Not sure I understand what rewinding has to do with anything here?
Even with box2d you will see strange behavior on collisions if you will do just a linear interpolation during those 10 units. It will look very glitchy and people will see that as a bug or bad physics.
Replying to da447m (#25792)
Not sure I understand what rewinding has to do with anything here?
It is in my OP :)
As an extra here, I always assume[d] jon blow did something like this with braid rewind to help compress state/pos, but to my surprise he said somewhere he recorded everything in 60hz!
So what is broken with such apparently simple pattern?
(sorry this Q looks a bit all over the place :)
Replying to mmozeiko (#25793)
To rewind you simply snapshot state after each simulate_physics() call, then you can reset it to any of previous states as you wish. For fixed timestep simulations this will be very simple and deterministic. I don't know what/how Braid did or what issues it solved or not.
There is this GDC video about the rewind in Braid: The Implementation of Rewind in Braid
Replying to mmozeiko (#25796)