Author: Pete R.
Date: 18:06:45 07/13/00
Go up one level in this thread
On July 13, 2000 at 19:47:39, Dann Corbit wrote: >This is run on a *very* slow machine (gets about 50K NPS on average), but it >runs 24x7 at ten hours each position. I am running a collection of interesting >positions, and found one where the suggested solution seems to be bettered by an >alternative choice. I would be interested to see if other engines find the >solution sooner. > >[D] r1n1kb1r/1b3pp1/1qn1p3/p2pP1Bp/Qp1P1N1P/5NPB/PP3P2/2R1R1K1 w kq - bm Nxe6; >id "Mike's computer test 2: 0032"; > >PFGA: EPD record: 32 ID: Mike's computer test 2: 0032 > clearing hash tables > time surplus 0.00 time limit 600:00 (600:00) > nss depth time score variation (1) > 7 3.00 0.75 1. Re2 Qc7 2. Rec2 Na7 3. Bg2 Rc8 4. > Nd2 > 7-> 4.98 0.75 1. Re2 Qc7 2. Rec2 Na7 3. Bg2 Rc8 4. > Nd2 > 8 6.99 0.80 1. Re2 Qc7 2. Rec2 Na7 3. Nxe6 fxe6 > 4. Bxe6 b3 5. Qxb3 > 8-> 21.45 0.80 1. Re2 Qc7 2. Rec2 Na7 3. Nxe6 fxe6 > 4. Bxe6 b3 5. Qxb3 > 9 27.45 0.80 1. Re2 Qc7 2. Rec2 Na7 3. Nxe6 fxe6 > 4. Bxe6 b3 5. Qxb3 > 9-> 41.79 0.80 1. Re2 Qc7 2. Rec2 Na7 3. Nxe6 fxe6 > 4. Bxe6 b3 5. Qxb3 > 10 1:32 0.92 1. Re2 Na7 2. Rec2 Qc7 3. Nxe6 fxe6 > 4. Bxe6 <HT> > 10-> 2:12 0.92 1. Re2 Na7 2. Rec2 Qc7 3. Nxe6 fxe6 > 4. Bxe6 <HT> > 11 4:01 0.85 1. Re2 Na7 2. Rec2 Rb8 3. Kh1 Be7 4. > Bxe7 Kxe7 5. Rc5 Qc7 6. Qc2 Rbc8 > 11-> 6:02 0.85 1. Re2 Na7 2. Rec2 Rb8 3. Kh1 Be7 4. > Bxe7 Kxe7 5. Rc5 Qc7 6. Qc2 Rbc8 > 12 10:29 0.89 1. Re2 Na7 2. Rec2 Nb5 3. Bg2 Nba7 > 4. Bf1 Qc7 5. Nd3 Rc8 6. Nc5 Be7 7. > Nxb7 Bxg5 8. hxg5 Qxb7 > 12 14:49 0.95 1. Nxe6 fxe6 2. Bxe6 Ne7 3. Bxe7 Kxe7 > 4. Bxd5 Rc8 5. e6 Kf6 6. Qd1 Be7 7. > Qd3 > 12-> 17:34 0.95 1. Nxe6 fxe6 2. Bxe6 Ne7 3. Bxe7 Kxe7 > 4. Bxd5 Rc8 5. e6 Kf6 6. Qd1 Be7 7. > Qd3 > 13 19:09 0.78 1. Nxe6 fxe6 2. Bxe6 Ne7 3. Bxe7 Kxe7 > 4. Bxd5 Rc8 5. e6 Kf6 6. Rc4 Be7 7. > Rec1 Na7 8. Bxb7 Qxb7 > 13 44:43 0.90 1. Re2 Na7 2. Rec2 Nb5 3. Bg2 Nba7 > 4. Qb3 a4 5. Qe3 Qc7 <HT> > 13-> 66:37 0.90 1. Re2 Na7 2. Rec2 Nb5 3. Bg2 Nba7 > 4. Qb3 a4 5. Qe3 Qc7 <HT> > 14 92:29 0.89 1. Re2 Na7 2. Rec2 Nb5 3. Bf5 Rb8 4. > Bd3 Nba7 5. Bf1 Rc8 6. Nd3 Qc7 7. Nc5 > Be7 8. Nxb7 Bxg5 9. hxg5 Qxb7 > 14-> 160:46 0.89 1. Re2 Na7 2. Rec2 Nb5 3. Bf5 Rb8 4. > Bd3 Nba7 5. Bf1 Rc8 6. Nd3 Qc7 7. Nc5 > Be7 8. Nxb7 Bxg5 9. hxg5 Qxb7 > 15 207:02 0.88 1. Re2 Na7 2. Rec2 Nb5 3. Bf5 Rb8 4. > Bh3 Nba7 5. Nd2 Qc7 6. Nxe6 fxe6 7. > Bxe6 Be7 8. Bxd5 Bxg5 9. hxg5 > 15 359:58 ++ 1. Bxe6!! Fritz 5.32 still likes Nxe6 at 15 ply. I think this will still win, so ultimately it may make no difference, but Bxe6 does look stronger since the knight can harass black more than the bishop. E.g. 1. Bxe6 fxe6 2. Nxe6 N8e7 3. Bxe7 Bxe7 4. Qc2! +-. 2...Be7 also fails to Qc2, and 2...b3 trying to prevent Qc2 holds out longer but is still dead meat in short order. The question is why the program chose Bxe6, did it see Qc2? Or will it change back to Nxe6 later (probably will flip-flop anyway).
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.