Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!uunet!mcsun!unido!rwthinf!ee.uni-sb.de
From: martin@ee.uni-sb.de (Martin Huber)
Newsgroups: comp.arch
Subject: Re: Optimal Computer Architectures
Message-ID: <3738@rwthinf.UUCP>
Date: 13 Dec 90 10:20:14 GMT
References: <1990Nov8.212412.14317@funet.fi> <3700@rwthinf.UUCP>
Sender: news@rwthinf.UUCP
Reply-To: martin@ee.uni-sb.de
Organization: Univ. of Saarland, Dept. of Electr. Eng.
Lines: 58
In-Reply-To: <1093@shakti.ncst.ernet.in>

I forward this for friend.   *Don't* reply to me, reply to
the address in the Reply-To header.

Here goes:

In article <1093@shakti.ncst.ernet.in> you write:
>In article <3700@rwthinf.UUCP> 
>
>>Norwitz Neal writes:
>>>I have a recursive algorithm that I would like to run as fast as possible.
>>>I am running a part of it on a SPARCstation SLC.  But at this rate, it might
>>>finnish in a couple of years!!
>
>and berg%cip-s01.informatik.rwth-aachen.de@unido.bitnet writes:
>
>>Why don't you have a look at the FORTH-chip (don't remember the manufacturer,
>>the part number should be something like 4016 if I recall correctly), the
>>processor has no general purpose registers, just a stack cache, the thing
>
>   The start-up that thought up the 4016 had start-up problems, and
>sold the technology to Harris. Harris cleaned up a lot of the hardware
>bugs in the Novix 4016, christened it the RTX 2000 (*R*eal *T*ime e*X*press),
>played with it for a little while, and decided to discontinue the line.
>
>   Sad. See other posts in this newgroup and comp.lan.forth for more
>details.
>
>   But your point on Stack Caching machines and recursive routines is
>well taken.
>
>   Just to comment, I think the original poster was talking about
>mature machines, and the Novix/RTX was yet no more than a chip and
>an architecture concept. (25ns RAM, three bank of it, no OS yet ...)
>
>-- shrikumar ( shri@ncst.in )

Another comment: I have been at the November '90 trade show ELECTRONICA
in Munich, Germany. I have visited HARRIS. They were perfectly willing
to sell the RTX2000. I got a databook, infos ... There was a Mr. Klaus
Flesch which is obviously something like general manager of the company
FORTH SYSTEMS, Germany. He *sells* RTX2000-based products. IMHO, the
RTX2000 is available. It *is* mature: 10MFIPS (FORTH micro-instructions
per second) is definitively state of the art in RISCS. The hardware is
capable of being micro-codable. I think this is better than many of
today's machines. Prices should be in the $500-$5000 range (please ex-
cuse my bad main memory, it is *not* a standard memory), in any case,
they *are* cheaper than a big useless number cruncher.

Take it or leave it, it's up to you!

Martin (mahu@ee.uni-sb.de)
Disclaimer: Mostly i wait for technology to improve, but sometimes
            improvements go by unnoticed. Moral: Watch out!
-- 
Sincerely,                 berg%cip-s01.informatik.rwth-aachen.de@unido.bitnet
           Stephen R. van den Berg.
"I code it in 5 min, optimize it in 90 min, because it's so well optimized:
it runs in only 5 min.  Actually, most of the time I optimize programs."