Author: Reinhard Scharnagl
Date: 00:24:03 01/03/04
Go up one level in this thread
HI Ed,
On January 02, 2004 at 21:18:38, Ed Trice wrote:
>Something is a little strange with the SMIRF values on that page.
please explain, why. Intutivly for me the possibilities of a queen in a 100x100
board seems to be grown where the pawns will have only a very small profit.
>The values of the pieces INCREASED moving from the 8x8 board to the 10x8 board,
>and they should DECREASE.
Corresponding to the value model descibed in Smirf, the pieces will profit from
the 10x8 board in getting more moves and possible threatening directions. And
because pawns gain only little from that, the piece values of the officers
consequently increase.
>For that reason alone, I think the model must be at least slightly flawed.
>
>Do you agree?
No. The values are defined relativly towards the pawn values. So the values from
8x8 board to the 10x8 board can not be compared directly. But looking to the
Smirf method itself, it really seems that they are slighly growing.
May be my method of calculating piece values is simpler than Taylors "safe
check" method, but I think the Smirf method produces more realistic values
especially for traditional chess and its pieces:
8x10 8x8
Piece SMIRF Vortex SMIRF Taylor
--------------------------------------------
Pawn 1.00 1.00 1.000 1.00
Knight 3.06 2.50 (!) 3.000 2.50 (?)
Bishop 3.60 3.00 (!) 3.375 3.03
King 3.72 ---- 3.750 ----
Rook 5.43 4.75 (!) 5.000 5.67 (?)
Archbishop 6.65 6.50 6.375 5.53
Chancellor 8.49 8.25 8.000 8.17
Queen 9.03 8.75 8.375 8.70 (?)
Additionally I have to state that a detail evaluation within the 10x8 board has
25% more importance than in an 8x8 board. Therefore the positional aspect in
10x8 chess will push back a little bit the static piece evaluation.
In fact stating 10x8 piece values today is somehow speculative, because there is
no sufficient gathering of experiences during centuries. But today there are
methods to verify such approaches by computer simulations or competitions
between different value implementations.
Such a competition then also would be a competition between the cited evaluation
models, what could be very interesting.
Regards, Reinhard.
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.