View Full Version : Particle Calculation Time...help!!!

07-01-2004, 11:59 AM
Spent the last few days wrestling with particles and on the whole I'm starting to win..I think...but I keep hitting a problem when I try to calculate the particle solution.

In simplicity my scene has an emitter and wind that generates 100 particles over 4000 frames using the wind 'path' option. The particles all follow the motion path of the wind...great. I now want each of those particles to emit more particles at frame 3500 to 4000 so they look like fireworks with particle flames coming out the back. Again I've got this working using another emitter for the flames, parenting it to the main emitter and setting 'parent-key' for playback. Fab dab. I've even added hypervoxels 'sprite mode'....brill.

OpenGL plays it back almost realtime and all is well. However, when I come to 'calculate' it the computer has a heart attack.....frames 3500 to 3530 go reasonably swiftly but then it starts to slow down on every frame and eventually at 3540 we're talking 15 mins per frame and slowing. The flame emitter paticles total 56687 and another 100 for the main emitter...With eight more of these setups to do things are looking bleak.

Is it just down to particle overload...why does the 'on the fly' OpenGL cope so well...how do I solve it??? Thanks for any help

07-01-2004, 01:22 PM
Take a look at your calculation resolution under the options! The realtime playback uses a resolution that is about 10 to 100 times rougher depending on the overall size of the effect; dunno how it is calculated exactly) than the actual calculation resolution so it is always interactive. So if your calculation resolution is set to 100 mm it would use ca. 1 m in the OpenGL preview. also check if you are using something like inter particle collisions, they slow down even the most simplistic setup.


07-01-2004, 02:56 PM
Thanks for the reply, options was set to 100mm so I've boosted it up to 5m:eek: . The certainly has speeded things up and will get the job done so thankyou. There is one collision object, a plane that the particles bounce off so thats probably much to blame. As the particles only need to be rough it's working a treat....10 mins and half way through already. Just a whopping great PFX file now for the render machines:rolleyes: