Knight Keystor

A record of my journey of chess improvement.

Wednesday, December 27, 2006

Update on Deep Fritz 10 Problem

I've been able to confirm the issue on a second PC. This one has a different processor and different OS, so I'd say the problem is definitely with Deep Fritz 10 (I can't comment on the non-deep version.) I've also been able to clarify it a little further. The problem arises in positions where black is able to castle long. There is no problem with white doing the same. I'm going to try to e-mail Chessbase about the issue. I'll put out another update if/when I hear back.

Saturday, December 23, 2006

Fritz 10/Deep Fritz 10 Problem?

I'm having a very strange problem with Deep Fritz 10. If anyone else has either version of Fritz 10, can you try to replicate this and let me know if you have the same problem? This is somewhat difficult to explain, so please bear with me. I have found that in some cases, infinite analysis mode gives incorrect information once a game has been saved in the database. It seems to happen in games where castling long is a possibility. I'll give two examples that I have been able to replicate 100% of the time:

In the first case, I paste the following moves into a new game:

1.e4 e6 2.d4 d5 3.Nc3 Nf6 4.Bg5 dxe4 5.Nxe4 Be7 6.Bxf6 gxf6 7.Nf3 b6 8.Bc4 Bb7 9.Qe2 Nd7 10.0-0-0 c6 11.Rhe1 Qc7 12.g3

Now, if I turn on infinite analysis mode, 12...0-0-0 is possible and is the first or second choice depending on the depth. If I save this game into a database, then close and re-open the game, 12...0-0-0 is not included in the analysis and Fritz WILL NOT let me enter the move at if it were illegal or something.

The second example is even more dramatic:

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Bxc3+ 6.bxc3 c4 7.Qg4 g6 8.Nf3 Qa5 9.Bd2 Nh6 10.Qh3 Nf5 11.g4 Nxd4 12.cxd4 Qb6 13.Bg2 Nc6 14.Qh6 Nxd4 15.0-0 Bd7

Now at this point, if I turn on infinite analysis, 16.Bg5 is the number one move at a depth of 16-ply and a score of 3.84 in favor of white. If I save this game into a database, then close and re-open it, then turn on infinite analysis, 16.Qg7 is the number one move at a depth of 16-ply with a score of 9.92 in favor of white!!! After a bit of digging, I realized that the reason for the difference is that after saving and re-opening the game, Fritz is assuming that 16...0-0-0 is illegal for some reason!?

If anyone can confirm whether this is a problem with Fritz 10 (or preferably Deep Fritz 10, since that is what I have,) I would greatly appreciate it. The two examples have been 100% repeatable for me, so if it is a problem with the program and not my computer, somebody should be able to replicate it.


Thursday, December 14, 2006

An Agonizingly Painful Loss... the club expert. He is the strongest player at my local club and his USCF rating has been as high as ~2050. What makes the loss so painful is that this is the second time (out of my multitude of losses against him) that I have had a devestatingly won game against him until the very, very end and then blown it. I'll dig up and post the first one some other day, along with the one win that I do have against him...

The game last night went for almost 4 hours and 73 moves. We played a line in the Ruy Lopez that I was quite familiar and comfortable with. I came out of the opening/early-middlegame up a full rook and a couple of pawns, giving me an overwhelming and "easy" win...but as you'd expect from a solid expert, he was a very tenacious defender. Slowly he was able to whittle down my lead...first winning the exchange, then winning my queen for a rook.

This put us into an endgame where he had a queen and a set of non-passed, doubled pawns and I had a knight, a rook, a passed center pawn, and three connected pawns, one of which was also passed. I played a very precise, winning endgame, always making sure that all of my material was sufficiently defended at all times...then it happened. I moved my rook to an undefended square with check, allowing him to move his king with a discovered check. I had seen the discovered check, but thought that his king only had two escape squares...and either of those two moves would have left me safe. What I missed was that in moving my rook, I had opened up a third escape square for his king which allowed him to win my rook. I played on for a few more moves, but it was hopeless. Here is the full game...let me know what you think:

Ron - Chris [C72]
Unrated Club Game - No Clock - 13.12.2006
1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 d6 5.0–0 Bg4 6.h3 h5 7.Bxc6+ bxc6 8.d4 Bxf3 9.Qxf3 exd4 10.Re1 Qf6 11.Bf4 g5 12.e5 dxe5 13.Bxe5 Qe6 14.Nd2 g4 15.hxg4 hxg4 16.Qd3 0–0–0 17.Qxa6+ Kd7 18.Bxh8 Qh6 19.Bxd4 Bd6 20.Be5 Nf6 21.Bxd6 Rh8 22.Bh2 Qxh2+ 23.Kf1 Qf4 24.Qd3+ Kc8 25.g3 Qh6 26.Qf5+ Kb7 27.Ne4 Nd5 28.Qxg4 f5 29.Qxf5 Rf8 30.Nc5+ Ka7 31.Qd3 Qh1+ 32.Ke2 Re8+ 33.Kd2 Qh6+ 34.Re3 Kb6 35.Nd7+ Kb7 36.Nc5+ Kb6 37.Na4+ Kb7 38.Qb3+ Kc8 39.Rae1 Nxe3 40.fxe3 Qh2+ 41.Re2 Rd8+ 42.Qd3 Qxg3 43.Qxd8+ Kxd8 44.Nc3 Qd6+ 45.Kc1 Qb4 46.e4 Ke7 47.a3 Qc4 48.Rd2 Ke6 49.Rd3 Ke5 50.b3 Qg8 51.Kb2 Qg1 52.Rd1 Qb6 53.b4 Ke6 54.Na4 Qa7 55.Nc5+ Ke5 56.c4 Qa8 57.Rd7 Qc8 58.Kc3 Qh8 59.Re7+ Kd6+ 60.e5+ Kxe7 61.Kd4 Qh4+ 62.Ne4 Ke6 63.a4 Qe1 64.Nc5+ Ke7 65.Nd3 Qd2 66.a5 Ke6 67.a6 c5+ 68.bxc5 Qa5 69.c6 Qb6+ 70.Nc5+ Ke7 71.Kd5 Qa5 72.Kd4 Qd2+ 73.Nd3 Ke6 Ron resigns 0-1

Monday, December 04, 2006

Back on Track

As mentioned in my 11/5 post, I went on vacation last was a relaxing chess-free, computer-free, mostly phone-free week. But now I'm back and ready to get to work again. I started Tactics Module-3 of PCT today and plan to stick to a M, W, F schedule for PCT while playing G/30 games on PlayChess on T & Th. I'm hoping that this will keep my play from getting rusty during tactics training since I have found that tends to happen when I'm doing strictly tactics. I'll probably also sprinkle some CPT and CTS time into the mix here and there. I'll let you know how it goes.

eXTReMe Tracker