Author: Robert Pope
Date: 11:10:57 06/04/99
Go up one level in this thread
On June 03, 1999 at 18:47:25, Prakash Das wrote: snip >> >>Please explain how any programmer can benefit by using any architecture other >>than Intel, if this event standardizes on Intel. Why would anybody ever program >>for a Mac, an Alpha, a parallel machine, etc., if all you get for your trouble >>is disqualified at worst and a time handicap at best? > > Again, I don't think I ever said anything about this being only an Intel >architecture. The platform could be anything. The key part was "uniform", not >what particular architecture. > But you are saying that it _has_ to be uniform! That is the problem. The uniformity requirement is about exclusion, even if the differences don't give a competitive advantage. So if the Pentium 450 64MB were chosen as the uniform platform, a program written in PowerPC assembly couldn't participate. What does that poor programmer do? Port the assembly code to Intel? Start from scratch? No matter what you choose as a "uniform platform", you are excluding all the programs written for anything slightly different. And you are discouraging potential programs from being written on other platforms: Suppose I find that I can double my program's speed by taking advantage of AMD's 3DNow multimedia extensions? Not all of these exist on Intel machines. Do I just ignore the discovery, since the "uniform platform" is not a K6-II? Suppose I want to experiment with a program that uses a memory intensive search? Do I even bother trying, since the uniform platform is limited to 64MB (or whatever)? After all, I would be spending perhaps years along a line of research that would prevent my program from competing. But what if it would work? And what happens when "Superchip, Inc." releases a 2Ghz chip, effectively destroying the marketability of the chip in our uniform platform? Do we stay with the old architecture, or switch to the new one, leaving behind all the programs that became too optimized for the uniform platform. snip >>This reminds me of that Kurt Vonnegut short story where strong people have to go >>around wearing weights so they won't have an unfair advantage over weak people, >>and smart people have to be distracted constantly by loud noises so they won't >>have an unfair advantage over less smart people. >> > > Interesting story, but not of relevance. >Or should we rather have giant sumo wrestlers beating up ants and calling >themselves world champions? It's very relevant. You are implying the following: "Even though I am better at programming Alpha computers, and they are better designed to run my program, and my program would play better on an Alpha, if I want to compete in the World Championship, I must program on platform X instead, because platform X is the uniform platform that has been designated as the only platform allowed in the World Championship." So either switch to platform X and produce a worse product, or don't participate. That doesn't even get into the issue of Big Iron. It's red ants versus black ants, not ants versus sumo wrestlers. In my mind, world championship should mean: - anyone who can play can come (or attempt to qualify to come) - the one who wins the most is the winner - the winner appears to be the best, so deserves to be called World Champion. Rob
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.