How to win KQ vs KRB, against Crafty, without EGTBs, in 50 m

[Slobodan R. Stojanovic]

Geschrieben von:/Posted by: Slobodan R. Stojanovic at 02 July 2004 01:59:01:

[Event "25´+3´´"]
[Site "SLOBODAN"]
[Date "2004.07.01"]
[Round "2"]
[White "Crafty-19.14"]
[Black "WildCat 4"]
[Result "0-1"]
[ECO "D93/16"]
[Opening "Gruenfeld with Bf4 e3, ...c5 Main Line"]
[Time "unterminated"]
[Variation "1500+3"]
[TimeControl "1500+3"]
[Termination "normal"]
[PlyCount "442"]
[WhiteType "program"]
[BlackType "program"]
1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Bf4 Bg7 5.e3 c5 6.dxc5 Qa5 7.Rc1 dxc4 8.Bxc4 O-O
9.Nf3 Qxc5 10.Bb3 Qa5 11.O-O Nc6 12.h3 Bf5 13.Qe2 Ne4 14.Nd5 e5 15.Bh2 Be6 16.Rfd1
Rfd8 17.Qc4 Nf6 18.e4 Rac8 19.Ng5 Nd4 20.Ne7+ Kf8 21.Nxe6+ Kxe7 22.Nxd8 Rxc4 23.Bxc4
Bh6 24.Nxb7 Qc7 25.Rc3 Nxe4 26.Rxd4 Nxc3 27.bxc3 Qxb7 28.Bxe5 Bc1 29.Bd5 Qb2 30.Rb4
Qd2 31.Rb7+ Kd8 32.Bd4 Bb2 33.Rxa7 Bxc3 34.Bb6+ Kc8 35.Rc7+ Kb8 36.Rd7 Qc1+ 37.Kh2
Be5+ 38.g3 Bxg3+ 39.fxg3 Qb2+ 40.Bg2 Qxb6 41.Rxf7 Qa6 42.Rxh7 Qxa2 43.Rd7 Kc8 44.Rd6
Qf7 45.h4 Kc7 46.Rc6+ Kd8 47.Rb6 Qf5 48.Rb4 Qh5 49.Rb7 Ke8 50.Ra7 Kf8 51.Rc7 Qe2
52.Rc8+ Kg7 53.Rc5 Kh6 54.Rd5 Qe3 55.Rd6 Qe5 56.Rd5 Qe2 57.Rd6 Qg4 58.Rf6 Qd4 59.Re6
Qf2 60.Re5 Qf7 61.Rd5 Qf6 62.Rc5 Qd4 63.Rd5 Qc4 64.Re5 Qc7 65.Rg5 Qe7 66.Bd5 Qe2+
67.Kh3 Qd3 68.Kg4 Qe3 69.Bf3 Qe6+ 70.Kf4 Qd6+ 71.Ke3 Qa3+ 72.Kf2 Qb2+ 73.Kg1 Qc2
74.Rd5 Kg7 75.Rg5 Kf6 76.Rg4 Qd2 77.Re4 Qb2 78.Rf4+ Ke5 79.Re4+ Kf5 80.Re2 Qd4+ 81.Kg2
Kf6 82.Kh3 Qd3 83.Bg4 Kg7 84.Re1 Qd5 85.Rd1 Qe4 86.Ra1 Kf8 87.Rf1+ Kg8 88.Bf3 Qf5+
89.g4 Qd3 90.Rf2 Kf7 91.Kg2 Kg7 92.Re2 Qd4 93.Re7+ Kf6 94.Re4 Qd6 95.Kh3 Qc7 96.Re3
Qf4 97.Rb3 Qd4 98.Bc6 Ke5 99.Bg2 Ke6 100.Bf3 Kd6 101.Ra3 Ke5 102.Ra2 Qc3 103.Kg2
Qc4 104.Rf2 Ke6 105.Kg3 Qc7+ 106.Kh3 Qc3 107.Re2+ Kf6 108.Kg2 Qc4 109.Rd2 Qf4
110.Rd8 Kf7 111.Rd1 Qf6 112.Kg3 Qe5+ 113.Kh3 Qc3 114.Kg3 Qc7+ 115.Kh3 Qc8 116.Be4 Qc3+
117.Rd3 Qe1 118.Bd5+ Kf6 119.Rg3 Ke5 120.Bg2 Qd2 121.Rf3 Ke6 122.Rg3 Qd4 123.Ra3 Qb6
124.Bf3 Kf7 125.Rc3 Qf2 126.Bd5+ Ke7 127.Rf3 Qe2 128.Ra3 Kf6 129.Bf3 Qc4 130.Re3 Qf1+
131.Bg2 Qc1 132.Rb3 Qg1 133.Rf3+ Ke6 134.Ra3 Qd4 135.Rf3 Qc5 136.g5 Ke5 137.Kg3 Qd4
138.Rf8 Qd6 139.Rf4 Qd2 140.Re4+ Kf5 141.Bf3 Qd6+ 142.Kh3 Qc5 143.Re8 Qa3 144.Kg3 Qd6+
145.Kh3 Kf4 146.Kg2 Qc5 147.Re2 Qc3 148.Re4+ Kf5 149.Re8 Qb4 150.Kh3 Qc5 151.Kg3 Qa3
152.Rd8 Qb4 153.Rd5+ Ke6 154.Rd1 Qc3 155.Rf1 Kf5 156.Kh3 Kf4 157.Kh2 Qe3 158.Kh3 Kf5
159.Kg3 Qe5+ 160.Kh3 Qc3 161.Kg3 Qe3 162.Kh3 Kf4 163.Kh2 Qd2+ 164.Kh3 Qd3 165.Kg2 Qd4
166.Rf2 Kf5 167.Kg3 Qd3 168.Kh3 Kf4 169.Kg2 Qd4 170.h5 gxh5 171.Bxh5+ Kxg5 172.Bf7 Qe4+
173.Kh3 Qg4+ 174.Kh2 Kh4 175.Kh1 Qe4+ 176.Kh2 Qc6 177.Bg8 Qc7+ 178.Kg2 Qb7+ 179.Rf3
Qg7+ 180.Kh2 Qe5+ 181.Kg2 Qg5+ 182.Kh2 Qd2+ 183.Kg1 Qe2 184.Rf2 Qe1+ 185.Kg2 Qe4+
186.Kh2 Qg4 187.Kh1 Qd4 188.Rh2+ Kg3 189.Rg2+ Kf3 190.Kh2 Qe4 191.Kh1 Qb1+ 192.Kh2 Qd1
193.Rg3+ Kf4 194.Kg2 Qe2+ 195.Kh3 Qf1+ 196.Kh2 Qe1 197.Rg7 Qf2+ 198.Rg2 Qe3 199.Bh7 Qf3
200.Bg6 Qd5 201.Kg1 Kf3 202.Kh2 Qe5+ 203.Kh1 Qa1+ 204.Kh2 Qh8+ 205.Kg1 Qd4+ 206.Kh2
Qf4+ 207.Kg1 Qh4 208.Bc2 Qe1+ 209.Kh2 Qe5+ 210.Kh1 Qd4 211.Bg6 Qd1+ 212.Rg1 Qd5
213.Rg2 Qc5 214.Kh2 Qc1 215.Bh5+ Kf4 216.Kh3 Qh1+ 217.Rh2 Qg1 218.Rg2 Qf1 219.Bg6 Qh1+
220.Rh2 Qf3+ 221.Kh4 Qg4# {50 moves rule} 0-1

[Dieter Bürßner]

Geschrieben von:/Posted by: Dieter Bürßner at 02 July 2004 02:10:55:
Als Antwort auf:/In reply to: How to win KQ vs KRB, against Crafty, without EGTBs, in 50 moves geschrieben von:/posted by: Slobodan R. Stojanovic at 02 July 2004 01:59:01:

This, and also the other case with KR-KB you pointed out recently can happen to Crafty surprisingly often without TBs. The problem is (I think) that Crafty knows, that it is draw usually (Bob: "99.99%"). So Crafty gives a 0.0 score or 0.01 score. But it evaluates all positions the same, and has no fear, to move with his K into the corner (because this gives the same evaluation), until it sees (too late), that it hit the exception of the rule. Of course this would not happen in these cases with TBs. But it could happen in KRBKRP for example. I think the later case is fixed now, after it was pointed out to Robert Hyatt a few times. Bob seems a bit difficult to become convinced of the problem.
Regards,
Dieter

[Dann Corbit]

Geschrieben von:/Posted by: Dann Corbit at 02 July 2004 03:12:17:
Als Antwort auf:/In reply to: Re: How to win KQ vs KRB, against Crafty, without EGTBs, in 50 moves geschrieben von:/posted by: Dieter Bürßner at 02 July 2004 02:10:55:

This, and also the other case with KR-KB you pointed out recently can happen to Crafty surprisingly often without TBs. The problem is (I think) that Crafty knows, that it is draw usually (Bob: "99.99%"). So Crafty gives a 0.0 score or 0.01 score. But it evaluates all positions the same, and has no fear, to move with his K into the corner (because this gives the same evaluation), until it sees (too late), that it hit the exception of the rule. Of course this would not happen in these cases with TBs. But it could happen in KRBKRP for example. I think the later case is fixed now, after it was pointed out to Robert Hyatt a few times. Bob seems a bit difficult to become convinced of the problem.

Seems like a lot of bother to fix it when the tablebase files are free.
It seems more logical to spend time on things that have no work-around.



my ftp site {remove http:// unless you like error messages}

[Slobodan R. Stojanovic]

Geschrieben von:/Posted by: Slobodan R. Stojanovic at 02 July 2004 03:12:36:
Als Antwort auf:/In reply to: Re: How to win KQ vs KRB, against Crafty, without EGTBs, in 50 moves geschrieben von:/posted by: Dieter Bürßner at 02 July 2004 02:10:55:

This, and also the other case with KR-KB you pointed out recently can happen to Crafty surprisingly often without TBs. The problem is (I think) that Crafty knows, that it is draw usually (Bob: "99.99%"). So Crafty gives a 0.0 score or 0.01 score. But it evaluates all positions the same, and has no fear, to move with his K into the corner (because this gives the same evaluation), until it sees (too late), that it hit the exception of the rule. Of course this would not happen in these cases with TBs. But it could happen in KRBKRP for example. I think the later case is fixed now, after it was pointed out to Robert Hyatt a few times. Bob seems a bit difficult to become convinced of the problem.
Regards,
Dieter

"Bob seems a bit difficult to become convinced of the problem."
Yes, I know how certain he was that Americans will find out Saddam´s WMD
I don´t know whether he is still so certain about this as before or not.
In this case, however, we have a fact: a chess game. And now someone interested should give an interpretation. I did my part, dectecting the fact and posting it.
Thanks for comments.
Regards. SL.

[Slobodan R. Stojanovic]

Geschrieben von:/Posted by: Slobodan R. Stojanovic at 02 July 2004 03:27:12:
Als Antwort auf:/In reply to: Re: How to win KQ vs KRB, against Crafty, without EGTBs, in 50 moves geschrieben von:/posted by: Dann Corbit at 02 July 2004 03:12:17:

This, and also the other case with KR-KB you pointed out recently can happen to Crafty surprisingly often without TBs. The problem is (I think) that Crafty knows, that it is draw usually (Bob: "99.99%"). So Crafty gives a 0.0 score or 0.01 score. But it evaluates all positions the same, and has no fear, to move with his K into the corner (because this gives the same evaluation), until it sees (too late), that it hit the exception of the rule. Of course this would not happen in these cases with TBs. But it could happen in KRBKRP for example. I think the later case is fixed now, after it was pointed out to Robert Hyatt a few times. Bob seems a bit difficult to become convinced of the problem.

Seems like a lot of bother to fix it when the tablebase files are free.
It seems more logical to spend time on things that have no work-around.

Even if EGTBs were not free, it shouldn´t be problem for Prof. Dr. Hyatt, I think.
Second, and the most importante:
There are people who don´t like to use EGTBs while tesating engines (I use them only when checking some positions for myself), and when they find out that an engine like Crafty make such kind of blunders, the overall impression is bad. From the aesthetic point of view in the first place, of course.
Regards. SL.

[Jose Carlos]

Geschrieben von:/Posted by: Jose Carlos at 02 July 2004 12:05:11:
Als Antwort auf:/In reply to: Re: How to win KQ vs KRB, against Crafty, without EGTBs, in 50 moves geschrieben von:/posted by: Slobodan R. Stojanovic at 02 July 2004 03:12:36:

This, and also the other case with KR-KB you pointed out recently can happen to Crafty surprisingly often without TBs. The problem is (I think) that Crafty knows, that it is draw usually (Bob: "99.99%"). So Crafty gives a 0.0 score or 0.01 score. But it evaluates all positions the same, and has no fear, to move with his K into the corner (because this gives the same evaluation), until it sees (too late), that it hit the exception of the rule. Of course this would not happen in these cases with TBs. But it could happen in KRBKRP for example. I think the later case is fixed now, after it was pointed out to Robert Hyatt a few times. Bob seems a bit difficult to become convinced of the problem.
Regards,
Dieter

"Bob seems a bit difficult to become convinced of the problem."
Yes, I know how certain he was that Americans will find out Saddam´s WMD
I don´t know whether he is still so certain about this as before or not.
In this case, however, we have a fact: a chess game. And now someone interested should give an interpretation. I did my part, dectecting the fact and posting it.
Thanks for comments.
Regards. SL.

Let's keep it on-topic, ok? There's no need for such comments.
Please, do not reply to this post.
José C.

[Dan Honeycutt]

Geschrieben von:/Posted by: Dan Honeycutt at 03 July 2004 04:20:42:
Als Antwort auf:/In reply to: [Moderation]Re: How to win KQ vs KRB, against Crafty, without EGTBs, in 50 moves geschrieben von:/posted by: Jose Carlos at 02 July 2004 12:05:11:

Let's keep it on-topic, ok? There's no need for such comments.
Please, do not reply to this post.
José C.

At the risk of getting myself warned I'm replying anyway. Good job José. CTF is available for those who want to talk politics, religion etc.
Dan H.