4 ms·
Comparison of compression efficiency between HEVC/H.265 and VP9 [pdf]
- rockdoe 12y ago"While trying to duplicate Touradj’s results, we discovered that the early version of libvpx ( November 2013) he used, had a bug that produced invalid vp9 bitstreams and encoder / decoder mismatches with the settings he used. " -- James Bankoski, Director at Google, former CTO of On2
- the8472 12y ago>Encoder JM 18.6 HM 15.0 >3. SUBJECTIVE EVALUATION Those are the reference encoders for the respective codecs which - to my knowledge - do not perform any psycho-visual optimizations or if they do not with the level of sophistication of state-of-the-art implementations such as x264. They're not meant for practical use While google's vpx library on the other hand is intended for practical use. Even for PSNR measurements (which are of limited relevance) this would be questionable. For subjective evaluation it's outright useless. These results should be taken with a ladle full of salt in my opinion.
- ZeroGravitas 12y agoI note it uses version 1.2.0 of the libvpx library. VP9 was first released in version 1.3 and even then the release notes state: "2013-11-15 v1.3.0 "Forest" This release introduces the VP9 codec [...] The VP9 encoder in this release is not feature complete. Users interested in the encoder are advised to use the git master branch and discuss issues on libvpx mailing lists." Seems odd that people were so keen to evaluate VP9 before it was actually released, and now that it's actually out and served in the tens of billions on Youtube no one is interested in publishing their findings on the actual production code. Has anyone got links to more recent results?