Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!zaphod.mps.ohio-state.edu!rpi!leah!bingvaxu!vu0310
From: vu0310@bingvaxu.cc.binghamton.edu (R. Kym Horsell)
Newsgroups: comp.arch
Subject: int x int -> long for * (was Re: F.P. vs. arbitrary-precision)
Summary: no
Keywords: arithmetic,arbitrary precision
Message-ID: <3984@bingvaxu.cc.binghamton.edu>
Date: 11 Sep 90 19:31:43 GMT
References: <3755@osc.COM> <4513@taux01.nsc.com> <119244@linus.mitre.org> <6837.26e7ee92@vax1.tcd.ie> <119612@linus.mitre.org> <3977@bingvaxu.cc.binghamton.edu> <119733@linus.mitre.org>
Reply-To: vu0310@bingvaxu.cc.binghamton.edu.cc.binghamton.edu (R. Kym Horsell)
Organization: SUNY Binghamton, NY
Lines: 52

In article <119733@linus.mitre.org> bs@gauss.UUCP (Robert D. Silverman) writes:
\\\
>I wish people would refrain from shooting their mouths off about
>multi-precision arithmetic until they've actually done a significant
>amount of it. Typically, only having 32 x 32 --> 32 slows down
>multi-precision arithmetic by about a factor of 10. Not having it
>can render a machine all but useless for performing certain tasks.
\\\

Well that ``factor of 10'', in my simple-minded way, sounded a bit
too far off the mark (of course it only _asymptote_ to less than 3 for sure).
So I ran my littul benchmark. The results for VAX 8750 and Sparc,
using a variety of compilers and optimization levels are as
follows:

Machine	Compiler	Optimization	benchmark 1	benchmark 2

VAX	
	cc		no		2.783		5.300
	cc		yes		2.566		5.133
	gcc		no		2.900		5.116
	gcc		yes		1.783		3.700

Sparc
	cc		no		1.833		3.733
	cc		yes		1.033		2.100

Admittedly there is a lost to be said for my lack of understanding
of experimental design, but I think we fail to see an
order of magnitude difference between benchmarks 1 & 2.

The difference? Benchmark 2 performed all calculations using
int x int -> int while benchmark 1 used both int x int -> int
and long x long -> long.

I reiterate -- since int x int -> long is not _logically
required_ then you must argue it's benefit from an _economic_ 
viewpoint rather than a convenience one. 

As the vast majority of computation (although an ill-defined 
quantity here) does not seem to suffer from its lack, and the 
admittedly special-purpose stuff here only runs a (binary) order of 
magnitude slower by using lower-precision, I can not see how
allocation of 1 barn to int x int -> long can be justified.

Of course, this situation may change -- technology has a habit
of throwing a monkey-wrench into things (e.g. the motor
industry and mating habbits).  

I await cogent arguments.

-Kym Horsell