Learning Forum

[Vu Vuong]

I want to run SBR+ simulation of some complex model in HFSS, but my laptop is obviously not up to the task.
So I’m trying to upgrading my PC  so that can run HFSS better.
Should I go for higher clock speed or the number of cores?. I have several options at the moment
1- Xenon gold 6148 with clock speed of only 2.4 Ghz in base and can be turbo to 3.7 GHz but it got 20 cores and 40 threads, 

2-core i5-13600K with clock speed of 3.9 GHz in based and can be turbo to 5.1 GHz but it got only 14 cores and 20 threads

3-dual xenon 6146 with clock speed of 3.2 Ghz in based and can be turbo to 4.2 GHz, each CPU has 12 cores (total 24 cores) and 40 threads

[GioF]

Hi Vu Vuong,

Setting up an appropriate workstation for your HFSS simulation may be a complicated task. Here is some more insight which will help you decide,

For the meshing stage of your simulation you should consider picking a higher clock speed in your CPU.

For the matrix solve calculations you should aim for the greatest amount of cores. This will significantly reduce your simulation time.

Last, since you refered to SBR+ simulation, I would like to mention that for SBR+ simulations GPU acceleration is available for complex model. So this may be another point for you to consider. You can find some more information on that in the following link,

Ansys Electronics 2023 R2 - GPU Acceleration