well, given there are already some existing hardware devices based on 
VLC, anything that VLC can handle could be handled in such devices (OK, 
they are used for streams, not CD/DVD/HD reading). We are trying to 
motivate some devs to make USF work with VLC, as DShow seemed too 
complex for the moment. It hope it works...

Yes, MP4's ttxt should be supported too someday in Matroska.

Vobsub is obviously a candidate for these profiles, as they are already 
handled by the DVDs. SRT and other processed text might not. Because you 
need for CPU for that. Probably why DivX chose a picture based format...

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From steve.lhomme at free.fr  Mon Mar 14 17:38:43 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Mon, 14 Mar 2005 17:38:43 +0100
Subject: [Matroska-devel] Checklist for our next pack
Message-ID: <4235BE13.30105@free.fr>

Hi,

Satsuki is OK to make our next pack, but he wants a detailed description 
of what we want. I actually have no definite idea of what we want. Here 
is what I can think of :

- Haali's splitter with only MKV reading enabled by default
- Disable MPC's internal and MatroskaSplitter
- FFDShow with H264 support, AAC, Vorbis and AC3
- vsfilter
- CoreFLAC
- CoreTTA
- RealMediaSplitter

+ should we ship MatroskaDiag ?
+ MatroskaProp with known bugs fixed ?
+ DScaler 5 for MPEG2 decoding ?

* the same system we had to detect the presence of Real DLLs
* new icon Betaboy has proposed

-- 
robUx4 on blog <http://robux4.blogspot.com/>


From aaa at aaaa.com  Mon Mar 14 19:07:49 2005
From: aaa at aaaa.com (Zen)
Date: Mon, 14 Mar 2005 19:07:49 +0100
Subject: [Matroska-devel] Re: Profiles v2
In-Reply-To: <4235985F.2080106@free.fr>
References: <4235985F.2080106@free.fr>
Message-ID: <4235D2F5.1090809@aaaa.com>

Steve Lhomme wrote:
> Hi,
> 
> I just uploaded a draft of hardware profiles that we could use. Any 
> comments are welcome, because it's an important part if manufacturers 
> consider using Matroska.

DTS : should be able to handle 6.1 or 7.1 streams. 7.1 audio is the 
future of audio, even more in computers (see HDA, Intel High Definition 
Audio in relpacement of old AC97)

Resolution : specify this is "up to" : what to do of video in 712*572? 
Should it be OK with players supporting 720*576? Video should be PAL 
(720*576) or NTSC (720*480) with no ratio problems

Are you sure that TTA, WavPack and Flac are necessary? This is not a 
mass usage... a lot less than DTS 6.1!!! Will it be used by Industry? I 
think not, Industry think to lossly compression...

"DVD Recorders" : name is bad : why not HD player?


From moritz at bunkus.org  Mon Mar 14 19:38:04 2005
From: moritz at bunkus.org (Moritz Bunkus)
Date: Mon, 14 Mar 2005 19:38:04 +0100
Subject: [Matroska-devel] Checklist for our next pack
In-Reply-To: <4235BE13.30105@free.fr>
References: <4235BE13.30105@free.fr>
Message-ID: <200503141938.07873.moritz@bunkus.org>

Hey,

> - Haali's splitter with only MKV reading enabled by default
> - Disable MPC's internal and MatroskaSplitter
> - FFDShow with H264 support, AAC, Vorbis and AC3
> - vsfilter
> - CoreFLAC
> - CoreTTA
> - RealMediaSplitter

Sounds pretty good to me.

> + should we ship MatroskaDiag ?

Definitely! It shouldn't be optional to install either. It's a pretty
important tool for bug testing, and it's small. So there's no harm done
in always installing it IMHO.

> + MatroskaProp with known bugs fixed ?

Dunno about the state of MatroskaProp, but I think that it's a good
thing unless it crashes a lot.

> + DScaler 5 for MPEG2 decoding ?

How important is that for us? Hmm... I'd say include it but make it
optional (?).

> * the same system we had to detect the presence of Real DLLs

Agree

> * new icon Betaboy has proposed

Agree

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: <http://lists.matroska.org/pipermail/matroska-devel/attachments/20050314/feb19fb7/attachment.pgp>

From steve.lhomme at free.fr  Mon Mar 14 21:06:08 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Mon, 14 Mar 2005 21:06:08 +0100
Subject: [Matroska-devel] Checklist for our next pack
In-Reply-To: <200503141938.07873.moritz@bunkus.org>
References: <4235BE13.30105@free.fr> <200503141938.07873.moritz@bunkus.org>
Message-ID: <4235EEB0.1000106@free.fr>

Moritz Bunkus a ?crit :
> Hey,
> 
> 
>>+ should we ship MatroskaDiag ?
> 
> 
> Definitely! It shouldn't be optional to install either. It's a pretty
> important tool for bug testing, and it's small. So there's no harm done
> in always installing it IMHO.

Good.

>>+ MatroskaProp with known bugs fixed ?
> 
> 
> Dunno about the state of MatroskaProp, but I think that it's a good
> thing unless it crashes a lot.

I think the 2.7 version had some serious pbs. Maybe Jory should have a 
look first. Also writing tags with it should be disabled. I think it may 
also produce bogus files. So extensive tests should be done if we plan 
to add it.

>>+ DScaler 5 for MPEG2 decoding ?
> 
> 
> How important is that for us? Hmm... I'd say include it but make it
> optional (?).

Using DMX, with just a double click you get your MKV files with DVD 
menus using the same codec as the original. The files are a bot smaller 
than DVDs, so it's still usefull (especially when we have DVD->MKV->DVD) 
so I think it's a good idea to allow this. Even if this use is not 
popular yet. And besides, it's a quality free software.



From steve.lhomme at free.fr  Mon Mar 14 21:19:25 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Mon, 14 Mar 2005 21:19:25 +0100
Subject: [Matroska-devel] Re: Profiles v2
In-Reply-To: <4235D2F5.1090809@aaaa.com>
References: <4235985F.2080106@free.fr> <4235D2F5.1090809@aaaa.com>
Message-ID: <4235F1CD.8000602@free.fr>

Hi,

And thanks for commenting :)

Zen a ?crit :
> Steve Lhomme wrote:
> 
>> Hi,
>>
>> I just uploaded a draft of hardware profiles that we could use. Any 
>> comments are welcome, because it's an important part if manufacturers 
>> consider using Matroska.
> 
> 
> DTS : should be able to handle 6.1 or 7.1 streams. 7.1 audio is the 
> future of audio, even more in computers (see HDA, Intel High Definition 
> Audio in relpacement of old AC97)

The number of supported channels is an indication for the player. 
Meaning that it should be the minimum to support. Same for the codec 
list. They will probably all support WMA and WMV9. But we don't had that 
to our list, simply because we don't support these codecs for the moment.

> Resolution : specify this is "up to" : what to do of video in 712*572? 
> Should it be OK with players supporting 720*576? Video should be PAL 
> (720*576) or NTSC (720*480) with no ratio problems

I used PAL as an example. Of course it's not a strict limit. And the 
main pb on resolution is the one for PocketPCs as most of them don't 
support VGA, maybe even some PMP ones don't. So the choice here (and 
everywhere else too) is do we want very few but very good profile 
compliance, or a more relax one that could be used in lots of devices.

I'm undecided on this. And that will probably be a team decision 
(Christian ? Betaboy ? Mosu ? Haali ? Toff ? etc).

> Are you sure that TTA, WavPack and Flac are necessary? This is not a 
> mass usage... a lot less than DTS 6.1!!! Will it be used by Industry? I 
> think not, Industry think to lossly compression...

Support for these codec is easy. But it's true that it looks strange 
that we support 3 lossless codec. Especially since some devices would 
not be compliant just because of that. Maybe I consider them as too 
important. If I had to choose one, it would be Wavpack, because it has 
this hybrid mode, which is good for portable players. So maybe I'll drop 
the other 2. Having at least one (again, that list in the minimum) is 
also a choice of quality vs ease of compliance... In the other hand TTA 
is a CoreCodec project and the format itself is very simple and 
efficient compared to FLAC. But that would be more a political choice.

> "DVD Recorders" : name is bad : why not HD player?

True. I thought about that afterward, but H was already taken (and D 
too). I'll rename the "HD player" ID to something else and use H for the 
HD profile.

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From paul at msn.com  Tue Mar 15 05:53:28 2005
From: paul at msn.com (Atamido)
Date: Mon, 14 Mar 2005 22:53:28 -0600
Subject: [Matroska-devel] Re: Profiles v2
In-Reply-To: <4235F1CD.8000602@free.fr>
References: <4235985F.2080106@free.fr> <4235D2F5.1090809@aaaa.com>
	<4235F1CD.8000602@free.fr>
Message-ID: <d15pp8$na$1@sea.gmane.org>

Steve Lhomme wrote:
> Zen a ?crit :
>> DTS : should be able to handle 6.1 or 7.1 streams. 7.1 audio is the 
>> future of audio, even more in computers (see HDA, Intel High 
>> Definition Audio in relpacement of old AC97)
> 
> The number of supported channels is an indication for the player. 
> Meaning that it should be the minimum to support. Same for the codec 
> list. They will probably all support WMA and WMV9. But we don't had that 
> to our list, simply because we don't support these codecs for the moment.

I would say that this is a bit of a special case.  The player itself 
doesn't need to 'support' a certain number of channels in DD or DTS. 
Basically all DVD players only output the raw encoded data to a digital 
out so that it can be decoded by a receiver.  My receiver supports a lot 
of different forms of DD and DTS, but the only thing my DVD player 
decodes is the different MPEG layer audio, which is always 2 channel.

So a hardware device like a DVD player should be required to support 
outputting DD and DTS of any channel amount to it's digital outs.  I 
don't think decoding should be required at all as it is so rare. (Some 
DVD players include DD decoders and an amp and 6 speaker outputs, but 
they aren't common.)

Other audio formats are harder though.  Vorbis for instance.  If a 
Vorbis file is in 5.1, what should be required?  That it be able to 
decode it and down mix to 2 channel?  That it be able to decode at least 
the front two channels?  AAC is the same situation.  WavPack and FLAC 
are a similar situation, although those are only likely to contain 2 
channel audio.

>> Resolution : specify this is "up to" : what to do of video in 712*572? 
>> Should it be OK with players supporting 720*576? Video should be PAL 
>> (720*576) or NTSC (720*480) with no ratio problems
> 
> I used PAL as an example. Of course it's not a strict limit. And the 
> main pb on resolution is the one for PocketPCs as most of them don't 
> support VGA, maybe even some PMP ones don't. So the choice here (and 
> everywhere else too) is do we want very few but very good profile 
> compliance, or a more relax one that could be used in lots of devices.

You could have a class of "DVD Player" certifications, and a class of 
"PDA" classifications.  Then you could also add an something like a 
"Ultra small" for something like cell phones in the future if it was 
required.

>> Are you sure that TTA, WavPack and Flac are necessary? This is not a 
>> mass usage... a lot less than DTS 6.1!!! Will it be used by Industry? 
>> I think not, Industry think to lossly compression...
> 
> Support for these codec is easy. But it's true that it looks strange 
> that we support 3 lossless codec. Especially since some devices would 
> not be compliant just because of that. Maybe I consider them as too 
> important. If I had to choose one, it would be Wavpack, because it has 
> this hybrid mode, which is good for portable players. So maybe I'll drop 
> the other 2. Having at least one (again, that list in the minimum) is 
> also a choice of quality vs ease of compliance... In the other hand TTA 
> is a CoreCodec project and the format itself is very simple and 
> efficient compared to FLAC. But that would be more a political choice.

Using FLAC may be considered a 'politically friendly' move.  I don't 
really know though.  I do think that FLAC is more commonly used than any 
of the other formats.  So if you wanted to go by what is in most common 
existence, than it would be FLAC.  Of course FLAC in Matroska probably 
isn't as common as other formats in Matroska.  Maybe something like "One 
of these three formats must be supported."  If you did that then players 
would have their option of which to support, whatever is easiest for 
them to implement in hardware.  And theoretically since they would all 
be for lossless, you could easily trans code between them.


Atamido



From aaa at aaaa.com  Tue Mar 15 15:16:19 2005
From: aaa at aaaa.com (Zen)
Date: Tue, 15 Mar 2005 15:16:19 +0100
Subject: [Matroska-devel] Re: Profiles v2
In-Reply-To: <d15pp8$na$1@sea.gmane.org>
References: <4235985F.2080106@free.fr>
	<4235D2F5.1090809@aaaa.com>	<4235F1CD.8000602@free.fr>
	<d15pp8$na$1@sea.gmane.org>
Message-ID: <4236EE33.9050903@aaaa.com>

Atamido wrote:

> existence, than it would be FLAC.  Of course FLAC in Matroska probably 
> isn't as common as other formats in Matroska.  Maybe something like "One 
> of these three formats must be supported."  If you did that then players 
> would have their option of which to support, whatever is easiest for 
> them to implement in hardware.  And theoretically since they would all 
> be for lossless, you could easily trans code between them.

Very bad idea : if you do this, you will not be able to know if your 
file will be playable on a player, because you don't have the free 
formats in your file for each kind of player!!!


From aaa at aaaa.com  Tue Mar 15 15:24:41 2005
From: aaa at aaaa.com (Zen)
Date: Tue, 15 Mar 2005 15:24:41 +0100
Subject: [Matroska-devel] Re: Profiles v2
In-Reply-To: <4235F1CD.8000602@free.fr>
References: <4235985F.2080106@free.fr> <4235D2F5.1090809@aaaa.com>
	<4235F1CD.8000602@free.fr>
Message-ID: <4236F029.1070709@aaaa.com>

Steve Lhomme wrote:

> Support for these codec is easy. But it's true that it looks strange 
> that we support 3 lossless codec. Especially since some devices would 
> not be compliant just because of that. Maybe I consider them as too 
> important. If I had to choose one, it would be Wavpack, because it has 
> this hybrid mode, which is good for portable players. So maybe I'll drop 
> the other 2. Having at least one (again, that list in the minimum) is 
> also a choice of quality vs ease of compliance... In the other hand TTA 
> is a CoreCodec project and the format itself is very simple and 
> efficient compared to FLAC. But that would be more a political choice.

This is already a political choice to have a lossless codec in a player 
specification, because nobody in the industry think this is the future, 
only geeks do this :)
But... If you think it is easy to implement this with every kind of CPU, 
yes, there is DTS and AC3 in DVD, why not 3 lossless formats? ;-)


And another thing :
why ID like P, X, T, H? Yes, H = HD player and do on... but : it is easy 
to know that 1>2, no T<H.
Maybe class 1, 2, 3, 4 should be better?


From moritz at bunkus.org  Tue Mar 15 15:57:33 2005
From: moritz at bunkus.org (Moritz Bunkus)
Date: Tue, 15 Mar 2005 15:57:33 +0100
Subject: [Matroska-devel] Checklist for our next pack
In-Reply-To: <4235EEB0.1000106@free.fr>
References: <4235BE13.30105@free.fr> <200503141938.07873.moritz@bunkus.org>
	<4235EEB0.1000106@free.fr>
Message-ID: <200503151557.38177.moritz@bunkus.org>

Hey,

> >>+ MatroskaProp with known bugs fixed ?
>
> I think the 2.7 version had some serious pbs. Maybe Jory should have a
> look first. Also writing tags with it should be disabled. I think it may
> also produce bogus files. So extensive tests should be done if we plan
> to add it.

Instead of extensive tests we should skip shipping it and build a pack
without it. We can always fix those bugs later and release another pack
then. I just think we should release soon, and having the time to test
it properly is probably lacking.

> >>+ DScaler 5 for MPEG2 decoding ?
...

> And besides, it's a quality free software.

:) Go for it.

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: <http://lists.matroska.org/pipermail/matroska-devel/attachments/20050315/8cd63444/attachment.pgp>

From steve.lhomme at free.fr  Tue Mar 15 21:59:43 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Tue, 15 Mar 2005 21:59:43 +0100
Subject: [Matroska-devel] Re: Profiles v2
In-Reply-To: <4236F029.1070709@aaaa.com>
References: <4235985F.2080106@free.fr>
	<4235D2F5.1090809@aaaa.com>	<4235F1CD.8000602@free.fr>
	<4236F029.1070709@aaaa.com>
Message-ID: <42374CBF.30709@free.fr>

Zen a ?crit :
> Steve Lhomme wrote:
> 
>> Support for these codec is easy. But it's true that it looks strange 
>> that we support 3 lossless codec. Especially since some devices would 
>> not be compliant just because of that. Maybe I consider them as too 
>> important. If I had to choose one, it would be Wavpack, because it has 
>> this hybrid mode, which is good for portable players. So maybe I'll 
>> drop the other 2. Having at least one (again, that list in the 
>> minimum) is also a choice of quality vs ease of compliance... In the 
>> other hand TTA is a CoreCodec project and the format itself is very 
>> simple and efficient compared to FLAC. But that would be more a 
>> political choice.
> 
> 
> This is already a political choice to have a lossless codec in a player 
> specification, because nobody in the industry think this is the future, 
> only geeks do this :)

Apple's ALAC ? Used for AirTunes ? It *is* the future in the home. At 
least until bandwidth is large enough to easily handle 96KHz/24bits 7.1 
channels uncompressed (OK, that's /only/ 2 Mbps but with video it's a 
lot more).

> But... If you think it is easy to implement this with every kind of CPU, 
> yes, there is DTS and AC3 in DVD, why not 3 lossless formats? ;-)

Because that makes compliancy harder to get ? We don't want profiles to 
be too simple but we don't want them to be too complex too. Maybe FLAC 
is better for manufacturers because it has a large user base, compared 
to TTA or Wavpack... The choice is still opened. But I like the Wavpack 
hybrid mode a lot (especially for portable devices, where it really 
makes sense).

> And another thing :
> why ID like P, X, T, H? Yes, H = HD player and do on... but : it is easy 
> to know that 1>2, no T<H.
> Maybe class 1, 2, 3, 4 should be better?

Nop, because we can't add profiles in the middle (profile 2.5 ?).
It's not that hard as long as the letters are easy to understand like 
for DVB :
- DVB-S for satellite
- DVB-C for cable
- DVB-T for terrestrial
- DVB-H for handheld

> _______________________________________________
> Matroska-devel mailing list
> Matroska-devel at lists.matroska.org
> http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-devel

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From moritz at bunkus.org  Fri Mar 18 22:38:12 2005
From: moritz at bunkus.org (Moritz Bunkus)
Date: Fri, 18 Mar 2005 22:38:12 +0100
Subject: [Matroska-devel] our Subversion repositories
Message-ID: <200503182238.16206.moritz@bunkus.org>

Hey,

I've just switched both of our Subversion repositories on Haali's server
(the "matroska" and the "www" repo) from Berkeley DB to the filesystem
based version. Maybe the problems we have with constantly needing to
recover the repos might go away...

If you notice any problems just tell me.

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: <http://lists.matroska.org/pipermail/matroska-devel/attachments/20050318/f3f8930c/attachment.pgp>

From chris at matroska.org  Sat Mar 19 17:45:47 2005
From: chris at matroska.org (Christian HJ Wiesner)
Date: Sat, 19 Mar 2005 17:45:47 +0100
Subject: [Matroska-devel] MKV DVD structure & player certification?
In-Reply-To: <42343DF4.9050101@free.fr>
References: <00bc01c5270f$d51d5830$1a00a8c0@marel.local>	<42341403.8050708@free.fr>	<002401c527bd$86b6b430$1a00a8c0@marel.local>
	<42343DF4.9050101@free.fr>
Message-ID: <423C573B.8050802@matroska.org>

Steve Lhomme schrieb:

> DigitalRise a ?crit :
>
>> For standalone players a content.xml in the main directory should be 
>> also
>> available which contains information about the files on the disc. 
>> These file
>> can be read from the player to take the right action (autoplay or 
>> show the
>> content for selection).
>
>
> That is a good idea. But then it should not be restricted to Matroska 
> (AVI, MP4, MPG, etc). 


Back from vacarion !

I guess you 2 are talking of different things. DigitalRise, you have to 
be aware that a single MKV file can contain very powerful Metadata about 
its content already, so there is really no need to have a separate XML 
file to describe the content on it, unless you are planning to have 
several different file types ( as Steve said, MKV, MPEG, AVI, MP4, etc. 
) on a single disc, and mixed. If you have only MKV files on the DVD, 
simply parsing the file will give you all the infos you need already.

Christian
matroska project admin


From steve.lhomme at free.fr  Sun Mar 20 11:34:02 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Sun, 20 Mar 2005 11:34:02 +0100
Subject: [Matroska-devel] Google Search
Message-ID: <423D519A.7060503@free.fr>

Hi,

As you know, it's always hard to find the video you're looking for on 
your HD except if you use very long file names and you are very organised.

Google has a desktop search tool that could make it easier to find 
videos on your HDD. Apparently there is an open API to add your own file 
type. Adding Matroska support there could be a good thing for users, and 
for the project in general. So if you have time to look at this it could 
be a great thing. Otherwise I'll have a look after I'm done with menu in 
VLC, various fixes here and there, a muxer in VLC (and before the editor 
we need).

Imagine searching on your system: "video using RealVideo with a 16:9 
aspect ratio and smaller than 30 MB" or "Tracks which producer is Trevor 
Horn".

http://desktop.google.com/developer.html



From steve.lhomme at free.fr  Sun Mar 20 11:42:43 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Sun, 20 Mar 2005 11:42:43 +0100
Subject: [Matroska-devel] Idea
Message-ID: <423D53A3.3070104@free.fr>

I saw this in the IRC log :

<nexus3>	IMHO this page needs a little images - in order to animate 
people downloading and trying out the files 
http://www.matroska.org/samples/matrix/index.html

Anyone ? Chris ?



From inuyasha.rules at gmail.com  Sun Mar 20 21:33:34 2005
From: inuyasha.rules at gmail.com (Tommy Pierson)
Date: Sun, 20 Mar 2005 13:33:34 -0700
Subject: [Matroska-devel] mkv to dvd?
Message-ID: <4b7332ee0503201233177c883f@mail.gmail.com>

i recently stumbled uppon your video format while obtaining anime
cartoons, and found a single flaw.  there is (as far as i can find) no
way to convert an mkv to avi or dvd video file, so as to view them on
a dvd player, or similar impliment.  i am pleased overall with the
format, but with as slow as my computer is, it is really frusterating
not being able to view the videos at my house( i currently burn them
to cd and go over to a friends house).

   thanks for considering adding this support to mkv format video, an anime fan


From chris at matroska.org  Mon Mar 21 22:34:13 2005
From: chris at matroska.org (Christian HJ Wiesner)
Date: Mon, 21 Mar 2005 22:34:13 +0100
Subject: [Matroska-devel] Wavpack Alpha DirectShow decoder
Message-ID: <423F3DD5.6040301@matroska.org>


Hi guys,

Toff has made a first alpha DirectShow decoder filter for Wavpack, you 
can find it here : 
http://www.matroska.org/~toff/WavPackDSDecoder20050321-2.zip

Get the latest splitter from Haali from http://haali.cs.msu.ru/mkv/ , 
register the new splitter as always with regsvr32.exe and it should play 
wavpack in MKA fine, but please note the given limitations right now :

------------------------------------------------------------------------
Status (2005-03-21) :

Only tested with 44100Khz 16bits stereo.

No wavpack splitter, so only works from matroska for now.

Need to be tested with any lossless / lossy / hybrid file with
sample_rates : 6000, 8000, 9600, 11025, 12000, 16000, 22050,
    24000, 32000, 44100, 48000, 64000, 88200, 96000, 192000
bitdepth : 8, 16
channels : 1, 2

more than 2 channels should not works
24 bits pcm and 32 bits float should require some little output 
mediatype modification
-----------------------------------------------------------------------------


Regards

Christian


From Liisachan at faireal.net  Tue Mar 22 02:27:58 2005
From: Liisachan at faireal.net (Liisachan)
Date: Tue, 22 Mar 2005 10:27:58 +0900
Subject: [Matroska-devel] Wavpack Alpha DirectShow decoder
In-Reply-To: <423F3DD5.6040301@matroska.org>
References: <423F3DD5.6040301@matroska.org>
Message-ID: <20050322102758LoRu%n@faireal.net>

Sorry, I used a wrong mail address to send the previous mail to 
the list.

This is what I sent:

Hi,

Great!

> Only tested with 44100Khz 16bits stereo.

Apparently 48000/16 too worked for me.

- (XviD + WV 44100/16) YES, Worked!
- (XviD + WV 48000/16) YES, Worked!

- (XviD + WV 44100/24) _NOT_ worked
- (XviD + WV 48000/24) _NOT_ worked

How can I test hybrid wv/wvc files?

Liisachan



Christian HJ Wiesner <chris at matroska.org> wrote:

> Toff has made a first alpha DirectShow decoder filter for Wavpack, you 
> can find it here : 
> http://www.matroska.org/~toff/WavPackDSDecoder20050321-2.zip
> 
> Get the latest splitter from Haali from http://haali.cs.msu.ru/mkv/ , 
> register the new splitter as always with regsvr32.exe and it should play 
> wavpack in MKA fine, but please note the given limitations right now :
> 
> ------------------------------------------------------------------------
> Status (2005-03-21) :
> 
> Only tested with 44100Khz 16bits stereo.
> 
> No wavpack splitter, so only works from matroska for now.
> 
> Need to be tested with any lossless / lossy / hybrid file with
> sample_rates : 6000, 8000, 9600, 11025, 12000, 16000, 22050,
>     24000, 32000, 44100, 48000, 64000, 88200, 96000, 192000
> bitdepth : 8, 16
> channels : 1, 2
> 
> more than 2 channels should not works
> 24 bits pcm and 32 bits float should require some little output 
> mediatype modification
> -----------------------------------------------------------------------------




From playful at gmx.at  Tue Mar 22 11:06:11 2005
From: playful at gmx.at (filewalker)
Date: Tue, 22 Mar 2005 10:06:11 +0000 (UTC)
Subject: [Matroska-devel] issues with Haali's parser
Message-ID: <loom.20050322T105525-24@post.gmane.org>

Hi Mike,

A lot of people have been demanding the support of yout parser in Zoomplayer.
So Blight(author of Zoomplayer) reacted...but he found some issues with your 
parser.

I quote from Zoomplayer forum:
[quote]
Blight:
"There's a bug in the haali filter. If you add directvobsub to the graph and 
then render the filter's subtitle pin, it won't connect to DirectVobSub...

This is why ZP isn't showing subs with the haali splitter.

I'm working to add support for stream-switching and such, but haali would need 
to fix his splitter for subs to work.

edit:
Arrgh, his splitter seems to be running over the entire filter list and creating 
each filter, this causes filters with debug-protection to kill ZP while I'm 
working on the code, not to mention, it's slowing down load times if your system 
is full of filters (like mine is).

Edit edit:
Ok, got the code in, next beta, haali's filter will be supported for stream-
switching and also with customized media profiles.

But he must fix that rendering thing and it would be nice if he removed that 
filter-scan thing, it causes a 1-2 slowdown for file load on my system (compared 
to the gabest filter). Not to mention that it might cause instability on some 
systems withs semi-stable filters.

(He should probably call it the "Haali Media Splitter" as it does more than 
matroska now)."[/quote]


Beside Blight some other people have problems to display subs with your filter 
although I don't have problems with subs.

I hope that these last issues can be solved.


Cu filewalker




From mike at po.cs.msu.su  Tue Mar 22 11:23:38 2005
From: mike at po.cs.msu.su (Mike Matsnev)
Date: Tue, 22 Mar 2005 13:23:38 +0300
Subject: [Matroska-devel] issues with Haali's parser
In-Reply-To: <loom.20050322T105525-24@post.gmane.org>
References: <loom.20050322T105525-24@post.gmane.org>
Message-ID: <423FF22A.3060708@po.cs.msu.su>

filewalker wrote:
> Blight:
> "There's a bug in the haali filter. If you add directvobsub to the graph and 
> then render the filter's subtitle pin, it won't connect to DirectVobSub...
> 
> This is why ZP isn't showing subs with the haali splitter.
> 
> I'm working to add support for stream-switching and such, but haali would need 
> to fix his splitter for subs to work.
This is news to me, as it obviously connects properly in other players.

> 
> edit:
> Arrgh, his splitter seems to be running over the entire filter list and creating 
> each filter, this causes filters with debug-protection to kill ZP while I'm 
> working on the code, not to mention, it's slowing down load times if your system 
> is full of filters (like mine is).
There is no such code in my filter. I suspect the FGM is doing that because the
output pin type is specified as wildcard in registration data.

On a side note, it'd be nice if blight could mail me directly or come to our irc
channel, then the issues could be resolved a lot more quickly.

/Mike


From paul at msn.com  Tue Mar 22 18:53:21 2005
From: paul at msn.com (Atamido)
Date: Tue, 22 Mar 2005 11:53:21 -0600
Subject: [Matroska-devel] Re: mkv to dvd?
In-Reply-To: <4b7332ee0503201233177c883f@mail.gmail.com>
References: <4b7332ee0503201233177c883f@mail.gmail.com>
Message-ID: <42405B91.8030606@msn.com>

Assuming you are in Windows and are using a DirectShow player to view 
the video, then you can use TMPGEnc to convert the file to MPEG-2 for 
use on a DVD.  OF course if I didn't know this, I would have searched 
Google and found the answer in the top results.

http://www.google.com/search?q=convert+mkv+to+dvd


Atamido

Tommy Pierson wrote:
> i recently stumbled uppon your video format while obtaining anime
> cartoons, and found a single flaw.  there is (as far as i can find) no
> way to convert an mkv to avi or dvd video file, so as to view them on
> a dvd player, or similar impliment.  i am pleased overall with the
> format, but with as slow as my computer is, it is really frusterating
> not being able to view the videos at my house( i currently burn them
> to cd and go over to a friends house).
> 
>    thanks for considering adding this support to mkv format video, an anime fan



From christophe.paris at free.fr  Tue Mar 22 19:31:43 2005
From: christophe.paris at free.fr (Christophe PARIS)
Date: Tue, 22 Mar 2005 19:31:43 +0100
Subject: [Matroska-devel] Wavpack Alpha DirectShow decoder
In-Reply-To: <20050322102758LoRu%n@faireal.net>
References: <423F3DD5.6040301@matroska.org> <20050322102758LoRu%n@faireal.net>
Message-ID: <4240648F.40509@free.fr>

Liisachan wrote:
> Great!
> Apparently 48000/16 too worked for me.
> - (XviD + WV 44100/16) YES, Worked!
> - (XviD + WV 48000/16) YES, Worked!
> - (XviD + WV 44100/24) _NOT_ worked
> - (XviD + WV 48000/24) _NOT_ worked
> 
> How can I test hybrid wv/wvc files?

Thanks for your test.

To make lossy or hybrid files, select the .wv in mmg and it will
automatically mux the correction data if available.
If not it will be a lossy file.

Regards Toff



From kurtnoise at free.fr  Tue Mar 22 23:46:44 2005
From: kurtnoise at free.fr (kurtnoise at free.fr)
Date: Tue, 22 Mar 2005 23:46:44 +0100
Subject: [Matroska-devel] Wavpack Alpha DirectShow decoder
Message-ID: <20050322234644.yzc4yxvzcn1a8w00@imp-test.free.fr>

Hi guys,


I made some tests with this decoder...So, here is my results :

Tools :
 - mkvtoolnix 1.4.1
 - wavpack 4.2 beta 3
 - WavpackDShow Filter 20050321-2
 - Haali Splitter 20050322
 - MPC 6.4.8.4


************* Sample 1 - (Audio only) ****************

1/ Lossless Mode :

-------------------------------------------------
|Sampling-Rate | Bitdepth | Channels | Playback |
-------------------------------------------------
|   6000 Hz    |  8 bits  |    1,2   |    Ok    |
|   6000 Hz    |  16 bits |    1,2   |    Ok    |
|   8000 Hz    |  8 bits  |    1,2   |    Ok    |
|   8000 Hz    |  16 bits |    1,2   |    Ok    |
|   11025 Hz   |  8 bits  |    1,2   |    Ok    |
|   11025 Hz   |  16 bits |    1,2   |    Ok    |
|   16000 Hz   |  8 bits  |    1,2   |    Ok    |
|   16000 Hz   |  16 bits |    1,2   |    Ok    |
|   22050 Hz   |  8 bits  |    1,2   |    Ok    |
|   22050 Hz   |  16 bits |    1,2   |    Ok    |
|   24000 Hz   |  8 bits  |    1,2   |    Ok    |
|   24000 Hz   |  16 bits |    1,2   |    Ok    |
|   32000 Hz   |  8 bits  |    1,2   |    Ok    |
|   32000 Hz   |  16 bits |    1,2   |    Ok    |
|   44100 Hz   |  8 bits  |    1,2   |    Ok    |
|   44100 Hz   |  16 bits |    1,2   |    Ok    |
|   48000 Hz   |  8 bits  |    1,2   |    Ok    |
|   48000 Hz   |  16 bits |    1,2   |    Ok    |
|   96000 Hz   |  8 bits  |    1,2   |    Ok    |
|   96000 Hz   |  16 bits |    1,2   |    Ok    |
-------------------------------------------------


2/ Lossy Mode :

-------------------------------------------------
|Sampling-Rate | Bitdepth | Channels | Playback |
-------------------------------------------------
|   6000 Hz    |  8 bits  |    1,2   |    Ok    |
|   6000 Hz    |  16 bits |    1,2   |    Ok    |
|   8000 Hz    |  8 bits  |    1,2   |    Ok    |
|   8000 Hz    |  16 bits |    1,2   |    Ok    |
|   11025 Hz   |  8 bits  |    1,2   |    Ok    |
|   11025 Hz   |  16 bits |    1,2   |    Ok    |
|   16000 Hz   |  8 bits  |    1,2   |    Ok    |
|   16000 Hz   |  16 bits |    1,2   |    Ok    |
|   22050 Hz   |  8 bits  |    1,2   |    Ok    |
|   22050 Hz   |  16 bits |    1,2   |    Ok    |
|   24000 Hz   |  8 bits  |    1,2   |    Ok    |
|   24000 Hz   |  16 bits |    1,2   |    Ok    |
|   32000 Hz   |  8 bits  |    1,2   |    Ok    |
|   32000 Hz   |  16 bits |    1,2   |    Ok    |
|   44100 Hz   |  8 bits  |    1,2   |    Ok    |
|   44100 Hz   |  16 bits |    1,2   |    Ok    |
|   48000 Hz   |  8 bits  |    1,2   |    Ok    |
|   48000 Hz   |  16 bits |    1,2   |    Ok    |
|   96000 Hz   |  8 bits  |    1,2   |    Ok    |
|   96000 Hz   |  16 bits |    1,2   |    Ok    |
-------------------------------------------------


3/ Hybrid Mode :

-------------------------------------------------
|Sampling-Rate | Bitdepth | Channels | Playback |
-------------------------------------------------
|   6000 Hz    |  8 bits  |    1,2   |    Ok    |
|   6000 Hz    |  16 bits |    1,2   |    Ok    |
|   8000 Hz    |  8 bits  |    1,2   |    Ok    |
|   8000 Hz    |  16 bits |    1,2   |    Ok    |
|   11025 Hz   |  8 bits  |    1,2   |    Ok    |
|   11025 Hz   |  16 bits |    1,2   |    Ok    |
|   16000 Hz   |  8 bits  |    1,2   |    Ok    |
|   16000 Hz   |  16 bits |    1,2   |    Ok    |
|   22050 Hz   |  8 bits  |    1,2   |    Ok    |
|   22050 Hz   |  16 bits |    1,2   |    Ok    |
|   24000 Hz   |  8 bits  |    1,2   |    Ok    |
|   24000 Hz   |  16 bits |    1,2   |    Ok    |
|   32000 Hz   |  8 bits  |    1,2   |    Ok    |
|   32000 Hz   |  16 bits |    1,2   |    Ok    |
|   44100 Hz   |  8 bits  |    1,2   |    Ok    |
|   44100 Hz   |  16 bits |    1,2   |    Ok    |
|   48000 Hz   |  8 bits  |    1,2   |    Ok    |
|   48000 Hz   |  16 bits |    1,2   |    Ok    |
|   96000 Hz   |  8 bits  |    1,2   |    Ok    |
|   96000 Hz   |  16 bits |    1,2   |    Ok    |
-------------------------------------------------



*************** Sample 2 (Audio + Video) ****************

1/ Lossless Mode :

-------------------------------------------------
|Sampling-Rate | Bitdepth | Channels | Playback |
-------------------------------------------------
|   6000 Hz    |  8 bits  |    1,2   |    Ok    |
|   6000 Hz    |  16 bits |    1,2   |    Ok    |
|   8000 Hz    |  8 bits  |    1,2   |    Ok    |
|   8000 Hz    |  16 bits |    1,2   |    Ok    |
|   11025 Hz   |  8 bits  |    1,2   |    Ok    |
|   11025 Hz   |  16 bits |    1,2   |    Ok    |
|   16000 Hz   |  8 bits  |    1,2   |    Ok    |
|   16000 Hz   |  16 bits |    1,2   |    Ok    |
|   22050 Hz   |  8 bits  |    1,2   |    Ok    |
|   22050 Hz   |  16 bits |    1,2   |    Ok    |
|   24000 Hz   |  8 bits  |    1,2   |    Ok    |
|   24000 Hz   |  16 bits |    1,2   |    Ok    |
|   32000 Hz   |  8 bits  |    1,2   |    Ok    |
|   32000 Hz   |  16 bits |    1,2   |    Ok    |
|   44100 Hz   |  8 bits  |    1,2   |    Ok    |
|   44100 Hz   |  16 bits |    1,2   |    Ok    |
|   48000 Hz   |  8 bits  |    1,2   |    Ok    |
|   48000 Hz   |  16 bits |    1,2   |    Ok    |
|   96000 Hz   |  8 bits  |    1,2   |    Ok    |
|   96000 Hz   |  16 bits |    1,2   |    Ok    |
-------------------------------------------------


2/ Lossy Mode :

-------------------------------------------------
|Sampling-Rate | Bitdepth | Channels | Playback |
-------------------------------------------------
|   6000 Hz    |  8 bits  |    1,2   |    Ok    |
|   6000 Hz    |  16 bits |    1,2   |    Ok    |
|   8000 Hz    |  8 bits  |    1,2   |    Ok    |
|   8000 Hz    |  16 bits |    1,2   |    Ok    |
|   11025 Hz   |  8 bits  |    1,2   |    Ok    |
|   11025 Hz   |  16 bits |    1,2   |    Ok    |
|   16000 Hz   |  8 bits  |    1,2   |    Ok    |
|   16000 Hz   |  16 bits |    1,2   |    Ok    |
|   22050 Hz   |  8 bits  |    1,2   |    Ok    |
|   22050 Hz   |  16 bits |    1,2   |    Ok    |
|   24000 Hz   |  8 bits  |    1,2   |    Ok    |
|   24000 Hz   |  16 bits |    1,2   |    Ok    |
|   32000 Hz   |  8 bits  |    1,2   |    Ok    |
|   32000 Hz   |  16 bits |    1,2   |    Ok    |
|   44100 Hz   |  8 bits  |    1,2   |    Ok    |
|   44100 Hz   |  16 bits |    1,2   |    Ok    |
|   48000 Hz   |  8 bits  |    1,2   |    Ok    |
|   48000 Hz   |  16 bits |    1,2   |    Ok    |
|   96000 Hz   |  8 bits  |    1,2   |    Ok    |
|   96000 Hz   |  16 bits |    1,2   |    Ok    |
-------------------------------------------------


3/ Hybrid Mode :

-------------------------------------------------
|Sampling-Rate | Bitdepth | Channels | Playback |
-------------------------------------------------
|   6000 Hz    |  8 bits  |    1,2   |    Ok    |
|   6000 Hz    |  16 bits |    1,2   |    Ok    |
|   8000 Hz    |  8 bits  |    1,2   |    Ok    |
|   8000 Hz    |  16 bits |    1,2   |    Ok    |
|   11025 Hz   |  8 bits  |    1,2   |    Ok    |
|   11025 Hz   |  16 bits |    1,2   |    Ok    |
|   16000 Hz   |  8 bits  |    1,2   |    Ok    |
|   16000 Hz   |  16 bits |    1,2   |    Ok    |
|   22050 Hz   |  8 bits  |    1,2   |    Ok    |
|   22050 Hz   |  16 bits |    1,2   |    Ok    |
|   24000 Hz   |  8 bits  |    1,2   |    Ok    |
|   24000 Hz   |  16 bits |    1,2   |    Ok    |
|   32000 Hz   |  8 bits  |    1,2   |    Ok    |
|   32000 Hz   |  16 bits |    1,2   |    Ok    |
|   44100 Hz   |  8 bits  |    1,2   |    Ok    |
|   44100 Hz   |  16 bits |    1,2   |    Ok    |
|   48000 Hz   |  8 bits  |    1,2   |    Ok    |
|   48000 Hz   |  16 bits |    1,2   |    Ok    |
|   96000 Hz   |  8 bits  |    1,2   |    Ok    |
|   96000 Hz   |  16 bits |    1,2   |    Ok    |
-------------------------------------------------


Comments : All seems to be fine. Good work Toff ;-) Just one issue during the
mux process with mmg. I've a warning at the end ( "wavpack_reader : non_audio
block found") for the files with a Sampling-Rate inferior to 44100 Hz. However
the playback works fine. But it's concern much more Mosu'stuff...;)


From christophe.paris at free.fr  Wed Mar 23 08:10:39 2005
From: christophe.paris at free.fr (Christophe PARIS)
Date: Wed, 23 Mar 2005 08:10:39 +0100
Subject: [Matroska-devel] Wavpack Alpha DirectShow decoder
In-Reply-To: <20050322234644.yzc4yxvzcn1a8w00@imp-test.free.fr>
References: <20050322234644.yzc4yxvzcn1a8w00@imp-test.free.fr>
Message-ID: <4241166F.5050605@free.fr>

Hi kurtnoise,

GREAT JOB !!!

Thanks a lot, it's exactly what I was looking for :)

Here is what are the next things to do in my
opinion (in order of priority) :
- a wavpack splitter
- more than 2 channels support
- 24 bits pcm/32 bits floats support

maybe an encoder and a wavpack muxer ?

Regards Toff

kurtnoise at free.fr wrote:
> Hi guys,
> 
> 
> I made some tests with this decoder...So, here is my results :
> 
> Tools :
>  - mkvtoolnix 1.4.1
>  - wavpack 4.2 beta 3
>  - WavpackDShow Filter 20050321-2
>  - Haali Splitter 20050322
>  - MPC 6.4.8.4
> 
> 
> ************* Sample 1 - (Audio only) ****************
> 
> 1/ Lossless Mode :
> 
> -------------------------------------------------
> |Sampling-Rate | Bitdepth | Channels | Playback |
> -------------------------------------------------
> |   6000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   6000 Hz    |  16 bits |    1,2   |    Ok    |
> |   8000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   8000 Hz    |  16 bits |    1,2   |    Ok    |
> |   11025 Hz   |  8 bits  |    1,2   |    Ok    |
> |   11025 Hz   |  16 bits |    1,2   |    Ok    |
> |   16000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   16000 Hz   |  16 bits |    1,2   |    Ok    |
> |   22050 Hz   |  8 bits  |    1,2   |    Ok    |
> |   22050 Hz   |  16 bits |    1,2   |    Ok    |
> |   24000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   24000 Hz   |  16 bits |    1,2   |    Ok    |
> |   32000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   32000 Hz   |  16 bits |    1,2   |    Ok    |
> |   44100 Hz   |  8 bits  |    1,2   |    Ok    |
> |   44100 Hz   |  16 bits |    1,2   |    Ok    |
> |   48000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   48000 Hz   |  16 bits |    1,2   |    Ok    |
> |   96000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   96000 Hz   |  16 bits |    1,2   |    Ok    |
> -------------------------------------------------
> 
> 
> 2/ Lossy Mode :
> 
> -------------------------------------------------
> |Sampling-Rate | Bitdepth | Channels | Playback |
> -------------------------------------------------
> |   6000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   6000 Hz    |  16 bits |    1,2   |    Ok    |
> |   8000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   8000 Hz    |  16 bits |    1,2   |    Ok    |
> |   11025 Hz   |  8 bits  |    1,2   |    Ok    |
> |   11025 Hz   |  16 bits |    1,2   |    Ok    |
> |   16000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   16000 Hz   |  16 bits |    1,2   |    Ok    |
> |   22050 Hz   |  8 bits  |    1,2   |    Ok    |
> |   22050 Hz   |  16 bits |    1,2   |    Ok    |
> |   24000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   24000 Hz   |  16 bits |    1,2   |    Ok    |
> |   32000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   32000 Hz   |  16 bits |    1,2   |    Ok    |
> |   44100 Hz   |  8 bits  |    1,2   |    Ok    |
> |   44100 Hz   |  16 bits |    1,2   |    Ok    |
> |   48000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   48000 Hz   |  16 bits |    1,2   |    Ok    |
> |   96000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   96000 Hz   |  16 bits |    1,2   |    Ok    |
> -------------------------------------------------
> 
> 
> 3/ Hybrid Mode :
> 
> -------------------------------------------------
> |Sampling-Rate | Bitdepth | Channels | Playback |
> -------------------------------------------------
> |   6000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   6000 Hz    |  16 bits |    1,2   |    Ok    |
> |   8000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   8000 Hz    |  16 bits |    1,2   |    Ok    |
> |   11025 Hz   |  8 bits  |    1,2   |    Ok    |
> |   11025 Hz   |  16 bits |    1,2   |    Ok    |
> |   16000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   16000 Hz   |  16 bits |    1,2   |    Ok    |
> |   22050 Hz   |  8 bits  |    1,2   |    Ok    |
> |   22050 Hz   |  16 bits |    1,2   |    Ok    |
> |   24000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   24000 Hz   |  16 bits |    1,2   |    Ok    |
> |   32000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   32000 Hz   |  16 bits |    1,2   |    Ok    |
> |   44100 Hz   |  8 bits  |    1,2   |    Ok    |
> |   44100 Hz   |  16 bits |    1,2   |    Ok    |
> |   48000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   48000 Hz   |  16 bits |    1,2   |    Ok    |
> |   96000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   96000 Hz   |  16 bits |    1,2   |    Ok    |
> -------------------------------------------------
> 
> 
> 
> *************** Sample 2 (Audio + Video) ****************
> 
> 1/ Lossless Mode :
> 
> -------------------------------------------------
> |Sampling-Rate | Bitdepth | Channels | Playback |
> -------------------------------------------------
> |   6000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   6000 Hz    |  16 bits |    1,2   |    Ok    |
> |   8000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   8000 Hz    |  16 bits |    1,2   |    Ok    |
> |   11025 Hz   |  8 bits  |    1,2   |    Ok    |
> |   11025 Hz   |  16 bits |    1,2   |    Ok    |
> |   16000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   16000 Hz   |  16 bits |    1,2   |    Ok    |
> |   22050 Hz   |  8 bits  |    1,2   |    Ok    |
> |   22050 Hz   |  16 bits |    1,2   |    Ok    |
> |   24000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   24000 Hz   |  16 bits |    1,2   |    Ok    |
> |   32000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   32000 Hz   |  16 bits |    1,2   |    Ok    |
> |   44100 Hz   |  8 bits  |    1,2   |    Ok    |
> |   44100 Hz   |  16 bits |    1,2   |    Ok    |
> |   48000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   48000 Hz   |  16 bits |    1,2   |    Ok    |
> |   96000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   96000 Hz   |  16 bits |    1,2   |    Ok    |
> -------------------------------------------------
> 
> 
> 2/ Lossy Mode :
> 
> -------------------------------------------------
> |Sampling-Rate | Bitdepth | Channels | Playback |
> -------------------------------------------------
> |   6000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   6000 Hz    |  16 bits |    1,2   |    Ok    |
> |   8000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   8000 Hz    |  16 bits |    1,2   |    Ok    |
> |   11025 Hz   |  8 bits  |    1,2   |    Ok    |
> |   11025 Hz   |  16 bits |    1,2   |    Ok    |
> |   16000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   16000 Hz   |  16 bits |    1,2   |    Ok    |
> |   22050 Hz   |  8 bits  |    1,2   |    Ok    |
> |   22050 Hz   |  16 bits |    1,2   |    Ok    |
> |   24000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   24000 Hz   |  16 bits |    1,2   |    Ok    |
> |   32000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   32000 Hz   |  16 bits |    1,2   |    Ok    |
> |   44100 Hz   |  8 bits  |    1,2   |    Ok    |
> |   44100 Hz   |  16 bits |    1,2   |    Ok    |
> |   48000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   48000 Hz   |  16 bits |    1,2   |    Ok    |
> |   96000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   96000 Hz   |  16 bits |    1,2   |    Ok    |
> -------------------------------------------------
> 
> 
> 3/ Hybrid Mode :
> 
> -------------------------------------------------
> |Sampling-Rate | Bitdepth | Channels | Playback |
> -------------------------------------------------
> |   6000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   6000 Hz    |  16 bits |    1,2   |    Ok    |
> |   8000 Hz    |  8 bits  |    1,2   |    Ok    |
> |   8000 Hz    |  16 bits |    1,2   |    Ok    |
> |   11025 Hz   |  8 bits  |    1,2   |    Ok    |
> |   11025 Hz   |  16 bits |    1,2   |    Ok    |
> |   16000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   16000 Hz   |  16 bits |    1,2   |    Ok    |
> |   22050 Hz   |  8 bits  |    1,2   |    Ok    |
> |   22050 Hz   |  16 bits |    1,2   |    Ok    |
> |   24000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   24000 Hz   |  16 bits |    1,2   |    Ok    |
> |   32000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   32000 Hz   |  16 bits |    1,2   |    Ok    |
> |   44100 Hz   |  8 bits  |    1,2   |    Ok    |
> |   44100 Hz   |  16 bits |    1,2   |    Ok    |
> |   48000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   48000 Hz   |  16 bits |    1,2   |    Ok    |
> |   96000 Hz   |  8 bits  |    1,2   |    Ok    |
> |   96000 Hz   |  16 bits |    1,2   |    Ok    |
> -------------------------------------------------
> 
> 
> Comments : All seems to be fine. Good work Toff ;-) Just one issue during the
> mux process with mmg. I've a warning at the end ( "wavpack_reader : non_audio
> block found") for the files with a Sampling-Rate inferior to 44100 Hz. However
> the playback works fine. But it's concern much more Mosu'stuff...;)


From kurtnoise at free.fr  Wed Mar 23 13:21:04 2005
From: kurtnoise at free.fr (kurtnoise at free.fr)
Date: Wed, 23 Mar 2005 13:21:04 +0100
Subject: [Matroska-devel] Wavpack Alpha DirectShow decoder
Message-ID: <1111580464.42415f3022ad9@imp4-q.free.fr>

Hey,

>
>Thanks a lot, it's exactly what I was looking for :)
>
>Here is what are the next things to do in my
>opinion (in order of priority) :
>- a wavpack splitter
>- more than 2 channels support
>- 24 bits pcm/32 bits floats support
Yes, a wavpack splitter will be great.

/me waiting... ;-)

>maybe an encoder and a wavpack muxer ?
Why not...

++
Kurtnoise



From agebosma at home.nl  Wed Mar 23 21:15:09 2005
From: agebosma at home.nl (Age Bosma)
Date: Wed, 23 Mar 2005 21:15:09 +0100
Subject: [Matroska-devel] W3C is considering Binary XML (!)(?)
Message-ID: <4241CE4D.4080901@home.nl>

Maybe I'm already running behind but still...if you guys didn't know it yet:
http://news.com.com/Faster+XML+ahead/2100-1007_3-5630957.html?part=rss&tag=5630957&subj=news

Prodoc


From steve.lhomme at free.fr  Wed Mar 23 22:40:10 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Wed, 23 Mar 2005 22:40:10 +0100
Subject: [Matroska-devel] W3C is considering Binary XML (!)(?)
In-Reply-To: <4241CE4D.4080901@home.nl>
References: <4241CE4D.4080901@home.nl>
Message-ID: <4241E23A.5050703@free.fr>

Age Bosma a ?crit :
> Maybe I'm already running behind but still...if you guys didn't know it 
> yet:
> http://news.com.com/Faster+XML+ahead/2100-1007_3-5630957.html?part=rss&tag=5630957&subj=news 
> 
> 
> Prodoc

Thanks Prodoc ! (long time no see)

Actually this format is not exactly new, but there is clearly a need. 
It' just XML with compressed headers. EBML is a bit too simple, but with 
a DTD-like system could be as powerful as a binary XML and remain smaller.

> _______________________________________________
> Matroska-devel mailing list
> Matroska-devel at lists.matroska.org
> http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-devel

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From agebosma at home.nl  Wed Mar 23 23:55:59 2005
From: agebosma at home.nl (Age Bosma)
Date: Wed, 23 Mar 2005 23:55:59 +0100
Subject: [Matroska-devel] W3C is considering Binary XML (!)(?)
In-Reply-To: <4241E23A.5050703@free.fr>
References: <4241CE4D.4080901@home.nl> <4241E23A.5050703@free.fr>
Message-ID: <4241F3FF.3010000@home.nl>

Steve Lhomme wrote:
> 
> Thanks Prodoc ! (long time no see)
> 

Aye, I know, don't realy have an excuse though...except the fact that I 
fried my mainboard so I had to get me a new one, re-install Windows, 
decided to switch to Firefox and Thunderbird instead of Mozilla and 
because of this I didn't have an IRC client anymore...if that counts for 
an excuse :-D

The other day I came to the conclusion that I seem to be semi-involved 
in too many different open source projects without actually doing 
anything tangible, not enough to my liking that is... I've been diving 
into the C++ maze lately though. I'm a bit fed up with all the PHP, 
HTML, CSS and JavaScript stuff I've been doing for the past couple of 
years so I desided to dust off my programming "skills" (of you can 
actually talk of skills) from the time before that (MSX Basic and 
Pascal). This might give me a chance to actually "create" something again.
I'm also looking for a OpenSource project that can be combined with my 
education at uni so I can use uni time for the things I realy want to 
do, which is helping out with OS projects I find interesting.
I might be rushing thing a bit now but one of my possible plans is to 
work on the MusicBrainz tagger. The way they are heading (or trying to) 
imo offers an opportunity to implement easy nested tagging. Eventhough 
the tagger isn't actually ment for this purpose, tags in formats like 
mp3 and ogg will just be stored the way it's being done now, it does (in 
a way) create a nested view if the info so if Matroska support is added 
this can be used to store the tags nested in just the Matroska container.

*wonders why typing all this because of a small comment*

*...*

*desides to send the mail anyway*

Age


From steve.lhomme at free.fr  Thu Mar 24 10:12:07 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Thu, 24 Mar 2005 10:12:07 +0100
Subject: [Matroska-devel] W3C is considering Binary XML (!)(?)
In-Reply-To: <4241F3FF.3010000@home.nl>
References: <4241CE4D.4080901@home.nl> <4241E23A.5050703@free.fr>
	<4241F3FF.3010000@home.nl>
Message-ID: <42428467.2090805@free.fr>

Age Bosma a ?crit :
> Steve Lhomme wrote:
> 
>>
>> Thanks Prodoc ! (long time no see)
>>
> 
> Aye, I know, don't realy have an excuse though...except the fact that I 
> fried my mainboard so I had to get me a new one, re-install Windows, 
> decided to switch to Firefox and Thunderbird instead of Mozilla and 
> because of this I didn't have an IRC client anymore...if that counts for 
> an excuse :-D
> 
> The other day I came to the conclusion that I seem to be semi-involved 
> in too many different open source projects without actually doing 
> anything tangible, not enough to my liking that is... I've been diving 
> into the C++ maze lately though. I'm a bit fed up with all the PHP, 
> HTML, CSS and JavaScript stuff I've been doing for the past couple of 
> years so I desided to dust off my programming "skills" (of you can 
> actually talk of skills) from the time before that (MSX Basic and 
> Pascal). This might give me a chance to actually "create" something again.
> I'm also looking for a OpenSource project that can be combined with my 
> education at uni so I can use uni time for the things I realy want to 
> do, which is helping out with OS projects I find interesting.
> I might be rushing thing a bit now but one of my possible plans is to 
> work on the MusicBrainz tagger. The way they are heading (or trying to) 
> imo offers an opportunity to implement easy nested tagging. Eventhough 
> the tagger isn't actually ment for this purpose, tags in formats like 
> mp3 and ogg will just be stored the way it's being done now, it does (in 
> a way) create a nested view if the info so if Matroska support is added 
> this can be used to store the tags nested in just the Matroska container.

Very interresting. Yeah, I don't think nested tags exist anywhere so 
far. One of the many things Matroska will have to push (and why it takes 
more time to become usable). Most of the audio apps are not designed to 
handle a whole album in one file or it's very hacky. That's currently 
the main problem. But a tag-only application could be a good first step.

I'm currently working on VLC and they have a playlist. I wonder if they 
have any plans to support this kind of feature in the future. Otherwise 
any iTunes-like audio player could be a good option. (maybe TCMP too).

Good luck with C++, it's a really good choice ;)



From paul at msn.com  Thu Mar 24 19:44:27 2005
From: paul at msn.com (Paul Bryson)
Date: Thu, 24 Mar 2005 12:44:27 -0600
Subject: [Matroska-devel] Re: W3C is considering Binary XML (!)(?)
References: <4241CE4D.4080901@home.nl> <4241E23A.5050703@free.fr>
	<4241F3FF.3010000@home.nl>
Message-ID: <d1v1p4$d0d$1@sea.gmane.org>

"Age Bosma"  wrote...
>
> Aye, I know, don't realy have an excuse though...except the fact that I 
> fried my mainboard so I had to get me a new one, re-install Windows, 
> decided to switch to Firefox and Thunderbird instead of Mozilla and 
> because of this I didn't have an IRC client anymore...if that counts for 
> an excuse :-D

Chatzilla for FireFox available here:
http://hacksrus.com/~ginda/chatzilla/revs.html


Atamido 





From chris at matroska.org  Sat Mar 26 16:22:44 2005
From: chris at matroska.org (Christian HJ Wiesner)
Date: Sat, 26 Mar 2005 16:22:44 +0100
Subject: [Matroska-devel] Checklist for our next pack
In-Reply-To: <4235BE13.30105@free.fr>
References: <4235BE13.30105@free.fr>
Message-ID: <42457E44.7000609@matroska.org>

Hi all,

i am ok with most of it, but here some additions :

- Wavpack decoder, even if alpha still ( it works great for me here, and 
Kurt Noise tested it extensively with very positive results ). Maybe 
Toff will even have the time to add a splitter until we release the pack
- AC3 filter, as always
- vsfilter, as always
- CoreAAC, as always
- CoreVorbis, as alwas ( note : i wouldnt trust ffdshow for audio 
playback yet )
- matroska muxer, as always

Mechanisms :

- detect old matroska splitter and remove it, or better set VERY low 
merit ( elder Zoomplayer versions will insist on matroskasplitter, 
because of Toff's DShow interface to read track header info, which is 
missing in Haali's splitter )
- detect MPC and disable internal splitter ( like in Haali filter )
- detect  Realplayer and send warning in BIG RED letters
- detect ffdshow settings and DISABLE wmv playback ( important !! 
installing matroska pack 1.0.3 will break WMV playback on some machines )

I hate to admit i have never used ffdshow for Vorbis and AAC audio 
playback, so maybe i am wrong in insisting to pack CoreVorbis and 
CoreAAC in the pack. On the other hand,  both  work great and arent too 
big, so whats the point in leaving them away ?

Other chances :

- we should leave mmswitch out now, its not needed anymore thanks to 
haali's splitter with built in audio switcher
- we should also leave matrixmixer out now, as most modern audio decoder 
filters have built in channel downmixing capability

Comment welcome

Christian

Steve Lhomme schrieb:

> Hi,
> Satsuki is OK to make our next pack, but he wants a detailed 
> description of what we want. I actually have no definite idea of what 
> we want. Here is what I can think of :
>
> - Haali's splitter with only MKV reading enabled by default
> - Disable MPC's internal and MatroskaSplitter
> - FFDShow with H264 support, AAC, Vorbis and AC3
> - vsfilter
> - CoreFLAC
> - CoreTTA
> - RealMediaSplitter
>
> + should we ship MatroskaDiag ?
> + MatroskaProp with known bugs fixed ?
> + DScaler 5 for MPEG2 decoding ?
>
> * the same system we had to detect the presence of Real DLLs
> * new icon Betaboy has proposed
>



From rumbah at gmx.net  Sun Mar 27 04:02:31 2005
From: rumbah at gmx.net (Rumbah)
Date: Sun, 27 Mar 2005 04:02:31 +0200
Subject: [Matroska-devel] Problem with Haali's splitter
Message-ID: <42461437.4030802@gmx.net>

Hi!

I muxed a Nero AVC Video with a 6ch VBR HE-AAC, a 2ch VBR HE-AAC and chapters 
with mkvtoolnix 1.4.1 to a mkv file. Now if I try to play the file the screen 
stays blank but the audio starts and after about 10 seconds, the video starts to 
play at the correct time, so audio and video are in sync, but the first 10 
seconds are black. But if I load the file an a player and hit stop and then 
play, the file starts at the beginning with video without problems. The first 
ten seconds are only missing the first time the file is played, after that it works.

Rumbah


From steve.lhomme at free.fr  Sun Mar 27 11:11:40 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Sun, 27 Mar 2005 11:11:40 +0200
Subject: [Matroska-devel] Problem with Haali's splitter
In-Reply-To: <42461437.4030802@gmx.net>
References: <42461437.4030802@gmx.net>
Message-ID: <424678CC.2080001@free.fr>

What version of Haali's splitter are you using ? And what decoders 
(video and audio) ? (most video players can give you this info)

Rumbah a ?crit :
> Hi!
> 
> I muxed a Nero AVC Video with a 6ch VBR HE-AAC, a 2ch VBR HE-AAC and 
> chapters with mkvtoolnix 1.4.1 to a mkv file. Now if I try to play the 
> file the screen stays blank but the audio starts and after about 10 
> seconds, the video starts to play at the correct time, so audio and 
> video are in sync, but the first 10 seconds are black. But if I load the 
> file an a player and hit stop and then play, the file starts at the 
> beginning with video without problems. The first ten seconds are only 
> missing the first time the file is played, after that it works.
> 
> Rumbah
> _______________________________________________
> Matroska-devel mailing list
> Matroska-devel at lists.matroska.org
> http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-devel

-- 
robUx4 on blog <http://robux4.blogspot.com/>



From rumbah at gmx.net  Sun Mar 27 13:11:44 2005
From: rumbah at gmx.net (Rumbah)
Date: Sun, 27 Mar 2005 11:11:44 +0000 (UTC)
Subject: [Matroska-devel] Re: Problem with Haali's splitter
References: <42461437.4030802@gmx.net> <424678CC.2080001@free.fr>
Message-ID: <loom.20050327T130238-624@post.gmane.org>

Steve Lhomme <steve.lhomme <at> free.fr> writes:

> 
> What version of Haali's splitter are you using ? And what decoders 
> (video and audio) ? (most video players can give you this info)


I tried version 19/03 and 22/03 of the splitter, the result was the same. The
video decoder was the Nero decoder, as aac decoder I used the MPC build in and
CoreAAC.
In addition to that, the MPC splitter has the same problem, version 6.4.8.4
normally works with Nero AVC, too.



From steve.lhomme at free.fr  Wed Mar 30 14:12:36 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Wed, 30 Mar 2005 14:12:36 +0200
Subject: [Matroska-devel] Re: Google Search
In-Reply-To: <423D519A.7060503@free.fr>
References: <423D519A.7060503@free.fr>
Message-ID: <d2e4vb$j91$1@sea.gmane.org>

Steve Lhomme a ?crit :
> Hi,
> 
> As you know, it's always hard to find the video you're looking for on 
> your HD except if you use very long file names and you are very organised.
> 
> Google has a desktop search tool that could make it easier to find 
> videos on your HDD. Apparently there is an open API to add your own file 
> type. Adding Matroska support there could be a good thing for users, and 
> for the project in general. So if you have time to look at this it could 
> be a great thing. Otherwise I'll have a look after I'm done with menu in 
> VLC, various fixes here and there, a muxer in VLC (and before the editor 
> we need).
> 
> Imagine searching on your system: "video using RealVideo with a 16:9 
> aspect ratio and smaller than 30 MB" or "Tracks which producer is Trevor 
> Horn".
> 
> http://desktop.google.com/developer.html

Well, Jory did it but didn't publish binaries yet (the code is in SVN).

One thing that could be done is to make the SegmentUID known to the 
search engine (via the displayed text if I understand correctly how it 
works). This way the engine could be later used to find matching segment 
UIDs on the system for linked files (like .part files being loaded for 
example ?). The same is usefull for Family UIDs.

I wonder if it would be possible to tweak eMule to look for a segment 
UID rather than a filename...

Maybe when I'll have Tiger (OS X.4) I could add a smiliar feature to 
Spotlight too.



From steve.lhomme at free.fr  Thu Mar 31 13:27:04 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Thu, 31 Mar 2005 13:27:04 +0200
Subject: [Matroska-devel] Japanese translations
Message-ID: <424BDE88.2070001@free.fr>

Hi Yudai,

Again, thanks for your great work. There is just a little thing you 
didn't notice. When you add a language page that was not existing, you 
need to update this page and the same page in other languages too. You 
add to add this line at the top (with the others) :

<!--#set var="jap" value="yes" -->

I'm going to add it for the ones you added yesterday.

cya

-- 
robUx4 on blog <http://robux4.blogspot.com/>


From Liisachan at faireal.net  Thu Mar 31 15:14:07 2005
From: Liisachan at faireal.net (Liisachan)
Date: Thu, 31 Mar 2005 22:14:07 +0900
Subject: [Matroska-devel] Japanese translations
In-Reply-To: <424BDE88.2070001@free.fr>
References: <424BDE88.2070001@free.fr>
Message-ID: <20050331152058.CC65B440052@p15097576.pureserver.info>

Hi,

This might be very trivial, I don't mean to be picky,
but it'd be advisable to use "jpn" instead of "jap" to avoid any 
possible misunderstanding.

ordinary people (including myself) usually don't care,
but technically, "jap" is *doubly* offensive slang
meaning:

 (a) "Japanese"
http://www.askoxford.com/concise_oed/jap?view=uk

and
(b) "Jewish woman"
http://encarta.msn.com/encnet/features/dictionary/DictionaryResults.aspx?refid=1861622917

I guess that's why the ISO 3-letter code is JPN,
for "political correctness."


Best regards,

Liisachan



Steve Lhomme <steve.lhomme at free.fr> wrote:

> Hi Yudai,
> 
> Again, thanks for your great work. There is just a little thing you 
> didn't notice. When you add a language page that was not existing, you 
> need to update this page and the same page in other languages too. You 
> add to add this line at the top (with the others) :
> 
> <!--#set var="jap" value="yes" -->
> 
> I'm going to add it for the ones you added yesterday.
> 
> cya
> 
> -- 
> robUx4 on blog <http://robux4.blogspot.com/>
> _______________________________________________
> Matroska-devel mailing list
> Matroska-devel at lists.matroska.org
> http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-devel



From steve.lhomme at free.fr  Thu Mar 31 18:46:06 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Thu, 31 Mar 2005 18:46:06 +0200
Subject: [Matroska-devel] Japanese translations
In-Reply-To: <20050331152058.CC65B440052@p15097576.pureserver.info>
References: <424BDE88.2070001@free.fr>
	<20050331152058.CC65B440052@p15097576.pureserver.info>
Message-ID: <424C294E.80300@free.fr>

Liisachan a ?crit :
> Hi,
> 
> This might be very trivial, I don't mean to be picky,
> but it'd be advisable to use "jpn" instead of "jap" to avoid any 
> possible misunderstanding.

The usual rule : if it ain't broken, don't fix it. This thing is not 
displayed, it's just some server-side code.

> ordinary people (including myself) usually don't care,
> but technically, "jap" is *doubly* offensive slang
> meaning:
> 
>  (a) "Japanese"
> http://www.askoxford.com/concise_oed/jap?view=uk
> 
> and
> (b) "Jewish woman"
> http://encarta.msn.com/encnet/features/dictionary/DictionaryResults.aspx?refid=1861622917
> 
> I guess that's why the ISO 3-letter code is JPN,
> for "political correctness."

I hate these 2 words together. I prefer the real world ;)



From Liisachan at faireal.net  Thu Mar 31 19:40:23 2005
From: Liisachan at faireal.net (Liisachan)
Date: Fri, 01 Apr 2005 02:40:23 +0900
Subject: [Matroska-devel] Japanese translations
In-Reply-To: <424C294E.80300@free.fr>
References: <20050331152058.CC65B440052@p15097576.pureserver.info>
	<424C294E.80300@free.fr>
Message-ID: <20050331174028.D5617440052@p15097576.pureserver.info>

Hi,

> The usual rule : if it ain't broken, don't fix it. This thing is not 
> displayed, it's just some server-side code.

Another usual rule: keep out of harm's way. Just because no one 
is likely to witness it now, doesn't justify it forever.

It might have been broken already if the translators had been 
more sensitive. Isn't this supposed to be an open-source project 
where anyone can join anytime? like memory leak, this could be 
potentially risky, at least indecent, even tho things may look 
like working now, ad hoc.

for the record i warned you: Telling someone from Japan to 
add the comment "jap" to his/her work is like asking someone 
from Mexico to identify himself as "wetback."

Personally _I_ don't mind, but I can foresee that nothing good 
will come out of this when more people from Japan happen to join 
you. It may happen if Matroska goes on for a few more years.
Or maybe not.

Liisachan


Steve Lhomme <steve.lhomme at free.fr> wrote:

> Liisachan a ecrit :
> > Hi,
> > 
> > This might be very trivial, I don't mean to be picky,
> > but it'd be advisable to use "jpn" instead of "jap" to avoid any 
> > possible misunderstanding.
> 
> The usual rule : if it ain't broken, don't fix it. This thing is not 
> displayed, it's just some server-side code.
> 
> > ordinary people (including myself) usually don't care,
> > but technically, "jap" is *doubly* offensive slang
> > meaning:
> > 
> >  (a) "Japanese"
> > http://www.askoxford.com/concise_oed/jap?view=uk
> > 
> > and
> > (b) "Jewish woman"
> > http://encarta.msn.com/encnet/features/dictionary/DictionaryResults.aspx?refid=1861622917
> > 
> > I guess that's why the ISO 3-letter code is JPN,
> > for "political correctness."
> 
> I hate these 2 words together. I prefer the real world ;)
> 
> _______________________________________________
> Matroska-devel mailing list
> Matroska-devel at lists.matroska.org
> http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-devel
> 



From steve.lhomme at free.fr  Thu Mar 31 20:00:17 2005
From: steve.lhomme at free.fr (Steve Lhomme)
Date: Thu, 31 Mar 2005 20:00:17 +0200
Subject: [Matroska-devel] Japanese translations
In-Reply-To: <20050331174028.D5617440052@p15097576.pureserver.info>
References: <20050331152058.CC65B440052@p15097576.pureserver.info>	<424C294E.80300@free.fr>
	<20050331174028.D5617440052@p15097576.pureserver.info>
Message-ID: <424C3AB1.7030001@free.fr>

Liisachan a ?crit :
> Hi,
> 
> 
>>The usual rule : if it ain't broken, don't fix it. This thing is not 
>>displayed, it's just some server-side code.
> 
> 
> Another usual rule: keep out of harm's way. Just because no one 
> is likely to witness it now, doesn't justify it forever.
> 
> It might have been broken already if the translators had been 
> more sensitive. Isn't this supposed to be an open-source project 
> where anyone can join anytime? like memory leak, this could be 
> potentially risky, at least indecent, even tho things may look 
> like working now, ad hoc.
> 
> for the record i warned you: Telling someone from Japan to 
> add the comment "jap" to his/her work is like asking someone 
> from Mexico to identify himself as "wetback."

I hope you don't go to every coders asking them to change their 
mnemonics because it might offense someone in a language (like matroska 
means sailor shirt in some).

If you have a "politically correct" dictionary, please review the rest 
of our code to see if we didn't offense anyone. Like on this page :

http://svn.matroska.org/svn/www/trunk/www.matroska.org/data/include/page-head.html

Each language is in english using the first 3 letters. Hopefully "fre", 
"rus", "spa", "por", "ger", "dut" and "ita" are not offense too.

BTW, "ja" means "yes" in german.

> Personally _I_ don't mind, but I can foresee that nothing good 
> will come out of this when more people from Japan happen to join 
> you. It may happen if Matroska goes on for a few more years.
> Or maybe not.

Aren't japanese coders using mnemonic too ? To make the code easy to 
read ? I'm sure anyone can understand the choice made here.

And finally in matroska we do french, german, texas, etc bashing all the 
time. No offense to anyone, just having fun.



From Liisachan at faireal.net  Thu Mar 31 21:57:50 2005
From: Liisachan at faireal.net (Liisachan)
Date: Fri, 01 Apr 2005 04:57:50 +0900
Subject: [Matroska-devel] Japanese translations
In-Reply-To: <424C3AB1.7030001@free.fr>
References: <20050331174028.D5617440052@p15097576.pureserver.info>
	<424C3AB1.7030001@free.fr>
Message-ID: <20050331195802.BA26044005A@p15097576.pureserver.info>

Steve,

Please understand that I didnt say all those things because I 
hate you or something, but because I'd like to kinda help you 
(you guys might really be in trouble because of this: quite a 
few Japanese instantly react more or less emotionally if you use 
the word jap--this problem is way more serious than you might 
think, for historical reason), and also because I didn't want to 
see the young person from Japan, willing to help you guys, 
discouraged all of a sudden because of a trivial thing.

Well, I'd say...you could have just said "Ok, this abbreviation 
is not good, but it's just naming convention and it's too late 
to change." or something, rather than like "Any word could be 
offensive. You felt insulted? So what?"

I already have said what I should.
I know you didn't mean to be rude.

Like I've been repeating, personally I don't mind at all
and I agree with you. Personally I even kinda hate those who 
insist you shouldn't abbreviate japanese as jap. They are kinda 
silly. But we can't ignore the reality.

Even in anime fandom, where ppl say they love Japan,
they often say jap for japan innocently
without knowing that is an offensive word and quite a few 
Japanese would get mad if they heard that.
Usually I say nothing in such a conversation,
because it's not my problem, and actually I don't mind that so 
much.

> Each language is in english using the first 3 letters. Hopefully "fre", 
> "rus", "spa", "por", "ger", "dut" and "ita" are not offense too.

We did the same thing for those long ago, but "may" for Malay, 
not "mal"
http://ld-anime.faireal.net/guide/mkv-may
"jpn" for Japan, not "jap"
http://ld-anime.faireal.net/guide/mkv-jpn

using the first 3 letters doesn't work to distinguish
Malayalam, Malagasy, and Malay.
But I think you know that's not the point...

> Aren't japanese coders using mnemonic too ? To make the code easy to 
> read ? I'm sure anyone can understand the choice made here.

Sure anyone understand jap stands for japan(ese), but I'm sure 
not everyone would say that is a good choice.

if you really want to know,
usually they use "j" or "ja" for "japanese" and "jp" for "Japan"
and "jpn" if you need 3 letters. Not always but usually.

I'm not blaming you. Just so you know.
Knowing it might help you to get along with someone from japan,
working together for a great project. Hopefully.
They are usually modest and polite (unlike me :P) and kinda shy 
and wouldn't say anything if you made them angry.


Liisachan