Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site hscfvax.UUCP
Path: utzoo!watmath!clyde!cbosgd!cbdkc1!desoto!packard!ihnp1!ihnp4!mhuxn!mhuxr!ulysses!allegra!mit-eddie!genrad!panda!talcott!wjh12!hscfvax!pavlov
From: pavlov@hscfvax.UUCP (840033@G.Pavlov)
Newsgroups: net.lang.f77,net.unix
Subject: Re: Wanted: Ultra-fast fortran compiler for UNIX
Message-ID: <131@hscfvax.UUCP>
Date: Sun, 28-Jul-85 22:30:55 EDT
Article-I.D.: hscfvax.131
Posted: Sun Jul 28 22:30:55 1985
Date-Received: Wed, 31-Jul-85 02:13:51 EDT
References: <621@astrovax.UUCP> <9871@Glacier.ARPA>, <426@carina.noao.UUCP>
Organization: Health Sciences Computing Facility, Harvard University
Lines: 20
Xref: watmath net.lang.f77:341 net.unix:5176


 Fortran (unfortunately ?) is important to us; we've looked at Fortran execu-
 tion times closely on our Sys III derivative, compared to BSD 4.2, Tops 20,
 and VAX VMS. 
 
 The primary problems are most definitely in the math libraries. A loop such
 as
    DO .....

     j = j+k
     r = a*b
       ......

will stay within the 30% performance range mentioned previously. But insert
cos(), atan(), etc, and Unix Fortran slows to a crawl.

I/O isn't much better .................


  greg pavlov, FSTRF, Amherst, N.Y.