< Earlier Kibitzing · PAGE 3 OF 3 ·
|Aug-07-06|| ||cavaleiro: <Larsker> Although he has an italian name, Vescovi is actually brazilian. He played this year's chess olympiad for Brazil indeed.|
|Dec-02-06|| ||M.D. Wilson: It's interesting to compare Capablanca and Alekhine on this point. When asked how many moves he saw ahead, Alekhine replied "one more than my opponent".|
|Dec-02-06|| ||Victor G.: Nice game by Kramnik. If he plays as well as he did on this game from move 44 to move 57 on game 5 against deep fritz 10,it will surely be 1-0 for Kramnik.|
|Dec-02-06|| ||fgh: One of my favorite Kramnik games since a long time...|
|Dec-02-06|| ||Honza Cervenka: One old Russian GM used to say "s molod'yozhu v endshpil" (with youngsters to the endgame) but it can be applied successfully on strong computers as well. This is really superb example of anticomp play.|
|Dec-02-06|| ||Microbe: This is an excellent game. I really enjoyed watching the rook and pawn endgame and slowly seeing Kramniks advantage expanding. Lovely.|
|Dec-02-06|| ||kevin86: I enjoyed how white used a rook as a surgical tool to gain an advantage and to help force in the king. I also like the strange pattern of exchanges earlier-by forks and discovered attacks. A great game!!|
|Dec-02-06|| ||TTLump: analyzing this game with my own "shallow" Genius-7 chess engine, reveals some interesting points that I think may be common to chess engines in general.|
Taking the position just before White's last move (after 56... Rd1), Genius-7 says that 57.Rxa5 is White's best move at 15-27 ply (15 ply for all legal variations, 27 ply for selected lines), scoring this at +2.12 for White after 3:47 of analysis. But then if I force the actual move, 57.Rd5+, Genius-7 scores the position at -6.39 for Black at 21-32 ply after only 1:32 of analysis.
Is this the "horizon" problem sometimes mentioned with reference to chess engines?
What is interesting is that when the Rooks are off the board (which is FORCED by White's move), the number of positions that have to be analyzed is drastically reduced, allowing it to go much deeper, much quicker, thereby greatly increasing the chess engine's "horizon".
Would this same type of weakness apply to deep Fritz ...
|Dec-02-06|| ||TTLump: just to clarify my last post, that "-6.39 for Black" actually is equivalent to "+6.39 for (in favor of) White"|
|Dec-02-06|| ||alexmagnus: <...one move...> Tartakower once said "I think two moves ahead but these are two good moves".|
|Dec-02-06|| ||tonsillolith: Here is an article that cites Capablanca as the one who said that he sees only one move ahead - the right one. http://sciam.com/article.cfm?articl...|
Also, I remember it being Bent Larsen as the one who said long variations are wrong variations, but I could be wrong.
|Dec-02-06|| ||Confuse: so kramnik was stronger in this match then he is today? or is his playing style today stronger then back then? (2807 vs 2760)?|
|Dec-02-06|| ||Rocafella: <Confuse> Ratings aren't always a measure of ability in a particular game, they usually just relate to the players current form. One other thing to be taken into account is that this Deep Fritz is weaker than the one in 2006|
|Dec-09-06|| ||Confuse: Thats true. So would that mean Kramnik was in "better form" back then then he was in the match recently? I guess that makes more sense. Thanks <Rocafella> : )|
|Dec-17-06|| ||Rocafella: <Confuse> No problem :)|
|Dec-10-07|| ||weisyschwarz: Would 30...Ba6 have been better?|
|Feb-27-08|| ||positionalgenius: <pawnsgambit>Look|
|Feb-28-08|| ||whatthefat: <TT Lump>
Yes, what you saw was an example of the horizon effect, and these types of weaknesses apply to all chess analysis software (and humans!).
|Apr-27-08|| ||TheDude: 30.Bc5!! WoW!!!
Great Game by "Deep" Kramnik.
|Apr-27-08|| ||sallom89: well played.|
|Jul-11-08|| ||apexin: Game Two
Vladimir Kramnik (2807) – Deep Fritz
Brains in Bahrain (2), 06.10.2002 [D27]
Kramnik has found a good strategy against Deep Fritz. As in his match against
Kasparov, he exchanges queens quickly to reach an endgame with just one or
two open files that are both firmly under control. This minimizes the danger of
tactical surprises by the computer and retains the chance of winning the endgame
with his fantastic technique. Well done Vladimir!
1.d4 d5 2.c4 dxc4 3.Nf3 Nf6 4.e3 e6 5.Bxc4 c5 6.0-0 a6 7.dxc5 Qxd1
This move looks modest, but Kramnik knows how to keep increasing the in
Did Kramnik want to bring Fritz out of his opening database? Here are some of
Kramnik’s previous games: 9.Nbd2 b6 10.Be2 Nbd7 11.Nc4 (11.Nb3 Be7
12.Nfd4 Bb7 13.f3 0-0 14.e4 Rfc8 15.Be3 Kf8 16.Nd2 Ne5 17.N4b3 Rc6
18.Rac1 Rac8 19.Rxc6 Rxc6 20.g4 h6 21.h4 Bc8 22.g5 hxg5 23.hxg5 Nfd7
24.f4 Ng6 25.Nf3 Rc2 26.Bxa6 Bxa6 27.Rxd7 Rxb2 28.Ra7 Bb5 29.f5 exf5
30.exf5 Re2 31.Nfd4 Re1+ 32.Kf2 Rf1+ 33.Kg2 Nh4+ 34.Kh3 Rh1+ 35.Kg4
Be8 36.Bf2 Ng2 37.Ra8 Rf1 38.Kf3 Nh4+ 39.Ke2 Rh1 40.Nb5 Bxg5 41.Nc7
Ke7 42.Nxe8 Nxf5 43.Bxb6 Kd7 44.a4 Rh3 45.Nc5+ Kc6 46.a5 Re3+ 47.Kd1
Re7 48.Rc8+ Kb5 49.Ne4 Rxe4 50.Rc5+ Ka6 51.Nc7+ Kb7 52.Rxf5 Be3
53.Bxe3 Rxe3 54.Rxf7 Re5 55.a6+ Kb6 56.Rxg7 Ra5 57.Kd2 Ra1 58.Kc2
Rh1 59.Kb2 Rh8 60.Kb3 Rc8 61.a7 Kxa7 62.Kb4 Kb6 63.Nd5+ Ka6
64.Rg6+ Kb7 65.Kb5 Rc1 66.Rg2 Kc8 67.Rg7 Kd8 68.Nf6 Rc7 69.Rg5 Rf7
70.Nd5 Kd7 71.Rg6 Rf1 72.Kc5 Rc1+ 73.Kd4 Rd1+ 74.Ke5 ˝-˝,
Kramnik,V-Kasparov,G London 2000) 11...Bb7 12.b3 Ke7?! (12...0-0 13.Bb2
Bd5 14.Rac1 Rfc8 15.Nfe5 b5 16.Nxd7 Nxd7 17.Nd2 Bb4 18.Nb1 Nb6
19.f3 Rxc1 20.Rxc1 Rc8 21.Rxc8+ Nxc8 22.Ba3 Bxa3 23.Nxa3 b4 24.Nc2
a5 25.Ba6 Nb6 26.e4 Bc6 27.Kf2 ˝-˝, Kramnik,V-Lautier,J Monte Carlo
1997) 13.Bb2 Rhd8 14.Ne1 b5 15.Na5 Be4 16.Bf3 Bxf3 17.gxf3 Rdc8
18.Nd3 Rab8 19.Nxc5 Rxc5 20.Rac1 Ke8 21.Bd4 Rxc1 22.Rxc1 Ra8 23.Kf1
e5 24.Bb2 Nd5 25.Rd1 Ne7 26.Nb7 Nc8 27.Rc1 f6 28.Ba3 a5 29.Rc7 b4
30.Bc1 Na7 31.Nd6+ Kd8 32.Rb7 Nc8 33.Nc4 a4 34.Bd2 axb3 35.axb3
Ra7 36.Rxa7 Nxa7 37.Bxb4 Kc7 38.Ke2 Kc6 39.Kd3 g6 40.Be7 Nc8 41.Bd8
Kb5 42.Kc3 Na7 43.Nd6+ Kc6 44.Ne4 f5 45.Ng5 h5 46.e4 Nb5+ 47.Kc4
Nd4 48.b4 Nb5 49.Be7 Nd4 50.f4 Nb5 51.exf5 gxf5 52.Nf7 exf4 1-0,
Kramnik,V-Karpov,A Frankfurt 1999).
9...b5 10.Be2 Bb7 11.Nbd2 Nbd7 12.Nb3 Bf8?
This move is unconvincing. A human would just not play it. Apparently the
computer believed that 12.Nb3 was to be followed by 13.Nd2, repeating for a
draw of all things! The Fritz team did not look happy.
A typical strategy to create squares for White’s pieces, along with weaknesses
in Black’s camp.
Kramnik immediately sends his knight to the newly available c4-square.
14...Bd5 15.f3 Bd6 16.g3 e5 17.e4 Be6 18.Nc4 Bc7 19.Be3 a5
Don’t play this kind of move cavalierly! The b5-square and the a5-pawn are now
very weak. White has a nice position, which is even nicer against a machine,
as Black has almost no counterplay. Black wants to castle and bring his
king’s rook into play, so Kramnik decides to prevent this and keep the initiative.
20.Nc5 (20.Rac1!?) 20...Nxc5 21.Bxc5 Nd7 22.Nd6+ Kf8
22...Ke7? 23.Nf5+ Kf6 24.Be7+ Kg6 25.Rac1±.
Deep Fritz cedes Kramnik the bishop-pair to ease the pressure, but the bishops
constitute a long-term advantage. 23...Ke7 24.Nf5+r.
24.Rxd6 Ke7 25.Rad1 Rhc8 26.Bb5 Nc5!
|Jul-11-08|| ||apexin: this looks like a tactical mistake, but computer programs do not commit such
errors. So think hard and you will see what Fritz had up his sleeve.
27...Ra6?? 28.Bxc5 Raxc6 29.Rxc6+i with check.; 27...Rxc6? 28.Rxc6 Nxa4
28.Ke1 Nd3+ 29.R1xd3 Bxd3 30.Bc5! (30.Rxd3?? Rxc6o) 30...Bc4
30...Bc2 31.Kd2 Bxa4?? 32.Rd5+ Kf6 33.Bxa4i.
31.Rd4+ Kf6 32.Rxc4 Rxc6 33.Be7+ Kxe7 34.Rxc6
The rook ending is better for White, as Black’s rook must passively defend the
a-pawn. The main principle of rook endings is: activity at all costs.
34...Rd8 35.Ke2 does not help.
Now Black has to stay passive. Notkin showed in Chess Today No.699 that
35...Rc8! is stronger, as the rook is activated after 36.Rxa5 Rc2. The pawn
endgame after 36.Rxc8 is drawn, but the points are very deep – so even Deep
Fritz has difficulties spotting them. The main line of Notkin’s analysis goes:
36.Rxc8 Kxc8 37.Kd2 Kd7 38.Kd3 Kd6 39.Kc4 Kc6 and Black has the opposition.
White can’t win the fight for tempi, e.g. 40.h4 g6 41.g4 h5 42.g5 Kd6
43.Kb5 f6! 44.gxf6 g5 45.hxg5 Ke6 46.Kxa5 h4 and the resulting queen
endgame is drawn.
36.Kd2 Kd6 37.Rd5+ Kc6 38.Kd3 g6?!
This weakens Black’s position. 38...Ra7 is better.
39.Kc4 g5 40.h3 h6 41.h4 gxh4 42.gxh4 Ra7 43.h5
Kramnik has all the time in the world and improves his position as much as
possible before playing Rd8.
Now Kramnik maneuvers to lose a tempo and thus bring Fritz into zugzwang.
He could do this by b2-b3, but this may waste an important reserve tempo.
Rubinstein and Capablanca would certainly be impressed!
44.Kd3, with the plan Ke3 and f4, also comes into consideration.
45.Kd5?? Rd8+ 46.Kc4 Rd4+o.
45...Kc6 46.Rd5 Zugzwang. 46...Kc7?!
46...Ra7 47.Rd8 Rc7 48.Rf8 Kd6+ should lose as well, but at least it is more
47.Kb5 b3 48.Rd3 Ra7 49.Rxb3 Rb7+ 50.Kc4 Ra7
The pawn endgame after 50...Rxb3 is lost because of White’s distant passed
pawn: 51.Kxb3 Kb6 52.Kc4 Kc6 53.b4i.
51.Rb5 Ra8 52.Kd5 Ra6 53.Rc5+ Kd7 54.b3 Rd6+
54...Ra7 55.Rb5 Ke7 56.Kc6 Ra8 57.Kb7 Rd8 58.Rxa5i.
55.Kc4 Rd4+ 56.Kc3 Rd1 57.Rd5+ 1-0
Simplifying into a pawn ending is the easiest way to win: 57...Rxd5 58.exd5
Kd6 59.b4 axb4+ 60.Kxb4 Kxd5 61.Kb5 Kd6 (61...f5 62.a5 e4 63.fxe4+
fxe4 64.a6 e3 65.a7 e2 66.a8Q+i) 62.a5 f5 63.a6 Kc7 64.Kc5 e4 65.fxe4
fxe4 66.Kd4 Kb6 67.Kxe4 Kxa6 68.Kf5 Kb6 69.Kg6 Kc7 70.Kxh6 Kd7
71.Kg7i. Until now Kramnik has managed to impose his will and reach simple
technical positions with everything under control. Let’s see if Fritz manages to
complicate matters in two days.
|Nov-06-08|| ||The Rocket: Kramnik spanks the calculating monster as white in a drawish opening. Very impressive.|
|Aug-02-11|| ||DrMAL: Beautiful game by Kramnik. 27.Bc6 is not sacrificing anything not sure why it is marked as such. 29.R1xd3 does temporarily go down the exchange for a few moves until 33.Be7+ recovers it, with position that looks at first drawn.|
However, white's rook is much more active than black's rook is, a key (along with active king) to rook endgames. The computer gets worse and worse position by failing to "understand" this elementary fact. With 46...Kc7? the second rule (active king) is violated, now white's king comes in to win.
|Mar-09-12|| ||LoveThatJoker: GUESS-THE-MOVE:
WE ARE NOW FOLLOWING
Kramnik vs Deep Fritz, 2002.
YOU ARE PLAYING THE ROLE OF KRAMNIK.
Your score: 122 (par = 100)
< Earlier Kibitzing · PAGE 3 OF 3 ·