Hello,
The short answer is the following:
The amount of memory on the video card determines the resolution of simulation you can run.
The single precision TFLOPS of the GPU primarily determine the speed.
For more information, please refer to the followng:
Discovery Live: “Thin” geometric features and relative GPU adjustments
http://www.pny.com/promo/ansys-discovery-live
You are navigating away from the AIS Discovery experience