From: Martin Stromberg <Martin DOT Stromberg AT lu DOT erisoft DOT se>
Message-Id: <200003141609.RAA19975@mars.lu.erisoft.se>
Subject: Re: Fastest integer type
To: djgpp-workers AT delorie DOT com
Date: Tue, 14 Mar 2000 17:09:34 +0100 (MET)
In-Reply-To: <38CE53D6.208D75E2@hotmail.com> from "Jesus Gil y Gil" at Mar 14, 2000 03:59:35 PM
X-Mailer: ELM [version 2.5 PL3]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
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

> You should run the benchmark that I posted 3 days ago to
> djgpp AT delorie DOT com, it will carify your dudes by yourself.
> The speed depends of the processor:
> - In 486 the fastest are 16 and 32 bits integers, chars are only a few
> slower, floats are more slow than integers, and double slowest.
> - In Pentium 8 bit integers are the fastest, 16 bits are faster and 32
> bits even more faster, floats and doubles the same speed, a bit slower
> that 16 bits integers.
> - In AMD K6 all types, 8,16,32 bits integer and float are equally faster.
> - In Pentium II and III, 8 bits are two times more slow than all the other
> types, 16 bits are much more faster, and 32 bits the fastest again, floats
> are equally fast than 32 bits integers, and doubles practically the same.

So if I use ints for all but FASTEST64_INT, we have the fastest, it seems.


Right,

							MartinS