Friday, May 21, 2010

Some things you might want to know about WebM and VP8

Right now, unless you're on the cutting edge of web development or in software/hardware video development, you don't really need to know anything about WebM and VP8. Better to let those on the bleeding edge play with it, and worry about browser adoption rates and things like that, and then in a year or so - if the water seems right - you can jump on in.

But incase you're curious, WebM is a project from Google to create an open video format for the web [subtext: we don't want to pay the H.264 patent holders]

The WebM Project consists of:
  • VP8, a high-quality video codec released under a BSD-style, royalty-free license
  • Vorbis, an already open source and broadly implemented audio codec
  • a container format based on a subset of the Matroska media container
Note that VP8 is the compressor, but WebM is the architecture that the codec plays within (so expect people to use the two names interchangeably.)

Several companies have already announced support for it, including Sorenson, who even has a VP8 Video Encoding with Squeeze 6.5 Beta: Video Tutorial.

But not everyone is so enthusiastic; when asked what he thought about it, Apple's Steve Jobs merely sent a link to an article by Jason Garrett-Glaser, a developer working on the x264 open source project (an H.264 encoder). Garrett-Glaser may be biased, but his article seems well reasoned (which means it's more technical than I can really understand.) He notes:
Overall, VP8 appears to be significantly weaker than H.264 compression-wise. The primary weaknesses mentioned above are the lack of proper adaptive quantization, lack of B-frames, lack of an 8×8 transform, and non-adaptive loop filter. With this in mind, I expect VP8 to be more comparable to VC-1 or H.264 Baseline Profile than with H.264. Of course, this is still significantly better than Theora, and in my tests it beats Dirac quite handily as well.
Which might be okay if, you know, we don't have to pay for it. But even that hope might be too much to wish for, as Garrett-Glaser notes:
Finally, the problem of patents appears to be rearing its ugly head again. VP8 is simply way too similar to H.264: a pithy, if slightly inaccurate, description of VP8 would be “H.264 Baseline Profile with a better entropy coder”. Though I am not a lawyer, I simply cannot believe that they will be able to get away with this, especially in today’s overly litigious day and age.
And MPEG-LA, the consortium that controls H.264 video standard seems to agree. They are already looking at "creating a patent pool license for VP8."

In the mean time, what should we do? Well, nothing. Sure Google owns YouTube, so don't be surprised if they move YouTube to V8, but you'll still be able to upload in H.264 or any other format. Let's not forget; as the owner of YouTube, Google may be the entity that's on the biggest hook for H.264 royalties, so their reasons for pushing another video format may be less than altruistic.

Oddly enough, I was reading a forum yesterday and someone wrote that "H.264 will become just an acquisition format" which - if true - is rather funny. H.264 was developed as a distribution format, not an acquisition one, and was only grudgingly adopted as a legitimate acquisition format as it was adopted by more and more low-end - and now prosumer - video cameras. Most serious users would tell you that H.264 isn't the ideal compressor for capture. It'll be ironic if that's all it becomes.


No comments: