mkhaw@teknowledge-vaxc.ARPA (Michael Khaw) (07/21/87)
What ever happened to Novix and their (RISC?) Forth chip?
Mike Khaw
--
internet: mkhaw@teknowledge-vaxc.arpa
usenet: {hplabs|sun|ucbvax|decwrl|sri-unix}!mkhaw%teknowledge-vaxc.arpa
USnail: Teknowledge Inc, 1850 Embarcadero Rd, POB 10119, Palo Alto, CA 94303mao@blipyramid.BLI.COM (Mike Olson) (07/22/87)
In article <14800@teknowledge-vaxc.ARPA>, mkhaw@teknowledge-vaxc.ARPA (Michael Khaw) writes: > What ever happened to Novix and their (RISC?) Forth chip? chuck haley (of forth fame) sold forth machine kits for a while. he may still do it. he designed (or helped design) the chip. he gave a presentation to the computer network of kansas city last year, when i still lived there. the demo was pretty impressive. his demo "machine" was a cpu board connected to a disk drive and a set of single-pole contact switches with springs. there were five switches, all mounted on a ring of wire about the size of your hand. this was his keyboard. i think the demo was to impress upon people that you could get the power of a vax for very little money. one other interesting statistic -- the novix chip ate one cycle doing subroutine calls and returns. the call could be combined with other operations and executed in a single cycle. mike olson britton lee, inc. ...!ucbvax!ucsfcgl!blia!mao
billk@crash.CTS.COM (Bill Kelly) (07/23/87)
>chuck haley (of forth fame)...
The little box w/ Novix and five-finger keyboard you described sounds like something I saw at the 1986 FORML.
Are you sure you didn't mean Charles H. Moore instead of "chuck haley" ?
Chuck Moore is definately of Forth fame (can you say, "invented Forth?") and
I believe was the designer of the Novix chip. (Also commonly called the "Chuck" chip, at least by people around here...)
Or was there more than one "chuck" who was involved with the Novix?
BTW, there have been a few messages about "The Winner" -- which chip does Forth threading the fastest, or something like that. (I'm not sure whether this
comparison was meant to include Forth's which are subroutine threaded, and
thus have no NEXT, or whether "The Winner" had to be an implimentation of
Forth that used an inner interpreter. Like indirect, direct, and token
threaded Forths.)
I would suspect that the Novix chip blows 'em all away. It is supposed to
take ONE cycle to do a call/return. One cycle for the call, _zero_ for the
return, because it can be merged with the previous op-code. That's going
to be pretty speedy... (or, I should say, *is* speedy).
--
--
Bill Kelly {hplabs!hp-sdd, ihnp4, sdcsvax}!crash!billk
"I hate operating systems!" --GMKwjb@moss.ATT.COM (07/23/87)
Come now, people, this discussion has drifted from anything related to the C programming language. Let's narrow the the newsgroup selection a bit, shall we? --Bill Buie
am@cl.cam.ac.uk (Alan Mycroft) (07/29/87)
In article <14800@teknowledge-vaxc.ARPA> mkhaw@teknowledge-vaxc.ARPA (Michael Khaw) writes: >What ever happened to Novix and their (RISC?) Forth chip? Dunno, but: A firm called Advanced Processor design, Newlands High Tech. Centre, Inglemire Lane, Hull HU6 7TQ, U.K. (0482 855927) (they were previously called Meta-Forth) has a VME card with a 3 times 780 performance which has Forth as its machine code. It is built out of standard ttl (about 50 chips), clocked at 20MHz. A single chip (+ memory) version is around, presumably faster, although I do not know about the speed. Their boards are badge engineered in the U.S.A. by Xycom, 750 North Maple Road, Saline, Michigan 48176. (313) 429 4971.