Saturday, September 27, 2014

Mate in 1 as Board Vision Exercise


White to move
#1
There are 3 substancially different Methods to solve such a problem.

I )  ( plain ) Calculation ( using the knowledge we are looking for 1# )
          1) Ra2, no check
          2) Rb1, no check
               .............
          n) Nd6 check. Is it mate? Can i take the rook? No.... Can i move away?  Ke7 still check Kd8, no check
              .......


II)  ( almost plain ) Pattern recognition. This is a typical KRvK Endgame posiition  ( if you can "substract" all other pieces! ). This method narrows the vision to a few pieces.

III)  ( plain ) Board Vision
Board Vision ( = Chess Board Vision = Chess Vision = Board Awarness ) is the recognition of "whats going on at the chessboard" now. Without "seeing" whats going on now you cant see "it" in the future either. 

We see that the black king has 3 squares d8,e8 and f8. Only 2 pieces can attack such squares at once, rooks and queens. Now you look along this rank and intersect with the possible moves of the white rooks and Queens. The intersection is at h8.

There is a wonderful explanation of the benefit of advanced Board Vision  --> here! <--

The method ( (motor) skill ) of "intersection" is explained in GM Maurice Ashley's Book: The Most Valuable Skills in Chess , which is a wonderful explanaition to his "Skill Trainer" , but i realy dont think that this book is interesting for the average chess beginner.

Wednesday, September 24, 2014

14 000 new "Board Vision" Exercises : Mate in 1

The 80 000 "Mate in 1" Exercises are seemingly to hard to generate a "quick" improvement. I already generated a useful tool to help to improve the "Mate in 1" - skills, thats the "Find all checks" Exercises ( see the links at the right top of this Webpage ). But the 80 000 Mate in 1 Exercises remain hard. They where generated from chess Problems. The Positions in Chessproblems are not "common", many moves are not legal because of pins... So there are too many things to learn to improve any skills ( quick ).

So i did generate 14000 new and "easy" Mate in 1 puzzles extracted from tactic-puzzles.

Play here
Explanations how to play 

Friday, September 19, 2014

The Myth of "Bad positional play of chessengines"

Its often said: Chessengines are bad in positional play and/or especially bad in closed positions. I think humans are not good enough to understand the brilliant positional play of the present engines.

 

See here a game Stockfish (Elo~3200+) vs. Cyborg Nakamura (Elo 2750+)  & Rybka (Elo ~3000)
A perfect closed position and Rybkamura was outplayed by stockfish!

I did check the "positional" Komodo 8, Houdini 4 and Stockfish 5 with the Strategic Test Suite 1-15, without Endgame-Tablebase, 2 sec / problem and 4 CPU on my i7 , Fritz 11


3) Stockfish: 82.5%  Average 0.45 sec/problem
2) Komodo: 84.3% Average 0.43 sec/problem
1) Houdini: 87.6% Average 0.42 sec/problem



Stockfish with twice as much thinkingtime = 4 sec/problem : 87.5% Average 0,57 sec / problem.











Wednesday, September 10, 2014

Chess Sever cannibalism

There are so many chess-online/trainings-server...

chess.com, chesscube, chessity, icc, fics, ideachess, chessgym, chess.emrald.net, chessgames.com, chessmoon.com, international chess school, chessclublive, chesslive, chessfriends.com, fide online arena, Remote chess academy, chess ebook, chessvideos.tv, red hot pawn, chesshere.com,
letsplaychess.com, gameknot.com, net-chess.com, caissa.com, chessworld.net, chess.net, chessanytime.com, thechesswebsite.com, improveyourchess.com, chessmagnetschool.com
... and hundreds more

They all offer more or less the same:

-Some educational texts
-Some educational Videos
-Online chess
-Puzzle Training
-Opening Explorer

And now i found the next one . chess24
Nothing new, nothing better..

Well chess24 offers few for free and try to charge a lot for few, thats a little uncommon.
The good aspect of the inflation of chess-server is the explosion of free offers. At youtube you find hundreds of free educational chessvideos, no need to pay for videos anymore, its impossible to watch all of them in one life. No need to pay for online chess, no need to pay for chesspuzzles... on a different server you find the same for free AND BETTER. Chess.com had the Tactic Trainer as premium feature, chesstempo (and many others!)  does offer that for free (AND BETTER), therfore the endgamet raining  is premium there ...
And as more server as less customers per server... silly "game".
They destroy each others chances to make any money..
Just chess server cannibalism

Monday, September 8, 2014

New Board Vision Exercises : Find all Checks


I was doing my "Mate in 1" Training  for several weeks now, the improvement is moderate :/ . These exercises are realy not easy. My skill-improvement-hypothesis is : it has to be ( almost silly ) easy to become a skill (  automatized thinking ).
So i did create an easier task and seemingly it helps :


The "Find all Checks" - Trainer

 Press the Start-Button and you will get Chesspositions with Black on top  where ou have to select all checks ( for white and! black )

 Qh2+ 
1 "Variation"
 Qb8+ Qc8+ Qd8+ Qd7+ ...
Rc7+ Rd5+
10 "Variations"

Nd6+ Nc5+....
Ooops: i missed Qxd5+
8 "Variations"

You may "cheat" by pressing "Help".

After some "Find all Checks" Training i made a nice improvement in  "Mate in 1" Performance.
Strongly recomended: make the Board Vision exercises of chessgym.net first

Play here.




Saturday, August 30, 2014

Break Through Plateaus

Most adult chess player don't improve in chess, they reach and stay at a rating plateau. So did i. But chess has so many "elements" : Calculation, Pattern, Board Vision, Visualisation, Positional Understanding, Tactical Vision, Opening, Endgames, Middle game, Attack, Defence, Concentration,....That is too much to learn at once. So i did put my main attention to Tactics. I solved hundred thousands of tactic puzzles and .. reached a rating plateau ( after a few thousand puzzles ). Tactic has so many "elements" : Mate, Material Gain, Pawn promotion, Pattern recognition, Calculation,... Seemingly still too much to learn. So i did put my attention to Board Vision. I did reach a "nice" level in the attack training at ChessGym but not a really high ( "GM" ) level. And here i made an experiment: i did force myself to use strictly a Thinking method: For 2 hours of training i was first looking for the queens ( is the white queen attacked, what pieces are attacked by the white queen ) then the rooks then the bishops then knights, pawns and finally the kings. After that i made a biiig improvement ( without strict thinking process ) in "attack" at chessgym and reached a "real high level".

I think there might be 3 possible reasons why this thinking process did help me improve further
  1.  A systematic thinking process improves the efficiency anyway
  2.  A thinking process dissect the "big" problem "Attack-Board-Vision" to smaller, easier Problems like "Queen-Attack-Board-Vision"... Such easier Problems should be easier to learn too.
  3. I'm thinking of an effect like the dual-n-back training. Thinking at something else simultaneously , did create "more skill"
  4. Maybe that's just a coincidence of some type
 Now i did start to implement a thinking process in my chess training as much as i can. It is really not that easy. My first step is, to constantly think about how i actually think during solving chess problems and try to improve that thinking step by step.

...

Well, i did break an other plateau at ~ 2060 in Tactic Trainer ( an Ipod App) .

 Is it my new way of thinking? Or is it my improved Board Vision? Did i get slower ( then the score improves anyway)? Luck? ...

I continue to write ( and use ) Board Vision Training Programs which support my thinking process. I think my "critical piece" training will help me a lot.
( A critical piece is as often attacked as defended. Such pieces can be successfully forked / double attacked.)