[Haskell-cafe] Yampa integral function

[Jerzy Karczmarczuk]

Le 17/04/2013 00:36, felipe zapata a écrit :
> Hello,
> I'm trying to write some simulator of
> physical systems using Yampa.
> The fundamental idea is to integrate
> the position and velocity, using
> the following algorithm:
> ... ...
Nothing to do with Haskell.
Where do you get this factor 0.5 in your formula:  v = v0 + integral ( 
0.5 * force * recip mass) ?
With constant acceleration v=v0+a*Dt =>  1.01, not 1.05

What is that:  let x = x0 + vdt2 ??
Again, a constant acceleration movement gives : x = x0 + v0*Dt + a*Dt^2/2.

Yampa or not Yampa, I suspect that your implementation of physics is 
simply wrong.
==

In general, even correcting all, you might have reasonable results in 
some trivial cases, but in general the extrapolating Euler schema is 
unstable, produces growing errors (e. g. in the oscillating case).

Jerzy Karczmarczuk





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.haskell.org/pipermail/haskell-cafe/attachments/20130417/3933085c/attachment.htm>