MathGroup Archive 2000

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

Search the Archive

Re: Why relatively slow on Apple G4?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg25490] Re: [mg25479] Why relatively slow on Apple G4?
  • From: Andrzej Kozlowski <andrzej at tuins.ac.jp>
  • Date: Thu, 5 Oct 2000 23:50:11 -0400 (EDT)
  • Sender: owner-wri-mathgroup at wolfram.com

Most of the speed of a G4 comes form Motorola's AltiVec technology and is
particulary noticable when using floating point operations. However,
software has to be written in a way that takes advantage of these features
(a good example is the latest Photoshop which is considerably faster on a
500 MHZ G4 than on a 1000 MHZ Pentium III. The same is true for several
encryption programs.)  and Mathematica 4 does not.

Andrzej

on 10/3/00 11:26 AM, Tom Burton at tburton at cts.com wrote:

> I'm a long-time but not-entirely-satisfied PC user. For several years, I
> glanced occasionally at "Karl's Mathematica Page", currently
> 
> http://fampm201.tu-graz.ac.at/karl/mma.html
> 
> and noted that the Power Macintosh was nearly as fast as or faster than the
> fastest PC. But the Mac OS kept me away.
> 
> With the public beta of Mac OSX looking good so far, I am fixing to switch
> from PC to Mac. But now I see on Karl's page and others, accessible from
> 
> http://smc.vnet.net/mathbench.html
> 
> that the G4/500, the fastest available Power Mac, is way down the list.
> Indeed, it appears that the Mac is slower than the 1000MHz PC by approximately
> the ratio of the nominal clock speeds (1000/450). In other words, the
> advantage of Mac/Motorola's RISC architecture seems to have disappeared.
> 
> What happened? Or is Mac's past speed superiority a figment of my imagination?
> Tom
> 



  • Prev by Date: Re: making a function linear
  • Next by Date: Re: Reading a file
  • Previous by thread: Why relatively slow on Apple G4?
  • Next by thread: making a function linear