Message-Id:	<3.0.6.32.19990205010153.00907790@pop.netaddress.com>
X-Sender:	pderbysh AT pop DOT netaddress DOT com
X-Mailer:	QUALCOMM Windows Eudora Light Version 3.0.6 (32)
Date:	Fri, 05 Feb 1999 01:01:53 -0500
To:	djgpp AT delorie DOT com
From:	Paul Derbyshire <pderbysh AT usa DOT net>
Subject:	Re: Question about long long math on intel archs
In-Reply-To:	<36BA52DD.98F5C749@mpx.com.au>
References:	<010501be5064$215c4780$1e2d87ca AT default>
Mime-Version:	1.0
Reply-To:	djgpp AT delorie DOT com

At 01:09 PM 2/5/99 +1100, you wrote:
>1000000 long-long's/long's    = 1.9375
>5000000 long-long's/long's    = 1.9
>10000000 long-long's / long's = 1.925
>50000000 long-long's / long's = 1.918
>
>I would have expected the 64-bit operations to take twice as long, but 
>in fact they were a bit quicker than that.  There seems to be no
>increased
>overhead when multiplying long long int's.

They DO take about twice as long. The nice thing is the multiplies don't
take FOUR times as long.
-- 
   .*.  "Clouds are not spheres, mountains are not cones, coastlines are not
-()  <  circles, and bark is not smooth, nor does lightning travel in a
   `*'  straight line."    -------------------------------------------------
        -- B. Mandelbrot  |http://surf.to/pgd.net
_____________________ ____|________     Paul Derbyshire     pderbysh AT usa DOT net
Programmer & Humanist|ICQ: 10423848|

- Raw text -