Author: Robert Hyatt
Date: 18:14:43 03/09/02
Go up one level in this thread
On March 09, 2002 at 11:05:45, Jon Dart wrote: >Phalanx seems to have an amazing talent for finding "quiet" but deadly moves. >Here is one example: > >[D] rn3rk1/pbp2p1p/1p2pp2/8/3PP3/q1P1QN2/P1P3PP/1K1R1B1R w - - > >Phalanx played Rd3 and the game continued c5 Nd2 Ba6 1-0 (White will play c4 and >Black will have to sac the Queen to avoid the deadly Qh6). > >There's probably a better defense but Rd3 is good. Arasan will play Rd3 after >about 2 minutes on an Athlon 1.3Ghz but doesn't give it a high + score. > >I couldn't get Crafty 18.10 to play it all. Crafty "hangs" for a long time >resolving the move d5. Here is the search output: > > clearing hash tables > time surplus 0.00 time limit 20:00 (20:00) > nss depth time score variation (1) > 7 0.97 0.00 1. Nd2 Qd6 2. e5 fxe5 3. Qg5+ Kh8 4. > Qf6+ Kg8 5. Qg5+ > 7-> 1.00 0.00 1. Nd2 Qd6 2. e5 fxe5 3. Qg5+ Kh8 4. > Qf6+ Kg8 5. Qg5+ > 8 1.75 0.00 1. Nd2 Kh8 2. Qf4 Qxc3 3. Qxf6+ Kg8 > 4. Qg5+ <HT> > 8-> 1.89 0.00 1. Nd2 Kh8 2. Qf4 Qxc3 3. Qxf6+ Kg8 > 4. Qg5+ <HT> > 9 3.18 0.00 1. Nd2 Kh8 2. Qf4 Qxc3 3. Qxf6+ Kg8 > 4. Qg5+ Kh8 5. Qf6+ > 9-> 3.64 0.00 1. Nd2 Kh8 2. Qf4 Qxc3 3. Qxf6+ Kg8 > 4. Qg5+ Kh8 5. Qf6+ > 10 8.14 0.01 1. Nd2 Kh8 2. Bc4 Nd7 3. Qg3 Rac8 4. > Rhf1 c5 5. d5 Rce8 > 10-> 22.20 0.01 1. Nd2 Kh8 2. Bc4 Nd7 3. Qg3 Rac8 4. > Rhf1 c5 5. d5 Rce8 > 11 34.17 0.00 1. Nd2 Kh8 2. Qf4 Qxc3 3. Qxf6+ Kg8 > 4. Qg5+ Kh8 5. Qf6+ > 11-> 45.47 0.00 1. Nd2 Kh8 2. Qf4 Qxc3 3. Qxf6+ Kg8 > 4. Qg5+ Kh8 5. Qf6+ > 12 1:09 0.00 1. Nd2 Kh8 2. Qf4 Qxc3 3. Qxf6+ Kg8 > 4. Qg5+ Kh8 5. Qf6+ > 12 10:01 0.06 1. d5 Qe7 2. Nd4 Kh8 3. dxe6 fxe6 4. > Bc4 Re8 5. Qg3 Bc8 6. Nb5 Na6 7. Rhf1 > 12-> 10:36 0.06 1. d5 Qe7 2. Nd4 Kh8 3. dxe6 fxe6 4. > Bc4 Re8 5. Qg3 Bc8 6. Nb5 Na6 7. Rhf1 > >The weird thing is, if I move Rd3 and then have Crafty search the resulting >position from the Black side, I get over +1 for White right away: Don't forget that Crafty is very asymmetric. You changed sides which changes a lot... > > clearing hash tables > time surplus 0.00 time limit 20:00 (20:00) > nss depth time score variation (1) > 6 0.43 1.18 1. ... Qa4 2. Qf4 c5 3. Rd2 Qc6 4. > Qxf6 <HT> > 6 0.69 1.05 1. ... Ba6 2. c4 Qe7 3. Rb3 c5 4. Qf4 > cxd4 5. Nxd4 > 6-> 0.72 1.05 1. ... Ba6 2. c4 Qe7 3. Rb3 c5 4. Qf4 > cxd4 5. Nxd4 > 7 0.98 1.36 1. ... Ba6 2. Rd1 Nc6 3. Nd2 b5 4. > Qg3+ Kh8 5. Qf3 Kg8 > 7 1.33 1.14 1. ... Qd6 2. Nd2 Ba6 3. c4 c5 4. d5 > Nd7 5. dxe6 Qxe6 > 7-> 1.96 1.14 1. ... Qd6 2. Nd2 Ba6 3. c4 c5 4. d5 > Nd7 5. dxe6 Qxe6 > 8 2.11 1.31 1. ... Qd6 2. Nd2 Ba6 3. c4 Nc6 4. > Qf2 Qb4+ 5. Rb3 Qd6 > 8-> 6.54 1.31 1. ... Qd6 2. Nd2 Ba6 3. c4 Nc6 4. > Qf2 Qb4+ 5. Rb3 Qd6 > 9 7.22 1.37 1. ... Qd6 2. Nd2 Ba6 3. c4 Nd7 4. > Qh6 Kh8 5. Be2 Rad8 > 9-> 15.49 1.37 1. ... Qd6 2. Nd2 Ba6 3. c4 Nd7 4. > Qh6 Kh8 5. Be2 Rad8 > 10 17.50 -- 1. ... Qd6 > 10 25.18 2.41 1. ... Qd6 2. e5 Qd5 3. exf6 Qb5+ 4. > Kc1 Qf5 5. Qh6 Qg6 6. Qh4 Qe4 > 10 25.44 ++ 1. ... Qe7!! > 10 33.78 1.62 1. ... Qe7 2. Nd2 Kh8 3. Qh6 Rg8 4. > Rh3 Rg7 5. Rh4 c5 6. dxc5 bxc5 7. Nc4 > 10 42.62 1.35 1. ... Ba6 2. Rd1 Nd7 3. Nh4 Rac8 4. > Qg3+ Kh8 5. Qh3 Bb7 6. Bb5 Rcd8 7. > Qf3 > 10-> 59.27 1.35 1. ... Ba6 2. Rd1 Nd7 3. Nh4 Rac8 4. > Qg3+ Kh8 5. Qh3 Bb7 6. Bb5 Rcd8 7. > Qf3 > 11 1:13 1.47 1. ... Ba6 2. c4 Qa4 3. Ra3 Qb4+ 4. > Rb3 Qd6 5. e5 Qe7 6. Qf4 fxe5 7. Nxe5 > Nd7 > 11-> 2:07 1.47 1. ... Ba6 2. c4 Qa4 3. Ra3 Qb4+ 4. > Rb3 Qd6 5. e5 Qe7 6. Qf4 fxe5 7. Nxe5 > Nd7 > 12 2:32 1.40 1. ... Ba6 2. c4 Qa4 3. Ra3 Qb4+ 4. > Rb3 Qe7 5. Nd2 e5 6. Qh6 Kh8 <HT> > >Note Crafty prefers Ba6 as Black's response, not c5 as in the game. > >--Jon
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.