3 ms·

I'm extremely skeptical of this. Analog on such small scales is overwhelmed by noise and very low precision. Analog computations aren't magic either, they still

I'm extremely skeptical of this. Analog on such small scales is overwhelmed by noise and very low precision. Analog computations aren't magic either, they still take energy and produce waste heat. You are fighting the laws of thermodynamics. I certainly don't see how it beats quantum computers at the things they are good at over classical computers.

adamnemecek 9y ago
> I'm extremely skeptical of this. It's ok, it took me some time to realize that the bit-copper orthodoxy need to be toppled as well. > Analog on such small scales is overwhelmed by noise and very low precision. But here we are making more and more photonic and optic things. Fiber optic is taking over. There quite a bit of work in silicon-photonic design (you can check this book https://smile.amazon.com/Silicon-Photonics-Design-Devices-Systems/dp/1107085454/ https://smile.amazon.com/Silicon-Photonics-Design-Devices-Sy... I mean there's noise yes, but it's actually somewhat manageable in a photonic setup. > You are fighting the laws of thermodynamics. Nowhere near as much as with copper. A fiber optic cable doesn't heat up. > I certainly don't see how it beats quantum computers at the things they are good at over classical computers. O(1) differential calculus is magic for a lot of purposes.