VP8 (WebM) submitted to the IETF

Earlier this month, it seems that VP8 was submitted as an informational document to the IETF:

VP8 Data Format and Decoding Guide
By J. Bankoski, P. Wilkins, Y. Xu of Google, Inc.

Naturally, Google's irrevocable, royalty-free license still applies.

If you are not familiar with the difference between VP8 and WebM, VP8 is the actual video compression format, while WebM is a multimedia container format based on Matroska, which consists of VP8 for video, Vorbis for audio.

It's important to note that this doesn't necessarily make VP8 a standard (at least yet), since informational IETF documents can basically be "anything".

Advertisements

7 thoughts on “VP8 (WebM) submitted to the IETF

  1. This document is less dependent on specific implementations of code than their bitstream PDF. Looks like the working group is on target for a guide release this summer.Given the current bugs in x86 VP8 Chroma, and certain other incorrectable implementation flaws, I wonder how much complexity is required to emulate these flaws in a dedicated decoder logic?Still, with even this much it should be possible for most browsers to view WebM content before the end of the year. Market penetration of embedded VP8 decoders will show us just how affordable the new chips really are.

  2. Originally posted by xErath:

    If this is a proper rfc, them I'm Nelson Mandela. Source code in the document ?

    One of the things that makes it difficult to write a 3rd party encoder, or to even be sure the code you're submitting for an update to the reference encoder is bitstream-correct 😉 You have to guess at what the output is supposed to be based on source code of the reference encoder. Why they have still yet to produce a proper bitstream definition I do not know.

Comments are closed.