Google & MPEG LA Reach VP8 Agreement

[XorEaxEax]

VP8 isn't as good as H.264 I'm afraid. We need VP9! But I guess they're waiting for WebM to be adopted everywhere.

It's certainly good enough for online video quality-wise. Just download youtube videos in webm (vp8) and mp4 (h264) and compare them visually, I can't decide which is which by just looking, and in terms of file sizes they are about the same, sometimes webm is a bit smaller and sometimes the mp4 file is a bit smaller. If you do test with ssim or psnr, or examine the video on a frame by frame basis you will most likely find that h264 retains better quality per bit than vp8 but it's not by some wide margin.

Like others said this should remove the last obstacle from making vp8 (and hopefully later on vp9) the official standard html5 video codec.

[XorEaxEax]

Still too little, too late I'm afraid. h265 already has a large consortium of licensees. MPEG-LA achieved it's goal of putting enough uncertainty in WebM until h265 was ready to go again making Googles efforts futile.

h265 is nowhere 'ready to go', you obviously have no idea of where it's at implementation-wise. The standard specification has just been finalized, the current encoder implementations are by all accounts unfinished and very unoptimized, not to mention that the amount of tuning that has gotten h264 encoders like x264 to reach the quality we are enjoying today is years away. Hardware decoding implementations in consumer devices won't show up until next year at the earliest.

VP8 is considered as a standard royalty free video codec for HTML5 online video, and it's the only real option at the moment as h264 is not royalty free. It's a great thing if VP8 becomes a HTML5 video standard, same goes for WebRTC. As for Google's part though, the most likely reason they bought On2 was so that they would have a codec they could specifically tune for their own online video services, particularly real-time video which is where they seem to have put alot of effort.

[deanjo]

h265 is nowhere 'ready to go', you obviously have no idea of where it's at implementation-wise. The standard specification has just been finalized, the current encoder implementations are by all accounts unfinished and very unoptimized, not to mention that the amount of tuning that has gotten h264 encoders like x264 to reach the quality we are enjoying today is years away. Hardware decoding implementations in consumer devices won't show up until next year at the earliest.

VP8 is considered as a standard royalty free video codec for HTML5 online video, and it's the only real option at the moment as h264 is not royalty free. It's a great thing if VP8 becomes a HTML5 video standard, same goes for WebRTC. As for Google's part though, the most likely reason they bought On2 was so that they would have a codec they could specifically tune for their own online video services, particularly real-time video which is where they seem to have put alot of effort.

Broadcom already is ramping up production of their BCM7445 as well as ViXS and others. The final spec has been around for a while now it was just the final ratification that just happened recently. Samsungs 2013 TV's also have h265 support (F8500). Sorry man but two years ago people were claiming how VP8 was gonna kill h264, that hasn't even come close to happening. Same thing is going to happen with VP9.

[XorEaxEax]

I never claimed vp8 was going to kill h264, nor have I seen anyone else say so. Why would vp8 have to kill h264? Or vp9 have to kill h265? Just as h264 is a technically better codec than vp8, h265 will likely end up being technically better than vp9, but I'm certain that just as in vp8 vs h264 the difference won't be huge by any stretch.

What matters is that vp8 (and likely vp9) is free for anyone to use, and free to be part of a royalty free standard for web video and be an official part of HTML5. Any company, individual, whatever can encode and distribute videos using vp8, implement it in any type of services, and if it becomes the HTML5 standard video format then it opens up a ton of opportunity, especially now as we are seeing massive hardware support for webm (vp8, vp9) from arm, broadcom, amd, nvidia etc.

Given that 'royalty free' is one of the key aspects of the 'standard' codec the HTML5 working group is deciding on I'd say vp8 is now, particularly in light of this this patent agreement to dispel uncertainties, the only logical candidate. With this and Opus as the official standard HTML5 audio codec we will have a royalty free top class video/audio combination which can be implemented everywhere with no licencing costs and be expected to work anywhere HTML5 is supported.

[Delgarde]

It's too late - but then, it was already too late before VP8 was released. The catch is the recording devices - if I film something using my camera, it's in H264, not VP8. And I'm *not* going to spend time transcoding it, just for the sake of using a codec with a better license.

[DeepDayze]

It's too late - but then, it was already too late before VP8 was released. The catch is the recording devices - if I film something using my camera, it's in H264, not VP8. And I'm *not* going to spend time transcoding it, just for the sake of using a codec with a better license.

True that is now, but now that MPEG-LA has backed off attacking VP8/9 at the moment there could well be incentive for video device makers to soon adopt vp8/9 as the standard as they'll avoid having to pay royalties per device produced to MPEG-LA for using h.264 encoding. This could be done by firmware updates in most devices.

[XorEaxEax]

It's too late - but then, it was already too late before VP8 was released. The catch is the recording devices - if I film something using my camera, it's in H264, not VP8. And I'm *not* going to spend time transcoding it, just for the sake of using a codec with a better license.

Too late for what? To become the HTML5 standard video codec? The way you repeat this 'too late' like a mantra makes me think you are just reflecting you employer's (Apple) dislike for anything from Google.

If vp8 is supported by every browser as part of HTML5, not to mention WebRTC, then it will be used by tons of online services as not only is it a standard, but also it is free of cost through royalties.

[Marc Driftmeyer]

Google settled rather than eat it hard in the courts from Apple, Microsoft, Panasonic, Sony, etc.

H.265 will be the defacto moving foward, as well.

[Delgarde]

Too late for what? To become the HTML5 standard video codec? The way you repeat this 'too late' like a mantra makes me think you are just reflecting you employer's (Apple) dislike for anything from Google.

If vp8 is supported by every browser as part of HTML5, not to mention WebRTC, then it will be used by tons of online services as not only is it a standard, but also it is free of cost through royalties.

Work for Apple? Hardly... I'm just being realistic about this. I have a device that records video, and like most such devices, it does so in H264. Ergo, any video content I post online is going to be uploaded in H264.

Nice as it would be to see a more open standard catch on, it's not going to happen.

[XorEaxEax]

Work for Apple? Hardly...

LOL sorry, I mixed you up with Deanjo whom I was discussing with right before, in my defence your screen names aren't entirely dissimilar.

As for your video statement, it doesn't matter which format you upload in, practically every online video site transcodes the video before they enable it for streaming, not only that but they also accept video in alot more formats than h264.

If vp8 becomes the HTML5 standard video codec that means that any site serving video can use royalty free vp8 to stream video to any HTLM5 supported browser, again no matter what the original format was in.