ONJava.com -- The Independent Source for Enterprise Java
oreilly.comSafari Books Online.Conferences.

advertisement

AddThis Social Bookmark Button
Article:
  Confessions of the World's Largest Switcher
Subject:   Srinidhi and the G5s
Date:   2003-11-12 18:12:36
From:   anonymous2
Response to: Srinidhi and the G5s

CCNA student for $ / Apple fan for everything else ( Newton - iPod - Ti PB - 1.42G DP PM ) --


I so wanted to be on staff/a student at VT when I saw pictures on VT's tech news site. You are so fortunate to work on this project.


Those who bash Dr. Varadarajan's G5 choice, lab - n/w design etc. do not understand his comments on this subject at the 2003 O'Reilly Mac OS X Conference ( several of my gurus attended one is reportedly accused of asking who received the frequent flyer miles attached to this on-line purchase -- a query which brought down the house ). ... " a lot of people get the math wrong when calculating the performance of the machines [ G5s ]. Each G5 processor has two, double-precision, floating-point units. Each is capable of a fused, multliple-add operation per cycle, so you get 2 flops per cycle. This means that 2GHz corresponds to 8 GFlops, so each dual G% can deliver a peak of 16 GFlops of double-precision performance. That is more than a modern
Cray node."


<http://www.macdevcenter.com/pub/a/mac/2003/10/29/osxcon_g5cluster.html>


Also, <http://www.macdevcenter.com/lpt/a/4319>


I am biased, but only to quality -- quality code and engineering. Call it a 'rose' as opposed to a PC or Mac, and for me if the computing device were to meet this standard for networking thru and thru, I'm a customer. I favor engineering in all products when price point is realistic because it has bought me, for example, a non-hacked network for nearly 4 years -- well worth the price differential Mac to PC.


Mac Fan in Los Angeles, CA