Author: Albert Silver
Date: 19:32:12 06/05/00
Go up one level in this thread
On June 05, 2000 at 21:23:14, Gareth McCaughan wrote:
>On June 04, 2000 at 19:33:45, Albert Silver wrote:
>
>[I said:]
>>> By not special-casing positions like this, you adopt the general
>>> rule that they're no different from other positions with the same
>>> balance of material.
>>
>> They're not different. If you're a pawn up you should play it unless it's
>> already blockaded and clear (or the program has 9-piece tablebases and
>> declares it a draw). What difference would it make if without special-case
>> knowledge the program judges this at +1.00 or with special-case knowledge
>> it lowers this eval to +0.4? In both cases it will believe it has the
>> advantage and in both cases will refuse to draw and play it out, only
>> in the second case you'll be slowing the program by giving it useless
>> special knowledge.
>
>I don't understand this argument. You could say the same thing about
>anything in the evaluation. Suppose the program has a choice between
>two moves; one leads to this position, the other leads to some other
>one-pawn-up position that *is* winnable. The other position evaluates,
>let's say, at +0.9. Then it matters whether this one evaluates at
>+0.4 or +1.0, no?
>
Sure, but not here. There is nothing obvious about stating that a rook ending
with pawns on one wing and a one pawn advantage can't be won. There are far too
many exceptions in my opinion for this to work, so that this knowledge is
useless (as it is incomplete), and dangerous (as it may lead to incorrect
decisions). In order to validate this knowledge, I am of the opinion that one
would have to include far too many conditions for it to justify the slowdown it
would inevitably cause.
>>> That's probably wrong more often than, say,
>>> deducting half a pawn in positions "like this one" (whatever that
>>> might mean).
>>
>> Who are you quoting when you say that?
>
>No one. The quotes were just to express dissatisfaction with
>my own vague terminology.
Ah ok. I had no idea what was meant by "whatever that might mean" or "like this
one" (which seemed as if you were quoting someone).
Albert Silver
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.