Author: Robert Hyatt
Date: 17:49:42 10/12/04
Go up one level in this thread
On October 12, 2004 at 17:13:45, Graham Laight wrote: >On October 12, 2004 at 15:11:27, Robert Hyatt wrote: > >>On October 12, 2004 at 13:11:27, Graham Laight wrote: >> >>>On October 12, 2004 at 13:07:22, Omid David Tabibi wrote: >>> >>>>On October 12, 2004 at 10:28:16, Graham Laight wrote: >>>> >>>>>Hydra seems to be a god-like system in its ability to pick out the profoundly >>>>>brilliant move. Is it available to play against online anywhere? If ever there >>>>>was a justification for charging a fee for a game of chess, then Hydra has it in >>>>>spades! In competition after competition, it just keeps on punching in those >>>>>brilliant performances. >>>>> >>>>>We have also known for many years that Fritz is outstanding - Franz and >>>>>Mattheius (sp?) have long had a subliminally fast system - but for the past >>>>>several years, they've also had a strategic system, with quite outstanding >>>>>positioning skills. For them to equal Hydra's Bilbao score with only a 1.9 GHz >>>>>processor represents consolidation of their position at the top of the >>>>>programmers' tree. >>>>> >>>>>Then we have Junior - which flopped. What was it doing in the computer team? It >>>>>was the only computer to get a negative score against the humans - and was well >>>>>short of what its team-mates achieved (see >>>>>http://www.ajedrezbilbao.com/cResultadosEN.htm). >>>>> >>>>>How does one explain such a poor performance by Junior, which had massively >>>>>superior hardware to Fritz? Should we ask the Junior programmers to forward >>>>>their program to to Franz Morsch for advice and improvement? >>>>> >>>> >>>>Reading comments like yours, I come to the conclusion that an Introduction to >>>>Statistics course should be made mandatory as a prerequisite for joining CCC... >>> >>>I obtained a university qualification in statistics 20 years ago (I did an >>>ancillary in statistics as part of my computing degree). >>> >>>-g >> >>You probably need a refresher course. :) >> >>I just picked four pennies out of my pocket and flipped each of them 4 times. >>Know what I found? One penny that will _only_ produce heads. It produced four >>heads in a row, and that is _clearly_ enough samples to say how it will do for >>longer tests, right? :) > >If this applies to Bilbao, then it also applies to any competiton of any form >where the number of games is less than enormous. After what you've written here, >you'd be wholly unjustified in ever getting excited about any competiton ever >again in your life (even drag racing). > >-g At a typical race night, we might run 16 times. Each run is timed, and the driver reaction time is accurately measured and reported as well. So we can draw all sorts of conclusions. But chess is not drag racing. And you might notice that I _don't_ get that excited about single events... But the main point is that 3 draws and a loss against 4 GM players is _not_ a bad result. 3 wins and a draw would be better, but 3 draws at that level is _not_ a bad result, particularly when humans are so good at reaching draws when they want to... > >>3 draws and a loss against GM players is _not_ a bad result. If you think it >>is, you have a highly exaggerated opinion of how strong computers really are. >>Hint: "think lower". >> >>based on that at least 1/2 of the GM players that enter tournaments produce very >>"poor" results... >> >> >>> >>>>>-g
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.