Path: utzoo!attcan!uunet!husc6!rutgers!iuvax!pur-ee!a.cs.uiuc.edu!p.cs.uiuc.edu!gillies
From: gillies@p.cs.uiuc.edu
Newsgroups: comp.arch
Subject: Re: Standard Un*x H/W architecture
Message-ID: <76700037@p.cs.uiuc.edu>
Date: 18 Jul 88 18:05:00 GMT
References: <980@garth.UUCP>
Lines: 7
Nf-ID: #R:garth.UUCP:980:p.cs.uiuc.edu:76700037:000:376
Nf-From: p.cs.uiuc.edu!gillies    Jul 18 13:05:00 1988


I once heard an expert on floating-point arithmetic state that CDC's
1's complement arithmetic was used JUST BECAUSE IT RUNS FASTER.  In
fact, the engineer estimated their 1's complement could always be
implemented to run 10% faster than IEEE arithmetic.  Since
speed-at-all-costs is the main selling-point of supercomputers, CDC
was very reluctant to abandon this advantage.