Sender: nate AT cartsys DOT com
Message-ID: <36BA6FF0.935A8B5D@cartsys.com>
Date: Thu, 04 Feb 1999 20:13:36 -0800
From: Nate Eldredge <nate AT cartsys DOT com>
X-Mailer: Mozilla 4.08 [en] (X11; I; Linux 2.0.36 i586)
MIME-Version: 1.0
To: djgpp AT delorie DOT com
Subject: Re: Question about long long math on intel archs
References: <3 DOT 0 DOT 6 DOT 32 DOT 19990202153421 DOT 0091a590 AT pop DOT netaddress DOT com>
	 <3 DOT 0 DOT 6 DOT 32 DOT 19990202173547 DOT 00915bb0 AT pop DOT netaddress DOT com>
	 <3 DOT 0 DOT 6 DOT 32 DOT 19990202223415 DOT 0094bda0 AT pop DOT netaddress DOT com>
	 <3 DOT 0 DOT 6 DOT 32 DOT 19990204024816 DOT 008d9720 AT pop DOT netaddress DOT com> <3 DOT 0 DOT 6 DOT 32 DOT 19990204163838 DOT 008bb180 AT pop DOT netaddress DOT com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Reply-To: djgpp AT delorie DOT com

Paul Derbyshire wrote:
> 
> At 10:10 PM 2/4/99 +1100, you wrote:
> >Hello,
> >
> >  Here's what my decaying 486DX2/66 running the program in a win95 dos
> >box had to say....
> >
> >Did 1000000 long long multiplies in 2.03297 seconds.
> >Did 5000000 long long multiplies in 9.61538 seconds.
> >Did 10000000 long long multiplies in 17.9121 seconds.
> >Did 50000000 long long multiplies in 98.341 seconds.
> 
> Interesting. My computer has only about 2.5 times the megahertz but does it
> 5 times as fast. Must be an extra factor of 2 in there just because it's a
> pentium, an this is using vanilla GCC and not PGCC.

Quite likely.  Pentiums execute many instructions in fewer clocks, have
bigger CPU caches, and some paralellize.
-- 

Nate Eldredge
nate AT cartsys DOT com