Author: Christophe Theron
Date: 12:24:46 10/20/02
Go up one level in this thread
On October 20, 2002 at 02:42:57, George Sobala wrote:
>Thank you Christophe, for this engine. (But while I've got your attention - will
>you PLEASE fix multivariation mode to allow proper analysis!)
Thanks for your post George. I'm still amazed that some people find Tiger too
passive, but I guess that they love Tiger so much that they just want it to be
both extremely strong AND very aggressive. :)
The multivariation mode problem will be fixed.
Christophe
>ChessTiger15 set to Gambit Aggressive mode plays wonderful attacking chess. It
>seizes the initiative - which many very "tactical" engines often fail to do.
>Witness the game below. If you liked yesterdays Acs-van Wely game (sub-20 move
>annihalation of a super GM), you will love this too. Gambit Tiger was running on
>a mere K6-450, pille was using Fritz 6 on a much faster machine. Take time to
>actually play through it: note that GT's evals were consistently strongly
>positive for Black from 19 .. Nf4 onwards. I have played this through and looked
>at numerous variations with lots of other engines and not found one that favours
>Black until the final crunch.
>
>What other engines would seize the initiative for a K side attack like this?
>
>What is readily apparent on playing through the hairy variations that this game
>produces (especially in alternatives to White's rapidly losing 29 Bxa6) is how
>FAST Gambit Tiger is in getting to the sharpest tactical lines compared to other
>engines. This is not unique to this game: I found the same with the Acs-van Wely
>game.
>
>
>[Event "ICC tourney 223 (10 6)"]
>[Site "Internet Chess Club"]
>[Date "2002.10.18"]
>[Round "1"]
>[White "pille"]
>[Black "redshift"]
>[Result "0-1"]
>[ICCResult "White resigns"]
>[WhiteElo "1377"]
>[BlackElo "2718"]
>[Opening "Benoni: classical with e4 and Nf3"]
>[ECO "A70"]
>[NIC "BI.20"]
>[Time "18:13:55"]
>[TimeControl "600+6"]
>
>1. d4 Nf6 2. c4 c5 3. d5 e6 4. Nc3 exd5 5. cxd5 d6 6. e4 g6 7. Nf3 Bg7 8.
>Bb5+ Nbd7 9. Bf4 Qe7 10. O-O O-O 11. Re1 Ng4 12. Bg5 Bf6 13. Bxf6 Ndxf6 14.
>h3 Ne5 15. Nxe5 Qxe5 16. Bc4 a6 17. a4 Bd7 18. a5 Nh5 19. Re3 Nf4 20. Qb3
>Rab8 21. Qb6 g5 22. Qc7 Rfd8 23. Ra3 g4 24. hxg4 Bxg4 25. Rg3 h5 26. Rb3
>Rdc8 27. Qb6 Kh7 28. Qa7 Rg8 29. Bxa6 h4 {White resigns} 0-1
>
>Now a test position after 28 .. Rg8: White played the losing Bxa6, but actually
>I am pretty certain that White is busted here anyway. I have analysed this
>extensively using multiple engines, and Black has a won game no matter what
>White does. How long before your engine indicates that White is lost?
>
>[D] 1r4r1/Qp3p1k/p2p4/P1pPq2p/2B1Pnb1/1RN3R1/1P3PP1/6K1 w - - 0 1
>
>However, Black's most sublime move was 27 .. Kh7: perhaps not the very best in
>this position but a typical Gambit Tiger attacking move (it gave eval of >2 to
>Black at this stage). When I saw this one over the board I felt that GT had the
>game in the bag. Would your engine play it (or Kh8?) By my analysis, after Kh7,
>both Qa7 and Bf1 lose - Bf1 taking rather longer to do so. Black always seems to
>have fresh initiative in the attack that ensues.
>
>[D] 1rr3k1/1p3p2/pQ1p4/P1pPq2p/2B1Pnb1/1RN3R1/1P3PP1/6K1 b - - 0 27
>
>Oh - and here's a bonus for you: that Acs-van Wely game:
>
>[Event "Essent Tournament"]
>[Site "Hoogeveen, Netherlands"]
>[Date "2002.10.19"]
>[Round "6"]
>[White "van_Wely(GM)"]
>[Black "Acs(GM)"]
>[Result "0-1"]
>[WhiteElo "2681"]
>[BlackElo "2591"]
>[Opening "Nimzo-Indian: 4.e3 O-O, 5.Bd3 d5"]
>[ECO "E48"]
>[NIC "QP.08"]
>[Time "05:16:58"]
>[TimeControl "7200+0"]
>
>1. d4 d5 2. c4 e6 3. Nc3 Bb4 4. e3 Nf6 5. Bd3 O-O 6. cxd5 exd5 7. Ne2 Re8 8.
>O-O Bd6 9. a3 Ng4 10. h3 Nh2 11. Re1 Nf3+ 12. gxf3 Qg5+ 13. Kh1 Qh4 14. Nf4
>Bxh3 15. Ncxd5 Re6 16. Nxe6 Bf5+ 17. Kg1 Qh2+ 18. Kf1 Bg3 {Black wins} 0-1
>
> Would your engine play 9 .. Ng4 or 10 .. Nh2? Was Acs' attack winning or could
>van-Wely have wriggled out?
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.