Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!elroy.jpl.nasa.gov!ncar!boulder!ccncsu!longs.LANCE.ColoState.EDU!kk881595
From: kk881595@longs.LANCE.ColoState.EDU (kevin knappmiller)
Newsgroups: comp.sys.mips
Subject: f77 on DECstation 5000
Keywords: FORTRAN DEC5000
Message-ID: <11219@ccncsu.ColoState.EDU>
Date: 14 Nov 90 04:35:06 GMT
References: <1990Nov13.184853.20015@groucho>
Sender: news@ccncsu.ColoState.EDU
Reply-To: kk881595@longs.LANCE.ColoState.Edu
Distribution: na
Organization: Engineering College, Colorado State University
Lines: 17

Does anyone have any recommendations of a good FORTRAN compiler for a DEC 5000?
We currently have the f77 supplied by DEC, which is not `good'.

My main use is a 3-D finite difference model.  When run on a Silicon Graphics
Personal Iris with the same MIPS CPU running at 20MHz instead of 25MHz the code
still runs 30% faster than the best I can get from the compiler DEC
sells when running on the 5000.

Using a test 400x400 matrix multiply the DEC supplied C compiler generates code
which runs 3 times faster than the code f77 generates.
(They both have the optimum loop ordering for the respective languages)


Kevin Knappmiller
Department of Mechanical Engineering
Colorado State University
Fort Collins, CO  80523