Weekly Edition
Return to the Development page
| |||||||||||||||||||
|
| |||||||||||||||||||
Dirac 1.0.0 released.
Dirac 1.0.0 has been released. The release tarball can be downloaded from http://diracvideo.org/download/dirac-research or http://sourceforge.net/projects/dirac Dirac Compression System ------------------------- Dirac is a video compression system that provides general-purpose video compression and decompression tools comparable with state-of-the-art systems. Dirac is a format and compression technology, not a specific codec. The most important part of the Dirac project is the specification which can be found at http://diracvideo.org/specifications Dirac Project ------------- The initial implementation of Dirac at resulted in a research implementation which is referred to as the Dirac codebase. It is now called dirac-research. It is a reference implementation and also a codebase for future algorithmic developments. Dirac 1.0.0 is the release of this codebase. Dirac website ------------- Information about Dirac and the associated projects can be found at http://diracvideo.org What's new in this release -------------------------- Several alpha and beta releases have preceded this release. The changes in 1.0.0 release are: - Compliance with Dirac Bytestream specification 2.2.2 - Adaptive GOP structure - Improved motion estimation. - Improved pre-filtering. - Major code refactor of encoder classes. - Added conversion utility for horizontal 3/4 filtering. - DirectShow Filter released to be able to play back Dirac v2.2.2 files raw bytestreams and Dirac wrapped in AVI in Windows Media Player and MPlayer Classic. Dirac website ------------- Information about Dirac and the associated projects can be found at http://diracvideo.org Dirac content creation and playback ----------------------------------- Information on the end-user applications that support Dirac can be found at http://diracvideo.org/wiki/index.php/Main_Page#End_user_a... (Log in to post comments)
the licence is GPL, LGPL, MIT, and MPL Posted Sep 17, 2008 12:34 UTC (Wed) by coriordan (guest, #7544) [Link] The licences used are GPL, LGPL, MIT, and MPL, and it seems all parts can be relicensed under GPL/LGPL. Here's the explanation from their FAQ:
What are the license conditions?
the licence is GPL, LGPL, MIT, and MPL Posted Sep 17, 2008 15:05 UTC (Wed) by stevenj (guest, #421) [Link] This seems a bit silly, although harmless. The MIT/X11 license is sufficient to be compatible with essentially all other free-software/open-source licenses, and is strictly more permissive than the GPL, LGPL, or MPL.
the licence is GPL, LGPL, MIT, and MPL Posted Sep 17, 2008 15:42 UTC (Wed) by zooko (subscriber, #2589) [Link]
So as I understand it, currently the Schroedinger implementation can be extended with proprietary extensions and the resulting product shipped to 3rd parties, but the dirac-research implementation cannot, because it is distributed only under reciprocal licences (Mozilla, GPL, LGPL).
This would be different if the dirac-research implementation were also available under MIT, as the schroedinger implementation is. In that case, people would be able to extend the dirac-research implementation with proprietary changes and redistribute the result.
the licence is GPL, LGPL, MIT, and MPL Posted Sep 17, 2008 16:11 UTC (Wed) by drag (subscriber, #31333) [Link]
Ya..
The MIT license doesn't address the patent issue in any manner. Which is possibly one of the reasons why it's chosen. (they want Dirac to be accepted by everybody and not just the open source crowd. That means working with proprietary software and that means patent issues)
If I remember correctly the mp3 codecs from Fluendo are under the MIT license, but the patent licensing that goes along with the binaries and source code effectively make them 'proprietary' codecs.
So by releasing the codec under MIT and GPL-related licenses (which loosely covers patent-related issues in a non-bulletproof manner) Dirac is effectively assuring us that they have a commitment towards Free media codecs and keeping Dirac open that you won't get if you do pure-bsd or pure-mit license then have end users convert it over GPL.
In other words, by releasing under the GPL they are making a commitment towards covering end users from related patents they own. This commitment is not expressed if they release it under a pure-MIT license.
the licence is GPL, LGPL, MIT, and MPL Posted Sep 18, 2008 7:57 UTC (Thu) by job (subscriber, #670) [Link]
It may be interesting to note that the Ogg codecs are under BSD license. I think Stallman adviced this, if I remember correctly, because it is reference code and stands to gain from use in proprietary software as well.
Stallman on Ogg Vorbis using permissive licence Posted Sep 18, 2008 9:05 UTC (Thu) by coriordan (guest, #7544) [Link] Yup. Here's Stallman's quote: I agree. It is wise to make some of the Ogg Vorbis code available for use in proprietary software, so that commercial companies doing proprietary software will use it, and help Vorbis succeed in competition with other formats that would be restricted against our use. That quote is from 2001. I wonder would it actually be better, in these rare cases when a permissive licence is best, for the developers to *also* release under the GNU GPL so that the patent and DRM non-restriction promises are clear. Hmm.
Stallman on Ogg Vorbis using permissive licence Posted Sep 18, 2008 13:24 UTC (Thu) by zooko (subscriber, #2589) [Link]
That appears to be what they are doing with the schroedinger project -- it is released under both GPL and MIT. Not so with the dirac-research project, which is released only under reciprocal licences.
Now that job and coriordan have pointed out these strategic considerations, it seems like they probably chose to do it this way because of these strategic considerations: they want the schroedinger implementation to be re-used in proprietary products for better market penetration of the codec itself, while they want the dirac-research implementation to have enforced reciprocality so that new innovations have to be open.
Just a guess.
Awesome! Posted Sep 17, 2008 12:36 UTC (Wed) by Tuxie (guest, #47191) [Link]
I'm really looking forward to some professionally made quality comparisions using real video sources such as DVD and BluRay rips, and encoding support in common video encoding applications.
Awesome! Posted Sep 17, 2008 16:14 UTC (Wed) by drag (subscriber, #31333) [Link]
Hopefully it will make a big splash.
It's designed specifically for running HD content over the Internet...
what does it *look* like? Posted Sep 17, 2008 16:25 UTC (Wed) by zooko (subscriber, #2589) [Link]
Yeah, if anybody has an actual video encoded in the Dirac codec, please share.
E-mail zooko@zooko.com, or upload it to the tahoe test grid:
Dirac 1.0.0 released. Posted Sep 17, 2008 19:11 UTC (Wed) by ttonino (subscriber, #4073) [Link]
Dirac encoded video can be found at http://dirac.kw.bbc.co.uk/download/video/maybefinal/
Three questions Posted Sep 18, 2008 5:15 UTC (Thu) by jd (guest, #26381) [Link]
First, will the BBC eventually use this for their iPlayer project? It would seem obvious to use a compression system that doesn't NEED to be deployed on 3rd party systems before it's used anywhere else, and the less bandwidth the BBC uses to serve iPlayer clients, the more clients they can handle, and ergo the more content they can make available.
Second, I have to wonder if this is being considered for the new Japanese ultra-high definition television. When you're talking about resolutions 24 or more times as great as high definition TV, you're obviously not going to be using MPEG II. At least, not if you still intend on supporting a thousand or so channels on the same fiber. Dirac might not get you all the gain you need, but any gain is good gain.
Last, it's nice that people are hosting some Dirac footage, but to get it "well-known", you really need a Youtube-like site that hosts it and (as far as necessary) plugins for browser/OS combos that don't support it. Any idea if anyone's doing anything like that?
|
|
||||||||||||||||||