Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!decvax!bellcore!ulysses!allegra!princeton!caip!seismo!umcp-cs!eneevax!hsu
From: hsu@eneevax.UUCP (Dave Hsu)
Newsgroups: net.micro.16k
Subject: Re: Re: National's 32332
Message-ID: <692@eneevax.UUCP>
Date: Fri, 30-May-86 08:57:37 EDT
Article-I.D.: eneevax.692
Posted: Fri May 30 08:57:37 1986
Date-Received: Sun, 1-Jun-86 05:37:05 EDT
References: <746@usl.UUCP> <253@spar.UUCP> <2793@sdcrdcf.UUCP>
Reply-To: hsu@eneevax.UUCP (Dave Hsu)
Distribution: na
Organization: Imperial Widget Research Center, Kingdom of Maryland
Lines: 44

In article <921@alberta.UUCP> bjorn@alberta.UUCP (Bjorn R. Bjornsson) writes:
>In <438@geowhiz.UUCP> larry@geowhiz.UUCP (Larry McVoy) writes:
>> I have a few comments & questions.
>>
>> 1) The story I have heard (albeit from sources that I don't view as gospel)
>>    is that the 32016 (in the symmetrics box) is s*l*o*w, about 1/4 a 68010.
>>    Is that true?
>
>You would be well advised to put even less faith in your "sources", for
>even though I cannot answer for the Symmetrics box, the Dhrystone results
>speak (with the usual truck load of salt) for themselves:
>
>|*----------------DHRYSTONE VERSION 1.0 RESULTS BEGIN-------------------------
>|* NSC ICM-3216 NSC 32016-10Mhz	UNIX SVR2	cc		1041	1084
>|		-------------

Actually, we had a Symmetrics rep over back in December or January and
we did get a chance to dhrystone their unit.  Although the figure escapes
me now, I believe that it was in the 7-800 range.  One major stumbling block
we did notice was that the compile times were enormous, certainly several
times as long as a Sun-2, and about as slow as my antiquated Callan.  The
line from the rep was that at the time, Symmetrics was still shipping an
old compiler based on some silly theoreticians papers, and that the code
it was generating was being threaded this way and that, generating subroutine
calls here and there with all the nasty activation record stuff, etc etc etc.

I was rather interested, though, by his claim that Symmetrics was on the
verge of releasing their 32032 version, soon to be followed by higher-
performance NS parts.  Also, he mentioned that a new compiler was in the
works.  Perhaps its time to hit them for fresh literature again.

Although this is completely incidental to the architecture, I probably should
add that somebody at Symmetrics was eating their Wheaties when they arranged
for the 4.2 port...it looked quite complete and happy, and was the first
machine that I saw anyway that included a variant on Chris Torek's ^T hack.

-dave
-- 
David Hsu  (301)454-1433 || -8798	"It was Dave, not me...honest!" -eneevax
Communication & Signal Processing Lab / Engineering Computer Facility
The University of Maryland   -~-   College Park, MD 20742
ARPA:hsu@eneevax.umd.edu  UUCP:[seismo,allegra,rlgvax]!umcp-cs!eneevax!hsu

"Filmed on location in space"