106 ( +1 | -1 ) Chunk value?I recently read a book by Hans Berliner a CC WCC from US contemporary of Fischer called The System, in which he has a chapter talkign about chunks and value that go along with them. Whats a chunk? Berliner says they are groups of pieces, or pawns, or both, that work together in coordination to create board control, attack and or defense, that is worth more than the added up value of the pieces. Batteries, connected rooks,castled king pawn trio , fianchettoed bishop, pawn structures, fortresses are all everyday examples where these chunks exist and are worth added value. What I would like to discuss is what do you understand about chunk's that has helped you win or draw a game in the past?
For example: One might say that they saw a way to trade to an ending with three connected pawns against a minor piece which is of course winning for the pawn side. That is of course a simple example. But do any of you have an example youd like to share.?!
40 ( +1 | -1 ) Berliner ...Sounds like a very authoritative source. Besides his CC play, he's met Fischer in OTB and played with the big boys. Also one of the first to pioneer computer Chess programming, back when the good ones were all on mainframes. It wouldn't surprise me if these "chunk" valuations he is making are part of the assessment technique used in his computer Chess programs !? I'd like to put this one on my Christmas List!
154 ( +1 | -1 ) Berliner ismy favorite! In fact, I got this book directly from him w/ his autograph when he first got it published. My teacher is/was Berliner's student in the 60's and we both still correspond with him fairly regularly. He taught at Carnegie Mellow for years teaching AI/robots type stuff before retiring to Florida several years back. I understand he recently had to have some heart surgery and haven't heard from him since. I hope he's OK. But, regarding chunking, I've read and re-read that section of the book over and over and I'm afraid it's either over my head, or Dr. Berliner just didn't explain it well enough. It's probably the former, but I still try to comprehend it. At the time he wrote the book, his lifetime score was 94 wins, 1 loss and 10 draws. A couple of years ago he played in the ICCF 50'th jubilee and unfortunately didn't fair so well. But he is, after all, approaching 80 years old. -> tables.iccf.com
He also developed a backgammon program that defeated the world champs in 1979. In 1984 he developed HITECH, a chess playing machine/program. Hitech dominated the computer chess scene until late 1988. During this time it became a US Chess Federation Senior Master (top 50 players in US), won the Championship tourney of the state of Pennsylvania 3 times, and beat former US Chess Champion Arnold S. Denker in a match by the score of 3.5 - 0.5
"The System" is a book I highly recommend to all serious chess players.
61 ( +1 | -1 ) I checked Amazon...to look at the material about "The System" and found that there's an out of print book by Berliner called "Using chunking to solve chess pawn endgames". There wasn't much info listed about it but if you're really dedicated you might be able to run it down.
For those interested, the 11 reader reviews about the "The System" give a pretty good indication of the thrust of the book. I recall that there are some reviews at Chesscafe and Jeremy Silman's site that are pretty negative. The Amazon reviews make it sound like a pretty entertaining read though.
37 ( +1 | -1 ) Thanks wschnitd for the info, ill look into it.
As for the book, The System, by Berliner, it is a very good book, i recommend it to all, especially any d4 players. If you do not like playing d4 this book still has a lot of information about positional values of pieces and pawns that was very instrictive and the principles of his system can be appreciated in any chess opening, however you wil see no e4 lines in it. But a great book.
71 ( +1 | -1 ) Ok here is an example of what i am talking about. Generally a Queen that is worth about 9 material points can mate against a knight and bishop. However if the knight and bishop coordinate in an endgame against a lone Queen. They can create a fortress with the king as an untouchable anchor. So in this case the knight, bishop worth about 6.25 material points, are worth at least equal to the queen that is 9. ANd if you are thinking, What about the king helping out shouldnt it be counted in? Perhaps it should, but hey, he's got a king too.
Here is the board position
White: King on c5 and Queen on d6 Black: King on a8, bishop on b7 and Knight on d5
Set it up and try to do anything as white.
Know of any other examples????? Merry Christmas Tom
28 ( +1 | -1 ) how boutRe4, pd4, pe3 vs a Q with kings anywhere, as long as they are on opposite sides of any rank or file attacked by the rook. Maybe say not in zugzwang. *** WT pawns on f7,g7,h7 and Kh8 vs any piece, that cant take on g7 or h7 for mate :) *** unattacked pawn pair on the 6th, with no K mate forthcoming, vs Rook & too distant king
8 ( +1 | -1 ) Oh, and the Ultimate ...Having Mate-in-one, with the move, vs any opposition! Um, except Tal ... who would hypnotize you :))
47 ( +1 | -1 ) Here is a nice endgame ideaSet up your board..... White pieces: King on h4, Bishop on h5, pawns on g4 and f3 black pieces: King on h6, pawns on c3 and g7
White to play and draw!!!
Before you look at the solution set this up for any chess engine that you know of and see if it plays as white to a draw. Can you find the draw before reading the rest of this post?
I will post the solution in a separate message to help any of you who actually want to try without seeing the solution first... Dont look ahead!
48 ( +1 | -1 ) SolutionI guess chess engines can not find the draw for white here. Could You? I guess that this is because the chess engine evaluation methods are flawwed.
And the say computers are better than humans at chess.
Here white can trap the king with a bishop and pawn fortress and the help of blacks g pawn. The Black Queen can not mate alone. See if any computer program you know of can find the draw here playing as white. I bet you your analyzer can not!!!
34 ( +1 | -1 ) Copyrightsthe material discussed in the above two posts is from -> membres.lycos.fr by Valentin Albillo, 1997 If interested I about other flawwed teqniques in computer chess engines i suggest you surf that site. Happy New year Tom
18 ( +1 | -1 ) Eating Crow,a computer played that exact sequence agains me first time i tried it..lol But the evaluation method is flawwed the computer doesnt understand or evaluate that it is a draw.