Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/3/84; site soph.UUCP Path: utzoo!linus!philabs!cmcl2!seismo!harvard!talcott!panda!enmasse!enms2!soph!dave From: dave@soph.UUCP (Dave Brownell) Newsgroups: net.arch Subject: Re: 68020 benchmarks (and normalization thereof)?? Message-ID: <156@soph.UUCP> Date: Thu, 23-May-85 19:26:15 EDT Article-I.D.: soph.156 Posted: Thu May 23 19:26:15 1985 Date-Received: Sat, 25-May-85 06:12:23 EDT References: <274@petfe.UUCP> <5598@utzoo.UUCP> Reply-To: dave@soph.UUCP (Dave Brownell) Organization: Enmasse Computer Corp., Acton, Mass. Lines: 21 In article freeman@max.UUCP (Jay Freeman) writes: > Thus if 16 MHz 68010's and 12 MHz 80286's happen to be the >best that's real this week, we should compare them, at design clock, with >components fast enough to support them. This also seem simplistic. If a given system (say the 80286) needs a memory system that's twice as fast and three times as expensive as another (say the 68010), I get no benefit from such a benchmark. Of course, if you can spend more money you can get better performance!!!!! What we really need is multi-dimensional benchmarks: compare the fastest available processors with two or three different standard speed memory arrays. This would give designers usable numbers on what kind of *system* performance (the only kind that matters) can be expected for a given cost or design complexity. -- Dave Brownell EnMasse Computer Corporation enmasse!dave@Harvard.ARPA {genrad,harvard}!enmasse!dave