Author: Ernst A. Heinz
Date: 10:30:07 10/26/98
Go up one level in this thread
On October 26, 1998 at 12:43:26, Bruce Moreland wrote: > > [...] > >One of the things you hear about if you talk to old-timers in this field is >that >the square of the pawn is a totally important evaluation term. I'm not sure >who >said it, but someone said that it's worth 100 Elo points. This was a long time >ago, back when programs would get killed in endings routinely, but I get the >idea that this wisdom hasn't been re-evaluated much since then. It is my >opinion that square of the pawn is worth almost nothing in practical play, >since >the number of K+P endings that actually occur, with play still in them, seems >to >be approximately zero. I am an avid proponent of including a sophisticated yet still static Pawn-race analysis in the evaluation (I deem the "square of the Pawn" rule to be too simplistic). But I cannot give you any quantitative figures about how much it is actually worth in practical playing strength if you do so. >8/6pp/4pk2/4p3/2PqP3/3Q2P1/4KP2/8 w - - 18 62 > >White to move and lose. My program played 62. Qxd4, with a score of about +1, >although the move apparently loses by force. I would be interested to know if >other programs avoid 62. Qxd4. If white can safely avoid this move it should >be >better, I think. "DarkThought" with sophisticated Pawn-race scoring as mentioned above and running on a 600MHz Alpha-21164a fails low on 62. Qxd4 in iteration #11 after roughly 2 sec. Thereafter, it wavers between 62. Qc2 and 62. Qf3+ which it both scores as slightly positive for White. >8/6pp/4pk2/8/2PpP3/6P1/4KP2/8 w - - 0 63 > >White to move. Here is the positions with the queens off, if this helps >anyone. > I think white is lost already here. If your FEN is correct, "DarkThought" disagrees. It suggests to play 53. e5+ and scores the position as roughly equal. PV = 63. e5+ Kxe5 64. Kd3 Kd6 65. Kxd4 e5+ 66. Ke4 Kc5 67. Kxe5 Kxc4 ... >8/6p1/5k2/2P4p/3pPP2/8/4K3/8 w - - 0 66 > >White to move is completely lost. Mine figures this out after a while, but it >takes too long, in my opinion (over a minute on a P6/200 to really start to >crater). I'd be interested in knowing if anyone knows they are dead quickly. Yes, now White is completely lost. "DarkThought" scores the position as -1.4 after a couple of seconds and fails below -6 after half a minute. Thanks for posting this interesting position. =Ernst=
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.