To give a relatively simple example, take this position from one of my games
White to move: Can I take the pawn?
(This position took me about 20min of calculation. I stopped only when the notebook kicked the power save on. I was at the third variation and I still see other lines.. My opponent thought I had hanged up and asked "U there?")
Some of the more important findings I gathered is:
1. At this stage, 2moves deep is the furthest I can see the clearest. Then it grows ever fuzzier. I start forgetting the placement of a pieces. The residual memories gets stronger. Goal: Advance this 1move deeper.
Before you name me weakling for seeing only two pathetic moves deep, note the operative word here is "Clear". I can move, say a knight 10moves ahead easily. I see the knight sure, but the positions of the other pieces are another matter.
If this seems just semantics, I think it's better understood by asking these questions. If I move the knight 10moves ahead, and see that it finally lands on a square that forks the king and the queen, will I do it? Yes, of course. But what if I forgot a pawn was there to capture the knight? Would I still do it? If I forgot, to my regret - yes of course. Big difference = Big Mistake.
As they say - There is seeing and there is seeing.
2. I move the pieces in my head too fast. When I was calculating, I was going through the variations lightning fast. Like in 5seconds I will go through something like Nxe4 Re8, f4 Ne8, Bg4 Qc2, Bxd8 Rxe8. But of course, somewhere along the way my opponent will make a totally different move that throws a monkey wrench on that pretty line. Going so fast, one is susceptible to miss something.
So now, remembering my name is not Fritz, I am trying to take it one move at a time. That is, Move a piece, Stop, Look around. Move a piece, Stop, Look around. This is really slow, almost like playing the actual game in my head. But I am willing to trade speed for greater accuracy.
Actually, I find this one of the beautiful parts of chess. Like a Jedi-master that sees the future and acts accordingly. And all those movies containing chess players, what do they have in common? They portray the chess master as this invincible person because he has predicted "all" the moves beforehand.
If I am truly a chess player, then I want to be able to say "I have predicted all your moves, prepare to die".
Otherwise, I'll just be another Tom, Dick, And Harry. And as any Russian school boy knows - that's to be avoided like the plague (Apologies to anyone named Tom, Dick, or Harry).
3. Stepping Stones - I've established that calculation is slow. But since the tree of variations can really grow to massive levels. If I were to start from the top for each line, I'd lose on time. So I started trying to use the stepping stone technique. That is, I fix an intermediate position in my head, and start from that instead of from move 1. This saves some time somewhat.
However at this point, 2moves deep is the clearest line so the effectiveness of this is hampered badly. I'm hoping that if I improve on point 1, this will improve also.
--
Well, these are the things I've been doing lately to better my chess. I've not been sleeping lately because of all those long games I've been playing, so I hope this losing streak ends soon. Otherwise my eyes are gonna pop out from their sockets, and I'll have a nervous breakdown from drinking too much coffee.
Time will tell if this path I've chosen is the correct one. Calculating my way to victory seems like making it harder than it has to be. But to someone who has completed the circles "its a piece of cake" :)
Besides I always did like doing difficult things.
Now excuse me while I go stock up on lattes.
I have gone the same road. Every tournament I came home with bad results I said to myself, if I only could calculate better then I would have crushed them. I have tried the ultimate to improve my calculation skills. And I dare to say I'm not too bad at it. But without the expected results. So my conclusion is that the human mind isn't shaped for error free deep tactical calculation in chess.
That's why I think it is reasonable to make an effort to get a better starting point. With more active pieces it is easier to start a tactic than with passive pieces.
That's why this position is so difficult to calculate. Your bishops do nothing, the f-rook does nothing, the knight is active but not from a good home. If you grab the pawn, you unleash his bishops, queen and rook. Is it worth it? Difficult to calculate.
But if you build a better starting point first, with more active pieces, things become simpler and the tactics become stronger.
Which could mean you had to avoid this position in the first place.
Of course, the computer says you can take the pawn. But then again, you are no computer.