Weird additions
After a little break, it is always a bit difficult to get back on track again. But let's give it a shot.
I was busy to reshuffle the tree of scenario's
It is logical to start with determining your targets. A target is always (nearly) outnumbered. What does that mean? It is either:
4r1k1/pqr1pp1p/p1Nn2p1/3P4/5QP1/8/PP1R1P1P/4R1K1 w - - 1 1
[solution]
We already talked about this position before. The point is, that the black queen has lack of space. But I did not see that. Only when I realised that b8 is a convergence point of the second order of the white rook and knight, I saw that the black queen is actually a target.
But that leads to a weird addition to the scenario which handles the lack of space (see the red ellipse in the first picture). It doesn't look elegant, but I can't do without the addition. The guide I'm developing must somehow lead my attention to b8.
Another position where I couldn't find the targets was this:
2R5/3P1pkp/5bp1/1q2N3/p7/6BP/5PPK/8 w - - 1 1
[solution]
The reason that I couldn't find the targets here, is because they are not on their target squares yet. A few preliminary moves are needed. I have it not worked out yet, but I smell a weird addition already.
I will update the post when I have worked it out. Feel free to comment already.
I was busy to reshuffle the tree of scenario's
It is logical to start with determining your targets. A target is always (nearly) outnumbered. What does that mean? It is either:
- Hanging (outnumbered, LPDO) or
- B.A.D. (Barely Adequate Defended, nearly outnumbered) or
- Of high value. A king is always outnumbered when you attack it
 |
| Diagram 1. White to move |
4r1k1/pqr1pp1p/p1Nn2p1/3P4/5QP1/8/PP1R1P1P/4R1K1 w - - 1 1
[solution]
We already talked about this position before. The point is, that the black queen has lack of space. But I did not see that. Only when I realised that b8 is a convergence point of the second order of the white rook and knight, I saw that the black queen is actually a target.
But that leads to a weird addition to the scenario which handles the lack of space (see the red ellipse in the first picture). It doesn't look elegant, but I can't do without the addition. The guide I'm developing must somehow lead my attention to b8.
Another position where I couldn't find the targets was this:
 |
| diagram 2. White to move |
2R5/3P1pkp/5bp1/1q2N3/p7/6BP/5PPK/8 w - - 1 1
[solution]
The reason that I couldn't find the targets here, is because they are not on their target squares yet. A few preliminary moves are needed. I have it not worked out yet, but I smell a weird addition already.
I will update the post when I have worked it out. Feel free to comment already.
Diagram 2 was easy for me
ReplyDeleteMaterial: Black is up a queen for a rook and a knight
What is the tactic about? Checkmate, Material or Promotion?
Promotion of the d7!
But a4 might promote too so d8:D Bxd8 Rxd8 is seemingly not enough
The bishop has to be removed from the d8 diagonal -> Ng4 is the first candidate and i saw the knight fork at f6 ( a typical endgame tactic )
Now i had to realize there is nothing else and so it did take me 1:05
OTB, and at first, I would be trying to win a pawn, but then I notice how problematic that is. I know there must be a solution, and so I solve it in about seven minutes or less.
ReplyDeleteIf I didn't know that this was a problem in need of a solution, I may not have discovered the answer (say, if playing OTB). This was a problem where just knowing there was an answer made it infinitely more easy to solve.
It is difficult to get any grip on diagram 2. It is a beautiful example of preliminary moves. The attacker must be brought into contact with the attacking square (f6) with tempo, and the targets must be lured onto their target squares (d7 and g8). Three preliminary moves are needed before the duplo attack becomes visible. The only idea I have discovered so far is that the target squares must be on the same colour, if the attacker is a minor piece or a pawn. Two desperado's are needed to get the targets in place, and one tempo move to get the attacker in place.
ReplyDeleteWhat characteristics of the position could help us out? What could help to guide our attention in the right direction?
I assume that the right place to look for preliminary moves, is in the "when everything else fails" section. That doesn't feel very satisfactory.
"What characteristics of the position could help us out? What could help to guide our attention in the right direction?"
ReplyDeleteThe primary difficulty is that parts of the solution, particularly the first move, are non-forcing and counter-intuitive.
Our general expectation for combinations (based on experience in solving lots of problems) is that the first move "should be" a tactical blow. Obviously, this is one of those general "rules" that do NOT apply in every situation. The process of elimination pretty quickly reduces to moving the Knight. Since the only potentially useful "attack" available is an "attack" on the Black Bishop, the first move is (somewhat) "telegraphed."
2. Ng4, the first move, (apparently) abandons White's "trump card" - the d7 Pawn. But, it "seals" the escape of the Black King. I quickly "saw" that the d7 Pawn could be captured by 2. ... Qxd7 (another apparently non-forcing move), and then the subsequent 3. Rg8+ Kxg8 4. Nxf6+, and White should be able to stop the a4 Pawn easily with the Bishop.
I then got sidetracked into considering 2. ... Be7, "preventing" that specific combination but allowing the d7 Pawn to promote. It took some time to finally realize that White can play 3. Be5+ f6 4. Bxf6+ Bxf6 5. Nxf6 (protecting the d7 Pawn). After 5. ... Kxf6 6. d8(Q)+ wins. 5. ... Qe5+ 6. g3 Qxf6 (hoping to capture on f2 with a perpetual check) 7. Kg2 and Black has no checking square; White will promote the d7 Pawn next move with threat of mate on f8. White should have no problem taking care of the a4 Pawn, especially if he can exchange the Queens.
This type of non-forcing problem is very difficult for me!
Puzzle #2, I spent a record 1hr 26 minutes on Chesstempo solving it, and spent time before that looking at the initial diagram. I did solve it correctly in the end, but started out by analyzing the imbalances, a host of wrong candidate moves, hanging pieces, etc. The Ng4 candidate move never jumped out at me, and it's surprising how long it took for me to even seriously consider it. Then I found the "tactic" yes. This is one of those puzzles, and OTB I imagine it's the same, where you could see it in a blitz game if you have an eye for the tactic, and miss it in a correspondence game if you don't.
ReplyDeleteThe funny thing is that I originally saw 1.Ng4 Qd7, 2.Bf4 QxR, 3.Bh6+! mates. Here, 2...h5 leads to the fork tactic in the game, but 2...g5 stops this pretty mate (but not the tactic) After finding the tactic, I still had to convince myself 1.Ng4 Be7, 2.Be5+ f6, 3.Bxf6+ (removing the defender) BxB, 4.NxB Qe5, 5.g3 Qf5, 6.Ng4 h5, 7.e8(Q), but finding the tactic was at least 95% (more like 97% or 98%) of the battle, as a strong calculator can see the line above fairly quickly.
Even close to the end, I was tempted to play 1.Re8 first, to take away 1...Be7, but 1...h5 would then take out 2.Ng4. It's a 1948 rated tactic, which is pretty high for Chesstempo.
ReplyDeleteA challenging problem. The tactic part of the solution was key, but there was still enough calculation involved to keep us calculators happy too!
Great analysis of the problem with finding the problem, Robert!
I hope that I have made clear what I try to accomplish. I use the tree of scenarios in practice. Then I try from the feedback to add scenarios for positions I fail. Then I reshuffle the tree of scenario's in order to make the order logical. In the end, redundant branches will be pruned.
ReplyDeleteIn diagram 1, I missed the fact that the black queen has a lack of space. Then I asked myself, what should Robert Cobles Chess Module shout in my ear in order to see that the queen has a lack of space? Answer: b8!
The moment that my attention makes contact with b8, it immediately pops up that the black queen has nowhere to go. Then I ask myself, which question leads to the answer b8? That question is, where do my pieces converge? The white rook and knight converge at b8!
In fact, the question has become: what are the convergence squares of the 2nd order? (points of pressure are convergence squares of the 1st order) That is awkwardly formulated, but since I know what I mean by it, I don't care.
I place that question in the space department. Which is a weird place. Hence the red ellipse in the scheme. But after some thinking, I now know where to put it, so that it has a logical place. I will tell you later.
Then I try to do the same with diagram 2. What should the RCCM shout in my ear? g8 and d7!
What questions would lead my attention there? g8 is a point of pressure. I already have a question for that. But what about d7? d7 is in fact a point of pressure for the enemy. If the target is not on the target square, it can be lured onto it when it is a point of pressure for him. So at some point, I must ask the question on which squares do I have a B.A.D. (Barely Adequate Defended) piece? Followed by the scenario "can I make that target square subject of a duplo attack?
Now this is a bit awkward and complicated, alas. That's why these questions should be asked in the section "when everything else fails". First I must interrogate the position for straight forward combinations. If that doesn't lead to a satisfying result, I must look for the preliminary moves which obscure the combination.
good Stuff. Thinking in Squares vs Pieces is a higher order of thinking that you are doing here.
ReplyDeletere: I spent a record 1hr 26 minutes on Chesstempo solving it..Wow!
That determination to solve a difficult problem regardless of how long it takes is one of the distinguishing characteristics of the potential master!
ReplyDeleteAs skills progress toward higher orders of thinking, the distinction between tactics and strategy will blur into irrelevance.
Please consider the following dissertation to be a supplementary abstraction related to the current investigation.
PART I:
Last night, I was re-reading portions of Thinking, Fast and Slow, Chapter 22: Expert intuition: When can we trust it?, Intuition as Recognition The following excerpts caught my attention. I'm going to paraphrase them somewhat, in order to move it from the given example context (firefighting) to chess.
". . . how the players could make good decisions without comparing options. The initial hypothesis was that players would restrict their analysis to only a pair of options, but that hypothesis proved to be incorrect. In fact, the players USUALLY GENERATED ONLY A SINGLE OPTION, and that was all they needed. They could draw on THE REPERTOIRE OF PATTERNS that they had compiled during more than a decade of both real and virtual experience to IDENTIFY A PLAUSIBLE OPTION, which they considered FIRST. They EVALUATED THIS OPTION by mentally simulating it to see if it would work in the situation they were facing. . . . If the course of action they were considering seemed appropriate, they would IMPLEMENT IT. If it had shortcomings, they would MODIFY IT. If they could not easily modify it, they would TURN TO THE NEXT MOST PLAUSIBLE OPTION and RUN THROUGH THE SAME PROCEDURE until an acceptable course of action was found."
"[Gary] Klein elaborated this description into a theory of decision-making that he called RECOGNITION-PRIMED DECISION (RPD) model, which applies to firefighters but also DESCRIBES EXPERTISE IN OTHER DOMAINS, INCLUDING CHESS. The process involves both System 1 [subconscious; intuitive; parallel] and System 2 [conscious; logical; sequential]. In the first phase, a tentative plan comes to mind by AN AUTOMATIC FUNCTION OF ASSOCIATIVE MEMORY—System 1. The next phase is a deliberate process in which the plan is mentally simulated to check if it will work—an operation of System 2. The model of intuitive decision making as PATTERN RECOGNITION develops ideas presented some time ago by Herbert Simon, perhaps the only scholar who is recognized and admired as a hero and founding figure by all the competing clans and tribes in the study of decision making. I quoted Herbert Simon's definition of intuition in the Introduction, but it will make more sense when I repeat it now: 'The situation has provided a cue; this cue has given the expert access to information stored in [long-term] memory, and the information provides the answer. Intuition is nothing more and nothing less than [pattern] recognition."
This process is very similar to GM Jonathan Tisdall’s variation processing approach which he contrasts to GM Kotov’s thinking process. Instead of identifying ALL candidate moves up front (Kotov), when attacking, one first eliminates as many “easy” solutions as possible, then picks the most difficult move (System 1) and then go forward, working out the variation to the end. When defending, one picks the apparently most critical move, and determines if it can be survived. If so, it is likely (NOT PROVEN) that the position can be defended. During this process, key points (potential tactical themes/devices) will surface in the consciousness. NOTE THESE for possible use in later calculations!
"Acquiring Skill"
"How does the information that supports intuition get "stored in memory"?
Aye, there's the rub, mates!
PART II:
ReplyDelete"Certain types of intuition [emotional learning] are acquired very quickly. . . . Little repetition is needed for [emotional] learning.”
“Emotional learning may be quick, but what we consider as “expertise” usually takes a long time to develop. [NO JOKE!] The acquisition of expertise in complex tasks such as high-level chess, professional basketball, or firefighting is intricate and slow because EXPERTISE IN A DOMAIN IS N-O-T A SINGLE SKILL BUT RATHER A LARGE COLLECTION OF MINISKILLS. Chess is a good example. An expert player can understand a complex position at a glance, but IT TAKES YEARS TO DEVELOP THAT LEVEL OF ABILITY. Studies of chess masters have shown that at least 10,000 hours of dedicated practice (about 6 years of playing chess 5 hours a day) are required to attain the HIGHEST levels of performance. During those hours of INTENSE CONCENTRATION, a serious chess player becomes familiar with thousands of configurations, each consisting of an arrangement of RELATED PIECES THAT CAN THREATEN OR DEFEND EACH OTHER. [The current investigation addresses this specific idea!] . . . In chess, recurrent patterns of interacting pieces play [a similar] role of letters, and a chess position is [comparable to] a long word or sentence.”
“ Acquiring expertise in chess is harder and slower than learning to read because there are many more letters in the “alphabet” of chess and because the “words” consist of many letters. After thousands of hours of practice, however, chess masters are able to read a chess situation at a glance. The few moves that come to their mind are almost always strong and sometimes creative. They can deal with a “word” they have never encountered, and they can find a new way to interpret a familiar one.”
PART III:
ReplyDeleteI disagree with this assessment of the chess “alphabet.” IMHO (following Momir Radovic), the DYNAMIC chess alphabet consist of four “letters.” Note that these describe the interacting “functions” occurring between pieces and squares.
A = Attack
B = Block
C = Constrain (Restrict)
D = Defend
DNA consists of only four “letters” but is capable of billions (trillions?) of combinations (not so odd that the same word describes dynamic chess “interactions”). The same complexity based on simplicity is true of the chess “alphabet.” Each complex combination is the multiplication of many simple combinations of a very simple 4-letter alphabet.
The virtue of the PoPLoAFun process is that it starts with the chess alphabet, and then rapidly progresses toward higher and higher levels of abstraction. At some higher level, there is a blurring or blending between tactics and strategy. The mind cannot help but AUTOMATICALLY try to make sense of a given position in terms that it can identify EASILY – and those terms are the four letters in the chess alphabet! First individual “words” are recognized, followed by tentative attempts to recognize phrases based on those words, then sentences, then paragraphs, then chapters, then books, then collections of books, then libraries. . . . Inherently, we are “meaning making machines” (to quote GM Rowson). This is why it is so important to FIRST begin with the vulture's eye view, without trying to immediately lock-in on a particular tactical theme/device. We need to give System 1 sufficient time to perform its recognition "miracle." If we jump past that recognition stage into calculating, we (usually) lose the ability to fly back up to the vulture's vantage point. I'm suggesting just simply "seeing" without trying to use or make sense of what you "see." The mind is more than capable of doing that without any logical assistance!
IMHO, the “problem” we often face when seeking to improve is that we (as adults) want to FORCE System 2 to logically “resolve” the situation, which causes us to ignore or shutdown any “advice” we might get from System 1 [the fabled Robert Coble Chess Module]. That’s not in any way intended to mean that we should shut down or refuse to use System 2. It’s just that we shouldn’t try to rely on System 2 exclusively. System 2 is relatively lazy, and would much rather not do any "heavy lifting" unless absolutely required to validate System 1’s “suggestions.”
We want to live in an idealized world. Our system 2 thinking tends to simplify matters in order to be able to make sense of the world. Hence we think that memory is good and forgetting is bad. Effort is good and laziness is bad. Out of your comfort zone is good, and into your comfort zone is bad etcetera etcetera.
ReplyDeleteBut the mind is a well balanced machine which balances the trade off of every mental activity very well. Every mental activity comes at the cost of energy. If that cost outweighs the benefit, your mind will not cooperate.
It is good to remember where you have left your car keys the last time, but it makes no sense to remember where you have left them three weeks ago. Hence forgetting is a blessing. If you remember the last 1000 times where you have left your car keys, all instances would interfere with each other and you wouldn't be able to make a choice. If you have ever walked to where you have parked your car two days ago in stead of to where you have left it yesterday, you will know what I mean.
Laziness has evoluted to save energy, so we shouldn't frown upon it so much.
Our comfort zone is developed for the very same reason.
The addiction to trial and error has shut down my ability to think logical about a position. The mind does so for economical reasons. If I want to circumvent that, I must find an intelligent way to tip the energy balance.
Using system II thinking to guide my attention is only a temporary solution meant for the study room only, of course. Like side wheels when you learn to ride a bike. Hopes are that system I takes it all over after some deliberate practice. Only then this method will have succeeded.
@Robert
ReplyDelete"The situation has provided a cue; this cue has given the expert access to information stored in [long-term] memory, and the information provides the answer. Intuition is nothing more and nothing less than [pattern] recognition" ( from part I )
The most active part of a master playing chess is their LTM. My current training is based on this idea. I experienced that "seeing a move" is really memory. I look at a position, dont have any clue, a move pops up i play the sequence and sometimes then i start to remember the puzzle. You can see this at many of these tactic puzzle solving videos at youtube. The ltm works as associative neural network, it detects pattern and creates "ideas". I think now that this is he reason why chess masters are playing still strong even when they play just blitz. Magnus Carlsen ( and others ) say that their first idea is usually the best move ( he says "always" ).
I don't buy this completely but i use it as an orientation : https://www.chess.com/blog/matzleeach/train-your-chess-skill-using-priciples-of-neuropsychology
I suspect that all the "chess philosophy" is just helping to memorise things better and quicker ( to the most ) and that the ltm can do that all by itself. Improvement in chess is then an improvement in the different aspects of chess memory.
That is the idea i am following at the moment, i hope to see a effect in maybe 2 years
@ Aox:
ReplyDeleteWould you please describe the process(es) that you are utilizing to train your LTM?
How do you define and select specific training patterns to be remembered?
Are you using training software? If so, is this commercially available software or your own creation?
Thanks!
I do something like this this http://chessok.com/?p=21207
DeleteEspecially at the moment only this : http://chessok.com/shop/index.php?main_page=product_info&cPath=7_26_29&products_id=220
Here a video about peshky : https://www.youtube.com/watch?v=Vn9zcsr_Ujw
( the software is a little buggy but usable )
Peshka allows multiple users and is rating each user separatly. I "am" 6 different users : u1..u6
u1 does see the tasks the first time with a rating of ~2200 ( i was puzzleing with this software before from time to time a little ) , u2 the second time with a rating of ~2500 ( i dont think it makes sense to go further, i cant understand anything at a level +300++ anyway ) ... u6 the 6. time with a rating of 2550. The ratings suppose to increase slightly. With easier courses i will look at the time spend for solving. I write in a text file what i did solve as u1 at each day, every day a new line. And i search in my library and at youtube if i can find these games with annotation and read/watch these, i dont find them that often though.
u1 is solving at day no 1, u2 is sloving at day no 3 ( one day in between ), u3 at day no 8... u6 at day 33. so i have the task at least for 33 days in my brain. This method is related to the method of chessimo, but chessimos workload is sometimes low and sometimes extreme. Read about supermemo at wikipedia. With my method the workload is every day about the same. I do the repetitions first (u2=repeating of what if learnd new 2 days ago..u6) and then according to the time left i do some new tasks as u1.
The tasks are : to find several requested moves in several lines in a middlegame of a famous player
I guess it might have been better to start with endgames? But i was more interested in strategy.. which is 80 ??? % tactics.
Its interesting that th real 2200 player in the article at chessok did learn the material at least!!! 4 times faster than i. So i guestimate now : that chess-ability is strong correlated in chess-learning-speed
I was reading about Manus Carlsons improvement. His ratingjump of several hundred elopoints to 2000 was with 9-10. Ath this age he had once a week a training with other kids analysing a master game. Soon Mangnus did get much better than the other kids in his group. Magnus did like learning things by heart, like the capitals of countrys. These days Magnus knows more than 10 000 mastergame by heart.. every single position.
I speculate that magnus did start learning chessgames by heart with 9 and that this was the reason of his improvement.
Well, fuzzy speculation, but i give it a try. I hope "the better method" can compensate "the better brain"
@ Aox:
DeleteThe only Peshka module that I have is CT-ART 4.0. I got it in anticipation of doing the Seven Circles. However, the Knights Errant convinced me that the MdlM "method" could not produce the desired results before I got started, so I've really not used it as much as I could have. Just for fun, I created a new player with an initial rating of 1800. After 35 problems, my rating is 1992. My original rating (way back when) is 2133. I don't think there is any question that this does NOT correlate to any OTB rating.
Are you following the SuperMemo2 algorithm for the spaced repetition cycles? When I did the calculations of the intervals, it appeared that you are using EF = 2.5 factor in the equation, based on the intervals for each player (u1..u6).
I have Anki, but it seems to me that the effort to breakdown and enter a sufficient number of training examples on any subtopic in chess is hardly worth the massive effort required. If there was a way of automating the process, I might be tempted.
Since the consensus for long-term improvement to the master level is correlated strongly with studying master games, I can't argue with the idea of "memorizing" master games per se. However, I think there is more "bang for the buck" for people with limited time for study in the ideas currently under discussion. I would love to see software that would automatically extract training positions from a large database of games based on the specific weaknesses of the individual player. That implies the capability to identify and extract useful patterns from games, and the capability to test and identify the weaknesses that need to be trained. [No, I don't have either the time nor the inclination to take on that task! I can assure you that those ideas are NOT an integral subcomponent of the Robert Coble Chess Module!]
Is there possibly more than one road to Nirvana, or is there only one true path? Beats me. When you don't know where you're going, any road will take you there - maybe.
CT-Art 4.0 is not checked by computer as far as i know but CT-Art 6.0 is ( but there are not many errors in CT-art anyway )
DeleteThe ratings of peshka are different even between different courses, i guess the rating of each problem was simply set by the author somehow.
The 7 circles are not working because the space between the 1. and the 2. pass is too high. An other problem is that CT-art does not have enough tasks / examples
and they might be to complicated for many. The transfer from a learned puzzle to a random other puzzle is microscopic, especially if you learn the puzzle without the right understanding.
I start u1..u6 with any random rating..after a while my u1 rating did stabilize at ~2200, ... u6 at ~2550.
The spaces between my repetitions where not calculated but are based on experience. I did want! the rating of every "u_x" to increase a bit. So this is not precisely any Supermemo. The spacing dont change, i see every puzzle at day 1,3,8 and so forth if i solved it easy or not. But i repeat my blunder at the same day until i can do them without error and i try to understand the task by eventually reading the annotation i can find or analyzing it with an engine
I want to have some "resitance" at every repetition because i hope to improve the search process and not the "picture of the position" to "movement of my arm". I dont want to memorize literally.
"I would love to see software that would automatically extract training positions from a large database of games based on the specific weaknesses of the individual player."
I think i know how that can be done. The training positions where the weaknesses are are the training positions where we are slow or blunder. So the program would do a guess the move for many games and special repetitions at personal weaknesses.
Of course a training of a special type of position is not possible this way. Thats why i do the strategy 3.0 course and will continue with other courses and some books with this method. I need to increase my learning speed somehow, https://chesstempo.com/chess-statistics/cmuroya17 for example did memorize puzzles muuuch better than i ( well he is muuuch younger too ;)
I feel like Ranting for a moment...I think that almost anything we know or learn can be labeled pattern recognition driving a car or walking to a destination. We dont do much that is pure unexplored thinking territory. Chess or anything else tends to be based on things we have done-seen in the past. So is finding the sitting duck comes as a intuitive bolt of lightning or a rather concious marching through steps like a Wheres Waldo puzzle. Does the concious marching through steps lead to the bolt of lightning intuition? This is the 7 circle arguement. Note this rant is not directed at anything specifically said here just things I am thinking about. How do we improve Chess intuition?
ReplyDelete@ Takchess:
DeleteJim, you're right that anything "can" be labeled as pattern recognition - which would render the phrase meaningless. However, I think the "conscious marching through steps" (a thought process, if you will) is a fundamental, preliminary requirement for developing intuition. As Temposchlucker has stated, this conscious adherence to a thought process is of value only in the study room, NOT WHEN PLAYING A GAME OTB. It is the repetition of those conscious steps that (eventually, hopefully) leads to an intuitive "sight" of the relevant patterns and potential moves in a given position, without loading STM/working memory and trying to figure out what to play using that conscious, logical step-by-step sequential S-L--O---W process. To use Temposchlucker's analogy of learning how to drive a car: you first have to apply conscious effort to "memorizing" (H/T to Aox) the logical sequence of steps to be done for each subtask of driving using system 2 (conscious, logical, sequential, step-by-step, therefore SLOW). To become an accomplished driver, a large number of those subskills must be practiced repeatedly until they become subconscious AND integrated together, so that they can be performed simultaneously by System 1 (subconscious, intuitive, massively parallel processing, therefore FAST).
That is one of the foundational assumptions underlying the current investigations.
One of the major problems facing adult players seeking improvement is the (erroneous) "assumption" that familiarity with the "concept" (knowledge) is equivalent to mastery of a particular skill or subskill. As long as you have to utilize any sequential step-by-step process and keep it in mind while trying to USE that skill, you have not acquired the skill. The capability to verbalize and regurgitate a DESCRIPTION of the "concept" on command is NOT the same as the skill to USE that knowledge to play better! The only way I know to grind a process into LTM/intuitive memory is to practice it over and over until it becomes a part of your DNA.
I'll repeat what I posted above (in part):
The acquisition of expertise in complex tasks such as high-level chess, professional basketball, or firefighting is intricate and slow because EXPERTISE IN A DOMAIN IS N-O-T A SINGLE SKILL BUT RATHER A LARGE COLLECTION OF MINISKILLS. Chess is a good example.
I know I have seen a significant increase in my "sight" as a result of breaking down the preliminary vulture's eye view into a quick scan of the board for Points of Pressure, Lines of Attack, and the Functions of each and every piece. (Once the ideas become ingrained, there is an automatic elimination [with NO conscious thought to do that] of irrelevant pieces/squares from consideration.) Those ideas evolved from a consideration and exploration of Dr. Lasker's notion of motifs being separate and preliminary to "looking" for tactical themes/devices. It sounds terribly inefficient when you write out a linear description of the process, but simply following the thought process consciously will very quickly habituate the mind to move along those steps without conscious thought. When you no longer consciously think about the process AS A PROCESS while using it, I would be so bold as to state that the "intuition" has been properly trained to "see" the important things in any given position. It is still surprising (to me) how often I can "see" the available tactical themes/devices concurrently once the PoPs, LoAs, and Funs begin to be identified. They "pop" into the conscious mind as if by "magic" (intuition).
I am still working on transitioning from conscious to totally subconscious "thinking" based on these ideas.
PART I:
ReplyDeleteI'd like to "connect the dots" between the "general principles" inherited from classical chess and the PoPLoAFun approach to the specific aspect of "pattern recognition" so beloved of all chess players, particularly psychologists and neuroscientists. It is by no means a complete connection; it is merely an observation I made last night while thinking about my previous comment.
Dr. Seigbert Tarrasch, in his book The Game of Chess : A systematic Textbook for Beginners and More Experienced Players, page 99, passes along a "general principle" regarding the juxtaposition of enemy pieces on the same line as the Queen.
"So ALWAYS be mistrustful of the massing of hostile pieces in the same line as your Queen! Away with her out of that line!"
Is this not an example of addressing the "line of attack" issue via a "general principle?"
Consider the following position, taken from Chess Life, June 2017, GM Andy Soltis's column "B2, or not B2?" (an obvious "play on words" based on Shakespeare's Hamlet soliloquy, "To be, or not to be?" He provides a bit of "received wisdom" regarding capturing the "poisoned Pawn" on b2: never, EVER take a Pawn on b2 with your Queen. He then proceeds to show why this is not always the case, quoting James T. Sherwin: "Why should I labor under antediluvian prejudices?" I digress.)
The position is from the game GM Leiner Dominguez Perez versus GM Levon Aronian, during the World Rapid Championships in Doha, Qatar, 2016.
[Game Link: Leinier Dominguez Perez vs Levon Aronian]
[FEN "4r1k1/2p1qp1p/6p1/2b5/PpQ1n3/1B2N2P/1PP2PP1/4R1K1 w - - 0 25"]
PART II:
ReplyDeleteLooking at the position, there are some immediate PoPs that jump into "sight" for both players. White has "pressure" against f7, e4, c5, c7, e7, and e8. Black has counter-pressure against f2, e3, and e1. Both Kings have "luft."
The LoAs are very interesting. White has the obvious line a2->g8 because the Black King sits on that line. White also has the not-so-obvious line e1->e8. Here's where Tarrasch's general principle regarding the Queen being "attacked" along a line comes into play. There is a concealed "threat" from the WRe1 against the BQe7 (and also against the BRe8 by extension). Obviously, there is a reciprocal "threat" against the WRe1 from the BQe7 and BRe8. (It is very important to "see" bi-directionally when identifying threats.)
The Funs involve the restrictions placed on the pieces. The WNe3 has the function of blocking the attack on f2 by the BNe4 and BBc5, which would capture a Pawn and fork the White King and Rook. The BNe4 prevents the line attack against the BQe7 and BRe8. There is a juicy square at f6 that would be nice for the White Knight, if it could get there, which would fork the BKg8 and BRe8. Opening up the line a7->g1 for Black by moving the WPf2 seems foolhardy.
All of the pieces except the WNe3 seem to be optimally placed at the moment. But, if the White Knight moves, there is that attack on f2 to consider. Perhaps there is a move that will maintain protection of f2 AND bring the possible "attack" against the Black Queen one step closer to fruition AND get the White Knight closer to the f6 square. These considerations (which occur very rapidly once the PoPLoAFun approach is internalized) lead to two potential candidate moves: 25. Nd5 (attacking the Black Queen) and 25. Ng4 (still protecting the f2 square). Both candidate moves move toward the f6 square, so that is not a differentiator between the two moves. I rejected 25. Nd5 simply because the f2 Pawn becomes pinned and the Black Queen is still mobile. So, 25. Ng4 seems to be the most promising line. Now there is a first-order threat to capture on e4, followed by the fork on f6. The tension suddenly spikes! The BNe4 cannot be protected by 25. ... f5 because the f7 Pawn is absolutely pinned. What can Black do? GM Aronian has the following FIDE ratings: Standard = 2793; Rapid = 2797; Blitz = 2753. He chose 25. ... Nd2?, counterattacking the White Queen. There followed 26. Qxf7+! and White will end up an Exchange and two Pawns ahead with an outside passed Pawn at a4 and 3:2 majorities on both the Queenside and Kingside. Given GM Perez's FIDE ratings: Standard = 2739; Rapid = 2803; Blitz = 2796, I think it is safe to say that he will convert the full point. Aronian resigned.
The “point” of this comment is NOT merely to show a nice combination. It is to show the intimate connection between fundamental concepts (PoPLoAFun) and the formulation of “general principles” (such as Dr. Tarrasch’s maxim to avoid allowing the massing of pieces on the same line as the Queen. Both approaches aim toward the same goal: to develop an intuitive “feel” for dangerous situations without calculating “I go here, he goes there; I then go there, he goes here; etc.” IN the case of the general principle, there is nothing to suggest specific moves in a given position. On the other hand, the PoPLoAFun approach DOES suggest specific moves AND narrows down the potential candidate moves.
Gotta love any approach that does all that!
We forget where we have left our car three weeks ago. Or even where we have parked it only two days ago. There is a reason for that.
ReplyDeleteMental resources are limited. It seems to work very much like a hard drive. When a file has been deleted, it will still be on the hard drive for some time. It isn't easy to access, but it is still there. Until that space is needed. Then the file will be overwritten. And the information in the file will be lost.
When you are young, the hard drive is mostly empty, and it can take a long time before a deleted file will be overwritten. I used to recall my latest game back then. Even when there was a half year between two games.
The reason we learn nothing from our efforts, is because the information is treated by the mind like where we have parked our car. Without a framework, we can't keep our learning experiences apart from each other. They soon will start to interfere with each other. After some time they are deleted, and even later they will be overwritten by new experiences. Only when we can hang them on a framework, we can save them, and the framework can be used for retrieval.
Spaced repetition is a nice idea, but without a framework it will be very difficult. And you have to be very careful with what you actually want to learn. The 7 circles of madness can be compared with wanting to learn to multiply fast by calculating sums like 1234 x 7654 = 9445036.
After 7 circles of 1000 calculations, you may have memorized 1000 sums, and still be lost when you have to calculate 9876 x 4356 because it is not in your database.
Memorizing 100k+ chess problems is not going to help either, since there are many more chess combinations. Even if you were able to memorize them, what you are probably not.
I once did play a vote chess game together with Kacparov, now a IM, back than a CM. Kacparov was extremly active at chesstempo ( and extremly improving) at the age of ~9 + a few years. At this vote chess we discussed the moves and i was asking kasparov often WHY this move is good in his opinion.. he could not answer, there was no framework. cmuroya17 is only a NM now.. should be soon a FM. He did learn several 10 000 puzzles ( he solves them now in 10-20 sec even if the average solving time for that puzzle is 5 or more minutes ). Now his standard rating is ~2800. He was able to learn the ct puzzles even with a space of 2 months between each repetition. I cant improve if the space is more than 8 days. He spend a lot of attention(=time) to each new puzzle. That helps to put it into the LTM, still, a fantastic memory.
DeleteI suspect that a framework is only a method to give a puzzle more attention and is pushing the puzzle that way a little deeper into the LTM. Well i use that method too. It simply feels better if you have the impression you understand the puzzle.
One framework of software like CT-art is that the puzzles are sorted by tactical motive. I do now strategy 3.0 now and learn atm the pros and cons of an Isolated Queen Pawn with way more than 100 examples. Thats a good framework imo.
I havent played-studied chess much in the last six years until recently. Now i just study and play 2 hours one night a week Otb at chess club. No internet games and no tournament play. After a short period of study, I feel my game and chess vision feels close to it did during my best days although it is hard to prove. In looking at my blog from 2006, there was a period where i did 40 plus circles of level 20 problems. It was not until I tried to understand more and think about why the tactics work did I see any real improvement. I am in agreement that looking for weakness >sitting ducks and looking around not just ahead + popaloafun is helping. That and study of tactics in gm annotated is helping the vision. That being said I still did not solve this first problem even though I read the previous blog post where it was featured. 8) interesting comments folks. I will have to set up Roberts position on my chess board and have a deep think. Cheers
ReplyDelete