Mathematica pricing gouging on Sun

Steve Christensen chrstnsn at uxe.cso.uiuc.edu
Fri Jun 23 07:00:49 AEST 1989


[In response to message from alfred at mcc.com:]

While I too do not like such a price for a piece of software considering
that the prices of workstations are dropping rapidly, I think that your
gouging remarks are a bit inaccurate.  I have been running Mathematica for
about two years, first as an alpha and beta tester and now as a developer
of a Mathematica package.  While it is true that the Mac price is much
lower than the Sun (or other workstation versions), the perfomance on a
Mac does not even come close to that on a Sun.  Most of the calculations I
and my collegues do will not run on even an 8 megabyte Mac II - the
programs bomb as the machine runs out of memory.  Further, every benchmark
I have runs from 4 to 20 times slower on a Mac relative to a similarly
priced SPARCstation 1.  So for similarly priced hardware, you get many
times the performance on a Sun over a Mac.  So I suggest that you divide
your price numbers (hardware plus Mathematica software) by the improved
performance for the higher priced version of Mathematica.  In my case,
since my programs will not run on a Mac II, the benefit I get from a lower
priced package is zero.

The front end feature will ultimately be solved for UNIX machine once
there is some sort of standard (probably X windows) for the buttons and
sliders.  I have found the notebooks to be fun and useful for many people
but are not of much use to me since I want to use my own windowing
interface more suited to my packages's needs.

New workstations software is almost always more expensive than similar PC
software.  As the SPARC chip machines start rolling out of Toshiba,
Solbourne, the Taiwan cloners and Sun itself, I suspect you will see
software prices go down.  While I dislike spending 1000's of dollars for
workstations software, I understand why many businesses must charge more
currently.  Once we all start buying more workstations and fewer
underperforming personal computers, we will help the situation.

Steve Christensen
NCSA, University of Illinois
steve at ncsa.uiuc.edu

[These are my opinions and not necessarily those of anyone else at NCSA.]



More information about the Comp.sys.sun mailing list