Author: John Merlino
Date: 08:40:18 01/03/02
Go up one level in this thread
On January 03, 2002 at 04:41:30, Uri Blass wrote: >On January 03, 2002 at 03:44:24, William Sprague wrote: > >>On January 02, 2002 at 19:51:17, John Merlino wrote: >> >>>On January 02, 2002 at 18:29:56, Roy Eassa wrote: >>> >>>>How long does it take your program/engine to find White's winning move (and to >>>>achieve a score that shows the program understands that White is winning)? >>>> >>>>[D] r4b1k/pp3Ppp/n1p5/2q1B3/2r5/6RP/PP2Q1P1/5R1K w - - 0 1 >>> >> >> >>John - I found it on CM-8000 using a P4 @ 1.3 gig >> >>Time Depth Score Positions Moves >>0:00 1/3 -2.31 2167 1. Rd3 Rc2 2. Qe1 Nb4 >>0:00 1/3 -2.14 2753 1. Re3 Rc2 2. Qd3 Nb4 >>0:00 1/4 -2.07 15833 1. Re3 Rc2 2. Qd3 Nb4 3. Qb3 >>0:00 1/4 -1.99 32362 1. Rb3 Rc2 2. Qe4 b6 3. Rd3 >>0:00 1/4 -1.78 38779 1. Rg5 Rc2 2. Qd3 Rxb2 3. Bxg7+ >> Bxg7 4. Rxc5 Nxc5 >>0:00 1/5 -1.78 63783 1. Rg5 Rc2 2. Qd3 Rxb2 3. Bxg7+ >> Bxg7 4. Rxc5 Nxc5 >>0:01 1/5 -0.56 143809 1. Re3 Rb8 2. Bxb8 Nxb8 3. Re8 Nd7 >>0:01 1/6 -0.65 189549 1. Re3 Rb8 2. Bxb8 Nxb8 3. Rb3 >> Rc2 4. Qe6 >>0:02 1/6 -0.60 333708 1. Rg5 Rc2 2. Qd1 Rf2 3. Bxg7+ >> Bxg7 4. Rxc5 Rxf1+ 5. Qxf1 Nxc5 >>0:05 1/7 -0.37 600219 1. Rg5 Rc1 2. Bxg7+ Bxg7 3. Rxc5 >> Rxf1+ 4. Qxf1 Nxc5 5. Qc4 b6 >>0:07 1/7 0.00 869620 1. Re3 Rb8 2. b3 Rb4 3. Bxb8 Nxb8 >> 4. a3 Rd4 5. Re8 >>0:14 1/8 0.42 1845104 1. Re3 Rb8 2. Rf5 Rc2 3. Bxg7+ >> Bxg7 4. Rxc5 Rxc5 5. Re8+ Bf8 6. >> Qg4 Rc1+ 7. Kh2 >>0:31 1/9 0.30 4295215 1. Re3 Rc1 2. Bxg7+ Bxg7 3. Re8+ >> Bf8 4. Rxa8 Rxf1+ 5. Qxf1 Kg7 6. >> Qf4 Nb4 7. Rb8 Qe7 8. a3 Nd5 >>1:38 1/9 2.36 13147444 1. b4 Rxb4 2. Rg5 h6 3. Qh5 Qxe5 >> 4. Rxe5 Rd8 5. Re8 Rbd4 6. Qe2 >> Nc5 7. Rxd8 Rxd8 >> >>Bill > >It seems that John marlino always insist to use the the default personality that >is not good for solving test suites. > >I believe based on experience that bigger selectivity is better for chessmaster >in most of the cases when tactics is involved and this is the reason that I >always use ss=10 in chessmaster6000(I have not chessmaster8000 and I cannot >increase the selectivity for more than 10 in chessmaster6000). > >Uri You are correct; I do always use the default personality (with the exception of using a 32MB hash table). This is easier for me, and is also easier for other CM users to duplicate (or test their processor/favorite personality against). If I used a custom personality, I would feel obliged to also post that personality's settings (as I wish other people would do when they post their CM findings), and I am just too lazy to do that every time. ;-) jm
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.