Author: Dann Corbit
Date: 12:29:44 05/30/00
Go up one level in this thread
On May 30, 2000 at 14:22:55, blass uri wrote:
>On May 30, 2000 at 13:49:25, Dann Corbit wrote:
>
>>Go right ahead.
>>
>>Every chess opening might be refuted one day. Every refutation might have
>>another refutation answer discovered. Maybe the optimal opening is 1. f4 (for
>>all we really know).
>>
>>There is no such thing as a proof of correctness unless it leads to irrefutable
>>checkmate. You won't be able to accomplish this for the Halloween attack.
>>Therefore, it is only one of the quintillions of possibly viable openings.
>
>I can be practically sure about some things without a proof.
>
>I am sure that 1.e4 Nf6 2.Qh5 is a wrong sacrifice and that black is winning
>inspite of the fact that you cannot prove a forced mate.
>
>I am sure about it more than I am sure about long proofs in mathematics because
>they may be wrong because of a mistake in the proof.
True enough. That's why they publish books like ECO. But what was once thought
sound many years ago sometimes becomes refuted. I was not talking about +500
centipawn or more clear evaluations [but computers and even GM's can surely be
wrong on these also]. I think the more polar the score, the more *probable* the
outcome can be determined. But there are no lead pipe cinches. I think
positions like the above are quickly abandoned {probably some of these are
wrong!} and so we don't even bother asking about them.
But for some gambit opening where the deep computer eval shows -150 or worse or
for some positional sacrifices, I think these are very unclear (but by no means
certain in any direction).
I also think that the vast majority of sound openings remain undiscovered.
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.