Wednesday, February 24, 2016

Salt mining in progress

 Latest update:  2016, august 11

This table contains the results of people who do the exercises (AKA the salt mines) listed in my sidebar. The results are registered in "average correct clicks per minute". If you want your results to be on this list, apply them via the comments. This list is added to my sidebar. The initial result is the first stable result after you have adapted to the exercise, screen layout, method of clicking and so on. Usually after a few hundred exercises. The current result is what you can maintain over a longer period of time, after some serious salt mining. In the beginning, it will be the same as your initial result.

Tempo Tomasz Aox Lain Munich

Init-curr Init-curr Init-Curr Init-Curr Init-Curr Init-Curr
Find all attackers 6 18 50 50 ?? 44
20 56
Find all defended pieces 7 18 84 84 ?? 52
15 57
FAC (Find all checks) 11 20 33 42

9 40
FAC BLACK K+P 21 35 46 55



FAC BLACK K+WHITE N 30 42 75 84 54 54


FAC BLACK K+WHITE Q 20 35 63 63 42 42


FAC BLACK KING ONLY 19 35 57 57



FAC NO QUEENS 20 25 37 46



Find all escape squares - Easy 25 35 49 69 40 42


Find all escape squares - Hard 18 24 38 52 22 22


M1-e (Mate in 1 easy) 12 16 31 39 15 15 8 8 26 33
M1-e BLACK K+P ONLY
23 27



M1-e BLACK KING ONLY
25 25



M1-h (Mate in 1 hard) 2 3 7 7



M1-h BLACK K+P ONLY
15 15



M1-h BLACK KING ONLY
17 17



M2-e Easy





Find All Pinning Pieces





Find All Protecting Pieces





Find All Blocking Pieces






In the salt mines

73 comments:

  1. HUGE thanks! :). Now we could have the instant access to the results and see how they are changing :).

    My comfort zone - Mate in 1 EASY - 33 MPM

    BTW. I changed my approach - now I will try to do it as correct as possible and AFTER I reach/build very solid algorithm (method) - I will do it faster. This way I will be able to see what are the problems with my solving. Now I understand why previously there was such a BIG gap between me any you guys (you have been solving correct and I have been shooting all the probable checks squares as fast as I could).

    ReplyDelete
  2. Mate in 1 - easy - BLACK K+P ONLY - 23 MPM (initial score, after 450 puzzles).

    As far as I analysed the puzzles while solving - there are "cutting" the lines mates - covering/interposing the line-pieces (Q, B or R) influence with a mate moves. These are quite hard to spot for me (many mistakes at these puzzles). And good news - I am getting better at the long (mating) moves with Q.

    ReplyDelete
  3. Mate in 1 - easy BLACK KING ONLY - 25 MPM (intial score, after 530 puzzles).

    This set is quite small one - I can memorize it within a week. That may influence the score a bit. Anyway I learnt a new trick at mating with a lot of pieces against bare King!

    ReplyDelete
    Replies
    1. Isn't that funny? You are doing the main task faster than the subtasks.

      Delete
    2. That could be explained by Tomasz improving at "m1 as a whole" and not by improving at subtasks

      Delete
    3. The M1-e black K+P forms a subset of only 2% of the whole set. These are the kind of tasks that hold Tomasz back at M1-e as a whole.

      Delete
  4. the initial value at many exercises is unknown as we did start with these exercises with a different rating system.
    And new saltminer will need more exercises (1000+?) to find their baseline.
    O god.. what have i done.. im mining salt again...

    ReplyDelete
  5. Attackers 44.45
    Defenders 52.32
    There was no initial value in attackers and defenders.. they where improvable and i estimatingly could still improve further

    ReplyDelete
  6. M1-e (Mate in 1 easy): 8 MPM after ~300 problems. I haven't tried other exercises more than ~50 problems.

    I got stuck in a lot of puzzles. I'm bellow 1500 skill and I'm interested in non-conventional/more efficient learning process. Sometimes I do tactics at chess.com (that's all my "training"), but I don't improve significantly, I'm too slow solving problems and being aware of board things. My chess experience is inexistent.

    That's explain my M1-e score. Ahahaha.

    ReplyDelete
    Replies
    1. 300 problems are not enough. I am restarting, made 1000+, and still forget sometimes that i dont have to click the piece i "want to move".
      The process of doing these puzzles with this GUI (graphical user interface) need to be automatic or it will take resources/robbing time.
      In "theory" you get ~~twice as fast if you gain ~~200 Elopoints ( i think you gain more speed in complex situations than in easy situations ) and there is some noise ( there is some delay to see the board.. ).
      Based on the idea that improvement is a matter of simplicity you estimatingly will have to do more and easier exercises than the better player.. but thats something we have to find more evidence.

      Delete
  7. @tempo,
    you had an improvement in m1-h, i wonder if it is possible to see anything changed in your tactic performance at chesstempo. IF you realy gained speed in an relevant subtask of m1-h this should have invluence at CT-Blitz.. i would be not astonished if these are now ..say 30 or even 50 points more, depending on the general importance of that skill. If this skill is too m1-h specific then..

    ReplyDelete
  8. Mate in 1 - hard - Mates / min: 6.650 (Done: 330)

    It is more and more clear that I have big difficulty to recognize (all) attacked squares, protected pieces and pins. The most painful is when I cannot see all the checks (!) or do not see how to attacked a few squares at once (like a blind spot). Most often I can work out which squares are the King's Escape Squares, but I have great problems to find the move that attacked them and makes the move LEGAL!

    Next tool (excercise) should be called: "Is this check legal?" (ITCL) LOL. It should be based on pinned pieces and recognizing between legal and illegal moves may fasten the process of finding the CORRECT moves!

    BTW. Was it Aox who mentioned than GMs are able to recognize if the King is in check in 145ms? In my case it would be 145s! LOL

    ReplyDelete
    Replies
    1. Which painfully demonstrates that being good at M1-e has very little effect on M1-h.

      I hope that Lain is making progress with the new exercises! Thus far he has done an excellent job!

      Delete
    2. There is the M1-h subsets FEN list:

      http://puu.sh/nmFct/1fcaabade8.txt ("raw list")
      http://puu.sh/nmF2k/cc56814e5c.txt ("parsed list")

      Some positions are duplicated so they are removed.

      We need to make the GUI and do more testing.

      Delete
  9. It looks like EST-hard is dedicated to me ;) :). Hard to say why I could score such good numbers, but it is probably NOT the place I can improve ;) :)

    Escape-squares Training - hard: Done: 418 (Squares / min: 38.24)

    PS. Back to the saltmines! I am not sure, but I am strongly convinced our group should be called "desperate saltimers and chess ideas testers".

    BTW. It is great we have so many tools (subtasks) to excercise, because sometimes I score such poor that I wonder if I am retarded or intelectually disabled!

    ReplyDelete
    Replies
    1. Evidently, M1-e has prepared you well for the FAES-hard! That's interesting!

      BTW, I recognize that feeling of being retarded and intellectually disabled!

      Delete
  10. Let's show a bit (a pinch) of excellence ;) :)

    Escape-squares Training - Done: 1027 (Squares / min: 52.16). Probably 60-65 may be a FM-level. I simply LOOOOOVE this excercise. It is like a fresh running around your backyard when you have already run the marathon ;) :)

    I suspect that reaching 50 SPM (by others) at this excercise requires some time and effort.

    I have big problems to express how much FAES gives me joy and pleasure! And at first I thought there were some mistakes. However after a while I spot the additional support "0-0 and 0-0-0" as moving King 2 squares from the initial square. I want to thank you once again my friends - these excercises are simply EXCELLENT even if at most I have such difficulties that I often think to give it up (as a progress at my case is close to nil/zero).

    BTW. It would be a good ideas to express and share the kinds of problems we face at these "salt miners". And discussing some ideas (and the ways how to correct bad habits) may be more than welcome! :)

    ReplyDelete
  11. Mate in 1 - hard - BLACK K+P ONLY - Done: 776 (Mates / min: 15.20).

    These puzzles are soooo amazing! The gold old days has come back. It bring memory - when I have been solving such puzzles (just about 800-1200, but the studies and etiudes are amazingly beautiful).

    Nice and funny fact: I though there are a "bunch" of puzzles at this case and I solved these a bit slower trying to remember (memorize) most of them. And after a while I realized that "the set is not going to repeat that much - let's look how many positions are there at the database". And I found out there are Total:
    13527 puzzles (LOL). Imagine solving puzzles with the concious intention to memorize these and realizing you were tricked by yourself :D.

    Anyway I LOVE these easier puzzles. My hope for a bit of progress light up! ;) :). As the athletes say: NO PAIN - NO GAIN! :)

    ReplyDelete
  12. Find all checks FAC - Done: 1727 (Check / min: 33.12)

    ReplyDelete
  13. Mate in 1 hard BLACK KING ONLY - Done: 511 (Mates / min: 16.92)

    ReplyDelete
  14. Now I am solving "Find all checks" (FAC) all the remaining types of puzzles. I want to have set the baseline (initial level) as fast as possible. After that we can see what subtasks are improved by my practice.

    If anybody wants to comment my numbers (scores) please just do it! I am curious if we can draw any conclusions related to the low/high scores I achieved at these puzzles. Especially as it is the FIRST step (to set up the initial level).

    And let me know how are you doing my friends :). Did you find a gold salt stone or we should be more patiente to discover the Holy Grail of our saltmining process?! What's your opinion? :)

    ReplyDelete
  15. Find all checks - FAC BLACK K+P - Done: 777 (Check / min: 45.79)

    ReplyDelete
  16. @TEMPO

    I have a big request to you: COULD YOU check out the links and headers to these? I think they need some correction. For example I clicked: Find all checks BLACK K+WHITE N ONLY - and the header at the website is: "Mate in 1 - easy; Select all squares where black can be checked WHITE KNIGHT ONLY". And at the Find all checks BLACK K+P ONLY there is "Check Training; Select all squares where white can go to give a check BLACK K+P ONLY".

    I am not complaining, but if there is an order we can avoid errors and misunderstanings. What is correct (as far as I noticed)? The links seem to have a good address and name inside it (chessex/fac_kp.html - I can figure it out as "Find All Checks - K+p"). Thank you very much in advance!

    ReplyDelete
    Replies
    1. The titles above the exercises used to be correct, but the latest updates because of the time calculation have overwritten these titles. I will correct it.

      Delete

  17. FAC - Find all checks BLACK K+WHITE N ONLY - Done: 910 (Mates / min: 74.74). As a initial point - it is not that bad ;) :). Although I could not mate the King - I have checked it a lot of times. Fantastic excercise to the kids! And it looks like I discovered a few patterns related to King and Knight distance :). Thanks a million for these amazing chess tools! :) Long Live Tempo! :)

    ReplyDelete
  18. FAC - Find all checks BLACK K+WHITE Q ONLY - Done: 868 (Mates / min: 63.05). The system (algorithm) of making checks when moving the queen... is in a progress ;) :). The excercises give me a lot of inspiration and it is a food for thoughts! :)

    ReplyDelete
  19. FAC - Find all checks BLACK KING ONLY - Done: 832 (Check / min: 56.70)

    ReplyDelete
  20. FAC Find all checks NO QUEENS - Done: 610 (Check / min: 36.83)

    ReplyDelete
  21. FAC Find all checks NO QUEENS - Done: 270 (Check / min: 45.65)

    ReplyDelete
  22. Find all checks FAC - Done: 488 (Check / min: 41.53)

    ReplyDelete
  23. FAES - Find all escape squares (Escape-squares Training) - Done: 488 (Squares / min: 49.23)

    ReplyDelete
  24. My new comfort zone - Mate in 1 EASY - 35 MPM (Done: 423; Mates / min: 35.11). It is the excercise that may be quite hard to break 40 MPM barrier. Anyway it is possible, but require a very good position comprehension :)

    ReplyDelete
  25. Mate in 1 - easy - BLACK K+P ONLY - 26 MPM (Done: 197 - Mates / min: 26.64).

    ReplyDelete
  26. FAES - Find all escape squares (Escape-squares Training) - Done: 955 (Squares / min: 61.60)

    ReplyDelete
  27. Still have problems with my salt allergie

    m1-e start = 15

    ReplyDelete
  28. FAES - Find all escape squares (Escape-squares Training) - Done: 656 (Squares / min: 67.27)

    ReplyDelete
  29. My tollerance for salt improves slowly i start feeling some flow at FAES-h, MAAYBE it becomes improvable? I will give the FAES-h- start value end of the week, estimatingly 17? 18?

    ReplyDelete
  30. Find all checks BLACK K+WHITE Q ONLY "Start" 42
    I did related exercises before
    The trick here is a strict thinking process, step along the rays of either the king or the queen.

    ReplyDelete
  31. FAES - Find all escape squares (Escape-squares Training) - Done: 463 (Squares / min: 69.45)

    ReplyDelete
  32. Find all checks BLACK K+WHITE N ONLY Start 54, i will never be much faster in any exercise.

    ReplyDelete
  33. FAC - Find all checks BLACK K+WHITE N ONLY - Done: 110 (Mates / min: 84.20). The max score has been 99.78, but it was just for a one minute or so. There is about 20-25% increase of speed when you change the method of solving - from clicking with the mouse - against clicking directly at the screen. I have been using tablet to achieve such score. If I will be able to borrow one for a few days I will try to "shot the score" up to 90 clicks/checks per minute with at least 200 puzzles ("...to maintain over a longer period of time").

    Aox - how could you comment this kind of progress? If we assume I started the initial level at 70 CPS and I would be able to increase it up to 95 CPS - is it any significant progress or just a statistical noise?

    ReplyDelete
  34. FAC - Find all checks BLACK K+WHITE N ONLY - Done: 1103 (Mates / min: 84.25).

    I scored this while using Chrome as Firefox is a few times slower. I think I could achieve even 100 CPS when using tablet. Should I score 100 at this excercise or it is not needed (necessary)? What are your suggestions? If I could score 110 - would it be a significant progress (starting wit 64 it would be 60% progress). What are your opinion? Aox? Tempo? Is it worth the effort or this excercise is not a good indicator of any progress?

    ReplyDelete
  35. Tomasz we do these Board Vision Exrcises to make m1 improvable.
    There is a lot of noise in the data, the hand - eye coordination for example, delays by slow browsers and so on
    But what we want to know, what we try to do is to improve in a "real" task, something with OTB relevance. We already know that ( some ? ) simple tasks are improvable.
    If you change the browser and use a tablet than we cant compare your data with the datas before ( it would be a better! training though in my opinion ).

    The key tasks are m1-e and m1-h. The Goal is to get quicker there. If we want to see/measure an improvement we need to keep everything else the same. So you MAY use a tablet with chrome.. but the datas at this page should be measured with mouse and your common browser.

    ReplyDelete
  36. At the moment it seems that FAES-h is improvable for me and it seems that it is sufficient to make m1-e improvable for me.
    I did a lot of bv exercises before, so this ( estimated ) experience cant be typical, not even for other tacticians of my strength.
    I hope my opponents dont read this blog ;)

    ReplyDelete
    Replies
    1. There are about 200 regular readers of this blog, and only two or so are from Poland. So little chance that it are your opponents.

      Delete
  37. @Tomasz, the goal is that a subtask no longer consumes time, energy and critical brain resources when doing the main task. Is your performance at FAC hampered by your ability to see knight checks? Probably that isn't your main problem any more.

    The main criterion is, what is going on in your head. Is it fluent (like speech) or not?

    ReplyDelete
  38. Aox Tempo - thanks for your answers.

    Aox: I just wanted to inform you that the system of testing the data (the tests) are not optimal - mostly due to the slow reaction tools like the browser, application, internet connection, mouse clicks/movement and that like. That means - most scores may be improved up to 20-25% from the highest scores achieved by me.

    The mate in 1-hard are too complex to me. They are a bit easier than a month ago, but I cannot see the real breakthrough. There are TOO MANY elements and it is extremally hard NOT to see these "disturbing ones". In other words - it is hard to focus only at the most important ones. The opposite is with the K vs N excercise. There is just two pieces and the goal is a very simple one - make a check (or max. 2 checks). That's why I can improve at this task up to 84 CPM.


    Tempo: I do not get how you want to achieve such a level that the task will no longer consume time or energy. Is it the same as I could say: "the energy and time level is the lowest?" (but not zero). Am i right? Or the only goal is to do these task with no effort at all? Is it the same as walking or breathing? Some kind of full automatic processes?

    My FAC is most often hampered by the "long moves" or illegal ones. Sometimes I can find "all minus one moves" and finidng the last one requires TOO much effort, time and energy. That's my problem. You can probably face similar problems as well.

    And what do you mean by saying: "what is going on in your head. Is it fluent (like speech) or not?" - all the subtasks or the specific one? The FAC, FAES or M1? Could you specify your question a bit more? Thanks in advance.

    BTW. I have improved my search algorithm with Knight checks (N vs K subtask) and most often it is the mouse that slows me down against achieving a score of 90 CPM.

    PS. Get back to the salt mines - maybe this time I would be able to find some golden nugget? ;) :)

    ReplyDelete
    Replies
    1. If a (sub)task becomes internalized, you are no longer consciously aware of doing it. That is the level of improvement we should be looking for. If you are talking, you are thinking about what you are going to say. You don't think about the process of speech itself, that flows automatic.

      When you drive a car, you think about where you are heading. You don't think how to shift gears.

      With chess, you are thinking about how to kill the king. You are not thinking about the rules, they are applied automatically. You once told me, that you did M1-e, without any thinking. You simply processed the positions automatically. That is the idea.

      I wrote a post about little disruptions. http://temposchlucker.blogspot.nl/2016/02/little-disruptions-with-big-effects.html That showed that certain elements were automated, but when shifting from one element to another, a conscious thought was needed. I gave the following example of a disruption:
      Rook h4 can take h5 with check but not e4

      What is fluent: I know the two places where the white rook can give check. I know I can take the knight with check.
      What is not fluent: I want to give check on e4, but bummers! there is a pawn on e4. I don't know what to do! Giving back the situation to the conscious thinking again, who decides after a few seconds of thinking to go on with the knight that can give a check.

      It is as if you are approaching a bend in the road, and the unconscious mind gives back the control to the conscious mind, and you consciously decide if you shift back the gears to 2 or 3.

      There is a lot in chess that is so basic, that it can be decided automatically. Just like the shifting of gears shouldn't require conscious intervention.

      When you attack a piece and it is defended, you should automatically look for how to attack the defender. There is no need to give the control back to your conscious mind. The conscious mind takes a fresh look and sees every position as "new". That's why it takes time before the conscious mind has adapted, and decides to go on with "look for a way to attack the defender".

      Delete
    2. The question "should I continue K+N to 95 CPM or 100 CPM?" should be replaced by: "what subtask of M1-h consumes the most time?". It is much more profitable to work on the weakest link.

      Delete
    3. Salt is mined in layers. These layers are deposited when seawater evaporates. When a cubic kilometre seawater evaporates, you get a layer of 300 meter salt. A cubic kilometre seawater contains 4-6 kilogram gold. That must be somewhere in the layer of salt.

      Delete
  39. Tempo said "The question "should I continue K+N to 95 CPM or 100 CPM?" should be replaced by: "what subtask of M1-h consumes the most time?". It is much more profitable to work on the weakest link."

    I say: !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

    We all might benefit if you could find an important subtask of m1-h which consumes a lot of your time at m1-h

    ReplyDelete
  40. At present I cannot break the score of 35 MPM at M1e. Anyway I was able to maintain the speed of 34 MPM all along 2000 M1e puzzles. It was not that easy, but I did it. In most of the case I could do these puzzles at the steady speed - the small difference is at the first 10-20 puzzles and the last 200. The initial process takes about 10-20 seconds to warm up my mind and the last 2-3 hundred - I had to slown down a little bit due to the fatigue and being very tired (not to mention a painful headache).

    I am the slowest at FAC (Find all checks) and I do not know what should be done to change it. Maybe I should do the chessgym puzzles (attackers and defenders). And if we talk about the Mate in 1 HARD - they are simply too COMPLEX to me - too many distractions all over the board. It is the same if I would be forced to speak simultaneuously with 20 people. Speaking with a few (3-4) in one time would be quite hard, but in one moment with 20 people? Close to impossible - at least to me!

    And how are you doing guys? What are your problems with the puzzles? Any conclusions? Any ideas? Any further problems or discoveries?

    ReplyDelete
    Replies
    1. I have some minor health problems, so i dont do that much for some days. But i improve in (very) small steps at FAES and M1-e. Now with the ney FAES-E puzzles i hope i can speed up a little?

      You may try to improve at every subtask of M1 and if that dont work at any of the subtasks of these subtasks. I have the impression that some of these exercises create synergies.
      The chessgym exercises are basic, we cant be too good in these, but there are more basic skills. One is the memory of the position. To give a check with a piece means: you need to recall where the king is, where the piece is, where the squares are where a check possibly could be given and what pieces might be blocking . Of course there might be many other and even more important subtasks of FAC.

      Delete
    2. There is one problem - I improved all the excercises (mentioned at the table) to the best score available (to my present abililties and skills). Maybe I should try another puzzles like memory, setting the pieces and aura vision. And probably attackers and defenders may be helpful as well.

      What are the priorities at the significant of improvement?
      1) Mate in 1 hard
      2) ?
      3) ?
      4) ?
      5) ?

      I mean - what are the next positions after M1h, that I could improve and we could call it "important and significant discovery". I know you want to improve at M1h, but if that would NOT be possible - what are other tasks (subtasks)? Can you list these (especially in my case - comparing at my results at the salt mine table).

      Delete
    3. I personally dont want to improve in m1-h , i want to improve in m1-e. M1-e is OTB relevant and m1-h is in the moment estimatingly to hard for me.
      M2 might be the next logical step and / or forks as a most common tatctics and/or pins because they have some relation to m1 too.
      Are you shure you cant get better in these exercises? I did need weeks to improve. I will work at m1-e for maybe months until i hope to reach 30+.
      And then i will analyse what the effect on my tactics is.

      The datas from Chesstempo will show me how my performance at different tactical motives will change

      Delete
  41. I made one important test. I tried to solve 2000 puzzles M1e. The average score I achieved is 30 correct clicks/mates (up to 32 Mates) per minutes. The current result is what I can maintain over a longer period of time, after some serious salt mining. What's more - I could do it even at the distanse of 4-5K in one session. And I do not use more than 5% of my overall energy (mental power). In other words I can say - I do not feel it is the (sub)task that tires me. I could do it simultaneusly with doing multiplication at the same time.

    Is it the level we could call it automatic? I mean - is it the level this excercise (task) is done at the subconcious level?

    I think a much better idea to me would be to test my skills at M2e. I do not know what could be done to improve significantly at M1h. What's more - no one has scored better performance (result) so far (at M1h). If this will be done - I can try to come back to this task and see how can I improve this score.

    My estimation is that I could solve ANY mate in 1 (easy) withing 2-3 seconds. There are about 2-3% of puzzles I have problems and I need about 10-15 seconds to solve it. What does it mean? Over 95% of these puzzles are cracked by me. I probably have solved about 100-120K of puzzles with the goal of memorizing and creating the best possible method (algorithm). It is not perfect, but a very efficient one - it does not use much energy or time and guarantees the correct solution(s).

    And to your last question Aox: for now I do not know what methods can improve my score (performance) at M1h. I do not think it is the matter of time, but rather the excercises that slow down my process of solving this kind of hard mate puzzles.

    I am looking forward to your performance and explaining what subtasks are needed to improve (significantly) at M1h (or even M1e - you are not yet an expert level).

    ReplyDelete
  42. I did the same test Mate in 1 - easy (BLACK K+P ONLY) - Done: 1701; Mates / min: 25.54. I could maintain the level of 24-25 MPM at the distance of 1700 puzzles. I had to use a bit more energy and time (about 10% of my mental energy), but in general I did not feel tired. I can do these puzzles and count some number at the same time.

    My performance of these puzzles is about 95-96% (4-5% I can have difficulty to solve withing 2-3 seconds). Anyway I do not find these puzzles difficult. What's more - I could do it even at the distanse of 3-4K in one session.

    It means mate in one easy is more or less perfect (fluent) in my case. There are others factors that stop me from reaching a high score at M1h. Or maybe it is very true with Robert Noble statement that "even masters would have BIG problems with such untypical puzzles". The level of complexity (difficulty) may be simply too high and the problems may be too much abstract. The more I think about it - the more I believe it. If we compare these puzzles to the REAL game positions - you can find just 3-5% of such positions at practical (real) chess games. That's why we may NEVER be able to solve such puzzles with the speed of 25-30 MPM unless we find extremally efficient methods (not a good ones).

    Now I try to see how I will cope with M2 puzzles. These puzzles should give me some kind of fresh view to the other ones and we will see if there is some correlations between fluency at M1e (32-35MPM) and these ones M2e.

    BTW. Tempo - did you intentionally left M2e puzzles outside your salt mine table scores? Or did you decide they are not relevant to our tests? I suggest they should have been added to the table, but it is up to you to do so.

    ReplyDelete
    Replies
    1. To become better at M1-h, you have to become better at the following four subtasks:

      Find blocking pieces
      Find escape square creating piece
      Find pinning pieces
      Find protecting pieces

      These exercises are added to my sidebar. Some of these subtasks are improvable, others are not. Find the pinning pieces looks improvable. Find the protecting pieces looks too complex. We must find out if we can dissect that one.

      Delete
    2. These four subtasks are essential for any tactic. Any attack except by the knight can be blocked. Any attack can be prevented by protecting the attacking square. Any attack can be spoiled by a pinned attacker.

      There are only two kinds of attack: a duplo attack which causes a time problem (must save two targets with one move), and a restraining attack which is based on a space problem. Only the subtask with an escape square creating piece is limited to a restraining attack. Of which mate is an ultimate example.

      Delete
    3. @Tempo:

      "To become better at M1-h, you have to become better at the following four subtasks:

      Find blocking pieces
      Find escape square creating piece
      Find pinning pieces
      Find protecting pieces"

      Could you EXPLAIN how to solve these (sub)tasks? I tried to find out by myself, but I do not understand how they work. I am going to test my skills using these puzzles, but I have to know how to handle them. Please explain these to me as simple as possible. I want to be 100% sure if I solve them in a correct way (to score a high result).

      "Any attack except by the knight can be blocked." - you probably missed the pawn attack - it cannot be blocked either.

      Thank you in advance for your answer :)

      Delete
  43. Tempo said "Find the protecting pieces looks too complex. We must find out if we can dissect that one."

    There are 3 possible methods to handle a check

    1. move king
    2. block attacker
    3. capture attacker

    If methods 1-3 are not possible its a checkmate.
    "Find all protecting pieces" is about check which would be a checkmate according to 1. and 2. but not 3.

    So some subtasks with already existing exercises are
    FAC
    FAES ( 1. = possible moves of king )
    Find all blocking pieces
    attacker
    defender
    ...

    If you improve at these and or other tasks you might make Find the protecting pieces improvable

    ReplyDelete
  44. M1-e (Mate in 1 easy) - Done: 1004; Mates / min: 38.52.

    I used Google Chrome as it responds a lot faster (8-10x) than Firefox. I am sure Find All The Checks tool (excercise) helped to break this score. I could even manage 40 MPM at the beginning. The biggest drawback of mine is the lack to see all the checks. If I fix this problem - I can break 40 MPM at the distance of at least 400-500 (as a stable result).

    It looks like solving puzzles FAC and FALM (first) should be obligatory to everyone who wants to make a high score at M1e. I do not know what range of influence it has at the M1h.

    ReplyDelete
  45. What is Falm?
    I will change to chrome too.. Firefox is realy slow :/

    ReplyDelete
  46. FAES-e ~42
    Faes-h ~22

    The difference between faes-e and faes-h is that i can "spot" the whole position at faes-e and work from memory and experience ( chunks ) while at faes-h i have to look and screen again and again.
    I "feel" that faes-e deepen my chunks ( = collections of pieces ) while faes-h forces me back to the single piece ( which might not be bad.. ). I see the factor position memorisation.

    ReplyDelete
  47. M1-e (Mate in 1 easy) - Done: 1013; Mates / min: 37.00.

    I practiced a bit "Find All Checks" and as far as I can notice - this puzzle helps to improve M1e. I do not have to search for all the possible checks and missing "long move" (especially done by Queen or Bishop) is now a very seldom.

    BTW. Tempo - you have not updated my result at M1e (now it is 37.00 and at 11th of March - 38.52). I underline it just to your attention.

    ReplyDelete
  48. Find all checks - FAC BLACK K+P - Done: 1008 (Check / min: 54.73).

    3 weeks ago I scored 46 only. Now (today) I improved up to 55 CPM. Is it a visible improvement or just a statistical noise (non-important change).

    ReplyDelete
  49. I am currently on holiday in Poland, and thus dont have access to my laptop back in England. So the following are not checked but from my memory. I hope I have it approximately correct.


    attackers (3 min sessions):
    initial (after 30 minutes) = 20 attacks per minute
    current = 56 (ATH)

    defenders:
    initial (after 30 minutes) = 15 undefended per minute
    current = 57 (ATH)

    finding checks:
    initial (after 30 minutes) = 8 or 10 or so checks per minute
    current = about 40 (ATH)

    M1-e:
    initial (after 30 min) =25 or 27 M1 per minut
    current = about 33

    (with a subset of just 3x300 M1-e (so 900 puzzles all together) I achieved about 38 as ATH. Not that much faster.

    I also trained "find all forks" and improved here about 50% more forks.
    Maybe aox remembers where to find this boardvision training?

    I had some transfer from M1e to M1h.
    When I did M1h, I mostly solved about 4 M1h. It depends - sometimes I got stuck and solved none per minute, on other occasions I solved 7 or 8 M1h.

    But after a lot of M1e - I felt some transfer, and was able to more often score regularly about 6-7 M1h, with less often being stuck. It is not that I could beat my old ATH in M1h, but I often had higher numbers on average. So a more stable "6 M1h per minute".

    I felt some transfer to chesstempo puzzles with my set "Mate in 2", and was able to indentify a few patterns I wasnt aware of previously.
    It is difficult to quantify "how much did I benefit", but I am sure I had some transfer. I felt it like "I know this pattern from the M1e-training". So I had active (conscious) recalls.

    I also felt a lot of transfer in the attackers and undefended pieces boardvision training. After this training I saw way more often backwards or sideways moves, and also spoted "hanging piece" problems very fast, too. So it had a positive effect on my CT Blitz rating. Again, it is difficult to quantify this, but the recalls in my thinking was definitively present, which serves me as prove that it had some effect.
    I trained it for some days, then plateaued and thus left it, then came back after some months and picked up training again - and usually was able to overcome my previous plateaus.

    So maybe, leave it for some time - but please do come back some time later?
    I see you trained in february, march, so maybe enough time has passed. Give it a try.
    Whenever I came back, it didnt take long to gain my old ATH, and the next day or so I already achieved a new ATH.

    And in case it really doesnt work for some or all of you: I myself improved about 130-150 CT Blitz rating points only. Other areas in chess can make you stronger, too. Again - at least other areas helped me a lot to improve my real rating and overall chess-ability. A few years ago, I would have said: "if you can not improve tactically - you can forget about improving."
    But if you ask me today - I disagree with this assumption today. You can improve without improved tactical vision, too. At least I think I did. 2 chess players who have a similar real rating like me are much better tactically than me. I assume so, because I lose most of the time in blitz games against them, and I assume that blitz is more about tactics than long OTB games.

    ReplyDelete
    Replies
    1. @Munich

      After 6 or 7 months break I came back to see what are the present results. So far I have seen I can make 36-38 MPM at M1-e. If others had solved the puzzles I could have joined them. Otherwise - it make no sense to me.

      Nowadays I set up the goal to solve 16K puzzles from workbooks and manuals (paper books only). This way I want to reach 2100-2200 strenght at tactics. I am going to read a few books related to attack and counting variations (f.e. "Forcing chess moves" by Hertan) at this process. My estimation is to finish solving at the year 2020. If anybody would be interested at my results - just remind me about it in 3,5 year :) ;)

      Good luck to others!

      Delete