UE5 by default uses a lot of flashy tech that is supposed to improve performance, but a lot of it only does so in scenarios that are already extremely unoptimized. Using more traditional methods tends to achieve the same fidelity at a fraction of the performance cost. But there’s no time for optimization, and these fancy options “just work”, so there ya go.
The end result is a poorly running blurry mess of a game, but at least it’s on schedule I guess.
Codilingus@sh.itjust.works 4 weeks ago
I looked up some videos from YouTube sleuths on why so many UE5 games suck. For any studio previously using UE3 or 4, they had to relearn/recreate nearly their entire workflow again. 5 very much changed damn near everything. But also that 5 has all this tech that everyone assumes works in all scenarios and is a miracle. When in reality it’s still software tech and has very real limitations and best use cases that studios ignore. Larger studios “should” be able to trial and error while burning through $ to figure it out, but usually management doesn’t give them enough time. Smaller studios can’t afford to have many many months of downtime learning to re-adapt everything. It’s just so damn complex that very few have had time and $ to just trial and error figure out its limitations and to work within them.
It SHOULD get better and better as time goes on, though. The tech pieces in 5 keep getting improvements, and theoretically people should eventually start to adapt to it correctly, and the knowledge should spread as devs move to different studios for new work.