[LinuxDVD] word of mouth about css is out
Michael Holzt holzt@multimediahaus.de
Thu, 15 Jul 1999 17:58:49 +0200 

See Slashdot. I'm unsure if this is really good.

-- 
mit freundlichen Grüßen,

Michael Holzt			Multimediahaus GmbH
Technik				D-58540 Meinerzhagen
holzt@multimediahaus.de		+49 2354 9296-0

[LinuxDVD] word of mouth about css is out
Paul Volcko pvolcko@concentric.net
Thu, 15 Jul 1999 10:43:34 -0400 

> See Slashdot. I'm unsure if this is really good.

It puts the anonymous supplier of the code on the spot. I also 
wonder why the supplier was left anonymous when his name is in 
the archives here anyway. 

It's being posted on Slashdot will likely draw the attention of 
someone either at Matsushita, the DVD Forum, or some 
copyright/license attorney. I hope that Mathew P. and the 
"anonymous" coder are ready.

I don't think the industry will make any kind of changes as has 
been voiced as a fear in the slashdot comments and here, I think. 
There isn't really anything they could do to change it without 
making the entire installed base incompatible. No one is going to 
do that.

Hopefully the response will be nothing more than a cease and 
desist order. I'd be ready for something a little more strenuous 
though.

[LinuxDVD] word of mouth about css is out
Andreas Bogk andreas@andreas.org
15 Jul 1999 16:46:22 +0200 

Michael Holzt <holzt@multimediahaus.de> writes:

> See Slashdot. I'm unsure if this is really good.

Especially as the information about CSS is not complete yet. The bulk
encryption routines are still missing.

Andreas

-- 
"We show that all proposed quantum bit commitment schemes are insecure because
the sender, Alice, can almost always cheat successfully by using an
Einstein-Podolsky-Rosen type of attack and delaying her measurement until she
opens her commitment." ( http://xxx.lanl.gov/abs/quant-ph/9603004 )

[LinuxDVD] word of mouth about css is out
Colin Davis cdavis@thepentagon.com
Thu, 15 Jul 1999 18:52:17 +0000 

Michael Holzt wrote:
> 
> See Slashdot. I'm unsure if this is really good.
> 

Damn it! This was a little soon. Well, this should help get the legality
issues delt with. The discussion is all ready up to 70 comments.

Johan, I believe your code is unique, yes? Where are you located? (USA,
Europe, ect.) as this will effect which copyright/patent laws your code
is effected by. 
Hadn't we decided that the Ac-3 decoder by Aaron Holtzman was fairly
legal as well? 

I believe at this point, all of the major _technical_ issues, have been
worked out in order to release a open sourced decoder.

Mpeg 2 decoding -> http://mpeg.openprojects.net/
CSS		-> Johan Addis's work OR the one posted on slashdot
(http://livid.on.openprojects.net/)
Ac-3		-> Aaron Holtzman decoder (http://www.engr.uvic.ca/~aholtzma/ac3/)

[LinuxDVD] word of mouth about css is out
Matthew R. Pavlovich mpav@purdue.edu
Thu, 15 Jul 1999 13:13:14 -0500 (EST) 

> I believe at this point, all of the major _technical_ issues, have been
> worked out in order to release a open sourced decoder.

This is all for a software decoder. There are a lot of hardware options.
That is what Johan's work helps with.

> 
> Mpeg 2 decoding -> http://mpeg.openprojects.net/
> CSS -> Johan Addis's work OR the one posted on slashdot
> (http://livid.on.openprojects.net/)

Same code. There are two parts to CSS. The drive unlocking and the
stream decoding. True hardware DVD decoders will decrypt the stream data
(Zoran 36710 for example).

> Ac-3 -> Aaron Holtzman decoder (http://www.engr.uvic.ca/~aholtzma/ac3/)
> 

Also done in hardware on a lot of decoders.

Matt

[LinuxDVD] word of mouth about css is out
Paul Volcko pvolcko@concentric.net
Thu, 15 Jul 1999 14:36:29 -0400 

> Hadn't we decided that the Ac-3 decoder by Aaron Holtzman was fairly
> legal as well? 

I didn't get that impression. Last I knew Aaron had contacted 
Dolby and requested their position on it. I haven't heard anythign 
since then. An update from Aaron would probably be good. :)

> I believe at this point, all of the major _technical_ issues, have been
> worked out in order to release a open sourced decoder.

> Mpeg 2 decoding -> http://mpeg.openprojects.net/

I've heard mixed reaction to the software decoder at this site. I 
can't remember what the specific problems with it were, but I do 
remember some concern about it. Giving it the benefit of the doubt 
and forgetting those potential problems... yes this technical 
problem is solved. 

I haven't looked at it myself but if it is free and legal then I'm 
guessing that it is only an implementation of a decoder, not a 
pieceof software that provides output to a video monitor (in other 
words it wil output a RGB or other format decoded signal set, but 
you still have to figure out how to send them to a monitor and all 
that). That classifies it as an implementation as such it can be 
free and freely distributed. Once it is put into something that will 
display to a monitor of some kind, though, then the $4 royalty 
comes into play (paid by the "system" producer, not the 
implementation producer).

> CSS -> Johan Addis's work OR the one posted on slashdot
> (http://livid.on.openprojects.net/)

They are both one and the same I think. Either way, it doen't do 
decryption and it may not actually do disc key passing correctly. 
It is an incomplete solution needing more work. This technical 
problem is *not* solved. Legalities of how this code was arrived at 
and actually using it in any given product are also up in the air right 
now.

A patch of sorts to the CSS code released here yesterday was 
sent out on the livid-dev list today. The assembly code was 
converted it to C according to the poster. I haven't looked at it yet 
so I'm not sure what level of recoding was done (an actual C 
algorithm implementation or simply a C function encapsulation of 
the assembly code). If the former, I'm impressed. That was a lot 
of assembly code to convert in under 24 hours on your own.

> Ac-3 -> Aaron Holtzman decoder (http://www.engr.uvic.ca/~aholtzma/ac3/)

This work entirely, from what I hear, but again licensing issues are 
not yet resolved. But yes, the technical problem of decoding AC-3 
streams has been solved in full.

Paul Volcko
LSDVD Project

[LinuxDVD] word of mouth about css is out
Aaron Holtzman aholtzma@ess4.engr.UVic.CA
Thu, 15 Jul 1999 12:25:15 -0700 

It would seem that Paul Volcko (pvolcko@concentric.net) said:
> I didn't get that impression. Last I knew Aaron had contacted 
> Dolby and requested their position on it. I haven't heard anythign 
> since then. An update from Aaron would probably be good. :)
> 
No news is good news right :) I've had no response from them, but from
experience they usually take around a month to reply. I keep having
these nightmares about armies of lawyers sharpening their pencils.

> > Mpeg 2 decoding -> http://mpeg.openprojects.net/
> 
> I've heard mixed reaction to the software decoder at this site. I 
*snip*
> problem is solved. 

This MPEG decoder isn't exactly fast per se. It don't think it was ever
designed to do realtime playback at any reasonable resolution. I think
it needs major work to be on par with similar Windows based decoders.

On the subject of performance, I'm wondering what kind of CPU utilization
people are seeing using ac3dec-0.5.3. On my Ultra IIi 300 I can play back
2 channel streams at about ~2% utilization. I heard some numbers as 
high as %40 on an AMD K-6 400. There might be something screwy with
the OSS driver that is causing that though. I'll be moving my development
to a dual celeron box soon, so it'll be a little easier to fix these
platform related problems.

cheers,
aaron