Then I got to 1050, and almost one of every 10 puzzles was smothered mate. A couple of years ago, when I myself was rated sub 1000, I had never seen a smothered mate puzzle. I have sometimes seen two identical puzzles (or just reversed sides), and their rating has been very few points apart. This turns out to be a quite accurate prediction. Each puzzle has it's own rating, and if you beat it, it drops, and vice versa.īecause of this, one could expect that puzzles that are similar to each other and have the same level of difficulty, will adjust it's rating to about the same as each other. The way TT is supposed to work (if they can get their formula right, it's almost there now), is that it's a self balancing system. Memorize all the tactics at your level and then move up. So tactics trainer has basically become a memorization tool. There were hundreds of thousands of puzzles so it seems odd to keep getting the same ones. On the very old site a few years ago you could see the distribution of ratings of all the puzzles. Some you can barely make the moves at that speed even without thinking. If you look at their recent problems you will see that they are solving everyone correct in about 3 seconds. They are gaming the system such that they top list now has 4000-5000 rated players. High rated players have so few puzzles to play they have clearly memorized them all. Instead you always "play" puzzles that have almost the same exact rating as you do and obviously that is a limited pool. you would get more points for harder puzzles and less points for easier puzzles. The way the rating system works you could be given a much wider range. Why are the puzzles given to you in such a small range around your rating? It creates a lot of repeated problems (even at the 1500-1600 level). I'm also aware of a few other websites, like chesstempo (where I believe multiple moves are allowed).I have so many questions about tactics trainer and there is so little information provided on the site. I actually prefer the problems on lichess, since they look (at least at the starting position) a little more realistic. (and possibly 1.) do not occur, but I'm not sure since I train rarely there. I found this old blog post on how tactics are generated on lichess, but I believe things have changed a lot since then and I think it's no longer relevant (I was not able to find more recent posts, and have not attempted to look at their code). make the puzzles less representative of real play. results in defensive play being neglected and 2., 3. It is easy to imagine why these features make the problem of identifying tactics from analysed games more tractable, but from the perspective of learning the game they are not ideal: 1. Then the computer will lose its queen and the tactic will be over much sooner than if it had tried to save the position.
#CHESS TACTICS TRAINER FULL#
For instance, say the tactic involves a mate threat, and that the opponent can choose between going down a full queen immediately (with no immediate mate), or trying escape with the king, with the latter alternative leading to forced mate in 7 (the line we should actually be calculating).
0 kommentar(er)
