Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
1998
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*November
*December
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 1998

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Mathematica 3 speed on a Sun Ultra1/200e ?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg14352] Re: Mathematica 3 speed on a Sun Ultra1/200e ?
  • From: Richard Easther <easther at het.brown.edu>
  • Date: Thu, 15 Oct 1998 00:28:55 -0400
  • Organization: High Energy Theory Group, Physics Dept, Brown University
  • References: <6vurf3$8l7@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

Jim Jennings wrote:
> 
> Is anybody out there running Mathematica 3 on a Sun Ultra1/200e ?
> 
> I have been running Mathematica on an old PowerMac 7100/80 and I am
> looking for a way to get more speed.  We have a Sun Ultra1/200e running
> Solaris 2.5.1.  I am considering buying Mathematica for it but I would
> like to get some idea of how much faster things will be.
 
[...]

> So it looks like the Sun will be only 2.3 times faster.  Can this be
> right?  Our Sun has 384MB RAM, will that make a significant difference?
> 
> It looks like I could do much better by just adding an upgrade card to
> my Mac...
> 
> Machine: Power Macintosh 7100/66, Newer MAXpowr G3 PDS, 88 MB, MacOS
> 7.5.5  Version = Power Macintosh 3.0 (October 5, 1996) Benchmark = 2.27
> 
> ... 5.4 times faster!
> 
> Can somebody explain to me what is going on here?  Are the benchmarks
> missleading?  Can someone running Mathematica 3 on a Sun Ultra1/200e
> with a memory configuration closer to mine tell me about their
> experience?  Could you run the benchmark for me & email me the result?

We have an Ultrasparc machine (quad processors, 250MHz(?) and 4Mb of
cache, plus 1Gb of RAM) here and it benchmarks at about 1.2 using the
same tests. 

I posted a similar query to yours to this group a little while ago, as I
was also puzzled by this apparently shoddy performance on a machine
which is intrinsically fairly fast. 

The difference is presumably in the amount of effort that Wolfram
expends optimizing their code for different platforms: and the Sparc is
clearly not a priority! (Assuming that the benchmarks are "fair" - does
anyone know of any alternatives?).

Richard


  • Prev by Date: Re: ListPlot (and many other questions about graphics)
  • Next by Date: Re: SOme problems with "Save"
  • Previous by thread: Re: Mathematica 3 speed on a Sun Ultra1/200e ?
  • Next by thread: Mathmatica Problem...