Slashdot Mirror
Google Enables VP9 Video Codec In Chromium
An anonymous reader writes "Last month, Google revealed that it was planning to finish defining its VP9 video codec on June 17 (today), after which it will start using the next-generation compression technology in Chrome and on YouTube. The company is wasting no time: it has already enabled the free video compression standard by default in the latest Chromium build."
I went to the Mozilla IRC (have Chatzilla installed before clicking) and typed this:
firebot: vp9 bugs
and got this:
Bug https://bugzilla.mozilla.org/show_bug.cgi?id=833023 , , nobody, NEW, Implement VP9 video decoder in Firefox
So it's not ASSIGNED to anybody yet, meaning "when" it'll be patched in isn't known.
And now you won't have to ask about a Firefox bug on slashdot ever again, because you know a more reliable place to ask.
> but last I heard Firefox didn't support webm/vp8 only ogg
It's been available for years now. They added support for webm/vp8 around 2010*
For a better comparison see the chart a few lines down: https://en.wikipedia.org/wiki/HTML5_video#Browser_support
*https://blog.mozilla.org/blog/2010/05/19/open-web-open-video-and-webm/
Most of the early results show that, while VP9 isn't better than h265, it's within a percentage point or two. That's not its problem.
Rather, there are two big issues towards VP9 adoption.
First, there is no hardware support for it at all so far, where as the next generation of mobile and desktop chips already have h265 support announced. And since both VP9 and h265 have order-of-magnitude higher processing power requirements then their predecessors. If you're software processing it will be noticeable even on a decent desktop. So a year from now all the latest phones will already support h265. And since any site serving to them will already encode for that, why would they double up for a codec that does not perform any better?
Second, Google may be selling this as a fully free and open codec, but that's what they said about VP8. And as soon as that was announced everyone yawned and bet that it walked all over h264 licensing. And a few months ago Google finally admitted it and paid out a big settlement to license h264 for their VP8 codec.
So when Google says 'you don't have to pay a license fee' what it has meant in the past is 'we haven't done our due diligence to see what licenses you need'. And anyone who cares about paying or avoiding licensing costs would rather pay up a small known fee than worry about massive liabilities from trusting Google's word. Again.
And it's an interesting fight for the 'free as in freedom' crowd, because while h264/5 are not 'free as in beer' they are entirely open specs and many of the best h264 tools have been open source right from the start. The professional tools don't care about the minor licensing costs and the hobbyist tools don't bother paying.
VP9 is still a work in progress, so no hard numbers as yet. One of its goals is to achieve 50% better quality with the same bitrate compared to VP8. Another goal is to provide a better encoding efficiency than H.265 which has the same approach on achieving a better quality around 50% compared to H.264.
Google actually did a direct comparison between VP9 and H.264 on a sample file at its recent I/O event and showed off a 63% reduction in file size. As for the quality, see the pic for yourself.
As for the licensing issue, Google cut a deal with the MPEG-LA consortium that controls H.264 to licence their patents for VP8 and VP9. So there is low possibility of any user of VP9 of being bogged down by patent lawsuits.
Why should you care? Unlike H.265, VP9 is free for commercial use . If your use is non-profit, there is no difference between the two.
Ehh, what 'big settlement' did Google pay? Google and MPEG LA announced an agreement, there's been no disclosure of any big settlement, and I seriously doubt there was one.
MPEG LA was actively looking (as in advertising for) any patents which could be used as a patent pool against VP8, and had they actually managed to create a strong portfolio then I don't think we'd ever seen this agreement take place. Also, given how long On2 (the company Google bought for their codec technology) have been active in video compression aswell as the patents they hold, it's not as if Google is just entering the video compression arena from scratch, and they may very well hold patents on which h264 and h265 could be found infringing.
And as far as licencing costs, there's no indemnification from patent trolls with MPEG LA licencing either, and MPEG LA's saber rattling turned out to be just that, no 'massive liabilities' ended up facing anyone.
This notion you try to sell that you would somehow be 'safe' with MPEG LA licencing, while opening yourself up to 'massive liabilities' if you use VP8/VP9 is just typical scare tactics as I see it.
Now I don't think VP9 will be quite as good as h265, but that's not really important. The important thing is that MPEG LA won't be able to corner the entire online video compression market, and that there is an actual competitive alternative (and that this competition is also open source and royalty free is a huge bonus).
Because the day there isn't, the companies who make up MPEG LA will start to collect heavily on their investments, massive-greed style. Which in turn will affect us end users as the increased cost will inevitably be shifted unto us, one way or another.
Furthermore it will lead to stagnation, as in: 'we will bleed this technology dry before we introduce the next generation', all in an effort to maximize profit with less effort.
So yay for VP9, may it (and it's later incarnations) live long and prosper.
From a technical point of view it's basically h265's peer. That's partially because it's largely based on the same tech as h265, in the same way VP8 was largely similar to h264. And is speculated that it has the same licensing issues that VP8 had, for most of the same reasons.
And the speed issue is entirely due to an almost complete lack of hardware support. And while h265 already has announced and demonstrated support, I am not aware of any VP9 support so far.
And doing VP9 decode in software has order-of-magnitude higher requirements than VP8. If YouTube serves up a VP9 video to your phone, you'll wish for the good old days of Flash video.
From the q&a afterward, it is mentioned that average vp9 quality is within 1% of h.265, but it didn't sound like h.265 was anywhere near ready to roll out, with the only available option being a horrifically slow reference encoder. As for speed, they claim it is about 40% slower than vp8, which is twice as fast as h.264. As such, vp9 should handily outperform h.264 in software.
The open source and royalty free vp9/opus combination sounds like an very compelling option for the html5 video tag, and may become a de facto standard before h.265 is widely deployed. Hardware support for vp9 is also in the works, so if the codec lives up to the claims, there no longer appears to be any good reason to put up with the MPEG LA.