http://arstechnica.com/security/news/2009/...er-security.ars
http://shaver.off.net/diary/2009/01/26/advancing-open-video/
http://www.0xdeadbeef.com/weblog/?p=977

Excellent news!!!!

Monty is probably quaking in his pants. Let's just hope the check is made out to Xiph.org! The only problem is the money is there, but who is going to actually put the developement efforts in? does this mean we will be seeing more "forks"? Does anyone know if Theora can scale to HD yet? They were still working on that last time I checked. Monty has a new working model called "Thusnelda" branch, which is suppose to take care of of the motion compensation problems inherient to transform coders. The output looks pretty good and by the looks of it they still have a ways to go if they want to get it on par with H.264. I am sure $100,000 will get things moving a little faster. . In terms of Vorbis we maybe seeing a lot of work being done on the side project "Ghost" which is just a prototype right now. The future looks bright indeed.

Future does look brighter, but only a bit. I don't know anything of the Foundation but is there actually a hard core team working on video and audio codec? Can just 100.000$ get the development wheel spin at a higher speed? Does the gift impact on their budget?

I hope Xiph will give something to Dirac too...

This has to be seen in the context of HTML5 development.

Mozilla wanted to make Ogg Theora/Vorbis a part of the standard as the "default" audio video format.

Apple rejected it because "it would open them to liability". Mozilla rejected Apple's suggestion of patented Quicktime+MPEG4.

I guess the main idea of this grant is to make the codecs more easily usable and widespread, so they become the de facto standard anyway.

I'm pretty confused about this issue now. I just did a bit of reading about this dispute, and came across the following quote:

This supports what I believed, about the prominence of MP4 codecs in the FOSS market. Certainly, I've used FFMPEG based software plenty; haven't we all? But later on that same page, someone posted this comment:

This was news to me, though I'm not all that surprised. Nokie and Apple have put forward a very good point though: part of HTML 5's aim is to unite standards across all mediums, including PCs, STBs, mobile phones and PDAs, but there are no hardware implementations of Theora or OGG (unless I'm misinformed?). Meanwhile, H.264 is well supported across all platforms. This is ignoring, for the moment, the patents issue.

Seems to me like this debate won't be resolved any time soon. The donation to Wikipedia/Xiph is very generous, and I hope it sees Theora (and Vorbis) through to much brighter days. However, without hardware support, it seems Xiph's codecs won't be winning over the HTML 5 working group. Perhaps Mozilla should donate some money to hardware decoder developers to see Theora/Vorbis support there?

That's my only question?. I am guessing that since there is $100,000 there that a team will be created or they will assist Monty in the development of Theora (I think there is already a small team of people working on it). Vorbis on the other hand I am unsure of. Nobody was really putting any development efforts into it outside of Aoyumi. Maybe they will be able to create a working group for that also.




Are you saying that the HTML 5.0 standard would have W3C tags that would consist of audio/video standalone? that's pretty interesting.




There are a few hardware implementations of Theora in terms of hardware. The only reason you don't see more manufacturers using it, because it's still in the beta stages. It's quality is somewhere between XVid and H.264, henceforth what the "Thusnelda" project is for. What are you talking about? Vorbis has plenty of hardware support on DAP that include Cowon and Sandisk for that matter. I am tired of hearing this argument that it's not well supported on hardware. When you say "well supported" do you mean on most players out there or are you really just referring to the IPod? Apple refuses to support Vorbis for example (this doesn't come as a surprise), because it's an open source (not a standard) and it's in direct competition with it. The same thing goes for FLAC.

http://dev.w3.org/html5/spec/Overview.html#video

What he means is that if you buy a modern SoC like for example a TI OMAP3 (which powers gazillions of mobile phone models), this chip has built-in hardware decoders for MPEG4.

You can still run the Theora and Ogg decoders in the software, but that'll eat more battery. The number of real chip solutions for Theora and Ogg remains low.

But the hardware manufacturers follow the software popularity as well. Get your codec widely distributed in a popular place (ahem, Wikipedia) and hardware follows.

Very cool. It appears as though the future of the web is going to be more multimedia driven? There is some cool stuff in this working draft here. It definitely would make things much easier if they had multimedia tags. I sympathize with them. Getting a majority of the web browsers to support the new spec. That's another task in and of itself.



I see what you are saying now.

Great news.
I think the development should be focused on video codec. Vorbis is very mature nowdays and is comparable to all modern audio codecs. While last version of Theora (thusnelda) isn't good even if to compare with MPEG-4 ASP http://www.0xdeadbeef.com/weblog/?p=977
http://vorbis.org.ru/phpBB3/viewtopic.php?f=2&t=589

And video's bitrates are 10x higher than audio's.

I think they have built in TMS cores, which are software programmable DSPs (which is why they "accelerate" everything from MPEG to MS codecs and more depending on what codecs you put in them). Looking at the TI website for the OMAP3, they even have software updates for the "hardware" decoders.



Out of curiosity, is it much higher for MPEG? Most "hardware" decoders I've seen are just DSP cores running ordinary software.