back
1 comments
> The deterministic way in which conventional computers operate is not well suited to dealing with the uncertainty and randomness found in many physical processes and complex systems.

Isn't this the point of overlaying statistical calculations on top of deterministic functions? What benefit is gained mathematically or performantly from having the basic unit of computing be fuzzy (I understand the 0's and 1's are a digital abstraction on top of analogue reality)?

> The attraction of using photonic components is that they operate much faster and are considerably more energy efficient, says Charles Roques-Carmes...

I am not a physicist. But the hangup for me seems to be not the components used, but the interactions between components. Don't photonic components have to interact with deterministic models or algorithms at some point? Or, if they themselves make up the model, doesn't the model interact with finite or deterministic elements when defining bounds?

> ...“The main advantage is that you could generate, in principle, very many random numbers per second...”

Generating very many random numbers isn't the problem. You can do that via a camera pointed at lava lamps [0]. It's getting the random numbers where they need to be, to interact with the models that use them. But maybe this is answered by answering my second point above.

[0] https://www.cloudflare.com/learning/ssl/lava-lamp-encryption...