REDUX problematic initiative
This diagram is a re-post from June 16, 2017.
It shows exactly why I don't make progress in chess. Back then I have pondered about this position for days. Now I tried it again, but I have forgotten almost all the details. Meaning that I don't learn anything by the way I treat such position.
I pondered about this position until system II was satisfied. System II is a logical thinking fetishist. It is satisfied with a theoretical explanation of the how of the position. That is of no use for system I, though. For system I you need a deeper understanding of the position. The how is not enough. You need an understanding of the why of the position. Why does it work? What can I learn from this position so that I can see the solution in the future within seconds? It doesn't make sense to do other puzzles when I haven't fully grasped the essence of this position. The how is totally clear by now. System II is totally satisfied. But there is still a whiff of magic hanging around this position. I can construct the solution, but I don't see it. Construction is typically something from system II. But in order to see it, matters must be simplified even more. That is what I mean by deep understanding. I'm not there yet, with this position. It doesn't make sense to go on with other positions before I have found the essence of this one. Satisfaction of system II is simply not enough. System I needs simplicity. Deep understanding sounds pretty highbrowed, but it means more simplicity.
r3rbk1/1p3ppp/2pn1qn1/p2p1b2/P2P3P/2PB1NB1/1PQNRPP1/4R1K1 w - - 1 1
[solution]
UPDATE Aug 24
I see three cannons pointing at the enemy:
Removal of the guard
Nd6 is clearly blacks most problematic piece. The immediate take of the defender doesn't work. The removal of the guard works only when you can do it with gain of tempo. It is one of the standard scenarios.
Exploiting the overloaded defender
Bxf5 doesn't work. It doesn't work because the move doesn't gain a tempo. It gives black the time to solve his problems. The white bishop doesn't threat anything from f5. That gives black the time to counter attack with Rxe2.
Main line
What baffles me the most in the main line is, where does white gain a tempo? Actually that is the wrong question. The question must be "where does black lose a tempo?".
The answer is: after Rxf8+ Nxf8. The problem is that Nxf8 (or Kxf8) has no additional punch. That gives white the time to cash in.
The vulture
I might have wanted to slaughter the vulture a bit prematurely. Apparently the vulture must get more education. It can fly for hours above this position, but it must know where to look at. My system II is so busy with its intellectual self-gratification, that it forgets to feed the poor beast.
The intellectual chunks of food are totally inedible for the vulture. System II must predigest them, taking into account the nature of the beast.
Initiative
Again the preservation of the initiative is the key to pruning the tree of analysis. I must predigest this further.
It shows exactly why I don't make progress in chess. Back then I have pondered about this position for days. Now I tried it again, but I have forgotten almost all the details. Meaning that I don't learn anything by the way I treat such position.
I pondered about this position until system II was satisfied. System II is a logical thinking fetishist. It is satisfied with a theoretical explanation of the how of the position. That is of no use for system I, though. For system I you need a deeper understanding of the position. The how is not enough. You need an understanding of the why of the position. Why does it work? What can I learn from this position so that I can see the solution in the future within seconds? It doesn't make sense to do other puzzles when I haven't fully grasped the essence of this position. The how is totally clear by now. System II is totally satisfied. But there is still a whiff of magic hanging around this position. I can construct the solution, but I don't see it. Construction is typically something from system II. But in order to see it, matters must be simplified even more. That is what I mean by deep understanding. I'm not there yet, with this position. It doesn't make sense to go on with other positions before I have found the essence of this one. Satisfaction of system II is simply not enough. System I needs simplicity. Deep understanding sounds pretty highbrowed, but it means more simplicity.
 |
| Diagram 1. White to move |
[solution]
UPDATE Aug 24
I see three cannons pointing at the enemy:
- Bg3
- Battery Qc2 Bd3
- Battery Re1 Re2
Removal of the guard
Nd6 is clearly blacks most problematic piece. The immediate take of the defender doesn't work. The removal of the guard works only when you can do it with gain of tempo. It is one of the standard scenarios.
Exploiting the overloaded defender
Bxf5 doesn't work. It doesn't work because the move doesn't gain a tempo. It gives black the time to solve his problems. The white bishop doesn't threat anything from f5. That gives black the time to counter attack with Rxe2.
Main line
What baffles me the most in the main line is, where does white gain a tempo? Actually that is the wrong question. The question must be "where does black lose a tempo?".
The answer is: after Rxf8+ Nxf8. The problem is that Nxf8 (or Kxf8) has no additional punch. That gives white the time to cash in.
The vulture
I might have wanted to slaughter the vulture a bit prematurely. Apparently the vulture must get more education. It can fly for hours above this position, but it must know where to look at. My system II is so busy with its intellectual self-gratification, that it forgets to feed the poor beast.
The intellectual chunks of food are totally inedible for the vulture. System II must predigest them, taking into account the nature of the beast.
Initiative
Again the preservation of the initiative is the key to pruning the tree of analysis. I must predigest this further.
My previous comment had errors, so I deleted it and re-wrote it.
ReplyDeleteI don't remember this position. (Obviously, I don't remember the solution either.) Here's my "stream of consciousness" approach made explicit.
1. Which side is moving? White.
2. First "intuition": Black Bishop on f5 is "attacked" twice, defended twice. Bf5 is B.A.D.
3. Second "intuition": Black Rook on e8 is "attacked" twice, defended twice. Re8 is B.A.D.
4. Common "connection" between the two intuitions: the Black Knight on d6 is part of BOTH defenses, therefore, is OVERLOADED.
5. Toss the problem to System 2: Figure out the order of captures on the 2 PoPs.
6. Try taking on f5 first with 1. Bxf5. (I don't know why System 1 suggested this, except that there is a "pin" on Bd3 against the White Queen, and I don't particularly like leaving THAT issue unresolved!). Black can capture on e2: 1. ... Rxe8 prior to recapturing on f5. This eliminates the second PoP on e8, resulting in an even set of exchanges. Not good, since the implication of "White to move" is that White can "win" something or checkmate. Terminate this branch unless nothing better shows up elsewhere.
7. That leaves 1. Rxe8. Chop that bugger off and put the onus on Black to figure it out! (H/T to GM Jonathon Tisdall - Improve Your Chess NOW!) Oh, crap, now I have to be Black and figure out a response!
8. What if Black captures on d3: 1. ... Bxd3, under the assumption of an Equal or Greater Threat (EGT)? Do I have anything FORCING in response (since Re8 is desperado)? YES! I can whack the Black Bishop on f8: 2. Rxf8+, WITH CHECK! I’ll "see" your EGT and raise you one. After any recapture on f8, I can then "save" my Queen with 3. Qxd3.
9. Okay, one variation works out okay, which is always a good sign that things are progressing in the right direction. What about 1. Rxe8 Rxe8? Oh, the Black Knight is still overloaded, defending both PoPs. 2. Rxe8! forces Black to choose which piece he's going to lose: the Black Bishop on f5 or the Rook on e8. Either 2. ... Nxe8 3. Bxf5 or 2. ... Bxd3 3. Rxf8+ (EGT again!) 3. ... Kxf8 4. Qxd3 and White is ahead in material.
10. Take a look around to "see" if we are quiescent: yep, don't "see" any additional threats from Black.
11. Total “thinking” time: less than 20 seconds.
The "point" is NOT to "solve" it nor to solve it faster than anyone else, but to illustrate the "thinking process," a mixture of System 1 (subconscious) and System 2 (conscious). HTH.
That has always been the problem with this blog. I show the diagrams that I have problems with, while other players, even with the same rating, have problems with totally different puzzles.
ReplyDeleteWhich makes it difficult to talk about what is going on, since it is hardly recognized.
Some positions trigger simply my system II, and no matter what, system I refuses to kick in. This is such position.
Even while I do understand the position fully intellectually, there is still an aspect that skedaddles me. I know it and I feel it, but I can't correct it.
My system II goes into administration mode, and I loose count before I know it. My free memory slots in Short Term Memory become occupied, and all of a sudden my head thinks like a tree.
You dont have the chunks for this type of tactics and i have doubts that its possible to get them by just looking at a position once
ReplyDeleteI did remember this position and parts of the ideas so i think i have some helpful chunks for this position. Still it did take me more than 20 sec to solve it, my thougts did jump around too much
I evidently miss the right chunks. Even worse, I have wrong chunks in stead.
Deletetry this one : https://chesstempo.com/chess-tactics/19873
ReplyDeleteI just solved this one, using a stop watch. Time: 1 minute, 6 seconds from the time I first saw the problem until I finished solving it successfully.
DeleteAox, this is a good followup to the problem originally posted. Rf6 is B.A.D. and Black created a new PoP wth 1. ... Re8?, leaving the Re8 as a B.A.D. piece. Removal of the defender (Bf7) with 2. Rxf7! Kxf7 (forced: 2. ... Rxe1 allows 3. Qxf6+, and there should be a checkmate shortly) 3. Rxe8 Kxe8 4. Qxf6, a Bishop ahead in material.
Aside: I lost 32 rating points on the next Chess Tempo problem because I forgot that I had to mark it as the last problem AND solve it, because I started writing this comment instead.
C'est la vie!
I did solve it at 2018-08-23 04:26:19 in 30 sec ( 8 sec after first move ) see https://chesstempo.com/chess-statistics/aoxomoxoa
DeleteI want to add some thoughts about this position from Aox:
DeleteLink: https://chesstempo.com/chess-tactics/19873
(Blogger doesn’t like secure links, so you’ll have to cut and paste. Sorry about that.)
LoAs of Qc3 and Rb7 converge on g7.
Consequently, the Rf6 and Bf7 “block” the contacts with this square. Prior to 1. … Re8?, there is 1:3 Black superiority on f7. I guess that’s why Black thought it was a good idea to try to get some exchanges going and bring the Black Rook to e8, thinking that the Bf7 was “protecting” it.
The Rf6 is in a “box”. It has no legal moves, even after Black’s first move. Dr. Lasker describes the situation this way while defining the “encircling motif”:
To name this motif, let us emphasize the TWO ideas underlying it: the idea of superior force at a given point [square], and that of immobility. WHAT IS IMMOBILE MUST SUFFER VIOLENCE. The light-winged bird will easily escape the huge dragon, but the firmly rooted big tree must remain where it is and may have to give up its leaves, fruit, perhaps even its life. Let us name it the motif of ENCIRCLING, since in this term the two ideas of violence and of immobility are blended.
This idea CAN be a “trigger” for System 1. Obviously, Re1 and Bf1 are not “attacking" anything of significance (the b-Pawn? Piffle!) in the Black camp. So, it’s up to the Rb7 and Qc3 to work up an “attack.” What better way to achieve a 2:1 superiority on f6 than with 2. Rxf7! 2. … R(f6)xf7 loses the Re8, so 2. … Kxf7 is FORCED. But then it becomes “obvious” (in some sense) that Kf7 has now taken on the FUNCTION of defending BOTH f6 and e8, so 3. Rxe8 FORCES Black to recapture 3. … Kxe8 allowing White to capture 4. Qxf6.
One of my besetting “sins” is a perverse attraction to grabbing material whenever and wherever possible. I finally arrived at a (partial; I’m still working on it!) solution, based on something from GM Kotov’s Think Like A Grandmaster, pg. 45:
“How then does a chess player choose which move to play in a given position? THERE IS NO EASY ANSWER; each player goes about choosing in his own way. The late Leningrad master Vitaly Chekhover used to say IN ALL SERIOUSNESS that he began his analysis when it was his turn by working out which piece he could put en prise! If he couldn’t SEE a way of losing his queen or his rook he would go on from there and only when he could SEE no advantageous way of giving up material did he start examining quiet moves.”
I’ll repeat something I’ve said before: when “looking” from the vulture’s eye view, ignore the standard material values until you can “SEE” the PoPs, LoAs and recognize the Functions for each RELEVANT piece in the area of the board which catches your eye.
Robert said:
DeleteThe Rf6 is in a “box”. It has no legal moves, even after Black’s first move. Dr. Lasker describes the situation this way while defining the “encircling motif”:
To name this motif, let us emphasize the TWO ideas underlying it: the idea of superior force at a given point [square], and that of immobility. WHAT IS IMMOBILE MUST SUFFER VIOLENCE. The light-winged bird will easily escape the huge dragon, but the firmly rooted big tree must remain where it is and may have to give up its leaves, fruit, perhaps even its life. Let us name it the motif of ENCIRCLING, since in this term the two ideas of violence and of immobility are blended.
The immobility to the rook is not of any importants at this tactic
lets move the pawn from d6 ro c4 ( now a wonderful protected free pawn ) the tactic remain the same
The Rf6 has now many extra squares and the save squares at a6 and d6
New game - Stockfish 9 64 7cpu
4r3/1R3b2/5rkp/1p1P1p2/2p2Pp1/P1Q3P1/q6P/4RBK1 w - - 0 1
Analysis by Stockfish 9 64 7cpu:
1. +- (5.19): 1.Rxf7 Kxf7 2.Rxe8 Kxe8 3.Qxf6 Qxa3 4.Qb6 Kf7 5.d6 Ke6 6.d7+ Kxd7 7.Qxb5+ Kd8 8.Qd5+ Ke7 9.Qe5+ Kd8 10.Bxc4 Qc1+ 11.Bf1 h5 12.Qf6+ Ke8 13.Qe6+ Kd8 14.Kf2 Qd2+ 15.Be2 Qd4+ 16.Qe3 Qd5 17.Qd3 Qxd3 18.Bxd3 Ke7 19.Bxf5 Kf6 20.Be4
2. +- (4.97): 1.Rxe8 Bxe8 2.d6 Qb1 3.d7 Bxd7 4.Rxd7 Qe4 5.Bg2 Qe6 6.Rb7 Qe2 7.h4 gxh3 8.Bxh3 Qd1+ 9.Kh2 Qe2+ 10.Bg2 Rf7 11.Rxb5 Kh7 12.Rc5 Qh5+ 13.Kg1 Qd1+ 14.Bf1 Re7 15.Rxc4 Qe1 16.Rc7 Qxc3 17.Rxc3 Kh8
Now lets give the RF6 another extra square g6 by moving the black king to g7
Neue Partie - Stockfish 9 64 7cpu
4r3/1R3b2/5rkp/1p1P1p2/2p2Pp1/P1Q3P1/q6P/4RBK1 w - - 0 1
New game - Stockfish 9 64 7cpu
4r3/1R3bk1/5r1p/1p1P1p2/2p2Pp1/P1Q3P1/q6P/4RBK1 w - - 0 1
Analysis by Stockfish 9 64 7cpu:
1. +- (#10): 1.Rxe8 Qb3 2.Qe5 Qxa3 3.Rh8 Qc5+ 4.Kh1 Qxd5+ 5.Qxd5 Kxh8 6.Rxf7 Rxf7 7.Qxf7 c3 8.Bd3 h5 9.Bxf5 h4 10.Qf8#
2. +- (#15): 1.Re6 Rexe6 2.dxe6 Kg6 3.exf7 Rxf7 4.Rxb5 Rf6 5.Rb7 Qb3 6.Rxb3 cxb3 7.Qc8 Kg7 8.Bc4 Rf7 9.Qe6 Rb7 10.Qd5 Rc7 11.Qe5+ Kg6 12.Qd6+ Kg7 13.Qxc7+ Kh8 14.Qf7 h5 15.Qg8#
3. +- (8.79): 1.Rxf7+ Kxf7 2.Rxe8 Kxe8 3.Qxf6 Qxa3 4.Qe5+ Kd7 5.Qxf5+ Kc7 6.Qe5+ Qd6 7.Be2 h5 8.Kf2 Kd7 9.Qxh5 Qb6+ 10.Kg2 Qf6 11.Qe5 Qg6 12.d6 Qxd6 13.Qxb5+ Ke7 14.Qxc4 Qd2 15.Qc5+ Kf7 16.Qh5+ Kf8 17.Qf5+ Kg8 18.Qe6+ Kg7 19.Qe5+ Kh7 20.f5 Qd8 21.Kf1 Qg5 22.Bd3 Kg8 23.Ke2 Kf8 24.Qd6+ Kf7 25.Bc4+ Kg7 26.Qd7+ Kh6 27.Qe6+ Kh7 28.Qf7+ Kh6 29.Qf8+ Qg7 30.Qxg7+ Kxg7 31.Ke3 Kf6
The tactic does still win
Now lets give the RF6 another extra free square f5 by moving the pawn f5 to f3 but still
New game - Stockfish 9 64 7cpu
4r3/1R3bk1/5r1p/1p1P4/2p2Pp1/P1Q2pP1/q6P/4RBK1 w - - 0 1
Analysis by Stockfish 9 64 7cpu:
1. +- (#14): 1.Re6 Rexe6 2.dxe6 f2+ 3.Kg2 Qb3 4.Rxf7+ Kg8 5.Qxb3 Rxf7 6.exf7+ Kxf7 7.Qxb5 Ke7 8.Bxc4 Kd6 9.Qb6+ Kd7 10.Be6+ Ke7 11.Qc6 f1Q+ 12.Kxf1 Kf6 13.Qd7 h5 14.Qf7#
2. +- (60.90): 1.Rxe8 f2+ 2.Kg2 Qb3 3.Qxb3 cxb3 4.Bxb5 f1R 5.Bxf1 Rf5 6.Bd3 Rf6 7.Kf2 Rd6 8.Ree7 Rf6 9.Rxb3 Kf8 10.Re5 Rd6 11.Rb8+ Kg7 12.Rb7 Rd8 13.Rf5 Rf8 14.Bc4 Kg6 15.Rb6+ Kg7 16.d6 Bxc4 17.Rxf8 Be6 18.Re8 Kf7 19.Re7+ Kf6 20.Rxe6+ Kxe6 21.d7+ Ke7 22.d8Q+ Kxd8 23.Rxh6 Kc7 24.Kg2 Kb8 25.Rg6 Ka7 26.Rxg4
3. +- (5.73): 1.Rxf7+ Kxf7 2.Rxe8 Kxe8 3.Qxf6 Qxa3 4.Qg6+ Ke7 5.Qe6+ Kd8 6.Qb6+ Ke7 7.Qc7+ Kf6 8.Qe5+ Kg6 9.Qe4+ Kf7 10.Qe6+ Kf8 11.Qxh6+ Ke8 12.Qc6+ Kf8 13.Qc8+ Ke7 14.Qc7+ Kf6 15.Qe5+ Kf7 16.Qe6+ Kf8 17.d6 Qa7+ 18.Kh1 Qa1 19.Qc8+ Kf7 20.Qf5+ Ke8 21.Qxb5+ Kd8 22.Qxc4 Kd7 23.Qc7+ Ke6 24.Qe7+ Kd5 25.Qe5+ Qxe5
@Aox: You missed my point: it was NOT that the position was about piling up on Rf6 because it is "in the box" (immobile) or that the encircling motif is the only or even the most important motif in the given position. It was (instead) that certain "intuitions" support saccades to the important area(s) of the board. It IS important to realize that 1. Rxf7! does "attack" Rf6 2:1, forcing Black to recapture with either the Rook or King. As I showed, grabbibg the unprotected Re1 with 1. ... Rxe1 leads to a quick checkmate. The "obscurity" in this variation is that if either player had a search tree depth set at 2 ply, then 1. Rxf7! Rxf7? obviously loses the Re8 to 2. Rxe8. On the other hand, it appears (at ply 2) that Black has covered both defensive tasks with 1. Rxf7! Kxf7. With this (premature) evaluation, Black would be happy for White to sacrifice the exchange on f7, and White in all probability would not play 1. Rxf7. But, that changes the problem (and the context) as given, and I'm pretty sure that everyone here has a search depth (usually) exceeding 2 ply!
DeleteOff topic: Your last two modifications change the position drastically; as GM Stockfish notes, there are two variations (in the modified positions) that are much more appropriate than 1. Rxf7!. Yes, the Rf6 is no longer "in the box". However, you gave no mention to the fact that with the Kg7, the Rf6 still has no moves: it is now ABSOLUTELY PINNED against the Kg7, and therefore IS immobile with NO moves.
I'm fairly certain that other interesting positions could be created by shifting various pieces of wood around the board to support a conclusion already formed before the analysis began. But what, if anything would it do to help us learn an approach to solving THIS SPECIFIC PROBLEM (or for that matter, any other position) relying on System 1 to point us in the right direction?
BTW, your suggestion of shuffling the pieces around to explore the position in depth reminds me of the excerpt from GM Nigel Davies's "The How and the What"article referenced by GM Jonathan Rowson in Chess for Zebras: Thinking Differently about Black and White. I included this excerpt in a comment on May 26, 2016 (in response to one of your comments), so I won't repeat the whole thing, but just the most important idea:
"The reality is that YOU'VE GOT TO MOVE THE PIECES AROUND THE BOARD AND PLAY WITH THE POSITION. Who does THAT? Amateurs don't, GMs do..."
I guess that puts me firmly in the amateur group!
Ciao!
robert said :
DeleteIt was (instead) that certain "intuitions" support saccades to the important area(s) of the board. It IS important to realize that 1. Rxf7! does "attack" Rf6 2:1, forcing Black to recapture with either the Rook or King.
while i think that the immobility of the Rf6 is completly unimportent ( we only need valuable material there which is not a knight and cant attack the Qc3 ) it is important That 1.Rxf7 is a threat of checkmate, so this puzzle follows the rule "most forcing move first" or in the new terminology of tempo : initiative. I wonder since a long time if the best move is the "most forcing move IN THE END".
On the first sight the most forcing move seems to be 1.Qxf6 but after the reply Kxf6 all force left white. I quess the idea to generate "keep the initiative chunks" should be beneficial.
Such chunks can be generatet by implementing a thoughtprocess.. or an aftermath.. during a game to skip such thoughtprocesses to run the chesstask with all the power of your processor you have.
Ever since our long discussion regarding PoPs, LoAs and Functions, I have worked very hard at identifying these aspects in any given position PRIOR TO calculating any variations. This information is supplemented by a thorough understanding and recognition of MOTIFS (in Dr. Lasker's vernacular). This initial phase is what I refer to as the "vulture's eye view." (Perhaps my notion is not congruent with your own definition of that term.) When I "feel" that I have a good grasp of these elements and System 1 starts silently yelling about what is really important to "see," then I resort to trying to use System 2 to determine possible variations using typical tactical DEVICES/THEMES which can be applied to what has been gleaned from the "vulture's eye" overview.
ReplyDeleteSince last December 2017, I have worked very hard on dogmatically using this approach, regardless of the source and type of the positions. As previously indicated, I work with a book and a highlighter, marking PoPs with squares and LoAs with "auras" extending out to the edge of the board in every direction where there is a potential target (including the possibility of checkmate). (The “downside” is that once you have completed a book, you have to go to another book. Sometimes I have bought another copy of the same book and plowed through all the problems AGAIN. Have I ever mentioned that I “might” have undiagnosed Asperger’s Syndrome?!?) I've "solved" hundreds of positions (perhaps thousands by this point in time; I haven’t kept track of the numbers) using this method. I have almost completed two extensive books on checkmate patterns for the first time, in addition to going back and repeating a lot of the exercises in some of my tactics books. I also have been working through Nimzovitch's My System (the Quality Chess edition) AGAIN for the umpteenth time, always applying the same method to every position, regardless of the type of position.
I note that I suggested the “desperado” motif as a key part of the solution in a comment to the original post. I also note that after using GM Stockfish to analyze this position when it was first presented, I still had a terrible oversight regarding the final material evaluation. So, maybe something has been “stuck” into my System 1 after all! Or, maybe (just maybe) some of that hard work over the last 8 months has stuck some usable “pattern recognition” patterns into my subconscious through repetition. Unfortunately, my System 1 is NOT “talking” to me, so I can’t be sure about THAT!
Or, it may be what Aox has diagnosed: I MAY just have certain “chunks” in my System 1 that you don’t have (and likely vice versa), which now make this position “simple” (in some relevant sense). I have no conscious way of knowing the reality.
Did you improve at chesstempo robert?
DeleteI really haven't used Chess Tempo much since I started trying to absorb the "thinking process". You can see a rise in average rating (with occasional dips) between DEC 2017 and JUL 2018. (Username: crazybob) Sometimes I get "called away" by family while I'm in the middle of solving problems, which tends to screw my rating. I also (usually) forget to set the "Last Problem" option, so I lose rating points for not finishing the last problem in a set.
DeleteBlitz mode will be a good test to see if I have "improved" as a result of my training. When I try it again, I'll post my results here.
I also haven't tried your problem above. I want to be very strict about how much time I use when I try it, in addition to monitoring the "impulses" I "feel" as I solve it. MY "problem" is finding uninterrupted time to focus on solving!
Update of the post in blue.
ReplyDeleteoff topics
ReplyDeletethere are 2 fundamental failures at chesstempo-tactics, to be too slow or to play the wrong move. Either i found the right idea to slow or i used to much time to find a refutation for a wrong idea or i did not see the refutation for the wrong move of me at all. So i did add some "refutation of a wrong idea of mine" in my training.
I looked at taking on d6, then taking on f5. When I realized those two didn't win anything, I went to looking at h5, when I realized that wouldn't lead me to h7 I looked at RxR and saw right away that I was winning a piece with the check, since the Nd6 is overworked (this is what I focused on first when I saw the puzzle).
ReplyDeleteI saw this puzzle last I was on Chesstempo and solved it in half-a-minute. This time I solved it in perhaps 4 minutes, but part of the reason is because I am not sitting on Chesstempo and looking at the easiest moves first, I'm approaching it more as if it were a normal position than if I were just trying to whip through problems. Still, 4 minutes to winning a piece is usually a good deal, OTB, as long as you don't give up on analyzing the position.
I just want to say one more thing. I've seen many tactics problems, the same one, multiple times, and it's no guarantee that I will do better the fourth or sixth time than I did the second time I saw the problem (sometimes it's harder seeing the same problem because then I have this problem with trying to use memory to come up with the answer!).
ReplyDeleteThat's why, I feel that the "pattern recognition" angle is a b*llshit assumption, and part of the reason I've left this board at one time. Pattern-recognition has been held too religiously at times on this board (although I know that that's not the main focus of the board, and for most members here the focus is on this methodology aspect, which I look more upon with interested amusement at times, and not with quite as much patience as the rest of you).
However, I do think that one could memorize this problem (be able to blindfold setup the pieces when someone tells you to set up this problem, and blindfold give the answer). That is NOT the same thing as this semi-mythical concept of the almighty "pattern recognition". If one memorizes in said way, they should hardly have to think beyond setting up the problem.
I'm not saying that I don't constantly use pattern-recognition to solve problems, indeed I do all the time, it's part of the experience of what makes a player stronger than another player in a given position. What I am saying is that "pattern recognition" is not how one solves a problem which has a certain uniqueness to the beholder - like you say, you might see a problem as hard because it requires a special problem-solving element for you, whereas someone else blindly solves the problem because they just saw that pattern yesterday.
I could go with the terms "pattern understanding", "pattern alertness", and even "pattern learning".
ReplyDeleteBut "pattern recognition" seems to herald some autonomous process, such as solving the problem with "love at first sight." There is an unconscious element, in the heat of the moment, to recognizing that something might be there, but to suggest that that also leads to the or an answer requires an additional step.
Some may be intuitively lead during the process of this second step at times yes, but that is not the same thing as saying to just recognize and you are done. I would say that one would need to absolutely memorize problems and their answers to make this recognition and your done step to even be possible.
To me pattern recognition is the key factor which helps me to FOCUS my attention towards important idea. Sometimes I can recognize the position very quickly, but the pattern alone is not enough to find the solution (at least immediately).
ReplyDeleteI have a dozen of positions I recognize their patterns, but I always have to create the solutions, because these specific positions are "too weird" to be memorized (with solutions of course). And there are some I can recall from memory very quickly, even if they might seem a bit difficult for others.
I think it will be a great idea to discuss "Woodpecker method" and share our opinion what we think of this.
I share the links to the Quality Chess blog related to The Woodpecker Method by Axel Smith and Hans Tikkanen:
Woodpecker Wednesday – Week 1 recap
http://www.qualitychess.co.uk/blog/6610
Woodpecker Wednesday – Week 2 recap
http://www.qualitychess.co.uk/blog/6624
Woodpecker Wednesday – Week 3 recap
http://www.qualitychess.co.uk/blog/6628
Woodpecker Wednesday – Week 4 recap
http://www.qualitychess.co.uk/blog/6651
What is the woodpecker method, other than to study a lot of tactics?
ReplyDeleteI'm still studying some opening books right now, so I only spend so much time with tactics at the moment, but I do still spend some time on them each week.
@ LinuxGuy asks: "What is the woodpecker method?"
ReplyDeleteLink: The Wooodpecker Method
There's a link to where you can preview part of the book.
Short answer: It's a reincarnation of the Michael de la Maza "Seven Circles" approach from Rapid Chess Improvement, endorsed by GM Axel Smith and GM Hans Tikkanen, co-authors. As you surmised, it's a "structured" way to study a lot of tactics repetitively in decreasing time increments. There's a free Excel spreadsheet to help keep track of your progress.
I really like Quality Chess books in general, but this is one I am leaving unpurchased. I already have MdlM's Rapid Chess Improvement. I think this approach has pretty much been "debunked" by the collective Knights Errant over time. (Go back to the start of Temposchlucker's blog to follow the attempts to make it work.)
Thanks, Robert!
ReplyDeleteI agree with your conclusion. When I lose a chess game, it's either I am just too tired and miss a check, I get over-powered by a _Master_ level attack/tactics, or I simply don't understand the position as well as my higher-rated opponent. This is why I liked what you said about moving from evaluation to evaluation of position, and how key a factor that is (in finding best moves)!
I am always trying to keep my chess book collection from growing out of control read-sell/donate, repeat. Right now, I am going through Averbahk's book on Tactics for Advanced Players. I am surprised this this book has a definite angle toward tactical studies, as if here were training us to be composers of tactical studies! So, long story short is, this is chess of course, so there will always be tactics that we are studying regardless of whether we are reading a book on endgames or openings, and if we are fussing over positional evaluations, then as you have alluded to, so much the better!
Yes, I have plenty of books on tactics already also. :-)
The performance on a a ste of tactical puzzles at a second circle improves depending on the
ReplyDelete-"complexity" of the puzzles
-the size of the set
-the amount of time spend for each puzzle ( intensity )
-the time in between these both attempts
The woodpecker method wants you to solve ~1000 puzzles ( thats ~MLDM) in 4 weeks, then in 2 weeks, then in 1 week,... ,1 day
Ok, a increase in speed by a factor 2 would be a increase in "performance ON THIS SET" by 100-200 elopoints, thats the result of empirical rabbits an my analysis. My personal experience is that my typical improvement on a puzzle seen after 8 days the second time is very small. After 28 days my performace would be virtually identical.
Someone who is able to be twice as fast at a set of 1000 tactic puzzles in the second circle 28 days later is already a fantastic tactician. They need to have already plenty of the necessary chunks.
The blogposts are done by International Master Andrew Greet. It will be interesting to see how he is improving at each circle. But as usual we dont have any hints how good he is at tactics in the beginning and how much he will be better in tactics after these circles.
He is now in the second circle and writes : "Another interesting point is that I’ve been using the board and pieces less that I expected, simply because my memory and recognition have kicked in such that I know almost immediately which part of the board I should be focusing on, and from there it’s just a matter of working out the details."
A tactician with 100-200 more tactical-elopoints is twice as fast in tactics and it seems he can memorize tactical puzzles twice? as good.
Wow, that is a short period of time. It sounds physically unhealthy to some extent, because one would have to go, I would imagine, long periods of time without breaks, and even then to keep those breaks short still.
ReplyDeleteI think that type of endurance should lead directly to a ratings improvement, but few have that type of endurance. I would think that few players work on their chess endurance, and many are held back because of that.
here is the key why we dont get better i think:
Deletehttp://www.qualitychess.co.uk/blog/6668#comment-370485
1st cycle: 1033 minutes; 88.9% accuracy
2nd cycle: 663 minutes; 93.7% accuracy
Andrew Greet does learn chesspuzzles much much much much better than i do
playing chess is in a way learning the positions of the game during the game. you need to be aware of plenty of pattern in the positions and step through your lines in your mind which is a mater of memorizing too
But how to improve the chess learning ???
I often compare learning chess with learning a language.
ReplyDeleteAs it happens, I am learning currently learning 2 languages:
polish and french.
In the past, I did several attempts to learn polish, meanwhile I am married 17 with my polish wife.
I made some progress, but really, when it gets really difficult in those language courses/books (around "year 2", when past and future is introduced), I never really made it.
I'd say, I am a "A2" in polish (A1 according to the "common european framework of reference for languages" = CEFR scale).
With french, I had it in school for 1 year, but that is almost 30 years ago.
Now, because of my son learning french at school, I started learning french with him, and restarted from scratch.
Again, after 2 years of learning and covering the same as my son covers at school, I might be a A2 in reading, but rather a A1 in listening, writing and speaking.
So what is the problem, and how does that apply to chess, too?
I learned a lot of vocabulary, like I learned a lot of patterns in 1500s rated blitz chesstempo tactics.
I can recall a lot of the vocab fast, same like I see patterns pretty fast, too.
But when it comes to construct sentences, when it comes down to speaking - I suddenly am very unsure. I have trouble to conjugate verbs accurately, and my grammar is a mess.
I tried to chat a little bit with a french boy after a game at lichess.org, but he had trouble to understand my simplest sentences. When the rather basic chat was over, I had more time to think where I had done mistakes, and I did a lot. If my chat-partner had been an adult, he probably had understood me better, but children are really a bit unexperienced when you dont use the language correctly, and thus a boy can hardly understand me.
I consider more difficult sentences about the same like difficult tactics. Patterns are certainly in place, sometimes, but other elements of the difficult tactics are pretty "weak" when it comes down to identify a clear pattern.
It is certainly not enough to know the vocabulary (to know the patterns). You really need to learn ready made sentences, and you need to understand these sentences with temposchluckers so called "deep understanding".
When I tried to look up complete sentences, my impression is, that there are too many sentences to be learnt. It really gets blurry as soon as your sentence is longer than 5 words.
I am afraid I hit a wall in french. But I do have the impression, that my son is able to improve. He isnt able to produce more sophisticated sentences, too. But different to me, he has better ideas to formulate his points with very simple sentences. Often I think "I could have had the idea to produce such a simple sentence", but the thing is, I did not, and instead tried to explain my points in a very rigid and complicated way.
Transfered to chess - we are probably trying to use to many rules and patterns when solving a tactic. The trick is to think of only the relevant bits, but since we do this unconciously, I have no cure. Which means I wont improve in chess tactics, and I will never speak french above A2 or B1.
It might be a matter of age, I suspect? We are unable to think simple, unable to adjust. We try to make sentences work, trying to find/recall the one word. However, when my son can not think of this one word, he finds quickly a simple way to express his ideas with other words. Whereas we stick to the same rigid sentence we have in mind. In chess, I am afraid we want to make tactics work which just dont work, and are unable to use a different pattern approach.
The problem with being an adult is that we've learned to demand quick solutions for problems, and so we presume that "getting better at chess" is learning all of these solutions, these "pattern tricks".
ReplyDelete“If I had an hour to solve a problem I'd spend 55 minutes thinking about the problem and 5 minutes thinking about solutions.”
― Albert Einstein
The adult thinker naturally wants to skip a step. Why should I spend 55 minutes, when all I want is the 5 minute answer? This works for many tasks, particularly those which don't involve us (call the plumber, problem solved). The thing is, in chess, we _are_ the plumber.
The reason that person is probably solving the problems so much faster in their subsequent circle, is that the problem solving is close together enough, and in any case the solver has evidently learned to memorize the solutions.
I think it's funny in their post how they talk about "solving time". Give me a break, it's more like recognition time. You are probably trying to resolve problems, where he knows that is a fools-errand and has gotten really good at memorizing problems.
I've gotten to this state as well, where I can use memory and recognition to avoid having to constantly resolve the same problems. This goes back to the advice of "memorizing 50 Master games". It's important to learn to memorize positions, move orders, sequences of moves, so that the recall of a seven move problem would appear automatic.
Solving a problem, once one is already proficient at the visualization part of chess, is about stitching together ideas. Ideas take time. If one remembers these ideas, the amount of time spent on a problem, relatively disappears to how much time was once taken.
Its already clear that a stronger tactician solves a tactical problem faster ( at average ). A tactician with 100-200 elopints more will solve the same set of puzzles in about half the time. They recognise more things ( and faster?).
DeleteBut what suprises me is, that this IM still has some imformation of these ~1000 puzzles he saw "28" days ago in his memory. He solves these puzzles the second time way faster und more precise. Both effects support nicely the chunk hypothesis.
The other question is: IF this method is good, then how to break it down for us mortals.
I guess the answer should be: smaller sets with easier puzzles and more repetitions
But that is what munich did, the irritating fact is, he did not improve in tactics, at least not a lot.
So.. we need to hope that tempo does find anything..
There are less than a thousand tactics diagrams (positions from famous games) that I see over and over again in various books. There comes a point when a "westerner" (in ancient east times, memorization was considered a more critical skill) realizes that it is far easier to memorize all of these positions than it is to re-solve them. Actually, if you have known people from India, you will see how they try to memorize when they learn. Chess visualization, IMHO, builds up memory anyway, so top players have built up their memorization ability.
ReplyDeleteStronger players calculate more quickly, but if only calculation speed is taken into account, then this is mostly relevant at faster rather than classical time-controls, except where one is already in time-pressure.
28 days is nothing, if the "pain" is still fresh. If you suddenly held the opinion that memorization is 10x more relevant than you think it is now, I can almost guarantee you'd memorize these problems much better, and for much longer.
There are words in my vocabulary that I rarely use, but I haven't forgotten them. The difference is that everyone knows that vocabulary has to be memorized, but they assume that tactics don't need to be, and that patterns are only subconscious.
If you don't try to memorize a tactical puzzle at all, then your brain will likely assume there was nothing new to understand, and then it will be just as lost the next time it sees the puzzle.
Aox (see above) is right. I did not improve in tactics (a lot).
ReplyDeleteInstead I improved by approaching chess with new knowledge, new insight. More guidance rules. A guidance rule is for instance: "A knight on the rim looks dim".
Together with my slightly improved tactic skills, I improved a lot. It is difficult to quantify which part of my training contributed how much.
By tactical pattern traing, I guess I imporved about 100-150 elo points.
The other 250-300 elo points I gained by playing openings, middle games and end games the "right way": here and there little rules that guide me through my games.
Examples: "to take is a mistake", "keep your options" (= threat is stronger than its execution), "rooks are far distance weapons, keep them away from the center", "best defender for the king is a knight", a lot of endgame rules based on pawn structures, chosing statistically promissing opening moves and I tried to figure out why some moves are more promissing than other opening moves, and if I am out of book - I have found rules that guide me how to find good opening moves.
I discovered lots of middle game rules, and - this is probably important - I am only thinking of those rules which are important in a certain position, and think of other rules when the position is a different one. I am never thinking of all rules, so I do have the ability to know what is relevant to think of, and what is not. However, if you ask me: "How do you know that your rule "rooks dont belong in the center" is not relevant? Sometimes the answer is simple: "Because in the current position there are no rooks which I could place into the center".
But sometimes it is not simple: "I can not tell for sure when to think of some rules and when not", and I can not even tell if I should have considered some rules during the game which I did not think of. But I should have? Lucky me, he blundered later?"