home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.arch
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!decwrl!concert!gatech!news.byu.edu!ux1!fcom.cc.utah.edu!ursa11.law.utah.edu!caperkin
- From: caperkin@ursa11.law.utah.edu (Charles Perkins)
- Subject: Re: 600 transistor CPU (was: MINIMUM instruction set)
- Message-ID: <1992Nov13.160442.9619@fcom.cc.utah.edu>
- Summary: Program the FLASC?
- Sender: Charles A. Perkins
- Organization: University of Utah College of Law
- References: <1992Nov12.164510.28248@dartvax.dartmouth.edu> <1992Nov12.220301.23915@fcom.cc.utah.edu> <1992Nov13.104602.10923@odin.diku.dk>
- Date: Fri, 13 Nov 92 16:04:42 GMT
- Lines: 36
-
- In article <1992Nov13.104602.10923@odin.diku.dk> torbenm@diku.dk (Torben AEgidius Mogensen) writes:
- >caperkin@ursa11.law.utah.edu (Charles Perkins) writes:
- >
- >>This caught my attention because of an idea I have been toying with, which
- >>I have decided to call the Fractal Logic Asynchronous Scalable Circuit.
- >>(Basically a 'memory' where every cell is a finite state autonoma 'conversing'
- >>with the cells next to it.) I figure eight thousand of the little guys on a
- >>chip, simple though they might be individually, could get some serious
- >>computation done.
- >
- >Have you considered how to program such a beast? And on the hardware
- >side, how do you load a program?
- >
- > Torben Mogensen (torbenm@diku.dk)
-
- It wouldn't be in C.
-
- Actually, the whole idea is to break out of the VN bottleneck.
- I imagine the architecture would be best suited to a wierd variant of prolog.
- Each cell, when queried, could (hypothetically) either return it's value
- (true or false) or (in conjunction with cells next to it) query other cells
- to determine what it's value ought to be, then report that.
-
- Thus there would be facts and rules. The more complicated a query, the more
- branching and parallel action occurs. (The organization of the cells
- would also follow a tree structure, rather than a linear or array structure,
- for several reasons. The main reason being the 'Scalable' in the acronym.)
-
- Programming would be through set, rather than query location.
-
- I would hate to port X to it, though.
-
- --
- .8Chuck .2AlexanderGrahmBell
- perkins@ursa11.law.utah.edu
- "I am not a potted plant."
-