My background is biology. Frankly neurons and cells just aren’t that fast.
The way we’re able to see high refresh rate isn’t based on neuron speed. (Ignoring input lag stuff). The eyes continuously sample as light hits them sending signals. Because it’s not a global shutter we’re able to perceive a displays refresh as it’s sending discrete images of a scene.
It’s far easier to notice an issue when there is a large change between two frames. The easiest way to reproduce this is in 3rd person games where you quickly rotate the camera as the large change between frames is very obvious to the eye. (There’s a limit obviously, but I can tell between 120Hz and roughly 240Hz, but I’m fine with 120Hz as I don’t tend to rotate the camera fast in games).
Google, Microsoft, and Amazon are making AI accelerators for their datacenters. Some are for training and others are for inference (running the trained models for services).
I’d be somewhat hesitant until the full benchmarks are out. AMD’s higher memory I think it was sounded neat, but that could be very fleeting in terms of advantages.