dcrookston at gmail.com
Fri Jun 30 14:50:48 MDT 2006
I'd really just like to build a proof of concept machine. I guess I
could just as easily create a virtual proof of concept machine but
where's the fun in that?
On 6/30/06, Levi Pearson <levi at cold.org> wrote:
> I don't personally have experience with this, but I've read a bit
> about Lisp Machines.
> Those 8051s are microcontroller chips, which means they're CPUs with
> general purpose instruction sets designed for embedded systems. In
> other words, making them execute Lisp code is essentially the same
> process as making a desktop CPU execute Lisp code. You write a Lisp
> interpreter or compiler for the chip, and there you go.
> Lisp Machines, on the other hand, had instruction sets and CPU
> architectures optimized for Lisp features. Typically a lot of the
> higher-level instructions were microcoded rather than being fully
> hardware-defined. There are some papers describing how they were
> designed, as well as some more recent designs for Lisp and Scheme
> CPUs, out there just a google search away.
> There are also a few Lisp Machine emulators that you might want to
> check out.
> If you really want to build your own Lisp machine, the cheapest/
> easiest way would be to create it in a FPGA. There are some
> companies like Microtronics that make little development boards with
> FPGA-based CPUs. This would involve a lot of research and
> programming work to complete, and there would be no performance
> benefit over Lisp on a fast general-purpose CPU.
More information about the PLUG