Live data from Hacker News

AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

videocardz.com

191–200 of 227 posts

Re: AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

#191
post #98

Earlier quoted context omitted.

Video codec is the most important though !

Are you deaf? I can't see how anyone else would be happy downloading a file they can watch fine but can't hear .

Audio patents are expiring and there shouldn’t be any difficulty decoding real time most audio codecs. If you can decode the video real time there’s a very good chance you can decode the audio real time too.

Re: AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

#192
post #143

Earlier quoted context omitted.

Are you deaf? I can't see how anyone else would be happy downloading a file they can watch fine but can't hear .

Not to mention that the codec barely tells you anything useful on its own. There are many parameters that affect the quality and file size. And heck it's not even guaranteed that you will be able to watch a video just given its codec. Like an old computer might be able to decode H.264 but only be able to do it in real-time at 480p due to processing requirements for going higher than that.

The goal is not to make something perfect its to make something better. AV1 / AV2 extension is better than MOV / MKV extension

Re: AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

#193

Earlier quoted context omitted.

Prior codecs all had film grain synthesis too (at least back to H264), but nobody used it. Partly because it was obviously artificial and partly because it was too expensive to render, since you had to copy each frame to apply effects to it.

AFAIK, FGS support is the exception instead of the rule. h.264 only had FGS as an afterthought, introduced years after the spec was ratified. No wonder it wasn’t widely adopted. VP9, h.265 and h.266 don’t have FGS.

Did they remove it from the specs? Wouldn't surprise me. But since it's just extra metadata and doesn't affect encoding you could still use the old spec.

Re: AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

#194
post #50

Will streaming services ever stop over-compressing their content? I have a top-of-the-line 4K TV and gigabit internet, yet the compression artifacts make everything look like putty. Honestly, the best picture quality I’ve ever seen was over 20 years ago using simple digital rabbit ears. You especially notice the compression on gradients and in dark movie scenes. And yes — my TV is fully calibrated, and I’m paying for…

4K Blu-ray is the top quality.

Re: AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

#195

Earlier quoted context omitted.

AFAIK, FGS support is the exception instead of the rule. h.264 only had FGS as an afterthought, introduced years after the spec was ratified. No wonder it wasn’t widely adopted. VP9, h.265 and h.266 don’t have FGS.

Did they remove it from the specs? Wouldn't surprise me. But since it's just extra metadata and doesn't affect encoding you could still use the old spec.

I don’t think it was ever part of the spec. It would surprise me if it was, because once a spec has been finalized and voted on, changing it is complicated. Getting agreement the first time is already difficult enough.

Re: AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

#196

Earlier quoted context omitted.

This sounds like what is called the soap opera effect: https://en.wikipedia.org/wiki/Soap_opera_effect Which is generally associated with excess denoisong rather than with excess grain.

Are you sure? That seems to be about motion interpolation . Not even about smooth motion, but about it being interpolated . Here the concern seems to be about how the individual, still images look, not anything about the motion between them.

Avatar 2 was particularly egregious with the poor interpolation

Re: AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

#197

Earlier quoted context omitted.

"A terrestrial (over-the-air) transmission carries 19.39 megabits of data per second (a fluctuating bandwidth of about 18.3 Mbit/s left after overhead such as error correction, program guide, closed captioning, etc.)," https://en.wikipedia.org/wiki/ATSC_standards#MPEG-2

But that's the whole multiplex, right? It could be a single channel, but usually you have many in the multiplex. I don't know how it works in the US, but for DVB-T(2) that's how it is.

Circa 2009 when analog TV was first shut-off in the US, each DTV station usually only had one channel, or perhaps a second basic one like a static weather radar on-screen. Some did have 3 or 4 sub-channels early on, but it was uncommon.

Circa 2019, after the FCC "repack" / "incentive auction" (to free-up TV channels for cellular LTE use) it became very common for each RF channel to carry 4+ channels. But to be fair, many broadcasters did purchase new, improved MPEG-2 encoders at that time, which do perform better with a lower bit-rate, so quality didn't degrade by a lot.

Re: AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

#198
post #113
post #110

Earlier quoted context omitted.

>Will streaming services ever stop over-compressing their content? Before COVID Netflix were at least using 8Mbps for 1080P content. With x264 / beamr it is pretty good, and even better on HEVC. Then COVID hit, every streaming service not just Netflix have excuses to lower their quality due to increased demand with limited bandwidth. Everything went down hill since then. Customer got used to lower quality I dont beli…

Most 1080p WEB-DLs are in the 6-8 Mbps range still, based on a quick glance.

Nice. There was a previous post on HN with showing the 5 to 6 samples he had were from 3 - 5Mbps.

Re: AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

#199
post #85

Earlier quoted context omitted.

Could help people on the hobby or lower budget FPGA side. H.264/5/etc never really made it

There is absolutely no way an FPGA would make sense. The requirements for AV1 and H265 far exceed the hardware resources of lower budget FPGAs. For the same process, FPGA logic density is about 40x lower than ASIC, and lower budget FPGAs use older processes. A h265 or AV1 decoder requires millions of logic gates (and DRAM memory bandwidth.) Only high-end FPGAs provide that.

There's mention that the decode could get a lot easier. Here's an H264 core that runs on older lattice chips and only takes 56k luts. https://www.latticesemi.com/products/designsoftwareandip/int... . Microchip's polarfires have a soft H.264 core as well taking under 20k. If AV2 will really be easier for hardware to implement, it might work out. Here's another example, H 264 decode in an artix 7, can do 1080p60 https://www.cast-inc.com/compression/avc-hevc-video-compress... . So with all due respect, what in the world are you talking about?

Re: AV2 video codec delivers 30% lower bitrate than AV1, final spec due in late 2025

#200
post #198
post #113

Earlier quoted context omitted.

Most 1080p WEB-DLs are in the 6-8 Mbps range still, based on a quick glance.

Nice. There was a previous post on HN with showing the 5 to 6 samples he had were from 3 - 5Mbps.

Some lower bitrate shows are animated cartoons where far less bitrate is really needed. I’m sure you could find some awful compression botches if you really looked on a public tracker though.
Post reply on HN