Registered: Jan 2008
Posts: 1

Hi all. Im a electronics systems engineer and im going to start doctor's degree this year.

Ive been making some hardware implementations of CAs into FPGAs (http://en.wikipedia.org/wiki/Field-...able_gate_array) .

With this implementations you can reach much higher speeds than using a computer because all the cells are implemented and all the CA evolutes in just one step, instead of analyse the state of just one cell at the same time (what a computer does).

But there is a problem: these implementations are only useful when you want to make A LOT of evolution steps, i mean, if the computer does all the work in 1 sec, implement the CA on a FPGA doesnt make any sense.

One of my objectives is make an application/Mathematica interface that, defining a CA behavior, it implements it automatically on a FPGA and give us the control of the CA in order to write initial states, make it evolute ect.

All this is a hard work (i already have some parts done), so the question is, can this be useful for someone?

Thanks!

Report this post to a moderator | IP: Logged

Registered: Feb 2004
Posts: 172

I have been wondering about hardware applications also. I have an FPGA chip with all the supporting hardware. I know that the latest video cards from certain manufacturers have some software "off-sourcing" capabilities. I think they are called "GPGPU" (General Purpose Graphics Processing Unit). You can use the chips for more than just graphics (like you might be able to off load some code to increase the power of certain genetic algorithms). I hope to hear more about your discovery of applications (something besides random number generators!).

__________________
P h i l i p . R . D u t t o n

Report this post to a moderator | IP: Logged