I think it would be great to display puzzle/problem type because most of us learn best from clearly defined problems. It can sometimes be frustrating and time consuming just to determine whether the puzzle is a mate problem or a material problem or a positional problem ( this is not the purpose of the puzzles I assume). So clearly defining the problem first will help us in learning and easily identifying positions that lead to a mate or positions that lead to material advantage.
an example is this position http://en.lichess.org/training/7774, After training on 5 consecutive mate problems, I get a non mate problem which I spent at least 2 minutes just to figure out that a mate is not possible and thus had to go for material advantage.
I understand the desire, but I think actually it's better training if you don't know.
Knowing exactly what the outcome is allows you to change the way you think about the position in ways that are not possible in real games.
For example, if you know you're solving a mate problem, and you look at a line that wins 2 pawns, you can reject it and keep looking.
That's not going to help in a real game, and is why the approach of just solving lots of tactical puzzles will only take one so far.
I know plenty of players who are decently competent (1900-2000 level) at solving tactical puzzles, but are only 1300-1500 in actual play, and that's why.
The more you know about the solution, the more your thinking changes in ways that don't reflect real game play. The very fact that you know it's a tactical puzzle means you get to reject moves that only lead to a positional edge, and if you know it's mate, then you get to focus your search even more.
Obviously this can still help you improve, and getting familiar with new tactical motifs is desirable, but it's definitely not a clearly superior way of doing training.
As an aside, a few chess engines have versions designed specifically for solving tactical position suites. Those versions invariably do significantly better on tactical suites, and significantly worse in normal play, for similar reasons as humans.
Honestly, I'd even prefer it if there were the occasional puzzle where there was no tactical solution, and you just had to make some non-losing positional move, but that's just me :)
Just my two cents.
Knowing exactly what the outcome is allows you to change the way you think about the position in ways that are not possible in real games.
For example, if you know you're solving a mate problem, and you look at a line that wins 2 pawns, you can reject it and keep looking.
That's not going to help in a real game, and is why the approach of just solving lots of tactical puzzles will only take one so far.
I know plenty of players who are decently competent (1900-2000 level) at solving tactical puzzles, but are only 1300-1500 in actual play, and that's why.
The more you know about the solution, the more your thinking changes in ways that don't reflect real game play. The very fact that you know it's a tactical puzzle means you get to reject moves that only lead to a positional edge, and if you know it's mate, then you get to focus your search even more.
Obviously this can still help you improve, and getting familiar with new tactical motifs is desirable, but it's definitely not a clearly superior way of doing training.
As an aside, a few chess engines have versions designed specifically for solving tactical position suites. Those versions invariably do significantly better on tactical suites, and significantly worse in normal play, for similar reasons as humans.
Honestly, I'd even prefer it if there were the occasional puzzle where there was no tactical solution, and you just had to make some non-losing positional move, but that's just me :)
Just my two cents.
This topic has been archived and can no longer be replied to.