Network Simulation question

Control and manage your RealFlow Nodes
Thomas Schlick
Posts: 178
Joined: Tue Aug 29, 2017 12:35 pm

Re: Network Simulation question

Post by Thomas Schlick » Thu Dec 21, 2017 1:33 pm

Thanks for the scene. I'll forward it our developers.

The crown is the culprit here and I'm pretty sure that your scene won't crash on the CPU. The problem here is that the crown (and most daemons) aren't optimised for the GPU. So, the computer always shifts the workload between the CPU and the GPU - and this slows down everything. With 2 or more crown daemons the situation becomes even worse and that's certainly the reason for these crashes.

The difference between standard and DY is mainly the GPU support. Standard particles asre only simulated on the CPU. But when you use DySPH, for example, only on the CPU then I'm pretty sure that RF won't crash. The main method behind DySPH is the same as with standard particles, but DySPH is faster by default. DyPDB, on the other hand, is a completely different technology and not as accurate as SPH, but in your RF version we've added some improvements.

And when you follow my last post's tips on the crown's sheeter settings you'll most probably be able to control the number of extra particles.

Shinya Akasaki
Posts: 17
Joined: Wed Nov 29, 2017 6:14 am

Re: Network Simulation question

Post by Shinya Akasaki » Mon Dec 25, 2017 6:31 am

Thomas Schlick wrote:
Thu Dec 21, 2017 1:33 pm
Thanks for the scene. I'll forward it our developers.

The crown is the culprit here and I'm pretty sure that your scene won't crash on the CPU. The problem here is that the crown (and most daemons) aren't optimised for the GPU. So, the computer always shifts the workload between the CPU and the GPU - and this slows down everything. With 2 or more crown daemons the situation becomes even worse and that's certainly the reason for these crashes.

The difference between standard and DY is mainly the GPU support. Standard particles asre only simulated on the CPU. But when you use DySPH, for example, only on the CPU then I'm pretty sure that RF won't crash. The main method behind DySPH is the same as with standard particles, but DySPH is faster by default. DyPDB, on the other hand, is a completely different technology and not as accurate as SPH, but in your RF version we've added some improvements.

And when you follow my last post's tips on the crown's sheeter settings you'll most probably be able to control the number of extra particles.

Thank you for the long post! I have solved alot of problem. And yes, Im sure the scene crashes because of the overlapping particles. I made my Crown on/off animation and now everything simulate without a problem. Thanks alot!

Post Reply