Recent content by OnesAndZeroes

I'll try it. I wonder how it will handle a threaded high memory bandwidth situation.

As an exercise I've coded a C++ fully connected neural network and am testing it on MNIST. The network is 784x512x512x10. I'm looking for the fastest reasonable CPU I can run this on. I've tried an i9-9900K with 4x 16GB RAM: 1 thread 3.8k training samples / second 15 threads...