Path: utzoo!attcan!uunet!seismo!sundc!pitstop!sun!decwrl!labrea!rutgers!cs.utexas.edu!natinst!brian
From: brian@natinst.UUCP (Brian H. Powell)
Newsgroups: comp.sys.mac
Subject: Re: Mac IIx
Summary: sometimes 30% faster
Message-ID: <1236@natinst.UUCP>
Date: 8 Oct 88 16:59:49 GMT
References: <23479@tut.cis.ohio-state.edu> <9791@cup.portal.com>
Organization: National Instruments, Austin, TX
Lines: 16

In article <9791@cup.portal.com>, Michael_J_Ward@cup.portal.com writes:
> ...BYTE Magazine's tests indicated a performance/speed improvement of
> zero percent to 20 percent depending on the class of problem undertaken.

     If you just happen to have optimized your code for the 68030/68882, you
can manage a 30% improvement.  Your code will look awfully funny, because
you'll have to arrange it to get the maximum 882 and 030 instruction
concurrency.  (The 882 is much better at executing more than one instruction
at once than the 881.)  In general (i.e., if you take a plain old Mac II
program and run it on the IIx), we agree with the benchmarks we've seen,
usually around the 15% faster range.

Brian H. Powell					National Instruments Corp.
	brian@natinst.uucp			12109 Technology Blvd.
	cs.utexas.edu!natinst!brian		Austin, Texas 78727-6204
	AppleLink:D0351				(512) 250-9119 x832