"The WebM project - a VP8 video stream and a Vorbis audio stream wrapped in a Matroska container re-branded as a WebM container - launched by Google, openly supported by every major chip maker, is going to be the major codec for Google's Chrome web browser. Yes, Google is dropping H264 support from the Chrome web browser."

http://www.osnews.com/story/24243/Google_To_Drop_H264_Support_from_Chrome

Come, gather around, and watch how a standard is being formed...

How? Dictated by Google?

Yes.

Not good.

As long as the codec is supported by ffmpeg, I can't care less (though H264 is currently a bit faster than VP8/WebM). :)

Quote:

Fortunately. By the way, I've been told that the guy who invented the Matroska container was on drugs when he did so... Apparently, it's a crazy format, compared to the well known AVI standard.

I have HD video I downloaded to my system to test play back that came in a Matroska container. Its works fine.

H.264 is a standard created by ISO (International Standards Organisation) MPEG and ITU (International Telecommunications Union). It was created by experts from various companies. Every aspect of it is documented although the specification is not free. Every feature is backed up by experiments and scientific proof that it adds value.

WebM is not a standard. It was created by one company who sold the patents and source to Google who have now tweaked it and released it as free. This does not make it a standard.

By dropping the support for H.264 from Chrome, Google is limiting our choice of formats we should use for video. But they still support Flash video which is H.264 in a Flash container so it's not that supporting H.264 is a problem...

Unless someone can show evidence that WebM provides higher quality for lower bit rate or easier coding/decoding then this is just politicking/marketing. Actually, by "dropping support for H.264" but still supporting Flash video then it's a completely null argument because H.264 in it's own container would be more efficient.

By the way, I was looking at adding support for Matroska to my own software but the overhead looks huge and the only place I could find that used Matroska was torrents; not good for commercial software.

Quote:
Theora comes from the same source btw.
Quote:
WebM doesn't provide any compression. It's a container format based on matroska.

VP8 is the video codec used. It has very similar compression ratio to H.264.

> WebM is not a standard. [...] This does not make it a standard.

While not being a de jure standard it may become a de facto standard.

http://videojs.com/2011/01/html5-video-statistics/

> not good for commercial software.

MKV support is offered by both commercial software and commercial hardware:

http://en.wikipedia.org/wiki/Matroska#Software_support
http://en.wikipedia.org/wiki/Matroska#Hardware_support

@koan

Chrome uses FFMpeg to decode videos, anyway. So removing H264 support is entirely artificial, from a technical point of view. It's just a way to enforce google's VP8/WebM (and possibly avoid paying H264 fees).

I've looked at Theora and I would not like to have to implement it. It looks a bit of a mess - many design decisions were not justified, simply trying to avoid using something like H.264 to avoid patent issues. But at the end of the day there are only so many ways to encode video efficiently that we know of, inevitably they chose similar but ever-so-slightly different schemes. With their hands tied in this way it is difficult to make something better.

WebM, http://www.webmproject.org/about/, specifies VP8 video and Vorbis audio so for all remarks concerning video quality WebM/VP8 are interchangeable. Iff VP8 quality is "as good as H.264" I would like to see scientific proof.

Google aren't avoiding H.264 fees because, as I said, if they continue to support H.264 video in a Flash container they are just as liable. It's politicking because as Fab said, it makes no difference. I prefer to wait for HTML5 and see if W3C set baseline format support for the video tag; they never specified what formats were supported for the img tag so it could be that they do the same thing again.

My point was that Matroska files are not supported by professional video editing tools.

Quote:

I think it has more to do with the latter!

Anyway, I think standards in general are good, and particularly when they are free, non-proprietary and open. Everyone will benefit from a free, common ground to stand on.

Of course everyone is totally free to try to build their own, proprietary solutions in parallel, but this WebM will be something every chip manufacturer supports, every gadget manufacturer supports, every sw manufacturer supports and every website using video.

@koan

AFAIK, Flash will now include WebM/VP8 support, and as you know, video isn't the only thing Flash is used for (in Chrome or any other web context).

Quote:

Which is exactly what is about to happen.

> Which is exactly what is about to happen. :-)

Yes, as indicated by the statistics I linked to :-)

Quote:

Relax Andreas! You just lay back and relax for a year or two, and watch it happen!

> Relax Andreas!

I'm very much relaxed already :-)

> You just lay back and relax for a year or two, and watch it happen!

I wasn't ironic. Did you even read that article and looked at the stats? I can see that WebM is on its way to become a de facto standard.

OK then!

> http://videojs.com/2011/01/html5-video-statistics/

Some more statistics:

http://www.html5test.com/videowars.png