Date: Fri, 17 Mar 2000 07:25:49 +0500
Message-Id: <200003170225.HAA00921@midpec.com>
From: Prashant TR <prashant_tr AT yahoo DOT com>
To: djgpp-workers AT delorie DOT com
In-reply-to: <14545.19777.671368.908549@mercury.st.hmc.edu> (message from Nate
	Eldredge on Thu, 16 Mar 2000 13:08:17 -0800 (PST))
Subject: Re: Fastest integer type
References: <Pine DOT SUN DOT 3 DOT 91 DOT 1000316115430 DOT 3117J-100000 AT is>
	<200003161405 DOT TAA00904 AT midpec DOT com> <14545 DOT 19777 DOT 671368 DOT 908549 AT mercury DOT st DOT hmc DOT edu>
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: dj-admin AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

Nate Eldredge <neldredge AT hmc DOT edu> writes:

> Any idea what the benchmark it uses is like?  For instance, if it does
> a lot of operations on a whole block of numbers, it may be important
> that more 8-bit ints fit into the cache than 32-bits, even if they
> aren't inherently faster.

It seems that a set of numbers is generated and the XOR and ADD functions
are performed on those set of numbers continuously.