Author: David Rasmussen
Date: 14:15:15 05/17/01
Go up one level in this thread
On May 17, 2001 at 16:01:05, Jesper Antonsson wrote: >On May 16, 2001 at 18:12:04, David Rasmussen wrote: > >>On May 16, 2001 at 18:00:22, Miguel A. Ballicora wrote: >> >>> >>>Excuse me, I am not a expert on this, but I am very curious about >>>a phrase that I have seen in all this discussion. "Chess is O(1)" >>>Is not the notation O() applied to algorithms? > >Yes, I should have said "(good) chess algorithms", but I was lazy, >obviously. :-) > >>Yes. >> >>>Well, chess is not an algorithm, it is game. So, should not we talk about >>>that alpha/beta is O(whatever), Minimax is O(whatever) etc.? >> >>Yes. >> >>That is exactly the problem. Some people thinks that chess together with >>alpha-beta is an algorithm in itself. But it is not. The extra stuff that is >>added to alpha-beta for it to play chess, is just stuff to describe the input >>to alpha-beta, the gametree. > >Nonsense. A chess program runs on a computer and terminates (in theory). Thus >the chess program is an algorithm. Is that your definition of an algorithm? It is not mine. If your "argument" stands on that definition, you can pretty much say anything you want without taking any responibility for it, because as everybody in computer science knows, there is (regrettably) no formal definition of what an algorithm is, that is usable in all the cases where we want to talk about this concept that we call algorithms. Some define an algorithm as something that can be expressed by a Turing Machine. Others do not. As this is purely a matter of taste, and a subject for philosophy, I can't really criticize what your definition. I can just mention that I find it a silly definition of an algorithm, and that a chess program as a whole is not an algorithm in my book, except in a very sick and deranged way.
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.