Tactics Trainer Opinions

I'd say at least a fifth of my fails are on winning moves. Of course there is the point that high-rated users come across more of such problems because they will tend to gravitate to the top percentiles.

Though this of course can't happen on the other side.

I don't see that as a case of having two solutions -- clearly a forced checkmate is a superior outcome to a forced material advantage no matter how decisive.

Yep, but if you end a problem a rook up and you lose points, then something seems incongruous, no?

It's very hard for a human to distinguish between say a mate in 6 and taking a free rook. Not to say it's completely pointless, in many cases the mate in 6 requires calculating some spite checks (which you usually don't do when playing).

No, I don't think so.  If it happens often enough the problems rating will end up finding an equilibrium at a point where those players that see the superior line can solve it and those that do not, cannot.

This is a totally bogus point. When there is a timer counting down and you see a win, you play it. It just happens that some people will see one win first, some the other.

It's not about 'ability' to see the correct line.

Consider a problem in which there is a forced mate in three and an alternate line in which there is a forced mate in four.  I've come across many such problems and don't consider seeing the mate-in-four line instead of the mate-in-three line a succesful attempt.  Neither does Tactics Trainer, and appropriately so.

I guess the key question is are you looking for the absolute best line or winning  line? I try and simulate tournament practice where winning is good enough--mate in 3 or 4 doesn't matter--easy material gain versus a mate that is a little longer that is fine. In a game you need to be more practical so TT has this flaw. Thought it is good practice and does sharpen tactical abilities I take the rating with a grain of salt.

Well, the rules for Tactics Trainer are to remove these ambiguities, so your view is a) in the minority and b) contrary to the site policy.

I can't believe anyone really believes a forced mate can be a fail. This is folly. If you see a forced mate in a game and you spend your time looking for one that's a move shorter, you're a fool, and that type of thinking will inevitably lead to time trouble in your games.

Are those problems actually removed?  I would consider that a problem with a single correct solution.

The Analysis board doesn't have any moves associated with it when it first opens up. You have to make all the moves yourself and then you can use the navigation at the bottom to move through them. You even have to make the 1st move when the Tactic doesn't have you moving first.

Of course. Or modified to admit unique solutions.

The point is that, once you find a winning sequence, it's useless to start looking for some other solution. The point is to win not to win with the fewest possible moves.

I agree with most of what you have said here, but not this. In many cases, winning in the fewest possible moves should be the objective.

Yeah, but honestly I get very upset when I solve a problem, find a winning continuation (say mate in 5) and TT says "nope, you got it wrong, you should have seen a mate in 4". By the way, a GM (So) just missed a mate in 2 at Corus with 20 mins on his clock!!

In a game, looking for a faster win, assuming the longer win you find is forced, isn't the point. The point is to win.

In a training program, especially this particular one, the point is to find the shortest winning line. I find it a bit disconcerting at times if the solution given isn't the one I found (though I don't find many that fit that criteria ). However, I would rather try and train myself to find the shortest solution to a mate puzzle in training and chalk it up to experience and learning if I find a slightly longer solution.

As long as the problems with multiple solutions with the same number of moves get fixed (found one of those and reported it) and ambiguous solutions are fixed or removed then I see no problem with that.

Well, in such cases you should be told something like "find a forced mate in 6". Otherwise you are pretty much left wondering in the dark, there are problems where you have say a mate in 6, a mate in 7 and two mates in 8. And they all might be based on the same idea, the only difference being the number of spite checks the opponent can throw in.

Tks. 

Forced mate should be credited in many cases. But, there are exceptions.

1.Ke7 wins, but misses the point.

While I wouldn't say that couldn't be an option, I personally don't like hints like that. There is an implicit "find the strongest move" in most of the Tactics and the fastest mate is the strongest option.

I guess it comes down to what you really want to get out of the training. If you want to just solve puzzles then it really doesn't matter if you find the forced mate in 3 versus the forced mate in 6. Or if you find mate or a theoretically won game. If you truly want to understand the position and hone your ability to calculate then trying to find the strongest continuation on your own works best, in my opinion.

We are already working with a different mindset during Tactics Training than we are in a game. You know there is something there and many times can find the right start just from that. Adding another layer of artificiality to that wouldn't be beneficial to me and, I assume, many others.

I completely get your frustration and understand where you are coming from. I know I wouldn't be against a modification of the trainer to allow multiple winning lines, maybe giving a bonus rating points for finding the most direct winning line. What I don't know is how hard that would be to do based on the way the system is currently coded. It could be done but would the time effort be worth the benefit?

Interesting Zyriab! So, would you like TT to contain such problems? Honestly I don't have a very clear opinion, because everything this problem tests can be done in a problem with an unique solution. There are many problems where underpromotion is necessary since promoting to a queen results in stalemate.