Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!olivea!apple!portal!cup.portal.com!mslater
From: mslater@cup.portal.com (Michael Z Slater)
Newsgroups: comp.arch
Subject: Re: 68300 speed request
Message-ID: <39104@cup.portal.com>
Date: 10 Feb 91 18:48:09 GMT
References: <27B456AA.3A8D@wilbur.coyote.trw.com>
Organization: The Portal System (TM)
Lines: 17

>How fast is one of the new 68300-series embedded CPU cores on
>32 bit integer code?  At 16 MHz with a DRAM memory system, would
>it be closer to a 68000 or a 68020?

Moto says the 68300 parts perform at about 80% of a 68020 at the same clock
rate. It's not clear what assumptions are behind, this though. It does assume
zero-wait-state memory on the 68300 part; I'm not sure what the memory 
assumption is on the 68020.  It will also vary with the application, as
always. Programs will small loops will benefit from the 68020's small I-cache,
and the multibit shifts are slower on the 68300 parts, which might affect
some programs considerably.

My gut feel is that with comparable memory systems and the same clock rate,
a 68300 part would be closer to an 020 than to a 68000, but may fall short
of Moto's 80% estimate.

Michael Slater, Microprocessor Report   mslater@cup.portal.com