question about time step between physics and (frames) process

:information_source: Attention Topic was automatically imported from the old Question2Answer platform.
:bust_in_silhouette: Asked By Asoth

Hello everyone,

is there practical case where you would avec a “physic delta” lower than target “display_delta”

otherwise :
if we consider that 60fps/15ms would be the standard target for pc games narrow-days, this involve that 15ms will always be the biggest timestep to be acceptable, right ?

would it be reasonable to have a 60fps/15ms target and and physics running at 30ms at the cost of 1 frame delay and tweening between 2 physics frames.

Thanks

(sorry if sound unclear, my battery running flat cannot polish the question)

:bust_in_silhouette: Reply From: MysteryGM

is there practical case where you would avec a “physic delta” lower
than target “display_delta”

You can lower the physics setting to be lower than the frame rate, this is possible because physics and rendering are separate systems.

Is it a good idea? No, it is a really bad idea.
The reason is in how physics calculate collisions. Less physics steps mean less time to find if something is colliding.
The result is things like the player falling through the floor etc. Really unstable physics.

would it be reasonable to have a 60fps/15ms target and and physics
running at 30ms at the cost of 1 frame delay and tweening between 2 physics frames.

If you set the physics frame lower you should see some stuttering. To remove this just tween the visual mesh with the position of the rigid body.
The main problem will be how inaccurate the physics get.

Thank you that what I thought but I needed a confirmation.
I wanted to know all possibilities to speed up my program if needed.
cheers

Asoth | 2018-11-30 09:53