I've noticed several signals that together may indicate a release of AV2 soon.First of all, its AVM's milestones are paused. The last milestone https://gitlab.com/AOMediaCodec/avm/-/milestones/20 lasted from Apr 8 to May 6. The previous milestone https://gitlab.com/AOMediaCodec/avm/-/milestones/19 lasted from Apr 2 to Apr 8, it immediately preceded the latest. However, there's (currently) no next milestone at https://gitlab.com/AOMediaCodec/avm/-/milestones/21 (error 404) for almost two weeks. That must mean something. That signal is stronger for every passing day.Then the open source of https://code.videolan.org/videolan/dav2d is out in the open.Then the compression performance results https://arxiv.org/pdf/2605.15800 are published (15 May 2026).
Can you not post works derived from Sisvel's patents without the relevant licensing package? Thanks
>>108859155I hope not, after all the kike shit being pulled off by proprietary video codecs (ie disabling hevc hardware decoding on computers that have said ASIC for it), AV2 needs to deliver TOTAL VVC DEATH not "10% better".
>>108859358av1 was already total vvc death while being 10% worse than vvcwhy the fuck would 10% better than vvc not be total vvc death as well? nobody would pay to use a patent encumbered AND less efficient codec.nothing is resurrecting vvc, no idea why you're so afraid of such extremely unlikely scenario.
>>108859358Being disabled in hardware is a good enough death for a proprietary codec.Also being 10% better is some old news. In my own opinion, SVT-AV1 has been already slighty better than VVCenc in Dec 2024.
>>108859155What is there to improve? We have hit a wall with video compression.
>>108861240The only wall hit is by your mom.
why do you use so many tripcodes and often take them off, daiz?fixlives,pixdaiz,jonsneeders,ss2,https://desuarchive.org/g/thread/108752241/#108752655
>>108859500I wonder if VVC is gaming objective metrics better than AV1 and thus the source of "look goys, vvc is 30% better than av1!" campaign. I didn't even know you could turn off the SSIM cheating in AV1.https://desuarchive.org/g/thread/108817526What makes me highly suspect VVC isn't really that much better than AV1 is you don't see 30% better compression when using the newest ssimulacra2 objective metric.
>>108862431Thanks for linking to your thread from the archive despite trannies hating the archive, Daiz. The third post led me to information you want scrubbed off the internet.
>>108862500Soldier we need you pronto, see >>108862353
>>108859155This won’t be usable for at least 2 years, unless you want to encode at 2 frames per second at 1080p
>>108861981Why have you tried to imitate my tripcode?You are not fooling anyone.
>>108863790I'm fine even with only one frame per second if the video is short (some 7 or 10 minutes), and the space to fit is tight (some 15 or 20 megabytes), and I'm not in a hurry.
