Game Development - Closed Source Stuff? FSR,DLSS,RT etc

Hello,

i try to find some infos/more indepth Infos about this big topic. If you are a vivid Gamer and read the news, Starfield looks like using AMDs techs and may not support dlss at release or ever. Its kind of heated for this game, as it looks like there is no playstation version, only pc and xbox.

some folks complained in the amd subreddit that these kinds of deals are shady etc.
maybe they remember, that nvidia did/does this practice with their brading in games and assume, that amd prohibits bethesda to implement this stuff. no one knows for sure, because we cant read the contracts.

so i want to know, why bethesda moved this way? lack of programmers doing one kind of tech three times for all of the names? maybe they want to produce a “bugfree” game, instead of supporting duplicated features? im not working in this sector and cant figure it out why this is kind of an heated argument. i mean nvidia/amd had alot of tech that wasn’t supported by many games and died out, like physx,gsync, amd’s old tesselation, dx 10.1 etc.

Guessing here, but the answer is probably because they’re writing for the console and the PC version is a port. Both the PS5 and whatever alphabet soup the current gen Xbox is are running RNDA gpus.

Edit, expanding: Both consoles and the Steam Deck run RDNA. FSR runs on those, plus also runs on any GPU you can throw at it on the PC side. It would save time to write once and run everywhere (despite how much of a meme Java made that idea…)

1 Like

na they use both amd hardware to the same extent. cant remember the exact comparable pc hardware for this, but they are nearly the same.

what i wanted to know, out of a developer perspective, why they went the amd way as pc exclusive partner. i can imagine they wanted to cheap out on workforce.

If it’s releasing on Xbox, then it isn’t PC exclusive, right?

Either way, to the extent that FSR and DLSS do the same thing, being able to just implement one of those (FSR, if my understanding is correct) and have it run on every supported platform, vs. having to implement both of them while not adding support for additional platforms, is a clear benefit. I don’t see the reason to implement DLSS in that case. No shady backroom deals required, just an aversion to wasted effort.

1 Like