The King Indian defense doesn't fit to Chess Computer

This forum is for general discussions and questions, including Collectors Corner and anything to do with Computer chess.

Moderators: Harvey Williamson, Steve B, Watchman

Forum rules
This textbox is used to restore diagrams posted with the fen tag before the upgrade.
Post Reply
User avatar
scandien
Member
Posts: 206
Joined: Mon Sep 12, 2011 1:15 pm
Contact:

The King Indian defense doesn't fit to Chess Computer

Post by scandien »

I have just finish a game versus the Super Forte C ( time control 40 move in one hour) , and with black i choose ( as usual ) the King Indian.
The Super Forte C select the 4 pawn lines and was badly defeated .

I have observe that most machine doesn't really understand the strategic moves and objectives of the very sharp King Indian Defense, and i have the feeling most machines are not tuned versus such opening.

May be we can find other opening where the machine can be outplayed !


[Event "Accession 2015 2016"]
[Site "Orsay"]
[Date "2016.07.14"]
[Round "1.2"]
[White "Super Forte C 6 MHz"]
[Black "scandien"]
[Result "0-1"]
[Annotator "nicolas,Normand"]
[PlyCount "51"]
[SourceDate "2016.07.14"]

{this game was plaid after a long pause of three month, as i lack time to play. I won
already the first game of my match versus the Super Forte C and my goal is at
least to draw with black to be able to win my match in only three gale , if i
am able to win with white in te next game.} 1. d4 Nf6 2. c4 g6 {
As all my gae versus d4 i am using the king indian defense} 3. Nc3 d6 4. e4 Bg7
5. f4 {a surprise , this is the first tie i face this variantion of the 4 pawn.
A really sharp opening were activity is the main point.} O-O 6. Nf3 Na6 7. Be2
e5 8. fxe5 dxe5 9. Nxe5 {white are gaining one pawn but ...} c5 {and the center
is under pressure. In such position the material is not really the more
important.} 10. Be3 (10. d5 Nxe4 11. Nxe4 Bxe5) (10. dxc5 Qxd1+ 11. Bxd1 Re8
12. Nf3 Nxe4 13. O-O Nxc3 14. bxc3) 10... Nb4 11. d5 (11. dxc5 Qe7 12. Qb3 Nxe4
13. Nxe4 Qxe5 14. Bf3 f5) (11. Rc1 Ng4 12. Bxg4 cxd4 13. Qxd4 Qxd4 14. Bxd4
Bxg4) 11... Nxe4 12. Nxe4 Bf5 {
many white pieces are under attack and black's pieces threat the c2 square}
13. Bxc5 $2 {too greedy} Bxe5 14. Bxf8 $2 {
this is not the best choice. Super Forte C aims a material advantage ...} (14.
Bxb4 Bxe4 15. Qd2 Bxg2 16. Bxf8 Qh4+ 17. Kd1 Bxh1) 14... Qh4+ {white's last
moves enable black to take advantage and to launch a strong attack on white's
king.} 15. Kd2 (15. Kf1 Rxf8 16. Qe1 Qf4+ 17. Bf3 Nc2 {was better for white})
15... Rxf8 {and white king is under attack... Blacks already won that game.}
16. a3 (16. Bf3 Bxe4 17. Bxe4 Qf4+ 18. Ke2 Qxe4+ 19. Kf1 Nc2 20. Qe2 Ne3+) (16.
Bd3 Nxd3 17. Kxd3 Qxe4+ 18. Kd2 Re8 19. Re1 Qd4+ 20. Ke2 Bxh2+ 21. Kf1 Qg1#)
16... Qf4+ 17. Ke1 Qxe4 18. axb4 Qxg2 19. Kd2 Qg5+ 20. Ke1 Qh4+ 21. Kd2 Bxb2
22. Qg1 Qf4+ {white king is nearly alone in face of black pieces} 23. Kd1 (23.
Qe3 Bc3+ 24. Kxc3 Qxe3+) 23... Bc3 24. Ra2 Qe4 25. Rb2 Bxb2 26. d6 {
Super Forte C resigns} 0-1

best Regards

Nicolas
User avatar
scandien
Member
Posts: 206
Joined: Mon Sep 12, 2011 1:15 pm
Contact:

Post by scandien »

hello Mike,

The 4 pawns variation is a very aggressive variation versus the King Indian, and very effective versus an unprepared player (see Gallagher book on King Indian). But a king indian player should not have any fear to face this variation. This is my view , and those of many king indian player ( gallagher, Bologan...)

Fro the Super Expert C view , the point is that the emerging position evaluation are mainly based on Mobility and activity rather than material. This is the weak point for the Kittinger programs which don't take into account mobility ( this is why they a better in speed game than at longer time control).
I thinh tha tmany machine will encounter difficulties with such position. Obviously program such as StokFisch or Komodo will crush any player in this varaition ( but i think in all variation of all opening).


For the line with 10.Bf3 gallagher gives the following analysis:
after 10.Bf3 Dxd1 11.Kxd1 Rd8+ 12.Kc2 (gallagher said that this is the best square for the king ) and then 12.... Nfxe4 leading to a force draw in gallagher analysis .

I am a (small) King indian player and this opening fit my playing style... this is probably why i have good result versus machine...

Best Regards

Nicolas
User avatar
scandien
Member
Posts: 206
Joined: Mon Sep 12, 2011 1:15 pm
Contact:

Post by scandien »

I refresh this topic with another game with the King Indian (this time this is the main variation). The game was plaid yesterday versus the MEPHISTO NIGEL SHORT at the time control of 1 hour per player for the entire game.


the game can be see here :
http://lechiquier-orseen.pagesperso-ora ... ort_r2.htm

as you can see the machine was not able to predict the threats on his king, and prefers to win a rook, giving black a winning attack !

I definitely have the feeling that this opening really doesn't fit Dedicated Computer Chess Machines, as well as any opening with long terms attacks on the king !
.

best regards

Nicolas
User avatar
dedicate computers
Member
Posts: 460
Joined: Wed Aug 01, 2007 2:13 am
Location: São Paulo

King Indian

Post by dedicate computers »

Hello

First I must say that you play very well. However, when I participated actively in the playchess machines tournament, several people said that computers do not understand the opening King Indian. I can't really explain why.
Regards
Oswaldo
Four.nine
Full Member
Posts: 911
Joined: Tue Mar 21, 2017 7:30 pm

Post by Four.nine »

Hello,

I am 4.9 and somewhat new. I saw your King's Indian. Just this morning, I posted on a KIng's Indian loss for Black....BUT two computers (Star Diamond and Mach4). If you are interested, here it is:


Game 17 Fool me twice, shame on me: SD repeats the Game 7 E76r: King’s Indian: Four Pawns loss and ….LOSES; no surprise. The surprise was SD REFUSED to Resign????

[Event "SD Mach4-33 match"]
[Black "Mach4-33"]
[Date "Mar 18, 2017"]
[Result "0-1"]
[Round "17"]
[White "SD"]

1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f4 O-O 6.Nf3 c5 7.d5 e6 8.dxe6 Bxe6 9.Bd3 Nc6 10.f5 Bd7 11.O-O Re8 12.Bg5 h6 13.Bh4 g5 14.Be1 g4 15.Nd2 Qb6 16.Qb3 Qxb3 17.Nxb3 Nb4 18.Bb1 Bc6 19.Rf4 Kh8 20.Bh4 Rg8 21.a3 Na6 22.Bc2 Nh5 23.Rf2 g3 24.Rf3 gxh2+ 25.Kxh2 Be5+ 26.Kh1 Rg4 27.Bf2 Rag8 28.Rg1 Nf4 29.Nd5 Bxd5 30.g3 R8g5 31.Rxf4 Bxf4 32.exd5 Bxg3 33.Bxg3 Rxg3 34.Rxg3 Rxg3 35.Nc1 Re3 36.Nd3 Re2 37.Ba4 Rd2 38.Ne1 Rxb2 39.Kg1 Ra2 40.Bd7 Rxa3 41.Bc8 Rc3 42.Bxb7 Nb4 43.Kf2 Rxc4 44.Kf3 Rd4 45.Bc6 Nxd5 46.Ke2 Ne7 47.Bb7 Nxf5 48.Nc2 Rh4 49.Kd2 a5 50.Kd3 Ne7 51.Bf3 f5 52.Ne1 d5 53.Ng2 Rd4+ 54.Kc3 a4 55.Ne1 Rb4 56.Bh5 a3 57.Nc2 a2 58.Na1 Rb1 59.Bf7 f4 60.Nc2 f3 61.Bh5 f2 62.Be2 Rc1 63.Kb2 0-1 I resigned for SD, Mate# 7

From Game 7:
E76r: King’s Indian: Four Pawns in book through White move 11. Queens trade at move 16/17, so I figure a more positional game: wrong. Some non-optimal M4 moves, 19. .. Kh8? (h5), 20. .. Rg8 (maybe h5?) and 22. Nh5? (Nc7) give SD a decent early lead of +1.7. That is, if SD calculates 23. Rxg4. SD calculates 23. Rf2? and: No more lead (+0.4). And M4-33 went on a roll: 23. .. g3!, 25. .. Be5+, 26. .. Rg4! and 27. Rag8. Bearing down on the Star Diamond King on h1 are M4-33: 2 Rooks, 2 Bishops and a Knight; something has to give. SD panics and messes up 29. Nd5? That just loses a Knight, because SD does not recapture, but calculates 30. g3?? GAME OVER. That was quick. When the dust settles, it is move 35 and SD is down the exchange B+N+4P vs. R+N+6P and -5.7 in eval!.
User avatar
scandien
Member
Posts: 206
Joined: Mon Sep 12, 2011 1:15 pm
Contact:

Re: King Indian

Post by scandien »

dedicate computers wrote:Hello

However, when I participated actively in the playchess machines tournament, several people said that computers do not understand the opening King Indian. I can't really explain why.
Regards
Oswaldo
The main point i think relies on the King Indian requires to play dynamically. In most variation, the material is not so important as other opening ( this is specially true for the main variation and the four pawn attack variation).

If the best program (such as stockfish) can predict the outcome after a great number of incoming moves, the weakest ( and most dedicated machine i think) are not powerful enough to play those opening correctly.
They just try to keep any material advantage (just as the Nigel Short do).
In the 4 pawn variation , the Super Forte C didn't consider the mobility of the piece and was disrupted after a few move.

I plaid another game versus the Nigel Short on the Ruy Lopez Exchange variation, and if i finally won the game , this is because the Nigel Short miss a winning plan in the endgame (it only have to push his queen side pawn)... The Nigel Shirt is far better in those solid position.

I think this is the main issue...

best regards

Nicolas
Post Reply