Time consuming trivialities
I'm soaring for the third day now above the chessboard. Trying to never let my mind engage in any activity that causes me to loose sight of the whole picture. What I found, is that about 50% of my time is wasted by trivial moves. Moves of which I am inclined to think that "people of my level" should see them a tempo. But apparently my level is in fact due to not seeing these trivial moves immediately. It must look dumb (as usual) if I give an example, but no worries, I'm sure you have your own trivial moves that you overlook. Looking dumb has never been an issue to stop me from publishing.
3q3k/pRr1r1pp/2nNbp2/2p5/Q7/3P2P1/P3P1BP/5RK1 b - - 1 1
solution
This position is in the middle of a sequence belonging to the solution. Apparently I had forgotten my holistic approach, since it took me two (!) minutes to find the right move. What did I do in those two minutes? Good question! I was totally engaged in solving the problem of my overworked queen. I wanted to take the bishop on c6, but couldn't find a way to do so without loosing a rook. I must be stubborn, I guess, to try for two minutes something that is clearly impossible, if I had looked at it from a distance. But even at the moment my tunnel vision was broken, it took me some time to realize that Rxb7 simply wins a piece.
Robert Coble said:
We gain proficiency (transfer knowledge into skill) in levels. We (usually) do not acquire complete mastery of each lower level prior to being introduced to the higher levels. Why not? Because we assume that we have proficiency when we really only have familiarity, and because of that "feeling" of familiarity, we try to move up one or more levels of abstraction before we are ready to do so; we lose interest in the "familiar" level because it no longer seems challenging. Because we "feel" that we have progressed beyond the "simplicity" of the familiar level (in spite of considerable contrary evidence, such as NOT "seeing" what is actually simple, like in the below-2200 puzzles on Chess Tempo), we convince ourselves that we are ready for more complicated stuff. Alas, it is the "simple" stuff that must be mastered before more advanced skills become ingrained into our subconscious as skill. WE KNOW THIS BUT WE IGNORE IT!
And that is exactly what this position is about. Imagine that this overlooking of trivial moves happens in 50% of the cases! The other 48% consisting of moves that are only a bit less trivial. Leaving 2% non trivial moves. The trivial moves are familiar, but don't jump you into the eye, which they should.
When I wrote about "thinking backwards" in 2007, I tried the same holistic approach as I do now. The difference is that I focussed on problems with a 2200 - 2400 rating. From the reasoning that I must exercise what I want to learn. But evidently, that was a bridge too far.
In the mean time, soaring around hasn't made clear yet how to solve this issue.
 |
| Black to move |
solution
This position is in the middle of a sequence belonging to the solution. Apparently I had forgotten my holistic approach, since it took me two (!) minutes to find the right move. What did I do in those two minutes? Good question! I was totally engaged in solving the problem of my overworked queen. I wanted to take the bishop on c6, but couldn't find a way to do so without loosing a rook. I must be stubborn, I guess, to try for two minutes something that is clearly impossible, if I had looked at it from a distance. But even at the moment my tunnel vision was broken, it took me some time to realize that Rxb7 simply wins a piece.
Robert Coble said:
We gain proficiency (transfer knowledge into skill) in levels. We (usually) do not acquire complete mastery of each lower level prior to being introduced to the higher levels. Why not? Because we assume that we have proficiency when we really only have familiarity, and because of that "feeling" of familiarity, we try to move up one or more levels of abstraction before we are ready to do so; we lose interest in the "familiar" level because it no longer seems challenging. Because we "feel" that we have progressed beyond the "simplicity" of the familiar level (in spite of considerable contrary evidence, such as NOT "seeing" what is actually simple, like in the below-2200 puzzles on Chess Tempo), we convince ourselves that we are ready for more complicated stuff. Alas, it is the "simple" stuff that must be mastered before more advanced skills become ingrained into our subconscious as skill. WE KNOW THIS BUT WE IGNORE IT!
And that is exactly what this position is about. Imagine that this overlooking of trivial moves happens in 50% of the cases! The other 48% consisting of moves that are only a bit less trivial. Leaving 2% non trivial moves. The trivial moves are familiar, but don't jump you into the eye, which they should.
When I wrote about "thinking backwards" in 2007, I tried the same holistic approach as I do now. The difference is that I focussed on problems with a 2200 - 2400 rating. From the reasoning that I must exercise what I want to learn. But evidently, that was a bridge too far.
In the mean time, soaring around hasn't made clear yet how to solve this issue.
https://youtu.be/Z2PiIciIvzg
ReplyDeleteI think I know where MY problem occurred while attempting to solve this problem. I started with trying to win the White Bishop (that was the "obvious" starting point), and, as you figured out, decided that it wasn't possible. I wasted at least a minute or two playing around with that idea. FINALLY, I started looking at that White Rook on b7. SHEESH! All I had to do was COUNT: it's attacked TWICE, and defended ONCE! Just take it, fer cryin' out loud! Then a sense of danger kicked in: IT CAN'T BE T-H-A-T SIMPLE! Here there be TIGERS! After 1. .. Rxb7 2. Bxb7 Rxb7, is there anything nasty that White can do at this point? OOPS! There's a nasty check 3.Qe8+ that WAS prevented by the (now missing) Black Rook on e7. PANIC?!? IS IT CHECKMATE?!? So, I went back to looking at the White Bishop again; no go there. Then I [FINALLY!] realized that I was dithering back and forth between the two alternatives, like Buridan's ass (see below for THAT proverbial conundrum). So, I made up my mind to go back to try "grab the Rook" until I made ABSOLUTELY sure that there was NO POSSIBILITY of getting checkmated on the back rank. FINALLY, I "saw" 3. ... Bg8 blocking the checkmate. WHEW! What a relief! And time passes, SO S-L--O---W----L-----Y. . .
ReplyDeleteWhat was MY problem? I started along the right path, then when encountering the first "difficulty," I abandoned the idea before trying to "see" if THAT discovered difficulty could be overcome. Consequently, like Buridan's ass, I "died" because I could not choose between the two alternative ideas. This is a psychological problem that must be rooted out of my thinking, if possible, or least kept in my awareness at some level if I can't rid myself of it.
I love this quote about choices, especially in light of the analogy of the vulture's eye viewpoint:
"If it feels like you're choosing between the lesser of two evils, DON'T. There is always a HIGHER choice."
Author: Michael Neill
Buridan's ass
@Aox: That was an instructional video! Love it!
My Aftermath strats usually with the question: Did you see "everything"? and if not: why not?
ReplyDeleteToo see is : the count at b7 is +1, the own backrankweakness and its protection Bg8
So: Did you see that( fast)? And if not: Why not (fast)?
I don't know if those questions were directed to me, but I'll answer anyway. I'm seriously interesed in others perception of their thought processes.
DeleteWhen first looking at the position, I "saw" [FAST] the 2:1 ratio at c6 A-N-D that if I started that exchange, I would (eventually) lose the Rook at e7. So I was aware of the functional relationship between those pieces/squares. My mistake was in pursuing various attempts to "make that work" instead of continuing to survey the entire position (from the vulture's viewpoint high above the prey) for geometrical and functional motifs [which I have been trying to incorporate into my "whole position status" surve] for some time). As a consequence, it very SLOWLY became clear that capturing on c6 was not going to work. It was only then that I returned to the vulture's spiral and started (again) looking at the geometrical and functional interrelationships between other pieces. The "solution" (as given on Chess Tempo) ended after 1. .. Rxb7 2. Bxb7 Rxb7. However, my "spider sense" kicked in and I "saw" (almost immediately AFTER begining to contemplate 1. ... Rxb7 that 3. Qe8+ M-I-G-H-T be a problem. My "screw up" was in rejecting that line before checking to see that 3. ... Bg8 prevents the back rank mate. The "clue" was obvious (in retrospect): Black Pawn at f6 opens the way for 3. ... Bg8, so there is no back rank mate threat AT ALL. I already explained why I ran off the rails by going back to the original investigation of taking on c6. It's so HARD to stop a mental train as it rushes into the tunnel!
Unrelated to the current topic (at least, not directly related): I just ordered Improve Your Chess Now by Jon Tisdall. I've had a PDF file containing the book for a while, but I always try to buy a copy of useful works when possible. It compensates and encourages the author for his hard work, and Tisdall obviously put a lot of effort into his book. It also is much easier for me to read, re-read and annotate a work when I have the "dead tree" in my hands. I'm somewhat of a Luddite when it comes to reading; I prefer the old way to the newer e-reader experience.
DeleteI also ordered Daniel Kahneman's book Thinking, Fast and Slow, mentioned on a previous post on this blog. I did a perusal of the book contents on Amazon, and think it will add to my (limited?) understanding of thinking processes.
My experiences and thoughts are an exact copy of Roberts'. With one additional remark: once I wended my attention to Rxb7, it took considerable time before I recognized that it would win a piece.
DeleteSo you did not see that the count at b7 was +1 "fast/early"
Deletereason was: yout attention was distracted by an other tactical weakness
( here my favorite awarenesstest video again :) https://youtu.be/jvKUqmEiA3E
I did see ( possibly by chance?? or by better tactical vision???? ) the weakness at b7 first, so i found the solution in hyperspeed and did not think about the weak bishop at all
Now
1) we can accept the result because by "statistics" you are sometimes faster than average because you start by chance with the right weakness/idea and sometimes you are slower than average because by chance you find the right idea "later"
OR
2) you still think this did take to long. Then you did find the refutation to slow you did stay too long at the wrong idea(s). Here you need to make further analysing why that happend. Estimatingly you where not agressive enough against your own idea??
OR
3) You do whats often suggested: you collect candidate moves first and check them on a low level first.
OR
something else
That is a "bear" of a problem to "see!"
DeleteAll kidding aside, I'd select "something else" but then I may be too close to the "problem" to clearly "see" either it or the solution. Hence my suggested experiment below.
Ad 1. That is what happens usually, but it is not acceptable.
DeleteAd 2. True, I investigated the wrong idea for too long. The problem is that you can't proof that something isn't there just because you don't see it. I entered a tunnel which I should have avoided.
Ad 3. I hadn't anticipated the move Bxc6, so I was somewhat surprised, and my attention was drawn to c6 because of that.
I simply entered a random line too soon, before I had scanned the whole board. I wasn't aware I had entered a tunnel, and I couldn't stop the train by myself.
Like you can see in a glance what the material balance is, you should see what is going on in the position. I'm not enough aware what the pieces do. For instance, that the counter mate threat could be stopped by Bg8 took a lot of time to see too. It is like "oops, Qe8+ might be mate", PANIC and FREEZING. Then the question arises "is there something I can do about it?" This question looks NEW, as if I have never prevented a mate before. After some time, I realized that I might put a piece in the line of attack, and I started looking for a possible piece.
I'd like to conduct an experiment, with your kind permission. Afterward, I'll bloviate on why I think your "vulture's eye view" is so important, and what insights it gave me, in conjunction with some thoughts regarding what I discovered from analyzing these three positions. I think it MAY be very important for helping to "solve this issue" of "seeing" the whole position.
ReplyDeleteConsider the following three positions. Please load them up and look at each of them "VISUALLY." However, please do NOT use an engine or endgame tablebases to analyze them! As you look at each one, think about what you "see" and your conclusions about each position (what SHOULD you "see"; what should the outcome of each problem be and WHY). Please write down your observations as you go through the positions so that you can share your conclusions here.
BTW, THIS IS N-O-T A TEST!
Position 1: 8/8/8/8/5k2/1p6/bK6/8 b - - 0 1
Position 2: 8/8/8/8/5k2/5pR1/1K5P/8 b - - 0 1
Position 3: 8/8/8/8/5k2/1p3pR1/bK5P/8 b - - 0 1
position 1: a lot of thoughts (between ""), but little seeing at first.
Delete"a lot of 8s in the FEN, probably an endgame"
"it's a draw, I cannot expel the king from the corner"
"wait, maybe I can sac the bishop at the right moment and get the opposition"
Forgetting to see, I just tried a few variations before the minds eye.
Bummer, I can't get the opposition hence it is stalemate no matter what.
It is a draw.
position 2:
it took me considerable time to see what colour I had and in what direction the pawns go.
"where is Luceena when you need him?"
Seeing: I can't promote due to a skewer on f8
"he can sac his rook when I promote, and I can't take back due to the rule of the square"
It is a loss.
position 3:
"is there a stalemate?"
"no"
right side of the board "I cannot make progress there"
Seeing: the bishop sac interferes on h7
left side of the board "same problem as before"
It is a loss
When I started the engine to check, I saw it was a win. Without even looking at the variation, I knew immediately how the win should be accomplished. I saw it.
The bishop check interferes with the skewer at f5. If the bishop is taken, I can promote with check.
I noticed a lot of verbal talking in my head, and little seeing during the study of the positions.
EXCELLENT!! Your commentary was EXACTLY what I was looking for! In particular, your last statement encapsulates one of the points of this experiment: ". . . a lot of verbal talking in my head, and little seeing. . .".
DeleteMy apology for any confusion regarding which side of the position you are supposed to take. I copied the positions from Fritz as FEN, so I had the "advantage" of "seeing" them and setting them up. I assumed (BAD ASSUMPTION!) that since it was Black to move in all three positions, that the positions should be viewed from Black's perspective. Obviously, it could just as easily have been viewed from White's perspective. What I found fascinating is that the Nalimov Tablebases gave the possibility of three results (win, draw, loss) for position 2. I can't imagine drawing or losing as White in that position, which demonstrates how limited MY imagination must be! (It also demonstrates how complete the Nalimov Tablebases must be.)
I'm curious: did you "see" (at a higher level, maybe up at the "vulture's eye view") any connection between the three positions? If so, did you think that the first and second positions were "combined" to make the third position OR did you think that the first and second positions were "extracted" from the third position? You obviously made a connection between position 1 and 3. Did the "solution" to position 1 "aid" or "interfere" in any way with the "solution" to position 3, before you started the engine?
I didn't understand the relevance of the statement in position 3: "Seeing: the bishop sac interferes on h7". Was that for stopping the White Pawn from promoting at some future point? Please explain that observation. . .
I'll wait to bloviate until I see if anyone else wants to take a crack at it. I'll write my thoughts about it tomorrow.
Thank you for the indulgence!
About the confusion of the colours I didn't express myself well. The confusion is caused by the fact that there are little references in the position, unlike a middlegame. It takes time to figure out in what direction the colours go and whether a pawn attacks you or not. Is the white pawn on the verge of promotion or is it standing on its initial rank? I have this orientation problem often in an endgame position.
DeleteI thought that the 1th and 2nd position were extracted from the 3rd, in order to prove that the whole is more than the sum of its parts. Hence I knew that there had to be some interaction between the left and the right side of the board in position 3. I only saw the stoppage of the white pawn at h7 by the black bishop (interfere), but I missed the interference of the bishop at f5 initially and I missed the promotion with check initially.
I also have that same orientation problem at times, even with positions that have markings around the edge of the board. I'll eliminate a possible line of attack because of a possible Pawn "capture" (which is impossible because the Pawn is moving in the opposite direction). It is a very weird feeling to discover that I'm moving in the wrong direction!
DeleteYou are correct that the 1st and 2nd positions were extracted from the 3rd position. Your summary of the situation is outstanding! I'll expand on this idea in another comment.
From my point of view.
ReplyDeletePosition 1: I learnt this position some time ago (maybe 4, 6 or 8 years ago) when I have been studying the basic endings. Besided that I tried to test if there is a hidden trick inside the position. Anyway I could not find the win because no matter when the Black King stands the Bishop sacrifice leads to a draw. And Black King cannot stand much closer (say, f6) due to the stalemate.
Position 2: I studied similar Rook vs pawn endings as well, but I noticed the pawn cannot promote with check. And the Rook can make a skewer. In addition the square rule exists. I could not find any type of draw.
Position 3: It is a combination of previous two positions, but I wanted to see if there is ANY additional "hidden" trick (connection) previously not possible at both positions. And the results was the same: I could not find any drawing chances for Black.
BTW. I am confused what Tempo means by: "...but I missed the interference of the bishop at f5 initially and I missed the promotion with check initially" and previous part: "...The bishop check interferes with the skewer at f5. If the bishop is taken, I can promote with check.".
1. ... f7 2.Rg8 Bb1 if the white king takes the bishop, black promotes with check, if white plays 3.Rf8+ then 3. ... Bf5 prevents the skewer, so black wins.
DeleteThanks for explanation Tempo. The only confusion now is the reversed chessboard (1... f2 2.Rg1, etc.).
ReplyDeleteI have just came up with the idea of broader sense:
1. HOW CAN this position be won?
2. WHY this position CANNOT be won?
3. HOW the position would have to be present (what position should be reached) - if it has to be won?
If we try to answer these questions we could discover the way to the Holy Grail of Chess ;) :).
What I mean is just the key (critical) positions we are analyzing and evaluating. If we PROOVE there is not way to reach the winning position - the position cannot be won. It is a very easy to understand this concept when analyzing all the fortresses.
What do you think about it? Any comments?
@Tomasz:
DeleteThank you for the ideas and suggestions!
My gut feeling is that logical thinkng would have to be used in order to PROVE a win or not a win. Since our concept of "proof" is based on using words, I suspect (without "proof") that this will not lead to an increase in skill. I lean more toward the "seeing" approach to solving problems. If I don't "see" the parameters within the position and the interrelationships between those parameters, I get lost digging around in the word salads, trying to find a scrap of meat, while the mental train hurtles toward the tunnel. GM Rowson describes this reliance on word descriptions as the means (for AMATEURS) to reduce the cognitive load associated with complicated positions. Grandmasters, on the other hand, dispence with the verbalization altogether. That would seem to contradict the usual descriptive annotations given by grandmasters in their books. IM Watson informs us amateurs that there is a serious danger here for the aspiring student. The annotators use words as a mere shorthand to summarize the most important points of a position, rather than killing entire forests trying to point out every detail and consideration that went into a specific decision. The danger is that we amateurs will (wrongly) infer that this is how the grandmasters think - in verbal reasonings and annotations as they are playing. The reality is that most of the time, the grandmasters are "seeing" complex interrelationships between pieces, squares and "chunks" of pieces in a kind of vague visual sense without any verbalization at all. They not only DON'T VERBALIZE, they don't even have a clear-cut "picture" of a board or pieces in their minds; it is all kind of non-descript abstract "painting." They toss around the pieces "visually" (sort of) in order to change those interrelationships. Most of their "thinking" (if it can even be called that in the conventional sense) has nothing to do with recalling general principles, "rules", etc. They don't think that way - but we amateurs do think that way! The principles, "rules", etc. must be arrived at by independently and individually examining a sufficient number of positions through individual hard work to make them part of our subconscious. (I am reminded of Temposchlucker's description of his skill at moving knights around the board LIKE A GRANDMASTER.) Glomming on to a set of principles, a thinking process, or anything else in chess (or any other field, for that matter) without "making them unique" to yourself (internalization) means that you may have an academic knowledge of them, but you don't have the skill to use them appropriately except in the simplest of situations. It is the exceptions that prove the rule!
"If only we could pull out our brains, and use only our "eyes." Picasso
I'll get off the soap box now. . .
PART I:
ReplyDeleteThank you to Temposchlucker and Tomasz for your participation! I apolgize for any confusion as to the orientation of the board and the direction of movement. FEN gives the position properly when it is loaded into Fritz, so it didn't occur to me that the board might be viewed from the opposite perspective.
The 3rd position was taken from Weteschnik's Understanding Chess Tactics, Chapter 10: Status examination. It can also be found on various web sites, but I could not find the actual game score. It is the ending of the game Borisenkov vs. Mezenev, USSR 1950, perhaps because it was a Correspondence Chess Championship. The 1st and 2nd positions were extracted from the 3rd position, as Temposchlucker surmised.
Temposchlucker's opinion that the experiment was about showing that "the whole is greater than the sum of its parts" is true, but not all of the idea. The idea was to demonstrate that our amateurish way of thinking about chess positions by using logical thinking based on words can lead to disasterous spillage of time and also to the wrong ideas. The reason is subtle, but can be cleanly separated out using those 3 positions. Weteschnik points out this difficulty in his discussion of this position.
I separated the two sides of the board into positions 1 and 2 to make the interference between the ideas clearer.
Position 1 is a draw, no matter who is to move. Both of you evaluated this position correctly. It was illuminating that Temposchlucker used words to “reason” through to the solution, whereas Tomasz relied upon his past experience from studying this type of position. Even if Black sacrifices the Black Bishop, the sacrifice cannot occur AND get the Black King close enough to get the opposition simultaneously to win the resulting K+P vs. K ending position. I was amused by Fritz’s antics. It evaluated the position as winning for Black [-+ (-7.49)] and kept running in circles. It shows that there is no “vulture” mode inside Fritz taking the big picture view! (I was not using endgame tablebases with Fritz.) The point is that sacrificing the Black Bishop does nothing to change the outcome; it is drawn.
Position 2 is a win for White, no matter who moves first. No matter when Black pushes the Black Pawn, White immediately goes to the 8th rank and skewers the Black King (if it is still on the f file) or just captures it if the Black King has moved off of the f file. The Black King has to get close to the queening square f1 to keep the skewer from happening and to win the White Rook when it captures the newly crowned Black Queen. However, in the interim White is advancing the White Pawn up the board for a new Queen because the Black King is outside the square of the White Pawn. The point is that White wins, no matter what Black does.
PART II:
ReplyDeletePosition 3 is the “prey” for the “vulture.” The point concerns the pitfalls of using a conscious logical thinking process based on words. Obviously, we have both position 1 and 2 combined. In the first case (position 1), the result is drawn. In the second case, the result is a White win. Yet, when combined, the result is a Black win! That seems contradictory to logical thinking based on words! Here’s how Weteschnik described the logical thinking process:
“At first glance it looks as if this position is winning for White. Both sides of the board seem to be under his control. The white king has a firm grip on the black bishop, and the rook might sacrifice itself (using a check on the f file) for the f-pawn if it advances, then White could win by pushing the h-pawn. THE ONLY WAY TO DEAL WITH THE ROOK CHECK WOULD BE TO PUT A PIECE BETWEEN THE KING AND THE ROOK WHEN IT TURNS UP ON f8 (emphasis added).”
[So far, so good. The logical thinking train is rolling merrily down the rails in the right direction. And then, it promptly runs into the tunnel.]
“But [why is there ALWAYS a “but”?] if the bishop moves it is taken. The prospects look rather grim for Black. The frustrating word zugzwang must ring in Black’s ears. If he moves his king, then White’s h-pawn will queen.”
[Notice how the logical reasoning process now tries to derail the train. Words are thrown on the tracks!]
“For example: 1. …Ke3 2. h4 Kf4 3. Rxf4+ and even if Black ttries to be clever and goes after the pawn with 3. … Kg4 instead of 3. … Kxf3, White answers 4. Rh3!”
“Again offering the rook while saving the pawn. Black would lose because of zugzwang as the black king can only temporarily block the white pawn.
PART III:
ReplyDelete“The individual elements of this analysis may be correct, but there is a crucial point missing: the intersection of the two sides of the board. Only if you see all of the individual pieces TOGETHER will you get the full picture [the vulture’s eye view] and this looks a little different: 1. …f2!! 2. Rg8 Bb1!”
“White has nothing better than taking the bishop, because 3. Rf8+ is met by 3. … Bf5 and the f-pawn queens anyway.”
“But by taking the bishop: 3. Kxb1 the status of the king is radically changed as it is now on the first rank. 3. … f1=Q+ the f-pawn promotes with check, winning a crucial tempo.”
“This is a good example that understanding all the individual elements of a position does not necessarily mean understanding the whole position.”
Thus begins Weteschnik’s description of the method of a status examination using checklists.
While thinking about this position and the description, it occurred to me that we often verbalize (perhaps even subvocalize) our logical thinking process. There are some serious problems with doing this. Obviously, if we reason ourselves into the wrong conclusions (failing to see the “forest” because we are too far down in the “trees”), we will either futilely expend considerable time trying to “make it work” or forget to fly back up to the vulture’s viewpoint. Notice that the “word salad” used by Weteschnik causes us to arrive at contradictory conclusions about the two sides of the board, and thus causes us to NOT see the interaction of the two sides, i.e., the position as a whole. Yet, when Temposchlucker put the position into an engine, he immediately “saw” the solution. There was no intellectualization, no verbalization, just the vulture mind “seeing” the prey and puncing on it.
There is a subtlety involved. We “think” that there is nothing to be gained in position 1 from sacrificing the bishop. Because we have already thought THAT, it is retained somewhat as a residual RESTRICTION on what works and what doesn’t work. That limits our logical thinking process because we have already articulated the restriction into words. It acts as a kind of filter to eliminate that from consideration. Without the words, just looking at the position brought the right course of action to mind: sacrifice the bishop immediately!
PART IV:
ReplyDeleteBut there is something else going on here. Because of the experience we have in the standard educational system, we are used to arriving at KNOWLEDGE as an accumulation of facts. Rarely if ever does the educational process teach SKILL. We “think” in words and stories. Once we have convinced ourselves that “it is just this way,” we no longer can accept that maybe, just MAYBE, “it really is NOT this way.” The words give us that old familiar “feeling” that we “understand” the position as a whole, while providing no “warning signs” that the words themselves are leading us into the tunnel.
I did an experiment using words with my youngest brother last night. My older brother and I are working with him on improvization on piano. I asked him to describe the relationship of the notes in the major scale, based on whole tones and half tones. He immediately rolled his fingers along the keyboard giving me the C major scale, up and back down. He stated that the half tones were between scale degrees 3-4 and 7-8. Absolutely correct regurgitation of KNOWLEDGE! He also knew that this relationship holds for every major key (all 15 keys, if you don’t consider enharmonic keys). Then came the crucial part of the experiment: I asked him to play exactly the same thing he had previously played, one half tone higher. He immediately flubbed it! We tried it with several keys. The only one he could actually demonstrate any fluidity on was C major. The KNOWLEDGE was immediately and directly available, but the SKILL was missing. How could this be? It is because we mistake “familiarity” (based on word salads) with mastery, knowledge with skill.
The words enable us to CONSCIOUSLY “think” about positions. But, given the empirical evidence so far, it appears that the bulk of skill resides in (perhaps) the fusiform face area of the brain. I’m pretty sure that the FFA is NON-VERBAL, so using logical thinking processes IN TRAINING may give us access to the WORDS as we try to solve some future problem, but the KNOWLEDGE has not been transformed into SKILL – and thus, we fail as ADULT chess improvers. The very words that we use to identify and categorize concepts into knowledge cause us to “think” that we understand, when we don’t. We can give a dictionary definition (just as my brother did for the location fo the half tones within a major scale) but we can’t play the tune. In this sense, “familiarity breeds contempt.” We become dismissive of the simple stuff and try to move on to more complicated concepts, when in actuality all we have is a crappy foundation for skill. How many times do we look at a “simple” position and dismiss it from any serious consideration because it is TOO simple? I know I do!
As a finishing point, consider this little anecdote from Weteschnik’s book.
“Another piece of advice! Even when you occupy yourself with such simple things as piece movement, important lessons can still be learned. A lot of people smiled when Mikhail Tal admitted that he liked to watch children’s chess programs on television. But, despite being a World’s Champion and one of the greatest tacticians of all time, he claimed that HE WAS STILL LEARNING FROM THESE SIMPLE PATTERNS.
Are we still learning skills from the simple stuff, or are we so convinced of our own knowledge that we think we have progressed beyond it? Are we hung up on our own knowledge petard, unable to progress because we confuse an understanding of the terminology with skill?
Thanks for allowing and participating in the experiment!
here an other example how to do it right ;)
ReplyDeletehttps://youtu.be/AGOzeJkoH_0
@Aox:
ReplyDeleteThat is a perfect illustration of what I was trying to say. Although John is using verbiage, he is simply describing what he is "seeing" AND why he is throwing the pieces around to different positions to take advantage of what he is "seeing." This very much reminds me of how Master Richard Bustamante (1975 US Armed Forces Champion) used to analyze with me watching: he would simply move the pieces around the board, give an evaluation (good, bad or needs more "looking") without any reference to general principles, "rules," etc. Everything was concrete to the specific position. He was mostly indifferent to any "theories" about chess. That brought to mind two different (and yet very similarly) stories from the literature.
GM Kotov (Think Like a Grandmaster, pg. 170: Another digression seems relevant. Once in a lobby of the Hall of Columns of the Trade Union in Moscow a group of masters were analyzing an ending. They could not find the right way to go about things and there was a lot of arguing about it. Suddenly Capablanca came into the room. He was always fond of walking about when it was his opponent's turn to move. Learning the reason for the dispute the Cuban bent down to "look" at the position, said, "Si, si," and suddenly redistributed the pieces all over the board to show what the correct formation was for the side that was trying to win. I haven't exaggerated. Don Jose literally pushed the pieces round the board without making moves. He just put them in fresh positions where he thought they were needed. Suddenly everything became clear. The correct scheme of things had been set up and now the win was easy.
GM Rowson (Chess For ZEBRAS) repeated something from GM Nigel Davies: In case neither of these two distinctions [between "knowing that" {KNOWLEDGE} and "knowing how" {SKILL}] means much to you, I came across a third saying much the same thing, in an article on Chesscafe.com by Nigel Davies called 'The How and the What'. Extracts are copied below with the author's kind permission: "I recently saw a newsgroup discussion about tournament preparation. Everything under the sun was mentioned from openings to endings and strategy to tactics with everyone having their own idea about how it should be done. I just commented that 'the how is more important than the what', leaving anyone reading this guessing as to what I meant. In fact the comment was deliberately enigmatic ... It really doesn't matter what you study, THE IMPORTANT THING IS TO USE THIS AS A TRAINING GROUND FOR THINKING RATHER THAN TRYING TO ASSIMILATE A MIND-NUMBING AMOUNT OF INFORMATION. In these days of a zillion different chess products, this message seems to be quite lost, and indeed most people seem to want books that tell them what to do. 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...
Yes, Masters see more and they are using bigger chunks, chunks of chess positions ("throwing pieces around to different positions"..) and chunks of move sequences
DeleteBut there is even more, they have better vision, they dont make "loops" ( come back to the same things ). They generate a stream of new information from the position, faster and sharper and more to the point. They can handle these big anounts of informations. They dont lose orientation in the tree of variations ( i for example often have to go back and reroute my path to see clearer )
I think there is no single effect which makes masters better! Like if you want to have a better car you need virtually everything to be better, a better engine AND a better transmission ( or the transmission will break soon ) AND better breaks ( or you will die soon ) AND better weels ( ot the whole energy will not go into the street ) AND ...
A perfect training should detect the weakest link(s) in the chain and i suspect that this one is not the same for everyone
I think there is no single effect which makes masters better!
DeleteIt revolves around the question: how many slack is there in our usual approach?
If I cut all the nonsense and waste of time out of my thinking, will I be a master level tactician?
So far everything points in the direction: yes!
It seems to me that you are overly worried about all the things where masters score better, according to scientific papers. You forget that the unconscious can work magic. If the brain is no longer overloaded with nonsense and bad habits, and if it has learned the appropriate pattern, or chunks, or whatever, there is room to work its miracles.
For a new car you only need money. Money will give you everything you want, concerning your car. In this metaphor, money is the equivalent for time. Cut the spill of time! If you are not a master by then, we talk again.
For a new car you only need money.
Deletenice rhetorical gesture the reply would be: for a good game/move you need ( money for ) a good engine.
But back to our discussion
You are right the concious work is only a tiny little fraction of the work we have to do when we play chess. We see that at exercises like m1. At m1 we ( at least i ) dont think at all. Board vision, tactical vision that what we need and that is what we dont get. Maybe your new method will help here? I have doubts but please go ahead, its worthid a try.
Masters simply spot the "most" things where we have to analyse.
All indicators seem to point in the same direction: see, don't think!.
ReplyDeleteThe past days I have been experimenting with this, and I found that puzzles can be solved by only looking at them indeed. It often takes an awful lot of time though, so my rating plummeted again, but after five days, it starts to normalize again.
The experiment of Robert showed that the habit of verbalizing is still quite strong in us. But the fact that it is even possible to solve a puzzle by mere seeing alone, is encouraging, and gives the training a clear direction. It works better when you know where you should aim at.
There are a lot of new patterns to see and to learn, mostly geared around the subjects focal points, initiative and overworked defenders. I hope and assume that assimilating these new patterns will speed up the seeing a lot.
"see, dont think" that sounds like saltmine, i see the loop closing ;)
Deleteverbalising is not necessary, that is used for example at speed-reading but verbalising dont harm as you can see when you watch masters solving puzzles and actually speaking their thoughts aloud into the mic: they are still quicker, sharper, better
@Aox:
DeleteI do not want to argue with you, BUT. . . (there's always a "butt" stuck in there somewhere!)
Could it be that the masters ARE actually slowed down by the verbalization, since the intent of the verbalization is to teach others, not to demonstrate the "master way" to analyze? As I watched the John Bartholomew video, I was able to "see" the point of the problem faster than he could talk about it. Have you noticed that in almost every problem, there are moments when he simply "runs silnt, runs deep" without saying a word, and then, sometimes after several seconds, when he comes back out of the "master trance," THEN he verbalizes what he was considering? In a couple of the problems, he went completely off the rails, and I was silently urging him to "look" again at the position. In at least one case, he truly went down the tunnel and crashed into the train. Not to worry; the train is fine. That IN NO WAY means that I could hold my own against him (or any other expert, never mind a master or higher level player) in any actual chess GAME or even in a "solving problems" contest.
Have you ever had the experience of listening to a speaker who read directly from his notes? Have you ever experienced the frustration of watching/listening to Bill O'Reilly on the Fox channel? (I rarely indulge THAT, since I find it so frustrating in many ways.) As he is speaking, the approximate text of his remarks is scrolled on the screen. I've finished reading before he usually begins to speak. When those kinds of situations occur, I want to throw something at the speaker. Why? BECAUSE I CAN READ A-N-D COMPREHEND MUCH FASTER THAN THEY CAN SPEAK! The master commentator must "convert" the rather vague concepts and interrelationships into a linear sequence of words that convey the specific point(s) he has in his mind; THAT MUST take a certain amount of time, relatively speaking MUCH MORE time, than if he just did the analysis silently. As a thought experiment, I considered an alternative way to transfer the ideas: the master simply shuffles the pieces around as he thinks through the problem, allowig a "visual" window into his mind. Again, I think it would slow him down, compared to when it all occurs only in his mind. So I conclude that the master IS being slowed down when he verbalizes, even though he still appears (usually) to grasp the essence of a position much quicker than we amateurs do.
But (another one of those "butts"!), for pedagogical purposes, we don't seem to have any other method for conveying ideas from one mind to another one except through either the written word or verbalization.
I'm beginning to think that the function motif (which includes the overworked defender) is one of the most important, if not THE MOST IMPORTANT, motifs for rapidly "seeing" the gist of a position quickly. It certainly focuses our sight on interrelationships between at least 2-3 (or more) pieces, thereby forming a related "chunk." (I hesitate to call it a "pattern" because that implies there is something that I already have in my subconscious LTM memory that I can recognize, and I'm pretty sure the requisite "patterns" are NOT in there yet.) Anything that allows us to "see" multiple things at once is a very good thing! Previously, I had given the highest priority to the geometrical motif, but (for some unknown reason) that motif now seems to "pop" into sight much quicker and easier for me than the function motif. I think (but cannot be sure) that it is connected with my decision to mentally trace the "aura" (rays of potential movement/capture) from each line-moving piece (Queen, Rook and Bishop) from the piece outward in all directions, through obstacles (same side pieces as well as possible captures), until each aura encounters the edge of the board; I also do something similar for Knights. I've been working on that specific thing for some months regularly, so (maybe) it's beginning to sink into my subconscious. The specific way I do it is to start with whichever piece grabs my attention first in a problem/position. I immediately trace out the aura for that piece, trying to be aware of potential targets (squares as well as pieces) while I am doing it. It has been surprising how many clues I get from that simple focusing process. The function motif comes into play in second place (for the time being), because I still have to "think" about it before I can "see" it; it is not even close to being in my subconscious.
ReplyDeleteI have no idea what benefit (if any) I am actually getting from this focus on "seeing" without verbalization, but I am more aware of potentialities than I was previously. It is showing up in the problems on Chess Academy. I still take a lot longer than I would like to solve some of the problems. However, my rating there went to 2038 (the first time I've ever gone over 2000) last week. I know I can't sustain or improve it unless I take my time and go through my "seeing" training regimen in each position, but it is encouraging, nonetheless. The hard part is cutting out the verbalization that usually accompanies the process. I keep reminding myself that I am not trying to absorb any knowledge from the position; not tactics, not strategy, nothing. The knowledge is irrelevant because it is unique to a specific position; "seeing" is my only focus at this point, and it should be relevant to ALL positions once it is burned into the subconscious.
How do you eat an elephant? One small bite at a time.
Function on the chessboard is recognized by its geometrical characteristics. An overworked rook is in contact with two of its own pieces, which in turn are in contact with two attackers. If the latter contact isn't already there, then the attacker and the target are connected to the same focal square.
DeleteInitiative is more complicated, yet the moves that maintain the initiative have certain geometrical characteristics too.
I agree with you Tempo.
ReplyDeleteHowever I am sure the BIGGEST GAIN (significant improvement) can be made when we know WHAT to see (to look at), WHAT is not important (!) and what are the goals (mate, winning material or promotion).
In other words - it is (close) to impossible to find the best variation (series of moves) that leads to gain an advantage when:
1. You do not know what to look for (lack of goal)
2. You want to analyse everything (lack of avoiding to analyse the noise)
3. You cannot see/discover the motifs (themes).
After it is done probably 90% of the tactical positions can be solved with using these 3 important points.
What do you think about this idea(my proposal)? Did these points cover with your findings?
Yes, but how to get there? And even with these steps you need to be sharp, quick and precise to calculate the critical lines.. without calculation chess dont work at all.
Delete@ Tomasz
DeleteIn an ideal situation, it would go as follows:
First step: At the beginning, almost before you start seriously looking at the position,you do a global assessment of the position: what is the material balance, and what is the position about (mate, gain of wood, promotion). Level 4.
Second step: Soaring. But what to look for? The patterns you have to look for are not in your system yet. Level 3.
Third step: Once you got the main idea of the combination (aha erlebnis), it is usually no problem to find the exact moves. They flow naturally from the main idea. Level 1 and 2).
This tells you what the main purpose of the training should be: to acquire the patterns you need for step 2.
Perhaps related to soaring above the board...as mentioned in the papers Aox has posted here, there's been a lot of academic work on the ability to reproduce positions from memory after viewing them for a short time. The performance is usually scored as the fraction of pieces in the correct position, or something like that. I'd like to assess how good my whole-position memory is, but I wasn't able to find any tool for doing that. Anybody have suggestions?
ReplyDeleteThere are a couple related tools in the sidebar, I think Aox made these. But if I remember right, one is subjective - it asks you if you feel you have the position in memory. As we all know, such feelings are unreliable. There's another tool where you're asked to specify the location of the missing piece. I found I could do fairly well just by guessing, without any use of memory at all. So I don't want to rely on this tool either, at least not for testing myself. --mfardal
@mfardal:
ReplyDeleteCrazy Bob's FREE Quick and Dirty DIY Whole-Position Memory Test Suite, Version 0.0.0!
Required Tools:
A Web browser (I use Chrome)
A chess position diagram editor (I use Fritz->File->New->New Position... and clear the board)
A stop watch (I use the Stopwatch tool on my Verizon flip phone)
Process:
1. Set up the chess position diagram editor to an empty board; leave it visible. Set up your stop watch to run.
2. Load up Chessgames.com in your browser. (Any other database of games will also work just fine. You can use Chessbase, Houdini/Chess King, whatever you have available.) Select a particular year (at random; set the Year is "=" YEAR) and "FIND CHESS GAMES!" Scroll down to a game selection at random (perhaps every 10th or 20th game). Select that game. Set "autoplay" and allow the game to proceed to the 10th, 20th, 30th, or 40th move (depending on how complicated you want the position to be). It really doesn't matter if you watch the position change or not while it is progressing, but I suggest keeping a close eye on the current move number. If the game ends before the preselected number of moves, pick another game.
3. When the game position reaches the preselected move number, turn off autoplay and immediately minimize the browser window.
4. Start the stop watch.
5. Fill in the position in the chess diagram editor. When you think you are done, stop the stop watch.
6. Use a screen copier (SNIP in Windoze) to copy the game position and your reconstructed position to MS Word or some other editor. Record your time from the stop watch. Figure out the number and type of diffferences between the original position and your reconstruction. There is literature online about what has been recorded and measured in past experiments of this type.
7. If you want a fancy statistical analysis, copy your times and other pertinent information (number of pieces, types of pieces, color to move, number of differences, whatever floats your boat) into a spreadsheet program like MS Excel.
8. Lather, rinse, repeat until you have collected sufficient data to give something other than statistical noise.
9. Analyze away to your heart's content!
The usual legal disclaimers apply. . ."as-is". . .no requirement to provide anything functional or of use to the end user. . . no warranty express or implied. . . no refunds of any kind. . . blah, blah, blah.
I have tried out Crazy Bob's software. Thanks for the suggestion! The program has actually been upgraded to v0.0.1 by now. The added feature allows you to fast forward to the end of the game and then back up a few moves (so as not to have an obvious checkmate that would make memory easier). This is useful because I was aiming for very simple positions. The program runs rather slowly, but that's mainly because it uses the outdated hardware between my ears for the evaluation step.
ReplyDeleteI have only tried 8 positions so far, with between 9 and 15 pieces. I gave myself about 30 seconds to absorb the position, about 30 to forget it, and a couple minutes to try to reconstruct it. It turns out the only part I'm good at is the forgetting. I used a scoring system where I gave myself 1 point for correctly placed pieces, 1/2 point for misplaced pieces, and -1 for extra ones, then divided by the number of pieces in the original position. (This is not based on any scientific papers, it just seemed to make sense.) Even with this "partial credit system", I only managed about 60% on average, give or take 10%. A couple positions I forgot completely: they just disappeared from my memory when I tried to rebuild them. Oddly enough, those were the ones with fewer pieces. At the other extreme, I managed one position almost perfectly, just placing one piece offset one square.
My hunch is that my memory is terrible compared to other players of my level, and I suspect it's holding back my tactical performance. But I don't have any concrete information to that effect. Nor do I know how to improve it. --mfardal
http://theinvisiblegorilla.com/blog/2012/02/15/how-experts-recall-chess-positions/
DeleteA master dont memorise the position of the pieces within 4 sec, they memorise what the pieces are doing to.
like: The queen attacks the Pawn at xy and defends the bishop at wz so she is sitting at ...
So i think the skill of "position memorisation" is strong relatet to board vision and visualisation.There are many tools out in the net to learn board vision and visualisation
a very nice tool ist the Chess ( Vision ) Trainer 1.5.1 from DKappe or the (E)books of Stephen Ward
Actually, I suspect that your "memory" is not nearly as bad as you think it is from your experiment. It confirms the results of previous experiments of this kind. Aox summarizes the point. The difference between amateurs and masters is chess-specific, not memory-specific. If positions of random assortments of pieces are done, the masters do only slightly better than the amateurs. As Aox suggested, there are tools available which will improve your board "vision." Chess-specific "skills" CAN be improved with practice!
DeleteCrazy Bob's software suite v0.0.2 includes a Python script to select positions of a specified number of pieces from a PGN file with multiple games, and save them as FEN codes in a text file. Then a FEN from this file can be pasted into SCIDvsMac and the "edit board" window used for reconstruction. Also included in 0.0.2 is a recommendation to use something closer to the original Chase and Simon protocol. These changes speed up the process, so that it's easy to get a decent test of memory. --mfardal
DeleteEvery morning i take a chessnewspaper ( Rochade ) in my bathtub and memorise a more or less random diagram ( until i can mentaly name each pieces position without looking ;) and then follow the moves in the paper trying to keep track of the position till the next diagram
Delete;)
Robert, by "memory" I meant position memorization, not my general memory. Though that is bad too, you can ask my family. I've read up on the board memory experiments so I know there's a difference.
ReplyDeleteAs you said in the other thread, it's not clear whether working on the position memory skill will help with tactical ability. Some say yes (Blindfold Chess authors), some say no (some scientists I forget which), I haven't seen any evidence one way or the other. I noticed a lot of 2011 on this blog was spent on visualization and memorization, but I'm not sure any definite answers emerged.
Aox, one reason I suspect it's important has to do with a recent experiment in the salt mines. Trying to absorb each board at the beginning turns out to help my performance--I see things faster and get stuck much less. So I'm thinking that the more I can absorb, the better I can think about the position as a whole.
I wonder though whether I should learn the skill by working on it intensively, or very gradually by studying position play and master games. And how to measure it accurately so I can detect even small improvements... --mfardal