Oh no no no no! AV1 dead and finished!
bbc.co.uk
VVC vs AV1 vs HEVC
Logan Moore
Aiden Murphy
S U I C I D E W A T C H
Hudson Kelly
The bitrate savings look higher with VV1 than with AV1. Surely that's what you want? VV1 doesn't save as much nitrate so will have higher filesizes
Jonathan Turner
Exact profile? Source material? Availability of hardware decoders? Objective quality metric? Decoding conplexity?
Angel Bailey
>save nitrate
>VVC CONFIRMED GRASSROOTS TERRORIST
David Harris
>PSNR metric
>encodes were made without -tune psnr
comparison done by retards
John Lee
>nitrate
filmfag confirmed, based projection booths exploding
>muh PSNR
only autist geeks thing PSNR is a relevant metric. Any AV professional knows better.
Thomas Lewis
How trustworthy are BBC articles like this? It looks very professional at a first glance, but once you read through it and see the results it looks rather shady.
No test files are given, quality assessment is purely done via PSNR (it also doesn't help that they call it a measurement for "objective quality"), no specific settings besides the pass count and speed preset are mentioned and those speed presets make no sense as aomenc doesn't use named presets like x264 or x265 but numerical values (although they don't mention what AV1 encoder they use either).
Also these results look biased. How realistic is it for a not even finished standard to have such a speed/quality ratio? Not even twice as resource intensive to encode as AV1 and yet already that much more efficient? Which video coding format in the past has managed to perform such a feat? None. They all were slow as dirt compared to the predecessors.
I also heavily doubt the AV1 vs. HEVC results (especially HEVC being more efficient for high bitrates), but that's a different story.
Nolan Sullivan
No codecs listed...
Henry Russell
Cope
Blake Cook
AV1 is in its infancy in terms of encoder maturity, just compare x265 during its first year of release compared to the quality it outputs today.
AV1 was finalized last summer, we currently have three open source encoders in development, in about a year they should give at least a decent estimate of what AV1 offers quality-wise, as of now there has been practically no meaningful tuning done.
Michael Moore
>How trustworthy are BBC articles like this?
BBC R&D is generally very trustworthy.
>I also heavily doubt the AV1 vs. HEVC results (especially HEVC being more efficient for high bitrates), but that's a different story.
You really shouldn't doubt it. It's been proven twice already that the license free alternatives to H.264 and H.265 cannot compete. VP8 and VP9 were both disappointments and were unable to 'beat' their licensed counterparts. So why would VP10/AV1 be any different really?
Alexander Jones
>So why would VP10/AV1 be any different really?
Because of the huge amount of patents at their disposal for VP10 compared to VP8/VP9, due to the forming of AOM (Alliance for Open Media) of which practically every major tech company is a member and pledge their patents.
Google, Amazon, Microsoft, Apple, Intel, AMD, Cisco, Facebook, Netflix, Mozilla, NVidia, Samsung, IBM, Realtek, Broadcom, ARM...
These are all companies behind AV1, and are all going to fully support it in software and hardware.
Easton Perry
>BBC R&D is generally very trustworthy.
They certainly don't appear like it with this particular article.
>So why would VP10/AV1 be any different really?
Because I've worked with all of them and those numbers go against everything I've experienced thus far. HEVC is really bad for high bitrate content, culminating at a much worse lossless compression than AV1 and AVC and there are more than enough reproducible (!) tests, which show AV1 as the clear winner for low bitrate content.
Also we're not looking at a situation like AVC vs. VP8, where the latter was missing important features, which made it clear from the beginning that it wouldn't be able to compete (the lack of a proper encoder also didn't help). From a technical standpoint AV1 is far more complex and represents the next generation of video coding formats (AVC/Theora/VP8 -> HEVC/VP9 -> AV1/VVC).
Cooper Roberts
In a year you'll be posting the same just wait tm pasta
Michael Myers
I bet you posted the same thing back when Xvid was king and h264 was just entering the scene.
So pointless.
Grayson Flores
Most of AOM joined the bandwagon in the last moment. They only contributed their logos.
The patent pool was significant enough to test eighty something but it turned out they had to ditch most of them.
In urban planning terms you can think of AV1 as Kowloon Walled City
Kayden Hall
PSNR
More like Penis NR (Nigger Rape)
Thomas Lopez
What an excellent post!
Landon Perry
OP works for MPEG
Nathaniel Reyes
I hope VVC is good. MPEG is really superior to Google+FOSS stunts at desigining the technology and doing it in a proper open/public review process. As opposed to AV1 where google's ideas went in without review and decisions were made by few google guys on a secret mailing list.
The risk here is that nobody will use the format and nobody will write encoders because all the free software guys are enamored with AV1/Google. So we may be stuck with that.
Leo Clark
Mpeg2 was the best codec, it would run 1080 easily even on toasty toasters.
Lucas Martin
Damn, open source just can't catch a break.
Evan Fisher
And now with a trustworthy metric.
streaminglearningcenter.com
James White
>The risk here is that nobody will use the format and nobody will write encoders because all the free software guys are enamored with AV1/Google. So we may be stuck with that.
I mean, if they want to offer it royalty free, like AV1, go right ahead, i'm sure that would help spur adoption.
But if they want to make it like HEVC and keep demanding $$ to use it, get fucked, AV1 will reign king for any sort of online media distribution. No one wants to pay for that shit.
Xavier Watson
Beware that this site is actually owned by an MPEG/Sisvel shill.
Jayden Jenkins
>No one wants to pay for that shit.
No one wants to pay higher bandwidth expenses, either.
Aaron Martinez
Eh, depends who.
It can often be cheaper just to lay a bigger pipe than it is to waste time (and money) using a codec with royalties.
Especially if AV1 offers the majority of the bandwidth savings.
I simply don't trust the OP graph at all, and it's fairly pointless to discuss until VVC gets finalized anyway
Robert Cooper
Here are also some more recent results regarding encoding times.
streamingmedia.com