Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!rutgers!mcnc!ecsgate!ecsvax!urjlew
From: urjlew@uncecs.edu (Rostyk Lewyckyj)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: Motorola vs Intel (who's faster?)
Summary: I want real data to compare
Message-ID: <1990Jul28.010116.11935@uncecs.edu>
Date: 28 Jul 90 01:01:16 GMT
References: <3384.AA3384@sosaria> <4303@azure10.UUCP> <20710@grebyn.com>
Distribution: comp
Organization: UNC Educational Computing Service
Lines: 32



  
So far in this (these) discussions everybody has studiously avoided
posting verifiable comparison program run times for Amigas vs. the
others. I want to see lots of data, verifiable !! data for differnt
programs. I propose that you all post detailed comparison runs.
Include the following:
  Machine, brand, model, specs
  System, compiler
  Program (with code if need be)
  Timing results.
I would like to see A2000, A2500, A3000 vs IBM PS/2 80 , and other
16, 20, and 25 mhz Intel 80386/80387 clones, on various floating point
intensive programs. 
 e.g. Linpack  300x300 , some fast fourier transforms without tabled
sines/cosines, eigenvalues of 50x50 matrices, exponential sums,
calculation with complex numbers.  
 ABSOFT or other Fortran for the Amigas, Lattice C, Manx C     VS
 Microsoft Fortran, Lahey Fortran, Microway Fortran for 386, MS-C,
 TURBO-C
Again! So far I have seen no postings of comparisons only meaningless
MIPS figures and statements that it depends on various things other
than clock speed. I want comparisons of the bottom line, results.
  
Please don't quote Dhrysotnes. They ar not even floating point.
However Specmarks would be welcome.
-----------------------------------------------
  Reply-To:  Rostyslaw Jarema Lewyckyj
             urjlew@ecsvax.UUCP ,  urjlew@unc.bitnet
       or    urjlew@uncvm1.acs.unc.edu    (ARPA,SURA,NSF etc. internet)
       tel.  (919)-962-6501