Sampletime on FMS 2.0 alpha 8 - What setting is the best ?

[Rekvin]

The sampletime might have a lot of impact for some people. I can choose anything between 0,005 and 10 and it gives more or less the same results. (I'm using an USB interface, AMD XP2000 and ATI Radeon 9800 pro).

The r/c USB controller is completely smooth even with extremely high sampletime (1 s and more). What I notice is just that the performance on the simulator is comprimised if I have too small sampletime (the FPS start drop when the sampletime become smaller than 0,001 on an XP2000 cpu). Then my processor is not strong enough, but it is no point since the controller is not affected by high sampletimes (for example 1 s.)

So I would assume that people with a slow cpu could use a very high sampletime (example 1 s) if the controller seem to work smoothly at that setting. Then they would just get better performance in the simulator (more fps).

What I don't know is if the sampletime affects anything else than the controller ?? The physics seems to be more or less the same at different sampletimes. You just get better peformance when the sampletime is high.

Anyone who have experienced something different with this ?

[fzr]

The sample time is (one divided by) the rate at which the physics is updated. In free flight, slow updates are quite tolerable as long as the forces on the plane do not vary rapidly.

However, when collision detection & resolution come into play, the forces and their rate of change become significant and a higher sample rate is essential to keep things smooth.

Best bet is to have it set as low as it will go (shortest sample time = highest sample rate) without ruining your framerate.