Thursday, December 31, 2015

A vision problem

In order to show you why we have such difficulties to improve, I show you a problem that is very simple. With hindsight, it took me 2:48 minutes to find the first move, and 2:08 to find the second move. That was two days ago. The problem has a rating of a measly 1472. You should expect that everybody who plays chess for longer than a year, would solve this problem in less then 10 seconds, so simple is the solution. So what's going on?

White to move


Today during a repetition session, I took another look at the position, and it took me again about a minute or two to see the solution. But now I was more alert for what was going on in my brain. Of course I saw that I could take the black bishop for free with fxe4. But my gut feeling told me that then my rook would be in trouble due to the pinned bishop on d3. The last move of black was bishop d5 to e4, and I believed that the sacrifice was not a true sacrifice, otherwise my opponent wouldn't plat that move. My focus was mainly aiming for h7. I tried to broaden my view and looked what the pieces were doing and how they could work together. ALL OF A SUDDEN it popped to my mind that 1.fxe4 connected the rook on h3 with the bishop on d3. It was easy to see that that wouldn't be of much help to free the pinned bishop though. While I was looking further into this, my attention was now more focussing around the region e2 and ALL OF A SUDDEN it popped to my mind  that there is no problem at all with the pin. I could easy unpin the bishop and save the rook on d1.

What does this story tell us? The tactical theme "pin" was seen immediately. The standard method for unpinning was not. Yet I have used this method hundreds of times in the past to free a pinned bishop and save the rook behind it. So why didn't it pop up immediately?

I have identified a few reasons for that:

  • I was biased that the bishop sacrifice would be correct, otherwise my opponent wouldn't do it.
  • I was focussing on h7 for an assault of the king.
  • I didn't formulate what my problem actually was.

Al these reasons can be eliminated by a disciplined mind. Yet matters are not to simple.
A bias is wrong once you know it is a bias. But more often a bias saves you time. You form an opinion about something without the need to go into the details every time. I have for instance the bias that the pawn on a6 is irrelevant for the solution. Since it is, I save a lot of time by not going into the details of pawn a6.
Focussing usually is good. It prevents that you come home with a new coffee machine while you went out to buy a pair of nippers. But focus comes at the price of ignoring the rest. It can lead to tunnel vision.
Usually I decide what a position is about before I start looking for answers. Is it about gaining wood, king assault, promotion or counter attack? But sometimes I forget to think about that.

All these reasons have in common that they are based on verbal reasoning. Which is time consuming no matter what. All reasons have a good side and a bad side. If you by accident choose the right implementation, you will save time, if you choose the wrong implementation it will cost you time. Another thing they have in common is that you will only know afterwards if you have chosen wisely. There is no way to know that beforehand.

The way I experienced it, the popping up of the themes had no relation with the verbal reasoning before. When the mind is occupied by verbal reasoning, there will be no popping up at all. In rare cases, the verbal reasoning can act like a cue, and trigger the popping up. But more often than not, all this verbal reasoning will cost you time. Little time if you are lucky enough that your biases prove to be right, and a lot of time when they do not. Like what happened to me with this easy problem.

The cue that triggered the pop up of the tactical theme "unpinning" was clearly visual. I don't know how it works, and I don't know how to train it. But somehow we must built in visual cues that cause "tactical theme-popping".

It's a strange problem. When we learn the looks of a kangaroo, we immediately are able to recognize it in a cloud, an ink spot, a tree bark or whatsoever. Without any additional training. But if we learn a tactical theme, we have great difficulty to recognize it in a position. Maybe because we have to recognize virtual properties, and not properties that are physical present. A tactical theme is a pretty abstract thing to recognize. What I try to accomplish, is to visualize the tactical themes. Learning to see them as if those abstract properties are really visible on the board.



  1. The funny thing is the low rating of this puzzle, lower rated player will simply have taken the bishop and than would have handled the problem how to unpin the bishop which is not difficult at move 2)
    So there was a weakness of vision, you did not see Be2
    And a weakness of thought controll ( thought process ) you did not look for a method of unpinning

    This are constructions i am working on at the moment
    Off couse its easier to think harder about 1. fxBe4 if you realise that black has no other serious tactical weakness..

  2. The investigation nowhere indicated that a thought process is necessary. With a bad thought process you suck terrible, while with a good thought process you suck less. Both need thinking, and while you are thinking, the brain is too occupied to pop up the tactical theme. Once you stop thinking and start looking, there is a possibility that something pops up. The example indicates that it should be possible to circumvent the thought process altogether. There is a time and place for thought processes, but that's not here.

  3. I disagree at the moment. The problem Tempo describes has NOTHING to do with vision - otherwise we can hammer all the things (objects) around us like they have been nails.

    More about vision (December 26, 2015) - it was THE SAME problem described. Tempo simply did not know how to defend against special kind of pinning. The solution? Learn how to UNPIN your piece(s) without loss!

    [quote myself]
    "I hope you will analyse it DEEPLY as just thinking about a minute is a serious sign where your problem is hidden. Of course you should set up by yourself some similar positions and draw the conclusions - now it is (a theme) about pinning and unpinning".

    It is like you could RECOGNIZE all the kangaroos, but when any of these is present "upside down" - you are simply lost. Unless you learn how to recognize these "upside down" kangaroos you will be lost at every (or most) puzzles related to such idea.

    Sorry if I sound harsh, but simply do not believe it is a problem with vision. It WOULD be a problem with vision if you could not imagine that the pinned piece can "come back to protect" to defend the "last in line" piece.

    The problem MAY be connected with the false idea of "always forward" (always attack) idea, but not with vision.

    BTW. Could you specify who is to move at your problem? (WTM or BTM will be enough). I tried hard to find the solution with black, only to realize (after reading next paragraph) it is white.

  4. " Of course I saw that I could take the black bishop for free with fxe4. But my gut feeling told me that then my rook would be in trouble due to the pinned bishop on d3. The last move of black was bishop d5 to e4, and I believed that the sacrifice was not a true sacrifice, otherwise my opponent wouldn't plat that move. My focus was mainly aiming for h7. I tried to broaden my view and looked what the pieces were doing and how they could work together. ALL OF A SUDDEN it popped to my mind that 1.fxe4 connected the rook on h3 with the bishop on d3. It was easy to see that that wouldn't be of much help to free the pinned bishop though. While I was looking further into this, my attention was now more focussing around the region e2 and ALL OF A SUDDEN it popped to my mind that there is no problem at all with the pin. I could easy unpin the bishop and save the rook on d1."

    That is a thinking process

    My question would be, what was wrong about that, or what could have been thought better

    One thing which seems to me was the biggest error : you din not pay enough attention to the last move of the opponent.
    Let me explain: A tactic is only possible because of tactical weaknesses, they may get more serious or more in numbers to make a tactic work. One move ago the tactic was not possible or you would have played it back then. ( You is the problemgenerator of chesstempo ;).
    So either your own last move OR the last move of the opponent did enable the tactic. So-> to think about the last move of the opponent payes off very often.
    Either it was not doing the right thing or it made even the wrong thing!

    So thinking : "otherwise my opponent wouldn't play that move" shows a substantial wrong thought its the opposit : very very often the last move das the error.

    1. If we go to extreme - all mind process are thinking ;) :)

      In general - I agree: the incorrect idea most often leads to nowhere (I do not discuss its creative value now).

      In my opinion there were TWO things wrong:
      1) The false assumption that your opponent have to play ONLY good moves - it is false. Otherwise you could have never win the game! (Perfect play from drawn position always leads to a draw).
      2) The lack of "homework done" - you probably have to break the old bad habit - to understand (with the examples help) how unpinning process is conducted (when and how you can defend your pinned piece). I can assure you - it is MUCH harder than most players think. At least - unless you study the pinning and unpinning process very closely.

      Now I want to quote NM Dan Heisman. As far as I remember he advised (in his article or to one of his students) to ASSUME that your opponent play a BAD move (a mistake). Now it is YOUR task to refute this statement. If you cannot find "the winning move" - it means that the position is safe (or you are blind to some tactic - then check out with the computer).

    2. I would make the definition that a thinking process in chess is a sequence of ( position related ) subjects you focus your attention too. That might be with reasons or without.

      You can easily miss any pattern if you dont put your attention to the related subject.

      I dont think that Tempo dont know the unpinning methods good enough, he did not look for it. Something you dont look for is hard to see.

      So we may ask ourselfs why did tempo not look for an unpinning method?

      1) he did trust the opponents move
      2) he was distracted by a "h7 weakness"

      So .. to continue the analysis of the thinking process we may ask, was thinking about h7 a good idea? And if not, why not?

      H7 is no weakness because white cant get enough pieces attacking this point in a reasonable time

      With this thought a new target has to be found.. and there is only the bishop..
      Now, with more attention the problem get solved quicker.

  5. You both miss the main point. ALL OF A SUDDEN the "unpinning theme" popped up into the mind. A theme cannot pop up if it is not already there. Further there was no relation to the verbal reasoning. The popping up was not caused by the thought process. Even worse, the thought process prevented the popping up since it kept the brain too busy. It was a visual cue that caused the popping.

    We can recognize kangaroos everywhere. Be it upside down or otherwise. But in order to do so, it must be TRIGGERED. Without a trigger, or cue, no recognition. The trigger mechanism is a kind of defence. Otherwise, our world would be infested by kangaroos. It isn't exactly clear how the trigger mechanism works. A little attention is needed. That's why the thought process is counter productive. A thought process consumes our attention. Only when it stops, there is room for attention for the trigger, causing it to fire.

    As far as I could see, the cue was visual. Since the unpinning theme is abstract, it is not a real physical vision, but more a mental vision.

    The question is, how do we manipulate these cues?

  6. Naa Tempo, you had a thought process but it was wrong. You are talking about concius controlling the process, which you did not ( or at least not correct ). You had your focus of attention to the wrong things for too long. You need to find a way to reduce wrong thoughts ( = objects of your attention ).
    What you are saying does mean you just want to stare at the board without looking for anything waiting for the whole movesequence to pop up by itself.

  7. If you shift gears in your car, you don't consciously decide to do so. Yet it is triggered when to do it. We know little about it. I just propose to have a closer look at how cues work by seeing them in action during problem solving. In order to learn to manipulate the cues to our advantage.

    Say, you know your brother or your friend is going to arrive at a crowded metro station. How are you going to recognize him? The pattern of his face is all too familiar, of course. The right way of scanning the crowd is going to help you out. No thought process is going to help you.

  8. Well, to a certain degree a thought process can help you. If you know he is going to arrive at platform B, you don't have to scan at platform A. For learning how to scan, that is irrelevant though.

  9. Funny to shift gear and to ride a bicicle are the standard excamples of motor skills. A process with thinking at the beginning which had become automated by concious repeating it over and over again. Without concioas training and feedback from car,gear, motor and teacher i think its impossible to r4each this stage.
    And with recognition of faces
    There might be a person you know but you dont recognise it ( right away ) because you see it at the wrong place and/or in the wrong context.
    By looking for this person you started a minimalistick thought process to put your attention to the image of that person.
    Same in chess at this puzzle:
    You did try to make the estimated weakness h7 "working" so you did not find 2.Be2
    This is like looking for the nice girl on track 7 instead looking for your friend, depending how much attention you give to that girl;).. you WILL! miss your friend

  10. I think (thought process AFTER the fact of solving the problem!) that there are four significant distractors at work in the problem: (1) Pin of the White Bishop [Rd8->Bd3->Rd1]; (2) Attack on c2 by two Black pieces [Rc8 & Be4 (THROUGH Bd3)]; (3) A very vague notion of attacking h7 [Qh5 & Bd3]; (4) Not thinking of moving Bd3 backwards.

    Temo is correct: the problem is looking for the cues, the visual "signposts" that have "NOW!" on them, with an "arrow" pointing toward "HOW?" to do (whatever). It is my opinion that identifying the tactical themes/devices is a necessary step in the visualization process, but it is NOT the FIRST step. The "NOW!" cues (motifs) precede the "How do I do it?" cues (tactical themes/devices).

    I know it is generally the case that terminology is sloppily applied in chess. One man's motif is another woman's theme. However, there is a subtle but distinct difference between the two concepts that helps distinguish between the "NOW!" cues and the "HOW?" cues. If you are always searching for the "HOW?" cues, you will never see the "NOW!" cues. In essence, the "NOW!" cues become the dancing bear. You can count all the "HOW?" cues you want, but it will not POP the appropriate thing to do into your vision; you simply will not SEE them!

  11. Regarding the conscious versus unconscious process:

    @Temo: "If you shift gears in your car, you don't consciously decide to do so. Yet it is triggered when to do it."

    This is only true of a process that has been completely internalized (subconscious). No one learns how to drive a manual shift car subconsciously. You start with "fits and jerks" and a very conscious thought process. You listen to your instructor (presuming you have one; otherwise, the process will take considerably longer because you will have to listen and look for all cues and then laboriously separate the useful ones (the sound of the engine as it revs up) from the useless ones (the rattle of the muffler). Each time that you internalize one of the multiple cues for the proper time to shift, you decrease the time to "do it right" and it becomes a smoother mechanical process. You will (may) eventually arrive at a point of skill in which you do not think about the cues at all; you will just know that "NOW!" is the right time to shift and you will have internalized the proper motions to do it smoothly.

    (An aside: my wife cannot use a manual shift car. She is not impaired physically or mentally, except having a mental "block" that tells her that she cannot do it. I have tried [in vain] to convince her to at least try. Not going to happen, not now, not EVER.)

    The problem, as I "see" it, is that our chess pedagogical methods do not correspond to the same process that we use to learn how to shift gears smoothly. Apparently, the enormous body of literature on chess does not correlate strongly with a good pedagogical method for enabling all students to learn how to "SEE!". Initially (during the initial phase of learning) the required thought processes must be worked through consciously, step by step. Only after those processes have been internalized (receded into the subconscious) can we progress. I think this is also one of the reasons that so few chess students (especially adults) actually progress beyond a certain (perhaps low) level of playing ability. The foundational thought processes (which should have been internalized from the beginning) are MISSING. Like a building with a shoddy foundation, very little can be built on it.

    I have noted many times that anybody (well, almost anybody) can throw together a dog house that will keep the dog (more or less) dry. However, it takes considerable more skill, knowledge and processes to build a 100-story skyscraper that will stand in spite of weight, wind shear and bad weather. There is a process through which one must progress in order to be an architect or structural engineer. Initially, it will all be conscious and very thought-intensive. Only after the fundamentals have been internalized can progress be made toward more advanced concepts.

    Perhaps I am "singing to the choir."

    The relevant question is and always will be: WHAT IS THE PROPER METHOD FOR TEACHING AND LEARNING TO "SEE" IN CHESS?

    (Apparently Dr. Lasker was not so far off the mark when he gave his opinion that our methods of chess pedagogy widely miss the intended target level of skill.)

  12. Robert there is the difference between real games, mixed puzzles and blitz puzzles. At blitz puzzles you know you will have at least +1.75 at the end.
    So you can ignore the threats of your opponent ( for a while ) and simply look for the goal, either you promote a pawn ( not possible, no pawns close to promote) or you checkmate ( not possible, not enough attackers ) or you win material.. and here you have to find : what material? The chances that the last move of your opponent was making the tactic possible is high..and what else than the bishop is to win?
    At the moment you are shure you have to play for to win the bishop you will find the method soon(er).
    Then you may look for problems the opponent may give

  13. Aox: I agree with you COMPLETELY! You have just described a process that MAY quickly give you a good result - in blitz. However, I contend (based on considerable personal experience and study) that the very process that brings you quick success (at least some of the time) in blitz is detrimental to your skill in "real" chess (whatever that is). As I stated before, YOU WILL PLAY THE WAY YOU TRAIN. Because the time factor is overwhelmingly significant in blitz, you can get away with minor "mistakes." However, that attitude will get you killed in "real" games. Why? Because at least once (and probably more than once) you will make a choice based on your blitz experience, in which the time factor has been internalized as being very significant, and make a sub-optimal move (or sequence of moves). In short, you will not "SEE" what you should see in order to play at a higher level.

    World Champion level players (such as Judit Polgar) did NOT gain their skills playing blitz! Instead, they honed a very explicit thinking process through many thousands of repetitions until it became subconscious. This subconscious thinking process (AND their accumulated knowledge and experience, AND generally high level of natural talent) then enables them to play blitz at a very high level, sometimes nearly at the same level as their slower games. "Slow and steady leads to fast and ready;" NEVER the other way around!

    NM Dan Heisman in his book The Improving Chess Thinker divided thinking into essentially three levels: Flip-Coin Chess, Hope Chess, and Real Chess. (Please note: these do NOT correspond to blitz, quick and tournament chess time controls.) He says the following on pg 156:

    "One key to graduating from Hope Chess to Real Chess is checking for upcoming danger on every move, and not just most of the time. For example, suppose you "only" play Real Chess on 95% of your moves but on the other 5% allow unstoppable threats. Assuming the average game is 40 moves, twice each game (5% X 40) you open yourself up to an immediate loss. If you allow these two oversights each game, your rating will be much lower than if you play Real Chess on every move. After all, it only takes one bad move to lose a game! If you play 1700 [USCF] strength for 38 moves but on two moves play at only a 500 level, what do you think your average playing strength will be for the entire 40 moves?

    NM Dan Heisman also stated the following in Elements of Positional Evaluation: How the Pieces Get Their Power, Introduction, pg 10:

    "Have you ever thought about why and when a player's rating levels off? Was it just lack of talent or did time and ambition enter the picture? I feel that these factors lie within the individual. But there is one aspect that can be controlled externally: the fundamental way that a person learns how to play; that is, the order he learns his chess skills and the priority (and emphasis) he puts on various aspects of these skills. Furthermore, the overwhelming majority of players attempt to improve in a haphazard manner, using a hodgepodge of learning techniques.

    Why is this? How can this best be remedied? In short, how does a player get rid of the dogma that can so easily hinder progress? It is not easy to do this if you scoff at aid or consider yourself to be "of sufficient strength." Most players do not follow suggestions until they are burned by practical experience. That is normal so long as they don't repeatedly make the same mistakes. Only the very gifted seem to learn by avoiding mistakes instead of making them.

    This leads to the [$64 MILLION] question, "What is the correct method?" A player interested in improvement draws, and enlarges upon, his and others experiences. I have drawn upon my experiences to form the basis of my theory of teaching.

  14. This comment has been removed by the author.

  15. An anecdote from my training days in martial arts. . .

    Tai Chi Chuan is often practiced at a very slow speed. It gives the (WRONG!) impression (to those who practice everything at breakneck speed) that it utterly useless as a practical method of self-defense. My Sensei had friends who were black belts in Tai Chi. We once had a joint session, with everyone practicing Tai Chi movements, s-l--o---w----l-----y. It turns out that it is very difficult to perform the movements correctly while moving so slowly. The real "eye opener" was when we switched from slow forms to practical self-defense. Those same Tai Chi black belts were amazingly fast when it came to blocking and counterattacking! They had ingrained the movements, so that when they were attacked at a high speed, they simply responded accordingly, with no wasted motion.

    On the other hand, I have also seen martial artists who made a fetish of always doing everything at the fastest possible speed, with emphass on "flash." In almost every case, their technique was clearly sub-par. Against someone who had trained the proper foundational skills slowly over considerable time, the "speed demons" usually (not always!) got their arses handed to them.

    There is a reason why it takes (usually) 4-5 years (10-15 hours per week) of slow and steady training to even approach a black belt level in a traditional martial art. And in those arts, a black belt is merely an indication that the student is now considered to be a "serious" student, N-O-T a "master." In my particular organization, one must reach the 5th degree black belt level to be considered a "Shihan" or Master Teacher. That usually takes 20-25 years of continuous study, practice, and teaching. A Master takes 30-50 years to reach (if even then) and only occurs at the 8th degree black belt level. I started martial arts training at age 40, studied and taught for 17 years under a 7th degree black belt, and reached 4th degree black belt, and was awarded the "Renshi" (Master Technician) title. On that basis alone, I know something about good teaching methods. I utilized that knowledge of teaching methods to teach myself how to play harmonica in a relatively short period of time (in less than 2 years I was playing with my own band). Those same methods will work for chess, or any other subject, for that matter. I will only provide one caveat: THERE IS NO "SECRET." As GM Savielly_Tartakower said, "The mistakes are there, waiting to be made." There is considerable food for thought in that aphorism. If you aren't making mistakes (AND LEARNING FROM THEM), then you will keep making the same mistakes over and over, and will not improve. It is not the making of mistakes that is important, but the LEARNING from them. As NM Heisman stated, "Only the very gifted seem to learn by avoiding mistakes instead of making them." Or as someone else stated, "When you learn to shave, try to learn on someone else's face first." Perhaps here we find the quest of Temposchlucker and so many others, so far unrealized satisfactorily, still suffering from the cuts to their own face. No offense to anyone; just my observation.

  16. Robert you said : "You have just described a process that MAY quickly give you a good result - in blitz."
    There is a misunderstanding.. i was talking about the so called blitz mode at chesstempo which has not much to do with the blitz games. The blitz mode is just a timed mode where the rating of the performance of the tactician is not independend of the time they use for to solve it. Some puzzle have an average solving time of 26 minutes which does mean that the correct solving of this puzzle will give the tactician an increase of rating even if they need 50 minutes or even more for this puzzle.
    My personal experience with unlimited thinking time for puzzles is negative, it just made me slower. It is easy for me to perform at master level at standard puzzles if i just think long enough about the puzzle, 1 hour, 5 hours.. a few days.. but then i start to recalculate lines over and over again which is at OTB an extreme bad habbit.
    It is necessary to have some timepreassure in the training ( at least somehow, somewhere ) because at OTB we have timepreassure to.

  17. @Aox: My apology for misunderstanding about blitz mode at Chess Tempo.

  18. I have already started solving special chess puzzles. They are rated 1900-2100 (from a paper workbook). They are really hard to me and that's why I decided to solve them in a SPECIFIC (original) way: I try to solve these as best as I can and after I write down the solutions - I analyse the book's solution with moving the pieces on the board. Beside that - I write down the conclusions and think over HOW to overcome the weaknesses that appearead from unsolved/incorrect solved positions.

    Let me know if any of you tried such approach and if yes - what form, how long and what was the overall effect of such activity.

  19. I have solved about 30 puzzles (these rated 1900-2100). Most of them I failed to solve correctly. What is my problem? I can see themes (pins, forks, double attacks, etc.), but I cannot merge them into the whole picture. And a few times I am working too sloppy (shallow) because I am too confused what is the correct solution.

    Anyway - working on correct solutions (from a workbook) gives me a lot of fresh ideas. Beside that I try to refuse the moves given as the answers from the original workbook. In one case I could not solve the puzzle and it turned out that the strongest engine was powerless as well ;) :).

    If anyone of you wants to try your hands at the most difficult one of the puzzles I have been struggling - you can copy the FEN string or the link to the diagram (as a picture).

    r3kbr1/pb1n1p2/2q1p2p/2P1P3/1p3Q1B/1Pp2N2/P3BPPP/R2R2K1 w q - 0 1

    The solution may be counted as "solved" if you give ALL the relevant variations - and do not stop counting (writing down the necessary moves) unless the position is quiet/clear (the advantage is clearly visible).

    I hope you may have at least 20-30 minutes of HARD work with this puzzle!

  20. PART I


    That was a tricky problem! Can I presume that it was from a game? If so, would you please post a link to the game moves (if possible)?

    I may have used over 2 hours(!) total to get a good idea of the resources available to both players. Unfortunately, I didn't do this all at once; I was on mobile patrol at work, and would stop periodically and take another long look at the position. I did all of my analysis from a diagram, and have not (yet) entered it into Fritz 11 (running Stockfish DD64) to find out everything that I DID NOT "SEE!" I'll try to describe the process whereby I arrived at the conclusion that White is winning in all lines. I'll point out a couple of tricky lines that gave me lots of trouble. Please keep in mind that once I was confident that White was winning, I would have played the first move without hesitation, and then worked out the specifics as the game progressed. I know that's a haphazard way to approach playing, but I've come to realize that everything does NOT have to be calculated to the nth degree; sometimes, the intuition (supplemented with sufficient concrete analysis and assessment of the resulting position(s)) is sufficient to proceed.

    I started with the relative King positions. This is tricky, because both players have considerable pressure on their opponent's King position. It appears at first glance that White has a serious problem on g2 the White Knight must not move on pain of a forced checkmate – by BLACK! Even Ne1 is not feasible because Black sacrifices the Black Rook on g2, and then if it is not taken, sacrifices it on g1 and then checmates with the Queen on h1), and that there are insufficient pieces to attack the Black King directly. Then I noticed the "killer" cue: the Black King (placed into that 3x3 box!) has NO MOVES! Oh goodie! A King stuck in the middle of the board is a very strong signal (to me) to throw everything at it. So, where is the "weak point"? I know it's hard to believe, but using the geometrical motif, I saw the coordinated attack of the White Bishop at h4 and the White Rook at d1, and extrapolated that I would really, really like to checkmate Black on d8! The attack motif says that we must not shilly-shally, so we need some way to force open the line to d8 for a Rook or Queen. Notice that there is another "weak point" at f7; it is defended only by the Black King, and the "Loose Pieces Drop OFF" motif says that the Black Rook on g8 is vulnerable, so adding another attacker against f7 looks promising!

  21. PART II

    1. Rxd7!

    Did I overlook that either the Black Queen or the Black King can capture the Rook? No! This is a forcing sacrifice: White has gained a Knight and threatens to capture 2. Qxf7#, so Black's replies are limited to two:

    1. ... Kxd7

    This "feels" wrong to me, simply because the White Queen pounces on f7 with check, and Black still has to figure out how to keep his King alive, stuck on the back rank with three White pieces bearing down on him.

    2. Qxf7+

    2. ... Be7 3. Qxe7+ Kc8 4. Rd1! and White has nasty threats in addition to being up in material.
    2. ... Kc8 3. QxRg8 and White has nasty threats and is a piece ahead.

    1. … Qxd7

    This looks stronger for Black than 1. … Kxd7.

    2. Rd1! Still “looking” at d8 for checkmate!

    Black now has 4 “reasonable” moves:

    2. … Qc8

    3. Rd8+ Qxd8 (forced) 4. Bxd8 and Black only has bad choices: (a) 4. … Rxd8 5. Bb5+ Rd7 6. Qd4 Bc8 7. BxRd7+ BxBd7 8. c6! and the Black King dies. (b) 4. … Kxd8 5. Qxf7 and Black is going to cough up a lot of wood.

    2. … Qc7 (At first, I thought this was a “refutation,” but there is a beautiful line of play: 3. Bb5+ Bc6 4. Qe4!! The Black Rook on a8 (a8!?) is now hanging. 4. … Rc8 5. Bxc6 Rxc6 6. Qxc6!! And if the Black Queen takes, then 7. Rd8# (SURPRISE! Our original target appears at the end!)

    2. … Qc6 (This also gave me some trouble, but then I found:) 3. Qf6! Threatening d8 again. If the Black Queen goes back to c7 or c8, then proceed as in the previous two lines.

    2. … Bd5 (This seems to be the hardest to crack)

    3. Qxb4 and White has sufficient material to compensate for the Rook sacrifice, plus the “attack” is not necessarily over; White threatens 4. Bb5, pinning the Black Queen. Of course Black can prevent this with 4. … a6. Then 5. Qxc3 is winning.

    Okay, Tomasz, it’s time to point out what I missed!

  22. Your comments and variations are really sufficient! You caught the idea and various motifs very well! However there is ONE drawback with the last variation of yours. And you called it right!

    2. … Bd5 (This seems to be the hardest to crack) - now please TRY really HARD - if you will do it correctly - I am going to give you even A+!

    (I suffered at this puzzle much more than you my friend - believe me).

    BTW. Finding 4. Qe4!! and explaining it in a right way - it was really impressive performance by you my friend. How could you find out if the Rook at the corner may be hanging?!

  23. For a long time I thought that it was a good idea to learn to solve problems with a much higher rating than your own. For our minds this problem it is a pretty Herculean task. We hardly can do this without making errors along the way. As long as you haven't automated certain subtasks, it simply is too difficult. And hence a waste of time and energy, which in the end might even be discouraging. It's like body building with a weight that is too heavy for you. In the end you get a tiny bit stronger if you are lucky. But it is not efficient.

  24. I will try to make it AS efficient AS possible. It should be a bit harder than my skills and abilities. Not all of the problems are that hard - most of these are about 3-4 moves deep without very hidden motifs or ideas. However I have to solve these problems too see WHERE are my biggest drawbacks (weaknesses). If I can draw serious and vaild conclusions - I will be able to recognize my weaknesses and fix them as fast as possible.

    In addition I am working on my overall level of play - reading some good textbooks. There is no other way to me. I cannot wait for the breakthrough that appears just because I want to play better ;) :)

    Improvement comes when you are able to work OUTSIDE your comfort zone!

  25. PART I:

    Tomasz, my friend! Thank you so much for encouraging further investigation of this position! I have (re-)learned so much while studying it!

    One of Dr. Lasker's maxims is: "If you 'SEE' a good move, 'LOOK' for a better one!" I wish I had taken that to heart before terminating my analysis of this fascinating position.

    During my first attempt, I terminated the analysis of 2. … Bd5 prematurely. Why do I say “prematurely?” Because the Black King was still in the “box,” stuck in the middle of the board, with his Rooks separated and only a couple of pieces to defend him. The assault motif says that we don’t want to give the opponent any time to bring pieces into play for defense, so we want to (a) keep the Black King in the “box” (until he dies of suffocation), and (b) we want to use forcing moves to maintain the initiative. As long as we can keep Black preoccupied with our threats, he has no time to execute any threats of his own.

    After 1. Rxd7!! Qxd7 2. Rd1!, I sloppily analyzed 2. … Bd4, and settled for a “less than best” move 3. Qxb4, threatening 4. Bb5, giving time to grab the c-Pawn with 4. Qxc3. Perhaps it was because I spent over two hours(!!) getting into the position, and “saw” that White had somewhat of an advantage in this line, and (like the amateur that I am) settled for less, without making any serious commitment to looking for the “best” move. (This is an obvious practical failure of demanding the highest level of play from myself; something that I need to work on.) Perhaps it was because I “saw” various attempts at taking advantage of the “criss-cross” attack of the two White Bishops, and thought (WRONGLY!) that 3. Rxd5!! Qxd5 would allow me to “win” the Black Queen (after 3. … Qxd5) with 4. Bb5+ Qd7 5. Bxd7+ Kxd7. It is true that I can win the Black Queen AND continue the attack against the Black King, but it misses the entire spirit of this position.


    What was my original “vision?” The Black King is in a “box” and there is a potential checkmate on d8 using Rook and Bishop. (Bear with me; this is only a "fantasy.") So, let’s revisit that lousy line, with a different attitude: we want only the “best” moves and we are going to throw everything we have at the immobile Black King!

  26. PART II:

    1. Rxd7!! Qxd7 2. Rd1! Bd5

    (I left Fritz 11 running Stockfish DD64 on this position overnight, and the esteemed silicon monster wants to take the White Rook on d1 with 2. ...Qxd1+! I haven’t had a chance to sift through its findings yet, but it does illustrate that I didn't use the computer for my analysis. I did actually consider this but thought - NO WAY I would try to fathom how to dig out of a +-(8.96) hole.)

    Let’s continue to apply the assault motif with a “smash and grab,” aimed at – d8, of course.

    3. Rxd5!!

    That certainly takes gonads, and White had better calculate this variation all the way out, if he wants to avoid getting killed. I note that at this point, Black’s “pressure” against g2 is rapidly disappearing. The only little Black Ninja left is the Black c-Pawn on c3, but I have no intention of letting him into the White “castle” for an assassination of the White King.

    We now get two variations, unless Black would like to just dump a piece. Either the Black Queen can recapture on d5 or the e6 Pawn can recapture on d5. I started with 3. … exd5 because I “saw” the loss of the Black Queen after 3. …Qxd5 4. Bb5 Qd7 5. Bxd7+ Kxd7. But taking the Black Queen is WRONG! 5. Bxd7+ is a “good” move, but it is NOT the “best” move! After I analyzed 3. … exd5 and found the beautiful forcing line hidden there, I retroactively applied the same idea to this line. I’ll show a different 5th move in this line after showing the 3. … exd5 analysis.

    3. … exd5

    I wasted a considerable amount of time (over a half hour) trying to figure out ways to get the White Bishop to b5, winning the Black Queen, but nothing seemed to work. I “saw” 4. c6! Qxc6?? 5. Nd4 Qc7 (any square but c7 allows the White Bishop to check on b5) but couldn’t keep “piling on.” I finally created a “stepping stone” diagram after 3. … exd5 and guess what POPPED up?!?

    4. e6!!

    What does THAT do? It keeps the Black King in the “box” AND it renews the attack against f7 AND it attacks the Black Queen. (Hello! How did I overlook THAT?) Again, Black has two answers, with the same idea in both lines.

    4. … Qxe6

    Too bad, so sad! The line from b5 to e8 is now opened, so the White white-squared cleric can slip the spear into the Black King!

    5. Bb5+ Qd7 (forced)

    And now comes the beautiful move:

    6. Ne5!!

    Wait! Isn’t the White Bishop en prise? Yes, BUT the White Knight is now keeping the lid on the “box!” The Black Queen now has two functions (the function motif makes an appearance): it must get rid of the threat of checkmate from the White Bishop AND it must now guard f7. OOPS – it is overloaded, so Black is about to die in the “box.”

    6. … Qxb5 7. Qxf7#

    6. … Rd8 7. Qxf7#!!

    6. … Rg7 7. Bxd7#

    6. … Be7 7. Bxd7+ Kxd7 8. Qxe7+ Kc6 9. Qd6+ Kb5 10. Nd4+ (stopping the Black c-Pawn from Queening and adding a piece to the King hunt) 10. … Ka5 11. Qc7+ Ka6 12. Qc6+ Ka5 13, Qb5#. There are some alternate ways to reach mate, so I didn’t bother to analyze them; two White pieces (especially Queen and Knight) coordinate quite well in mating the Black King, and I'm sure that I could find the right sequence to drive the Black King into mate.

    4. … fxe6 5. Ne5!!

    Now the White Knight is maintaining the “box” around the Black King, hitting the f7 square with Knight and Queen, and forcing the Black Queen to move off of d7, opening the way for the cleric assassin on b5 – to divert the Black Queen!

    5. … Qb7 6. Bb5+! Qxb5 7. Qxf7#

    I now began to have fantasies about using this same “idea” in the 3. …Qxd5 4. Bb5 Qd7 line.

    5. e6!!

    Our “box” around the Black King is still intact! There are the two answers:

    5. … Qxb5 6. Qxf7#

    5. … fxe6 6. Ne5! And the Black King is going to be annihilated again.

    That was an absolutely WONDERFUL study position!

  27. I have to learn to stop writing up this analysis in the wee hours of the morning after work.

    There is an error here: 6. … Be7 7. Bxd7+ Kxd7 8. Qxe7+ Kc6 9. Qd6+ Kb5 10. Nd4+ (stopping the Black c-Pawn from Queening and adding a piece to the King hunt) 10. … Ka5 11. Qc7+ Ka6 12. Qc6+ Ka5 13, Qb5#. There are some alternate ways to reach mate, so I didn’t bother to analyze them; two White pieces (especially Queen and Knight) coordinate quite well in mating the Black King, and I'm sure that I could find the right sequence to drive the Black King into mate.

    I apologize for that oversight on the 8th move; I got ahead of myself: I have to get the Q to f7 FIRST!

    I'll try to fix it tomorrow, if I have time.

  28. I cleaned up the really sloppy “analysis” (such as it is) of the 2. … Bd5 line, using a board to physically SEE the actual piece positions as the variations unfold, instead of trying to do this from memory. It’s amazing how much I tangled up everything while trying to rely on memory of analysis from hours earlier. Back to the “visualization salt mine” for me! In any event, it seems clear to me that White is winning in all lines. (If you find some alternative that I screwed up, please bring it to my attention.)

    1. Rxd7!!

    1. … Kxd7 3. Qxf7+ Kc8 4. Qxg8 and White is winning.

    1. … Qxd7 2. Rd1!

    (Stockfish DD64 “thinks” that 2. … Qxd1+ 3. Bxd1 is the “best” alternative out of lots of bad alternatives. I think Black is just lost, down Queen & Knight for two Rooks, and no compensating attack against the White King. Suprisingly, 2. … Bd5 is pretty far down on the list.)

    2. … Bd5 3. Rxd5!!


    (a) 3. … exd5 4. e6!!

    (a1) 4. … Qxe6 5. Bb5+ Qd7 (forced) 6. Ne5!!

    (a11) 6. … Qxb5 7. Qxf7#

    (a12) 6. … Rd8 7. Qxf7#!!

    (a13) 6. … Rg7 7. Bxd7#

    (a14) 6. … Be7 7. Bxd7+ Kd8 (7. … Kf8 8. Qxf7#) 8. Bxe7+ Kxe7 9. Qxf7+ Kd8 10. Qxg8+ Kc7 11. Qxa8 and the White Knight prevents the Black c3 Pawn from queening with 12. Nd3 (when needed). There are alternative moves for Black in that analysis, but White has a crushing attack against the Black King.

    (a2) 4. … fxe6 5. Ne5!!

    (a21) 5. … Qxb5 6. Qxf7#

    (a22) 5. … Rd8 6. Qxf7#!!

    (a23) 5. … Rg7 6. Bxd7#

    (a24) 5. … Be7 6. Bxd7+ (Same idea as [a14], just one move earlier)


    (b) 3. … Qxd5 4. Bb5+ Qd7 5. Bxd7+ Kxd7 6. Qxb4 is winning.


    I think there are several things I “learned” from this problem. (1) If there are still forcing moves and some way to continue using the assault motif, then continue analyzing to quiescence; (2) There are “hidden” resources in most non-trivial positions but we have to search diligently for the “best” moves; and (3) I need to work on increasing the accuracy of my mental visualization of the positions as the variations unfold.