Author: Chuck
Date: 21:54:13 03/16/01
Go up one level in this thread
On March 16, 2001 at 22:07:17, Brian Kostick wrote:
>Chuck,
>
> Interesting enough but if I interpret the help file correctly then Fritz
>would not even have marked giving a Queen away as a blunder, or bad move, when
>Threshold is set to 1800?? B.K.
>
>Help file snip:
>Threshold: Here you specify when Fritz should consider a move a mistake. If you
>enter a high value (e.g. 300 = three pawn units), then only grave blunders will
>be considered. If you set a very low value the number of “Better is…”
>commentaries and variations will increase.
>
Yes, I suppose you are right. So much for "intuitive".
>
>On March 16, 2001 at 21:05:02, Chuck wrote:
>
>>My copy of Fritz 6 just finished performing what it calls Full Analysis on a
>>high-level correspondence game. Since this was a correspondence game between
>>such strong players, I expected any tactics Fritz might find would be very deep.
>>So I set the Calc. Time to 1200 and the Threshold to 1800. The result was over 2
>>days of thinking by Fritz.
>>
>>I was curious as to whether Fritz 6 might reveal anything on the game that we do
>>not normally see. I chose a high-level correspondence game because I didn't want
>>to see a bunch of tactical variations, but was hoping Fritz would come up with
>>some strategic insights. The end result was: No, computers aren't up to this
>>task yet. As they have little strategic awareness, they aren't much help here.
>>Fritz made 2 comments about gaining space on pawn pushes. In any case, I figured
>>the resultant analysis might be of interest to many people here, so I decided to
>>post it. I pulled the pgn Fritz wrote, so I guess if you save this pgn to a file
>>and open it with a Chessbase product, you'll get a nice view of it. Perhaps I
>>should have tried this with Hiarcs 7.32, which I have, maybe he would give more
>>strategic-type comments, but I would be surprised if it would amount to much.
>>For me, it was an interesting test and I hope that some programs will make
>>strides in this area in the near future. At least Fritz didn't put a bunch of
>>"?"'s on the players' moves. : )
>>
>>Chuck
>>
>>[Event "Arne Henriksen Memorial 2000"]
>>[Site "?"]
>>[Date "????.??.??"]
>>[Round "?"]
>>[White "Larsen, Soren"]
>>[Black "Mousessian, John"]
>>[Result "0-1"]
>>[ECO "A00"]
>>[WhiteElo "2303"]
>>[BlackElo "2466"]
>>[Annotator "Fritz 6 (1200s)"]
>>[PlyCount "68"]
>>
>>1. Nc3 {A00: Irregular Openings} 1... d5 2. e4 dxe4 3. Nxe4 e5 4. Bb5+ {
>>last book move} 4... c6 5. Bc4 Nf6 6. Qe2 Nxe4 7. Qxe4 Bd6 8. Nf3 O-O 9. d3 (9.
>>Nxe5 $2 {is inferior since it leads to} 9... Bxe5 10. O-O Re8 $19) 9... Nd7 10.
>>Bg5 Qb6 11. O-O-O Nc5 ({Instead of} 11... Qxf2 12. Be3 Qxg2 13. Rhg1 $18) 12.
>>Qe2 Qc7 13. d4 exd4 14. Nxd4 b5 (14... Bxh2 $2 {doesn't solve anything} 15. g3
>>Ne6 16. Bxe6 (16. Rxh2 $6 {is no comparison} 16... Nxg5 17. f4 Nh3 $16) 16...
>>fxe6 17. Rxh2 $18) 15. Bd3 a5 (15... Bxh2 $2 {doesn't work because of} 16. g3
>>Nxd3+ 17. Qxd3 $18) (15... Nxd3+ $6 16. Qxd3 h6 17. Bh4 $11) 16. Qf3 Nxd3+ (
>>16... Bxh2 $4 {capturing this pawn is a mistake} 17. Bxh7+ {
>>Demolishes the pawn shield} 17... Kxh7 18. Bf4 $18) 17. Rxd3 Ra6 (17... Bxh2 $2
>>{is worthless because of} 18. g3 f6 19. Bf4 $18 (19. Rxh2 $6 fxg5 20. Qe4 h6
>>21. Nxc6 Bb7 22. Ne7+ Kh8 23. Ng6+ Kg8 24. Qe6+ Qf7 25. Nxf8 Rxf8 26. Qxf7+
>>Rxf7 $18 (26... Kxf7 $4 27. Rd7+ Kg6 28. Rxb7 $18)) (19. Qxc6 $6 {
>>is easily refuted} 19... Qxc6 20. Nxc6 Bxg3 21. fxg3 Bb7 22. Ne7+ Kf7 $14)) 18.
>>h4 a4 {Black wins space} 19. Nf5 Bxf5 20. Qxf5 a3 21. b3 c5 22. Rd5 c4 {
>>This push gains space} 23. Rxb5 cxb3 24. Rxb3 Rc8 25. Rd1 g6 26. Qd3 Be5 27.
>>Kb1 Rd6 {Threatening mate... how?} 28. Qe2 Re6 29. Rdd3 (29. Rxa3 $4 {
>>the pawn must be left alone, otherwise White will be punished} 29... Rb6+ 30.
>>Kc1 Bb2+ 31. Kd2 Bxa3 $19) 29... Bg7 {The mate threat is Rxe2} 30. Be3 Rc6 {
>>c2 becomes the focus of attention} 31. Qd1 Rxc2 32. Rd7 Qe5 {
>>Do you see the mate threat?} 33. Rd8+ Bf8 {Threatening mate: Rb2} 34. Qd4 Qe7 (
>>34... Qe7 35. Qd7 Qe4 $19 (35... Qxd8 $6 {is the less attractive alternative}
>>36. Qxd8 Rb2+ 37. Rxb2 Rxd8 38. Rb6 $11) (35... Rxd8 $6 {is much weaker} 36.
>>Qxe7 Rb2+ 37. Rxb2 Bxe7 38. Rb3 $11)) 0-1
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.