David Rodriguez

Running a simulation with PDG it can be decomposed in about 800 tasks. Using the 10 cores of my CPU does not help too much. I would like to run the simulations as fast as my full pc can afford.

Would it be possible to throw that 800 processes to the GPU cores instead of the CPU cores?

If it makes sense, can somebody give me some steps or guidances about how to implement it?

Thanks in advance
and appreciate any information.

Thank you so much Tomas! Had no idea that that could be a problem.

Appreciate!

Hello everyone.

I am trying to get a collision between Tommy and the grains inside a POPNetwork. I tried plenty of different combinations: increasing substeps, compute velocity with trails and so on… but can not get any collission.




Am I doing something wrong? Can I not define grains inside PopNet?

Appreciate any kind of help because have been struggling with it for more than a while.

I attach the file so it will be easier to take a look.

Thanks in advance.