@michael-3397 wrote:
In several threads, I’ve read that people get different results depending on whether their Topaz product is run on CPU only, or whether they run on an external graphics processor.
I don’t understand this. Why would the computation result depend on the computational platform that runs it? I would expect a fully determined computation to run more slowly on CPU-only execution compared to an eGPU. I would not expect to get different answers from an eGPU vs. CPU.
Would appreciate it if someone would explain. Thank you.
Posts: 2
Participants: 2