Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!apple!agate!ucbvax!hplabs!hp-sdd!megatek!spot!mark
From: mark@spot.megatek.uucp (mark thompson)
Newsgroups: comp.arch
Subject: More SPARC floating point
Message-ID: <626@megatek.UUCP>
Date: 14 Jul 89 17:51:42 GMT
Sender: news@megatek.UUCP
Reply-To: mark@megatek.UUCP ()
Distribution: na
Organization: Megatek Corporation, San Diego, Ca.
Lines: 14

Well, I'm still trying to learn about floating point speeds on SPARC. My
new question is this: Sun claims that the 4/330 is a "2.6 MFLOP machine."

I wonder just where that number came from... it doesn't look like that can
be the cannonical double precision linpack result (given the FPC and 8847
hardware). 

Could somebody clarify how the 2.6 number was arrived at?

-mark
-- 
mark thompson						uunet!megatek!mark
  <Opinions expressed herein are not necessarilly those of Megatek Corp>
--