Inconsistant force readings in NL

I've recently noticed that often no limits, especially when running slowly (with low fps), give much higher force readings than normal. The speed is the same, but the forces go from intense to deadly. Whats up with this?

In game glitch: tell it to (i think he's named ole?) so he does not have that glitch in nolimits2. [;)]

It isn't a glitch, those forces are there. However, they happen so quickly that they might not even show up in 25%. This usually means you have a bump or a jerk where the train isn't rolling over a smooth radius so the quick but small change creates a spike. Real rides can do this too, but since it isn't sustained it doesn't matter.

I wouldn't worry about it unless you can clearly feel a jerk or bump where it happens, in which case you need to smooth it over a bit more unless it's intended.

No,I mean that the forces in places are higher than when the program is running well. For example, i get yellow g forces instead of the normal green that i get most of the time, with no change to that area of the track.

[Mikey]

It's a known glitch in the calculation routines for the game.

Originally posted by musicmonkey

No,I mean that the forces in places are higher than when the program is running well. For example, i get yellow g forces instead of the normal green that i get most of the time, with no change to that area of the track.

NL uses some sort of differencing scheme to calculate the forces, as all simulations do. If the fps is low, the simulator may choose to increase the "time step" in the physics simulation. If that is the case, the innaccuracies can increase by a power related to the time step difference.

For example. If the simulator is poop and uses the euler method, it will calculate the gradient at one point, extrapolate to the next time step, calculate the gradient there, and so on. The bigger the time step, the larger the error. However, the smaller the time step, the greater the processing overhead. See this graph for an example.



Put simply, the simulator is attempting to divert computer resources to the graphics, and it is felt in the physics engine.

Yeah, forces do change based on the fps...I've seen a clear difference in spots between a ride say with a lot of 3ds in view and the same ride with no scenery. Even on 25% I get that problem occasionally.

Interesting info blythy, thanks.

I have also noticed the G force is different at slower speeds. I like blythy's mathimatical explanation of the problem. But I'll have to admit, it's over my head. If the NL designers are trying to save computer resources for better graphics I think they are wasting their time because the new PCI Express video cards crank out some awesome FPS now. With my PCI Express card, I'm getting 100 FPS most of the time.