Google to develop new open-source video compression codec - VP10

[ahhming]

Google is developing a more efficient video compression codec called VP10 that will use 1/4 of bandwidth resources than H.264 at the same quality.

 

Google product manager James Bankoski said that VP9 uses half the bandwidth needed to deliver the same quality video as the popular H.264 format -- and with VP10, the company is trying to cut it in half again.

Google claims that VP10 should also provide shaper images, better color, and a better dynamic range between lights and dark. Another strong point for VP10 is its open source.

 

Still, this is a battle that isn't yet won, and the dominance of YouTube means that VP10 will have a fighting chance against HEVC. The fact that VP10 is open source is another point in its favor; the licensing for HEVC could add red tape and costs for developers and devices makers. It'll be a while before this conflict comes to a head, however: Google said it hopes to hit its performance targets for VP10 by the end of next year.

 

 

Source:

http://www.engadget.com/2015/08/31/google-vp10-4k-video-data-usage/

http://www.cnet.com/news/inside-googles-master-plan-for-faster-sharper-streaming-video/

[Bob Jim]

Huh. Still don't understand codecs.

[RuLeZ]

Damn, VP9 is already heavy for the CPUs, VP10 will be crazy

[ahhming]

Huh. Still don't understand codecs.

[Bob Jim]

....I have searched for that video... At least 5 times.... lol thanks

[Sauron]

dis gon b gud

[LAwLz]

Google is developing a more efficient video compression codec called VP10 that will use twice as less bandwidth resources than H.264 at the same quality.

It is actually 1/4 of the bandwidth needed for H.264.

VP9 files at the same quality takes up around 60% as much space as H.264,and VP10 will try to make the same kind of leap again.

I was rooting for Daala for the next "codec war" but I would be equally pleased if VP10 wins (if it is technically superior of course). Kind of tired that proprietary codecs like H.264 keeps winning.

[Guest]

Could be interesting

[Eroda]

h264 dominates because all the anime pirates woke up one day and said lets all start using x264 which int urn started a chain reaction now most pirated stuff i have seen is h/x264

[qwertywarrior]

its awesome that there is an opensource set already available

OPUS for Audio

VP(x) for video

[qwertywarrior]

h264 dominates because all the anime pirates woke up one day and said lets all start using x264 which int urn started a chain reaction now most pirated stuff i have seen is h/x264

well to be fair H264 is old and at the time it was better than other video codecs

also x264 helped

[silberdrachi]

h264 dominates because all the anime pirates woke up one day and said lets all start using x264 which int urn started a chain reaction now most pirated stuff i have seen is h/x264

 

Wasnt part of it that VPX wasnt supported nearly as well for hardware decoding? So it was a lot easier on the computer to use 264 than it was any of the VP versions?

[LAwLz]

h264 dominates because all the anime pirates woke up one day and said lets all start using x264 which int urn started a chain reaction now most pirated stuff i have seen is h/x264

I really doubt that was the reason. If everyone just followed the anime community (the best when it comes to codec and technology support) then we would see 10bit H.264 everywhere.

[DEDRICK]

H265(x265) takes a MASSIVE amount of CPU power, I can only imagine what VP10 will take.  

 

It looks like they are focusing on RGB(or i444) Full Range encoding too (better color, and a better dynamic range between lights and dark).

 

Limited Range = 16-235 (TV/movie Standard)

Full Range = 0-255 (Full RGB)

 

RGB takes a lot more space but with the compression improvements it will actually be worth doing

[Guest]

Feed the WebM power!

VP9 was great!

[Misanthrope]

I'm ready for it, however I worry this is too far ahead in the future: most video playback appliances like tablets and smart tvs just won't have the CPU power.

[Eroda]

H265(x265) takes a MASSIVE amount of CPU power, I can only imagine what VP10 will take.  

 

It looks like they are focusing on RGB(or i444) Full Range encoding too (better color, and a better dynamic range between lights and dark).

 

Limited Range = 16-235 (TV/movie Standard)

Full Range = 0-255 (Full RGB)

 

RGB takes a lot more space but with the compression improvements it will actually be worth doing

 

wont be long before all GPU's have inbult encode/decode hopefully then it should be a non issues

[LAwLz]

wont be long before all GPU's have inbult encode/decode hopefully then it should be a non issues

Decoders? Sure no problem. HEVC is already seeing decently widespread support for hardware decoding so hopefully the next standard will only take a year or two to get decent support as well.

 

Encoders? GPU encoders are much worse in terms of quality compared to CPU encoders. For game streaming and stuff it's okay, but for archival purposes you want CPU encoding, and that's slow as balls.

[LucidMew]

Uh, hasn't this already been known for a while? This CNet article says they were developing it back in Middle of September last year, and said 
 

He wouldn't comment on VP10 arrival times, but said Google is working hard to speed up development. The gap between VP8 and VP9 was 24 months, but he hopes Google can get iteration cycles down to 18 months.

 

-snip-

 

If Google can make its 18-month goal, that would mean VP10 would arrive about the end of this year, with another 18 months or so for that version of the codec to arrive in chips. Google expects the first VP9-equipped mobile devices to appear in 2015.

[Eroda]

Decoders? Sure no problem. HEVC is already seeing decently widespread support for hardware decoding so hopefully the next standard will only take a year or two to get decent support as well.

 

Encoders? GPU encoders are much worse in terms of quality compared to CPU encoders. For game streaming and stuff it's okay, but for archival purposes you want CPU encoding, and that's slow as balls.

 

 

my main point would be if nvidias next GPU had hardware encode/decode and they added 265 support in shadowplay it would lead to faster adoption i know alot of people who use shadowplay to record stuff now over 3rd party software  granted they havent got the best computers so the performance hit without it is bad. i use it as an example simply because i couldnt play new games and record @ the same time without it