Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!elroy.jpl.nasa.gov!decwrl!pa.dec.com!rwood@pa.dec.com
From: rwood@pa.dec.com (Richard Wood)
Newsgroups: comp.sys.hp
Subject: Re: HP 9000 s 700 questions ????
Message-ID: <1991Mar28.131436@wsl.dec.com>
Date: 28 Mar 91 21:14:36 GMT
References: <1991Mar13.121715.25269@qut.edu.au> <31480011@hpcuhe.cup.hp.com> <1991Mar16.014605.11367@mcs.kent.edu> <7401@titcce.cc.titech.ac.jp>,<1991Mar19.221804.22673@riacs.edu> <1991Mar20.210940.27785@ni.umd.edu>
Sender: news@pa.dec.com (News)
Reply-To: rwood@pa.dec.com
Organization: DEC, Palo Alto Worksystems
Lines: 27

From: mike@UC780.UMD.EDU (Mike Santangelo):
Subject: Re: HP 9000 s 700 questions ????
>
> Model#    Mhz   MIPS  MFLOPS    SPECmarks  Enclosure-type  Purpose
> 9000/720  50     57     17         55      desktop         workstation


From: irf@kuling.UUCP (Bo Thide')
Subject: HP9000/720 vs. Sun SS2. SSBA Benchmarks (long).
> HP9000/720 (Cobra):
> -----------------
> Identification : HP9000/720 HP-PA 89 HP-UX 8.01 50 MHz 64 MB 2*200 MB
> linpackrs(single precision,with optimization,rolled):8803 Kflops
> linpackus(single precision,with optimization,unroll):13733 Kflops
> linpackrd(double precision,with optimization,rolled):8803 Kflops
> linpackud(double precision,with optimization,unroll):13733 Kflops


According to the press release and Mike Santangelo's post, the 50MHz
Cobra clocks at 17 MFLOPS.  However, the benchmark ran on a 50MHz
Cobra at the AFUU (the SSBA series) shows the machine at 13.7MFLOPS.

Quite a difference - so what's the explanation?

-- ---------------------------------------------------------------------
Richard Wood     Corporate Worksystems Team      Digital Equipment Corp.
========================================================================