Author: walter irvin
Date: 11:33:09 01/06/00
Go up one level in this thread
On January 06, 2000 at 13:22:41, Ricardo Gibert wrote: >On January 06, 2000 at 11:58:00, walter irvin wrote: > >>what does it take to defeat a GM in a 6 game match?????? what ply does the >>computer start pushing the GM back ????? to unify the guess it is nessisary to >>stick with 1 program .also to pick a program most people have , so i choose >>crafty .since every one is trying to figure out 40/2 hrs.lets figure what mhz is >>needed to reach the different ply at 40/2 hr . >> >>ply elo mhz available score vs GM >>1 1350 ? now 0-6 >>2 1400 ? now 0-6 >>3 1450 ? now 0-6 >>4 1500 ? now 0-6 >>5 1550 ? now 0-6 >>6 1600 ? now 0-6 >>7 1800 ? now 0-6 >>8 2000 ? now 0-6 >>9 2100 ? now 0-6 >>10 2200 ? now 0-6 >>11 2300 90 mhz + now 0-6 >>12 2400 133 mhz + now 1-5 >>13 2450 300 mhz + now 1-5 >>14 2500 450 mhz + now 2-4 >>15 2550 600 mhz + now 2-4 >>16 2600 2 ghz + now w/multi proc 3-3 >>17 2650 ? ?? 3-3 >>18 2700 ? ?? 4-2 >>19 2750 ? ?? 4-2 >>20 2800 ? ?? 5-1 >>21 2850 ? ?? 5-1 >>22 2900 ? ?? 6-0 >> >>this is my (guess) i think beyond 22 ply its over for man vs machine . > >I don't believe your chart is close to accurate at all. For instance, every >doubling of cpu speed is worth about 50-70 rating points depending on who you >ask and if you assume an EBF of 3 or 4, then... > >Besides that, extrapolating to the extrema is "extremely" dubious. i could be way off ?? i was just putting how hard the program was for me to beat in the earlier ply's after that i had to go on how the program did vs other programs . also i have noticed that the program jumps in elo at certain ply's . if anyone has any views as to what the elo's for the different ply's i'd like to see them to compare how close i am .
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.