This recent thread Artificial Neural Networks using RC networks and ICs led me to informally start exploring the area of analog computation and that led to me learning about Chua's Circuit.

Turns out this is a rich area, lots of interesting stuff going on, for example a true randomness generator based on chaos is described here.

There's also a neat website that explains the basics (which I'm about to read) so I wondered if any others here have built such circuits? Interestingly the circuits are often expressed in terms of an abstraction called a Chua Diode, has anyone else here ever heard of this beast?

Here's a little about Leon Chua, very interesting dude.

 

I remember trying to build the Chua circuit back in highschool. I didn't have very precise methods at the time, so it didn't end up working, but I definitely remember trying to figure out how it worked. Chua's other work with memristors and similar devices is fantastic, and formed the basis of some of my research in college.

 

I remember trying to build the Chua circuit back in highschool. I didn't have very precise methods at the time, so it didn't end up working, but I definitely remember trying to figure out how it worked. Chua's other work with memristors and similar devices is fantastic, and formed the basis of some of my research in college.

I wish I could really grasp the circuit and the differential equations its based on, I love this kind of stuff but I'm very rusty, spent too much time writing code for a living, it is refreshing to see mechanisms that are not algorithmic, that's the main thing behind what I've been saying here.

These days many of us tend to look at everything digitally, perceiving all kinds of functions as algorithms, but it's a narrow lens and I think we miss a lot, other ways of looking at things.

Its similar (in a sense) to the distinction between Turing Machines and Lambda Calculus, two very different but (proven to be) equally effective ways of describing "computability".

Lambda Calculus (well, functional languages) are much much closer to mathematics than is imperative programming, so here too its often refreshing to look at problems through a different lens.