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 27th August 2016, 11:09   #81  |  Link
mzso
Registered User
 
Join Date: Oct 2009
Posts: 930
Quote:
Originally Posted by ls1dreams View Post
When does everyone think we'll get hardware decoding from intel chips?

I'm really frustrated with my laptops getting super hot just playing the most simple videos, google hangouts, etc.

For example, VP9 came out in 2013, and even today, 3 years later, we still don't have a chip that does full hardware decoding (only partial). Kaby Lake will resolve this, but it will have been 4 years. The same goes for HEVC.

Having a slightly smaller stream hardly matters to me if it means my cpu is spinning up to 50% utilization or higher.

AV1 seems to have everyone on board, but when will we get an igpu (macbooks/etc) that supports it? Kaby Lake is obviously out. Cannon Lake too soon?

I'm overdue for upgrading my laptop, but don't really want to wait 2 more years for this! I just want to play 4k videos on a macbook retina without the machine overheating. (Hell, even 720p or 1080p).
The solution is simple. Don't buy an intel based computer, if you want stuff supported.
mzso is offline   Reply With Quote
Old 27th August 2016, 12:15   #82  |  Link
mandarinka
Registered User
 
mandarinka's Avatar
 
Join Date: Jan 2007
Posts: 729
Make sure you are using FFVP9 decoder. Chrome uses the libvpx instead (bad, inefficient). I think Firefox adopted FFVP9, but I have no idea if it is used on the Apple OS X.

No idea if the Intel drivers that implement the partial acceleration even work/are available on the platform (or Linux). If they don't, you could try under Windows and see if the partial acceleration works under 4K - not sure there.
mandarinka is offline   Reply With Quote
Old 27th August 2016, 12:37   #83  |  Link
IgorC
Registered User
 
Join Date: Apr 2004
Posts: 1,315
I have tried https://media.xiph.org/video/derf/we...10bit_420.webm chrome was the fastest.


Quote:
Originally Posted by mzso View Post
The solution is simple. Don't buy an intel based computer, if you want stuff supported.
Intel was first to support VP9 decoding (at least hybrid). Anyway I don't see anyone doing better. "not buying Intel" doesn't solve anything.
IgorC is offline   Reply With Quote
Old 27th August 2016, 16:10   #84  |  Link
hajj_3
Registered User
 
Join Date: Mar 2004
Posts: 1,120
Quote:
Originally Posted by ls1dreams View Post
When does everyone think we'll get hardware decoding from intel chips?
I think the AV1 codec is supposed to have hardware decoding available starting at the end of 2017 iirc. That will likely be from ARM and Qualcomm i suspect. Hardware decoding is way more important for those companies than it is for laptops and desktops. I suspect that by mid-2018 all intel and amd apus will have it and all amd and nvidia gpus.
hajj_3 is offline   Reply With Quote
Old 27th August 2016, 17:02   #85  |  Link
wiak
Registered User
 
Join Date: Jul 2003
Location: somewhere north
Posts: 260
Quote:
Originally Posted by hajj_3 View Post
I think the AV1 codec is supposed to have hardware decoding available starting at the end of 2017 iirc. That will likely be from ARM and Qualcomm i suspect. Hardware decoding is way more important for those companies than it is for laptops and desktops. I suspect that by mid-2018 all intel and amd apus will have it and all amd and nvidia gpus.
i suspect amd navi and nvidia voltra will support it and chips based on them (it takes 2-3 years from planning to release of a new gpu)
if you want to play modern codecs, you should get a proper pc not some underpowered laptop
laptop are made to be portable and low power
__________________
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; 27th August 2016 at 19:13.
wiak is offline   Reply With Quote
Old 29th August 2016, 03:56   #86  |  Link
Nintendo Maniac 64
Registered User
 
Nintendo Maniac 64's Avatar
 
Join Date: Nov 2009
Location: Northeast Ohio
Posts: 447
Since AMD's Zen cores are intended to scale all the way down to power consumption levels that are similar to their Jaguar cores, it's theoretically possible that AMD will be more proactive in implementing AV1 hardware decoding in their Zen APUs since it'll be much more important for such low-power form-factors.

Attached Images
 

Last edited by Nintendo Maniac 64; 29th August 2016 at 04:30.
Nintendo Maniac 64 is offline   Reply With Quote
Old 29th August 2016, 15:18   #87  |  Link
mzso
Registered User
 
Join Date: Oct 2009
Posts: 930
Quote:
Originally Posted by Nintendo Maniac 64 View Post
Since AMD's Zen cores are intended to scale all the way down to power consumption levels that are similar to their Jaguar cores, it's theoretically possible that AMD will be more proactive in implementing AV1 hardware decoding in their Zen APUs since it'll be much more important for such low-power form-factors.

Yeah... That attachment will probably never be approved.
mzso is offline   Reply With Quote
Old 30th August 2016, 10:53   #88  |  Link
hajj_3
Registered User
 
Join Date: Mar 2004
Posts: 1,120
Quote:
Originally Posted by Nintendo Maniac 64 View Post
Since AMD's Zen cores are intended to scale all the way down to power consumption levels that are similar to their Jaguar cores, it's theoretically possible that AMD will be more proactive in implementing AV1 hardware decoding in their Zen APUs since it'll be much more important for such low-power form-factors.

zen apus are supposed to be out in mid 2017, can't see av1 hardware decoding being available in zen apus, possibly hybrid decoding if you are lucky. Zen+ will likely have it though.
hajj_3 is offline   Reply With Quote
Old 6th September 2016, 07:20   #89  |  Link
Nintendo Maniac 64
Registered User
 
Nintendo Maniac 64's Avatar
 
Join Date: Nov 2009
Location: Northeast Ohio
Posts: 447
Some presentations from VideoLAN Dev Days 2016 that may or may not be relevant:

VideoLAN Dev Days 2016: Daala

VideoLAN Dev Days 2016: Update on VPX

Quote:
Originally Posted by mzso View Post
Yeah... That attachment will probably never be approved.
Then stand back and watch as I cheat the system!


Last edited by Nintendo Maniac 64; 6th September 2016 at 07:24.
Nintendo Maniac 64 is offline   Reply With Quote
Old 6th September 2016, 12:27   #90  |  Link
dapperdan
Registered User
 
Join Date: Aug 2009
Posts: 201
Another related video talk, on the Eve VP9 encoder:

https://www.youtube.com/watch?v=t_z52-CBut0
dapperdan is offline   Reply With Quote
Old 7th September 2016, 17:29   #91  |  Link
littlepox
Registered User
 
Join Date: Nov 2012
Posts: 218
I'd guess AMD spend half of their R&D budget on PPT design.
Really, I'd be interested to buy some templates from them for my own use.

Last edited by littlepox; 9th September 2016 at 05:10.
littlepox is offline   Reply With Quote
Old 11th September 2016, 06:38   #92  |  Link
Jamaika
Registered User
 
Join Date: Jul 2015
Posts: 697
New web codec AOM. Codec VP10 turned into to AV1.
https://chromium.googlesource.com/we...px/+/nextgenv2
https://github.com/mbebenita/aomanalyzer
Websites obsolete:
https://github.com/jmvalin/aom
https://github.com/smarter/aom

Last edited by Jamaika; 11th September 2016 at 09:45.
Jamaika is offline   Reply With Quote
Old 30th September 2016, 15:52   #93  |  Link
Clare
Registered User
 
Join Date: Apr 2016
Posts: 61
I do hope they will rework the encoder completely instead of relying on the current libvpx one because that encoder is way too slow.
Clare is offline   Reply With Quote
Old 20th October 2016, 19:58   #94  |  Link
sneaker_ger
Registered User
 
Join Date: Dec 2002
Posts: 5,565
What problem? Error message?
Please stop filling AV1/Daala/VP9 threads with your ffmpeg and color troubles. Make new threads in the newbie section.

Last edited by sneaker_ger; 20th October 2016 at 20:00.
sneaker_ger is offline   Reply With Quote
Old 20th October 2016, 22:34   #95  |  Link
Jamaika
Registered User
 
Join Date: Jul 2015
Posts: 697
Problem with conversion
https://www.sendspace.com/file/2qnw7j

Last edited by Jamaika; 21st October 2016 at 07:52.
Jamaika is offline   Reply With Quote
Old 31st October 2016, 08:58   #96  |  Link
Jamaika
Registered User
 
Join Date: Jul 2015
Posts: 697
Test Clare codec AV1 vs X165 is underestimated because preset is good, bpg has veryslow.
Preset for X265 is:
0.ultrafast
1.superfast
2.veryfast
3.faster
4.fast
5.medium (default)
6.slow
7.slower
8.veryslow
9.placebo
Preset for VPX/AOM is:
0. rt + 37% quality
5. good (default) + 50% quality
9. best + max.63% quality
I associate it mistakenly with the parameter VPX "cq-level". So you could set for veryslow quality VPX best + max.63%.

No one probably doesn't know what effect it has value of quality "min-q". The first frame is always a big mistake for added value for codec VPX and AV1.

Interesting is also why each encoder VPX&AOM has a different parameter of quality.
FFmpeg has CRF.
What possesses Adobe plugin 1.003? Must enter function "cq-level". Changing the value of quality means a change bitrate and the number of frames P. 37% is 2frames, 50% is 4 frames, 63% is 6 frames. It is a pity that libvpx isn't such fuction. I could make a mistake because I did not use the same function {--tile-columns=4 --arnr-maxframes=7 --arnr-strength=5}.
Approached by a characteristic size of the frame to the X265. Changing good to best doesn't change the size of the bitrate, only the number of reference vectors in frame.
Edit: The function pass 2 is a falsification. There is no dynamic frame B. Analyzer video Elecard doesn't show that it has been implemented.

Last edited by Jamaika; 20th December 2016 at 10:25.
Jamaika is offline   Reply With Quote
Old 1st November 2016, 14:39   #97  |  Link
benwaggoner
Moderator
 
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,750
Quote:
Originally Posted by Jamaika View Post
No one probably doesn't know what effect it has value of quality "min-q". The first frame is always a big mistake for added value for codec VPX and AV1.
In the VPx series, "min-q" always set the minimum quality. No frame would be encoded with a quality less than that IIRC, depending on other settings, this could lead to either dropped frames or increased bitrate (historically there wasn't a VPx "VBV" to violate, but that is changing with AOM).
__________________
Ben Waggoner
Principal Video Specialist, Amazon Prime Video

My Compression Book
benwaggoner is offline   Reply With Quote
Old 1st November 2016, 14:58   #98  |  Link
Jamaika
Registered User
 
Join Date: Jul 2015
Posts: 697
I didn't notice dropped frames. However, the bitrate of the first frame I is 2x larger or less than others frames I. It depends on the aspect ratio video. Why is it like that? Only the creator knows.
Quote:
Originally Posted by benwaggoner View Post
(historically there wasn't a VPx "VBV" to violate, but that is changing with AOM).
We'll see how Elecard release free analyzer. It is only known that the AOM is based on VP9

Last edited by Jamaika; 1st November 2016 at 15:50.
Jamaika is offline   Reply With Quote
Old 1st November 2016, 16:43   #99  |  Link
benwaggoner
Moderator
 
Join Date: Jan 2006
Location: Portland, OR
Posts: 4,750
Quote:
Originally Posted by Jamaika View Post
I didn't notice dropped frames. However, the bitrate of the first frame I is 2x larger or less than others frames I. It depends on the aspect ratio video. Why is it like that? Only the creator knows.
Bigger perhaps due to initial buffer fullness assumptions. This was a VPx parameter.

Why it would depend on aspect ratios is baffling. Was this different content aspect ratios in square pixel, or different SARs? If you have a repro-able bug, it should get passed on to the AOM folks.
__________________
Ben Waggoner
Principal Video Specialist, Amazon Prime Video

My Compression Book
benwaggoner is offline   Reply With Quote
Old 1st November 2016, 17:12   #100  |  Link
Jamaika
Registered User
 
Join Date: Jul 2015
Posts: 697
Thanks for hints. Indeed AOM fixed problem first frame I. No need to to use min-q for the FullHD. To test this, I used the analyzer mkvinfo.
mkvinfo -s av1.webm >> av1.txt
Unfortunately, I can't create (profile 3) movies 4K. Encoder requires 16MB RAM. I've got too old for computer.
Code:
Track 1: video, codec ID: V_AV1, mkvmerge/mkvextract track ID: 0, pixel width: 2048, pixel height: 1152, yuv422p10le (profile 3), min-q=0
I frame, track 1, timecode 0 (00:00:00.000), size 166818, adler 0x00b4b52f
...
I frame, track 1, timecode 2000 (00:00:02.000), size 151294, adler 0x7c901f15
...
I frame, track 1, timecode 4000 (00:00:04.000), size 153129, adler 0xf2b726d9
...
I frame, track 1, timecode 6000 (00:00:06.000), size 163401, adler 0xb73093d4
...
I frame, track 1, timecode 8000 (00:00:08.000), size 184388, adler 0x19ddc916
Code:
Track 1: video, codec ID: V_AV1, mkvmerge/mkvextract track ID: 0, pixel width: 3840, pixel height: 2160, yuv420p10le (profile 2), min-q=0
I frame, track 1, timecode 0 (00:00:00.000), size 483939, adler 0x9517be61
...
I frame, track 1, timecode 2000 (00:00:02.000), size 379716, adler 0x3066620a
...
I frame, track 1, timecode 4000 (00:00:04.000), size 385909, adler 0x409e46ec
...
I frame, track 1, timecode 6000 (00:00:06.000), size 402404, adler 0xfe265630
...
I frame, track 1, timecode 8000 (00:00:08.000), size 380221, adler 0x36dd028a
...
I frame, track 1, timecode 10000 (00:00:10.000), size 498620, adler 0x16fbcc9d
...
I frame, track 1, timecode 12000 (00:00:12.000), size 414623, adler 0xa940b1b2
Code:
Track 1: video, codec ID: V_AV1, mkvmerge/mkvextract track ID: 0, pixel width: 3840, pixel height: 2160, yuv420p10le (profile 2), min-q=5
I frame, track 1, timecode 0 (00:00:00.000), size 346569, adler 0x2f11e1bc
...
I frame, track 1, timecode 2000 (00:00:02.000), size 340183, adler 0x2be6ca69
...
I frame, track 1, timecode 4000 (00:00:04.000), size 341161, adler 0xcdc0e26d
...
I frame, track 1, timecode 6000 (00:00:06.000), size 356605, adler 0x8767bc55
...
I frame, track 1, timecode 8000 (00:00:08.000), size 369449, adler 0xa6319ee4
...
I frame, track 1, timecode 10000 (00:00:10.000), size 369914, adler 0xb9adbcbc

Last edited by Jamaika; 2nd November 2016 at 11:08.
Jamaika 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 17:13.


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