With the cooperative matrices starting to standardise I'm kind of starting to wish for a cross-studio cooperation to see an open ML upscaler good enough to kill proprietary solutions. I don't have the power to start something like that unfortunately. - ThreadSky

About ThreadSky

newincpp.bsky.social • 1 day ago

With the cooperative matrices starting to standardise I'm kind of starting to wish for a cross-studio cooperation to see an open ML upscaler good enough to kill proprietary solutions.
I don't have the power to start something like that unfortunately.

Comments

meuns.bsky.social•1 day ago

Dp4 was already there and nothing happened.

newincpp.bsky.social•1 day ago

Access to technology is only about 10% of the complexity of the endeavour here. The hard part is getting people or companies to work together.

aschrein.bsky.social•1 day ago

ML in gfx is quite a switch, you need to align a lot of figures to begin getting it to work. There's a clear percolation threshold - a lot of the times it's unclear it should work at all, and on top of that there's not a ton of papers out there(for realtime), almost tribal knowledge at this point.

meuns.bsky.social•1 day ago

Convincing managers to put money inside a large enough team to achieve this when IHVs are giving upsamplers is difficult... ^^

newincpp.bsky.social•1 day ago

Something completely open where the training process would be shipped with it and that would be designed to be retrained relatively rapidly so the results could be handcrafted per-projects.
A bit like the process mentioned in here: https://teamdeakins.libsyn.com/jo-plaete-vfx-supervisor

newincpp.bsky.social•1 day ago

Saying this "per project" but if it's along the line of what it's mentioned in the podcast. There is a universe where a game could switch models like colour grading. Loading them and switching them depending on what the scene needs.

Posting Rules

Be respectful to others
No spam or self-promotion
Stay on topic
Follow Bluesky's terms of service

Comments

Posting Rules

Reply