Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion.

Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules.

 

Go Back   Doom9's Forum > Video Encoding > VP9 and AV1

Reply
 
Thread Tools Search this Thread Display Modes
Old 30th September 2015, 11:47   #21  |  Link
nevcairiel
Registered Developer
 
Join Date: Mar 2010
Location: Hamburg/Germany
Posts: 10,336
Dolby Vision is annoying, they should stick to the HDR metadata that UHD Blu-ray favors (SMPTE ST- 2084/2086). It needs far less decoder changes.
__________________
LAV Filters - open source ffmpeg based media splitter and decoders
nevcairiel is offline   Reply With Quote
Old 30th September 2015, 14:50   #22  |  Link
kolak
Registered User
 
Join Date: Nov 2004
Location: Poland
Posts: 2,843
Dolby Vision won't catch up, as it cost to much money in licensing.
kolak is offline   Reply With Quote
Old 4th February 2016, 16:41   #23  |  Link
dapperdan
Registered User
 
Join Date: Aug 2009
Posts: 201
Apparently work has begun:

https://chromium.googlesource.com/webm/aom/
https://chromium-review.googlesource...oject:webm/aom

Mostly based on VP9/10 to start with.
dapperdan is offline   Reply With Quote
Old 6th February 2016, 00:24   #24  |  Link
benwaggoner
Moderator
 
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,738
Quote:
Originally Posted by nevcairiel View Post
Dolby Vision is annoying, they should stick to the HDR metadata that UHD Blu-ray favors (SMPTE ST- 2084/2086). It needs far less decoder changes.
The 2086 metadata is static across an entire title, so it still requires a lot of prediction and MIPS for a display's tone mapper to adapt to per-scene changes in color volume or contrast.

And it's really not that valuable. A single flash frame from a camera flash can double the MaxFALL the whole rest of the title would use, and thus be very misleading for a tone mapper.
__________________
Ben Waggoner
Principal Video Specialist, Amazon Prime Video

My Compression Book
benwaggoner is offline   Reply With Quote
Old 5th April 2016, 22:01   #25  |  Link
Nintendo Maniac 64
Registered User
 
Nintendo Maniac 64's Avatar
 
Join Date: Nov 2009
Location: Northeast Ohio
Posts: 447
AMD, ARM, and Nvidia have joined, and because it apparently wasn't established before, this new upcoming in-development video codec will be an open-source project:
http://aomedia.org/press-release/the...and-announces/

Hopefully AMD will finally implement VP9 hardware decoding on their upcoming APUs and GPUs...

Last edited by Nintendo Maniac 64; 5th April 2016 at 22:06.
Nintendo Maniac 64 is offline   Reply With Quote
Old 5th April 2016, 22:18   #26  |  Link
mandarinka
Registered User
 
mandarinka's Avatar
 
Join Date: Jan 2007
Posts: 729
All reference encoders/decoders are open source (H.264, H.265, VP8/9), so that was obvious. The distinction was never in open versus closed source anyway.

After all the development model is more or less the same, but probably with faster development and less testing/reviewing than the MPEG guys do, if libvpx traditions are any sign. So possibly there will again be few bugs in the frozen specification, like VP9/VP8 had

The only difference between these "libre" (or whatever term is currently in) codecs and H.264/H.265/... is really in the patent licensing requirements and usage royalties.
mandarinka is offline   Reply With Quote
Old 6th April 2016, 12:55   #27  |  Link
dapperdan
Registered User
 
Join Date: Aug 2009
Posts: 201
I don't think they're announcing that it's open source, but that it is available and mentioning in passing that it's open source. Though I think it's actually been available already, so this is just an official announcement that it's available.

In other oddities, the new members are being given "founder member" status, even though they're joining a while after the others.

Finally, the link seems to point just to the patent licence info (possibly the page will be updated to point to the actual code) but it appears to live here:

https://aomedia.googlesource.com/aom/

A notably non-impartial place to host the code I would have thought, though it's all git in the end.
dapperdan is offline   Reply With Quote
Old 7th April 2016, 01:08   #28  |  Link
Nintendo Maniac 64
Registered User
 
Nintendo Maniac 64's Avatar
 
Join Date: Nov 2009
Location: Northeast Ohio
Posts: 447
Quote:
Originally Posted by dapperdan View Post
In other oddities, the new members are being given "founder member" status, even though they're joining a while after the others.
Well they're all chip hardware designers...Intel was the only chip hardware designer that was a member, so there could have very well been several more "vacancies" for other chip hardware designers as well.

In other words, maybe because Intel was previously the only chip hardware designer, the group as a whole left the door open for other chip hardware designers?
Nintendo Maniac 64 is offline   Reply With Quote
Old 10th April 2016, 03:47   #29  |  Link
BadFrame
Registered User
 
Join Date: Jun 2013
Posts: 98
Quote:
Originally Posted by Nintendo Maniac 64 View Post
AMD, ARM, and Nvidia have joined,
Interesting, seems like they got the hardware support largely covered with these new members.
BadFrame is offline   Reply With Quote
Old 13th April 2016, 13:44   #30  |  Link
mandarinka
Registered User
 
mandarinka's Avatar
 
Join Date: Jan 2007
Posts: 729
http://www.streamingmedia.com/Articl...ticleID=110383

Quote:
When will the codec ship? The group is aiming to freeze the bitstream sometime between the end of the 2016 and March 2017. Expect to see browser-based support soon thereafter, with the first hardware support within 12 months after that.
On the one hand that is soon, on the other hand this is not much time for research/development. Hopefully it won't have bugs in spec this time, but this probably means they will have to stick with most of libvpx since that is what AV1 its currently built upon, and other IP will only be patches on the top.
mandarinka is offline   Reply With Quote
Old 13th April 2016, 18:34   #31  |  Link
Clare
Registered User
 
Join Date: Apr 2016
Posts: 61
Quote:
Originally Posted by mandarinka View Post
http://www.streamingmedia.com/Articl...ticleID=110383



On the one hand that is soon, on the other hand this is not much time for research/development. Hopefully it won't have bugs in spec this time, but this probably means they will have to stick with most of libvpx since that is what AV1 its currently built upon, and other IP will only be patches on the top.
Both Daala and AV1 are already well advanced if one looks at AWCY:
https://arewecompressedyet.com/?r%5B...&s=ntt-short-1

Also, the VP10 basis it is built upon does not seem to take all development happening in the nextgenv2 branch. So they could potentially add more feat. already developed from VP10 later.

I'm more concerned at the processing power needed for encoding.
Clare is offline   Reply With Quote
Old 13th April 2016, 21:20   #32  |  Link
mandarinka
Registered User
 
mandarinka's Avatar
 
Join Date: Jan 2007
Posts: 729
Speed is inconsequential IMHO, even if it was 2-4x+ as slow as optimised HEVC or VP9 encoder, that would be soemthing that would be solvable - if by nothing else, by throwing more CPU cores there.

My fear is that the problems are going to be the same as with VP8/VP9 and (to a lesser degree, it has matured somewhat) x265: the encoder will be too fresh, too crude, without good rate control and lacking in utilisation of the new compression tools. The advances of format will be probably negated by poor quality encoder. This problem is hard and practice shows it takes years to improve. I think CPU requirements are nothing compared to that.
mandarinka is offline   Reply With Quote
Old 25th April 2016, 19:53   #33  |  Link
wiak
Registered User
 
Join Date: Jul 2003
Location: somewhere north
Posts: 260
did some testing
https://nwgat.ninja/test-driving-aomedias-av1-codec/
http://screenshotcomparison.com/comparison/170492
__________________
Woah! Ninja?! http://nwgat.ninja/ (AV1 Overview)
"Not available in your region" has now been redefined as "Go Pirate, you filthy scum" Nwgat

Last edited by wiak; 25th April 2016 at 19:57.
wiak is offline   Reply With Quote
Old 25th April 2016, 21:21   #34  |  Link
mandarinka
Registered User
 
mandarinka's Avatar
 
Join Date: Jan 2007
Posts: 729
That nicely noisy output, is that with PVQ enabled? (IIRC Daala guys worked on porting it to AV1.)
mandarinka is offline   Reply With Quote
Old 26th April 2016, 16:20   #35  |  Link
wiak
Registered User
 
Join Date: Jul 2003
Location: somewhere north
Posts: 260
Quote:
Originally Posted by mandarinka View Post
That nicely noisy output, is that with PVQ enabled? (IIRC Daala guys worked on porting it to AV1.)
i dont think so, i did a mirror of the aom googlesource git repo to github for simpler reading
https://github.com/nwgat/aom/search?...r+Quantization
https://github.com/nwgat/aom/search?...=PVQ&type=Code
__________________
Woah! Ninja?! http://nwgat.ninja/ (AV1 Overview)
"Not available in your region" has now been redefined as "Go Pirate, you filthy scum" Nwgat
wiak is offline   Reply With Quote
Old 29th April 2016, 08:13   #36  |  Link
Jamaika
Registered User
 
Join Date: Jul 2015
Posts: 696
Could someone compile the latest codec AOM?

PS Is the job suspended on the development of the codec? I don't see updates.
Jamaika is offline   Reply With Quote
Old 29th April 2016, 13:40   #37  |  Link
dapperdan
Registered User
 
Join Date: Aug 2009
Posts: 201
The main repo is currently here:

https://aomedia.googlesource.com/aom/

which has more recent commits, though nothing in the last couple of weeks
dapperdan is offline   Reply With Quote
Old 2nd May 2016, 21:10   #38  |  Link
mzso
Registered User
 
Join Date: Oct 2009
Posts: 930
Isn't there some information about the internals of this codec? Does it have something innovative like lapped transfrom (was supposed to be) for Daala.
mzso is offline   Reply With Quote
Old 3rd May 2016, 11:04   #39  |  Link
BadFrame
Registered User
 
Join Date: Jun 2013
Posts: 98
Quote:
Originally Posted by mzso View Post
Isn't there some information about the internals of this codec? Does it have something innovative like lapped transfrom (was supposed to be) for Daala.
I haven't seen any documentation, likely because the development is still in a 'let's try everything' phase, with the bitstream being locked down by the end of the year (at the earliest).

Also there are a lot of different branches where work is being done, here is a branch with Daala techniques being tested:

https://chromium-review.googlesource...oject:webm/aom

Here is what I believe is the official AOM branch:

https://aomedia.googlesource.com/aom/

And here is the 'nextgenv2' branch, where there's a lot of work done at a rapid pace:

https://chromium.googlesource.com/we...+log/nextgenv2
BadFrame is offline   Reply With Quote
Old 3rd May 2016, 13:22   #40  |  Link
mzso
Registered User
 
Join Date: Oct 2009
Posts: 930
I was thinking more in the vein of the Daala demos. About roughly what novel/innovative ideas they have in the air. (If they have any)
mzso is offline   Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 11:30.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.