Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 (Tek) 9/28/84 based on 9/17/84; site mako.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!harvard!talcott!panda!genrad!decvax!tektronix!orca!mako!jans
From: jans@mako.UUCP (Jan Steinman)
Newsgroups: net.arch
Subject: Re: Re: RISC (really on multiplication d
Message-ID: <869@mako.UUCP>
Date: Mon, 22-Jul-85 19:05:44 EDT
Article-I.D.: mako.869
Posted: Mon Jul 22 19:05:44 1985
Date-Received: Thu, 25-Jul-85 03:23:28 EDT
References: <149@mips.UUCP> <600005@pbear.UUCP> <493@mmintl.UUCP> <378@utastro.UUCP> <2397@ut-sally.UUCP>
Reply-To: jans@mako.UUCP (Jan Steinman)
Organization: Tektronix, Wilsonville OR
Lines: 20

In article <2397@ut-sally.UUCP> Jim Crandell writes, quotes:
>>>I would guess, conservatively, that well over 90% of the integer multiplies
>>>that are done generate a result with no more bits than the larger of the
						^^^^
>>>two operands.
>> 
>>I didn't know that over 90% of multiplication operations were multiplying
>>something by 1.  Wow.  You learn something new on the net every day!
>
>They're not.  Actually, 45% multiply by 1, and another 45%+ multiply by 0.
>
*** Sarcasm Mode Off ***

One would assume the original poster meant atomic units, such as machine
words, rather than (the obviously wrong) bits.  It should be stressed that
such determination is essentially a run-time operation, and the additional
overhead of overflow detection MUST then be done.
-- 
:::::: Jan Steinman		Box 1000, MS 61-161	(w)503/685-2843 ::::::
:::::: tektronix!tekecs!jans	Wilsonville, OR 97070	(h)503/657-7703 ::::::