[ODE] Re: Measurement System

[David Whittaker]

Just remember to choose base units such that, for your particular
simulation, all the values are as close to 1.0 as possible to reduce
numerical inaccuracies.  From my somewhat limited knowledge of fpu
calculations, you could probably choose another range, as long as all the
numbers are close to *each other*, but the standard advice is to keep it
around 1.0.

David

> <lukasz@datafast.net.au> skreiv:
>
>> Just to clarify, does this mean that stepsize in the dWorldStep
>> function is equal to your choice of unit of time? That is, if I choose
>> seconds then dWorldStep(id, 0.01) will step the world one hundred'th
>> of a second?
>
> Yes, but that of course requires you to use the same time unit for all
> other sizes, i.e. if you were to use minutes in dWorldStep, you'd have
> to calculate your gravity constant, typically 9.81 m/s2, to m/minutes2,
> if you understand. It would probably be very confusing...
>
>
> _______________________________________________
> ODE mailing list
> ODE@q12.org
> http://q12.org/mailman/listinfo/ode