I've reached the level where the tactics are tough. I can spend a lot of time figuring them out and usually get them BUT is it better to do them that way, no matter how much time I spend, just by brute force, or is it better after 30 seconds to just go ahead and look at the answer and try to 'learn' the concept and hope to recognize it the next time, thus saving time and hope this is a more efficient way to learn them?
Thanks for any input.
I don't spend any more than 5 minutes on a tactic. If I don't get it by then I will view solution.
I think it really depends on what you are trying to develop by doing the tactic problems.
If you are trying to improve your calculation skills then keep on brute forcing your way through to the solution.
If you are looking to improve pattern recognition, then give yourself 30 to 60 seconds per puzzle. If you haven't solved it, click for the answer. Once you see the answer, go through the puzzle again 3 times to try and get it in to your memory bank.
Thanks all. Makes sense. Think I'll do them both ways then to improve both areas. Doing it 3 times is a great idea; I'm used to just doing them once.
On chesstempo you can do both. Spend as much time as it takes on tactics in standard mode plus solve a tactical puzzle as fast as you can in blitz mode. For you have two ratings you'll know better how good you are at each. Greetings
I just want to approve what
@jameslatham says, I do it the same way. Unfortunately I have no patience, so my focus is on pattern recognition but I would say the further you are away from the national master level the more it is important to strengthen calculation skills before moving to patterns.
And
@teerdurchzogen is also right, chesstempo is the best site for training tactics, lichess is the best site to play chess!
thanks again; I should calculate then.
Every now and then a pattern will jump up I know, but not often...like I love seeing the Greek B Sac
Set up the position on a chessboard (real one), take your time and once you're sure write down the solution, then check/compare with the computer.
Good idea. I usually just look at computer screen.
> I've reached the level where the tactics are tough.
> Tactics rating: 1747
Lol, really? Have you ever seen 2400+ tactical problems? That is a hell full of traps.
I don't think tactics is just about calculation or patterns. You should analyze it just like you do with a usual position: finding weak points at your and opponent camp and try to use their weak points, not allowing them to use yours while they defend (unless your pattern recognition finds the solution immediately, which should not be the case when your tactics rating stabilized).
The mere brute-force is useless, you never use it in real game. What you always should use is some version of alpha-beta pruning, when you start analyzing lines with the most natural moves and if the most natural move already refute the line, you can safely drop the previous move without analyzing other responses of the opponent. In tactics the most natural moves are (in the order of decreasing of naturality): checks, captures, mating threats, queen-trapping/capturing threats, other threats.
The problem with lichess tactics is that it is automatically generated, so there are a lot of problems with some weird computer lines where you can't be sure that you find a solution. But I think you should try to find the best move anyway. Even if it seems that some move is just good and not forces opponent to lose material, you can often make it and then after the response see you can really win material. So just spend reasonable amount of time (5-10 min) and choose the move you think the best. If it is correct, spend some amount of time again etc.
