Bob H." wrote:

>
> The galaxy collision program I was trying to convert over from Basic to POV
> a while ago does gravity interaction in a sense.  Each particle is the same
> mass but I'm sure that is a trivial matter.  The stumbling block was due to
> my not knowing how to get the DIM (dimension) arrays changed, among other
> things.  Was intended for 2D plotting of the 3D motions.
> Really needs to be more a orbit simulator.
>
> Pretty neat web pages here:
> http://burtleburtle.net/bob/physics/orbit101.html
>
> Bob H.

When I did a simulation, it looked exactly like the next to last one in the
link you shared here. My simulation was simply my flocking algo:  at time t,
calculate the gravitational forces on every particle, and use this acceleration
to modify the current gravity.  I spent weeks and never got a stable system, I
think, because I didn't conserve energy. This, I think would take calcs at
infinitely small delta t's.   Or maybe something about potential energy,  but
I'm confused.  Your model would never decay, and allow a stray "Jupiter" to
wander into it?  If so, what's the algo?

Post a reply to this message