Path: utzoo!utgpu!attcan!uunet!husc6!bloom-beacon!tut.cis.ohio-state.edu!rutgers!ucsd!nosc!helios.ee.lbl.gov!pasteur!ames!elroy!mahendo!wlbr!scgvaxd!trwrb!aero!venera.isi.edu!raveling
From: raveling@vaxb.isi.edu (Paul Raveling)
Newsgroups: comp.sys.hp
Subject: Re: HP 9000/500 vs. Vectra with 386 CPU and 387 co-processor?
Message-ID: <6019@venera.isi.edu>
Date: 3 Aug 88 15:12:06 GMT
References: <592@rocksanne.UUCP> <830016@bgphp1.UUCP>
Sender: news@venera.isi.edu
Reply-To: raveling@vaxb.isi.edu (Paul Raveling)
Organization: USC-Information Sciences Institute
Lines: 19


	[This was an attempted email reply, but it didn't make it
	through the mailers]


	Interesting results...  I'll cast a vote for posting the
	benchmarks (including the box filter benchmark -- I must
	have missed it the first time around).  If you don't post,
	I'd appreciate copies to try on assorted machines here.

	BTW, a multi-CPU approach seems to me to be the appropriate
	architecture for purportedly high performance workstations,
	but anything using a Unix kernel will be fighting OS architecture
	to utilize multiple processes (and processors) efficiently.


---------------------
Paul Raveling
Raveling@vaxb.isi.edu