Google and Mozilla to create a joint next gen video codec?

[hajj_3]

here is some discussion on brandon eich's blog (mozilla ceo):

user:
What is happening with Daala and VP9? Have you asked google if they will join forces with you and just make a single codec otherwise we will have the same
issue that we had with h.264 vs webm, we need a single next gen video codec.

Brendan Eich says:
October 15, 2012 at 4:50 pm
Daala is a way out still. I’ll let Robert O'Callahan and others respond at greater length.

Robert O'Callahan says:
October 15, 2012 at 4:57 pm
The next-gen free codec issue is being worked on. I don’t want to say much about it here, but we are organizing a group at the IETF to standardize it.

It looks as though google, mozilla and xiph may join forces to create a single next gen patent-free video codec of similar quality to HEVC. This would be wonderful news:) Google had previously said that VP9 was scheduled for Q3 2013 btw, so we shouldn't expect a codec from anybody until atleast then.

[Bloax]

This would be wonderful news.

[J_Darnley]

That sounds like quite a challenge if they are going to actively avoid patents. Lets hope the standard they produce is a better quality document than VP8 and that they ask for "public" input before they make the bitstream final.

[iwod]

I am hoping we will get a "Opus" equivalent for video codec, dont know if that is possible or not.

[oibaf]

It looks VP9 may be near to be released: http://git.chromium.org/gitweb/?p=webm/libvpx.git;a=shortlog;h=refs/heads/experimental

[iwod]

It looks VP9 may be near to be released: http://git.chromium.org/gitweb/?p=webm/libvpx.git;a=shortlog;h=refs/heads/experimental

Would they have a standard this time around or just reference implementation again?

[schweinsz]

It looks VP9 may be near to be released: http://git.chromium.org/gitweb/?p=webm/libvpx.git;a=shortlog;h=refs/heads/experimental
How to download the experimental VP9 source code?

[easyfab]

git clone http://git.chromium.org/webm/libvpx.git
git checkout experimental

[schweinsz]

git clone http://git.chromium.org/webm/libvpx.git
git checkout experimental

I open the git bash, and type "git clone http://git.chromium.org/webm/libvpx.git" and enter,
the program stop at the line "cloning into libvpx", and nothing is downloaded.
Why?

[oibaf]

I am hoping we will get a "Opus" equivalent for video codec, dont know if that is possible or not.

Would they have a standard this time around or just reference implementation again?

I think VP9 will be released in a similar manner of VP8. However there is an ongoing working group (WG) for an Opus-like IETF standardized video codec. The participants have mail domain name of Cisco.com, Xiph.org, Juniper.net, Mozilla.org, Google and others. See more info about the first meeting (2012-11-05) here:
https://datatracker.ietf.org/meeting/85/materials.html#wg-videocodec
Note the Google VPnext presentation, Google is likely interested in the standardizing effort, however this is probably a long term solution, VP9 should be completed sooner (just guessing from what I read on the net). :)

To keep updated see also the video-codec WG mailing list:
http://www.ietf.org/mail-archive/web/video-codec/current/maillist.html

[oibaf]

I open the git bash, and type "git clone http://git.chromium.org/webm/libvpx.git" and enter,
the program stop at the line "cloning into libvpx", and nothing is downloaded.
Why?

It works fine here:
$ git clone http://git.chromium.org/webm/libvpx.git
Cloning into 'libvpx'...
remote: Counting objects: 27071, done.
remote: Compressing objects: 100% (6849/6849), done.
remote: Total 27071 (delta 21752), reused 25040 (delta 19929)
Receiving objects: 100% (27071/27071), 12.72 MiB | 642 KiB/s, done.
Resolving deltas: 100% (21752/21752), done.

Make sure your Internet connection works and use git 1.5.4 or newer as specified here: http://www.webmproject.org/code/build-prerequisites/

[schweinsz]

It works fine here:
$ git clone http://git.chromium.org/webm/libvpx.git
Cloning into 'libvpx'...
remote: Counting objects: 27071, done.
remote: Compressing objects: 100% (6849/6849), done.
remote: Total 27071 (delta 21752), reused 25040 (delta 19929)
Receiving objects: 100% (27071/27071), 12.72 MiB | 642 KiB/s, done.
Resolving deltas: 100% (21752/21752), done.

Make sure your Internet connection works and use git 1.5.4 or newer as specified here: http://www.webmproject.org/code/build-prerequisites/

Yesterday I type the same command line and it works. Perhaps there is some problem with my network. Thank you.

[dapperdan]

Google is likely interested in the standardizing effort, however this is probably a long term solution, VP9 should be completed sooner (just guessing from what I read on the net). :)

Seems like it, someone from Google's already talking about VP9 + Opus in WebM:

https://groups.google.com/a/webmproject.org/d/topic/webm-discuss/hQAGZeKQK-E/discussion

[iwod]

I think VP9 will be released in a similar manner of VP8. However there is an ongoing working group (WG) for an Opus-like IETF standardized video codec. The participants have mail domain name of Cisco.com, Xiph.org, Juniper.net, Mozilla.org, Google and others. See more info about the first meeting (2012-11-05) here:
https://datatracker.ietf.org/meeting/85/materials.html#wg-videocodec
Note the Google VPnext presentation, Google is likely interested in the standardizing effort, however this is probably a long term solution, VP9 should be completed sooner (just guessing from what I read on the net). :)

To keep updated see also the video-codec WG mailing list:
http://www.ietf.org/mail-archive/web/video-codec/current/maillist.html

By Opus like, i was referring to its technical superiority rather then patents free, IETF or whats not. Opus perform better then all other popular codec as a real time voice codec and also as a lossy music codec. And there is still a lot of improvement and tuning to be made for such a young codec!

I am not sure if such technical superiority is possible with x264 as a competitor.