I am wondering how feasible / common it is to do player physics / movement in Update rather than FixedUpdate. In the past I have either used a system to check & cache input in update and then use it in fixedupdate, or at least scale forces by deltatime to be framerate agnostic when used in update. However the current tech stack I have ended up with seems like I will have to use update and the framerate noticeably changes the player movement.
It seems to me that if the game can consistently achieve 60fps this willl not be a problem - and from watching some gdc talks it seems that this is a pretty common goal. And anyways many other elements of the game don’t feel very good below 60fps so i’ve been thinking maybe my goal should be to just get the game to run steadily at the target framerate on everyones machine rather than ensuring consistent physics simulation across a range of framerates.
Is this a foolish approach?
For background I am working on a networked multiplayer fps using Opsive’s UltimateCharacterController () and fixing this fps dependent physics seems like it will be fairly labor intensive.