Re: [glob2-devel] To Erik: My Profiling Information

[Leo Wandersleb]

> > Looks like you use a small Taylor polynomials for cos and calculate it
> using the
> > Horner scheme.  I don't think this can be made faster.

um yet there is not trigonometrical function in the clouds calculus.
i was only refering to possible improvements. also this c/s would only be 
needed once per timestep!
In my example it was to calculate dx so that x+dx lies on the circle around x 
and circles with time.

> We could use a faster random generator.

also random numbers are only needed once _per game_.

> We could reuse random numbers from the game part of glob2,
> because dependence between game and graphics wouldn't make
> clouds uglier.  Only game random numbers must not rely on graphics
> random numbers, but the other way round it's OK.

that should be done to let clouds influence game later on.

> We could use a queue (may with limited maximum size).
> When we create a game random number, we put it also into that
> queue.  If we pull a graphics random number, we pull it from that
> queue.  Unless the queue is empty.  In this case we have to create
> a graphics random number.  We can't use this for game decisions,
> because it is not synced and must not be, because not everyone
> uses high graphics details.

wind would not need clouds (high grafix).

greetings, leo wandersleb