von Neumann architecture

But we had to reinvent the entire chip implementation flow to deal with diagonal wires and ended up with 150 plus patents as a result, invented all kinds of stuff.

And at Tabula, the problem we're trying to solve is to help folks to design high performance chips quickly and inexpensively with the underlying technology we're trying to develop of building a scalable post-von Neumann computer and its associated software.

So, let me touch on a couple of the lessons that at least I've learned from the experience to date.

I think it's valuable to try to make connections that haven't been made before.

I try to explain memory, which is one particular aspect of cognition.

Computers were largely modeled on the understanding of how the human brain works, even at the time of von Neumann and, and at the very early stages, I believe.

So what I'm saying is that in the computers, you have basically the operating system, which is not something you can normally touch, uh, which, which runs the, the entire system.

Mm-hmm.

But it's nothing like the computers that are designed and built.

I think it's a completely different thing, and we haven't actually figured it out yet.

And we tend to apply modern, uh, information theory, which is based on von Neumann architecture and Shannon information, uh, to trying to quantify information in the brain.

Uh, but what we're really doing there is characterizing the potential limits of the brain.