Forums

Why I hate Tactics Trainer

Sort:
0ort
woodshover wrote:
dpruess wrote:

i think the long endgame problems are tactical; the ones i have encountered have been some of the most delightful problems for me.

i think you are right, there may be an issue where the pass rate on the first move skews problem ratings somewhat. but i don't know how many problems suffer from this. i mean, what percent of problems you encounter are mis-rated?

and maybe we can have a rated mode where the time is ticking down but you don't have to see the bar. that might be mildly useful, but i don't know-- how many people want that?


 There IS a rated mode where the time is ticking down, but you don't have to see the bar!


Personally I highly recommend giving this mode a try. Your rating may take a hit at first but in the long run I think this is a much better way to do TT.

Natalia_Pogonina
Kacparov wrote:

the only thing TT needs is multiple solutions, this is why I don't use it. but your other points don't make much sense to me


I do use it, but agree with this point. TT makes one adopt a somewhat perverted way of thinking. Often you instantly see a win, but become afraid of overlooking a faster way to mate, or mate instead of winning a piece... And it's very hard to rate these adequately. E.g. if the player has lousy technique, he may fail to win a knight up (so his only chance is to mate on the spot). A more skilled person may convert a pawn. Thus, partial credit won't do the trick either. I don't want to receive, let's say, 50 percent of the credit for winning a queen instead of checkmating. Who cares? The opponent will resign anyway at this point...

rooperi

 The opponent will resign anyway at this point...


Not on this site, they won't :)

repecmps

The only problem I find with the tactics trainer is as follows:

I very often solve a problem, in the time allowed, but because i was a little slower than the average I still lose points. (the problem is marked as passed)

Sometimes I even lose less points by not solving a problem than I do by solving a problem slowly...

tjfjerstad

There also some problems that I disagree with what they say is the best move.

JuicyJ72

I also am not a huge fan of the timer, I'm no kid so unless I know the pattern it can take a little time.  Other than that it's pretty good, it can be frustrating to get a Mate in 3 versus Mate 2 but it's good practice on seeing all the escape routes and the moves to squeeze the king.  But in a perfect world a win is a win and getting 0 points on those is frustrating

Kupov3
repecmps wrote:

The only problem I find with the tactics trainer is as follows:

I very often solve a problem, in the time allowed, but because i was a little slower than the average I still lose points. (the problem is marked as passed)

Sometimes I even lose less points by not solving a problem than I do by solving a problem slowly...


No that's impossible unless you got a percentage of the problem correct earlier. 

brokennb

I would like some wisdom about some of the tactics where it is not obvious.  I read the posts, but chess.com could be proactive and tell us up front:  This allows a pin on the night on f6 after... allowing ...

brokennb

I consider Chess.com to be my chess training site.

In order to make Chess.com more effective, I would like Chess.com to have suggested training (mentor lessons or videos) based on my tactics results.  For example, I could be always missing a typical mating pattern with the rook and the knight.  In this case, Chess.com could recommend the mating patterns video that discusses exactly this.  

How to accomplish this:  Tag tactics trainings with some code to identify what it's testing.  Have a web page that shows where you need the most training, sorting the counts of misses for each tag.  Have a link to go to suggested videos or mentor lessons to practice, along with a link to similar tactics to allow for testing afterwards.

I love the site.  I hope that it evolves and becomes more effective.  If you can do these things, you could get more premium members.  Focused training on weaknesses allows people to utilize their time effectively.  Allow them to do that and you've won the game.

Patrick

bondocel
brokennb wrote:

I would like some wisdom about some of the tactics where it is not obvious.  I read the posts, but chess.com could be proactive and tell us up front:  This allows a pin on the night on f6 after... allowing ...


The analysis and source often contain answers to such questions. Sometimes the comments for a specific problem might be worth reading. 

bondocel
brokennb wrote:

I consider Chess.com to be my chess training site.

In order to make Chess.com more effective, I would like Chess.com to have suggested training (mentor lessons or videos) based on my tactics results.  For example, I could be always missing a typical mating pattern with the rook and the knight.  In this case, Chess.com could recommend the mating patterns video that discusses exactly this.  

How to accomplish this:  Tag tactics trainings with some code to identify what it's testing.  Have a web page that shows where you need the most training, sorting the counts of misses for each tag.  Have a link to go to suggested videos or mentor lessons to practice, along with a link to similar tactics to allow for testing afterwards.

 

Repetition is the solution to what you describe above. Do as many tactical quizzes as you can. Until the mating patterns or some other patterns become so familiar that you can spot them even when playing a one minute game.

repecmps
Kupov3 wrote:

No that's impossible unless you got a percentage of the problem correct earlier. 


 Right, so it's possible to lose less points for a failed problem than you do for a solved problem. We agree. Thanks for taking the time.

Gert-Jan
dpruess wrote:

i think the long endgame problems are tactical; the ones i have encountered have been some of the most delightful problems for me.

i think you are right, there may be an issue where the pass rate on the first move skews problem ratings somewhat. but i don't know how many problems suffer from this. i mean, what percent of problems you encounter are mis-rated?

and maybe we can have a rated mode where the time is ticking down but you don't have to see the bar. that might be mildly useful, but i don't know-- how many people want that?


 We already have a rated mode where the time is ticking down andyou don't see a bar. You can remove the v from  the 'show timer' settings at settings. Then you have rated problems and dont see the timer.

shiro_europa
Kotomitsuki wrote:

There are several Tactics Server and several Free Tactics Problem-Collections in the Net. Choose the one you like. They all have some pros and cons. Some have no timelimitations for a move, most dont allow any error at the line at all.The most are free of charge.

To complain about TT is like going to Paris and complaining that the people dont speak spanish.


 You hit it right on the head.

dpruess

appreciate the suggestions. in particular i like nimzo's 4th suggestion to be able to dl a pgn of the tactic you have just failed. i am happy to hear we already have the "no timer bar" mode that was requested by OP. i also would like to accomplish the tagging of problems, but it's a massive undertaking.

-X-
brokennb wrote:

I consider Chess.com to be my chess training site.

In order to make Chess.com more effective, I would like Chess.com to have suggested training (mentor lessons or videos) based on my tactics results.  For example, I could be always missing a typical mating pattern with the rook and the knight.  In this case, Chess.com could recommend the mating patterns video that discusses exactly this.  

How to accomplish this:  Tag tactics trainings with some code to identify what it's testing.  Have a web page that shows where you need the most training, sorting the counts of misses for each tag.  Have a link to go to suggested videos or mentor lessons to practice, along with a link to similar tactics to allow for testing afterwards.

I love the site.  I hope that it evolves and becomes more effective.  If you can do these things, you could get more premium members.  Focused training on weaknesses allows people to utilize their time effectively.  Allow them to do that and you've won the game.

Patrick


 A great suggestion. I know it would be a big job, but if chess.com ever decided to do this, I would be very pleased.

planeden

thanks for the responses dpress.  a pgn would be nice, then we can put it into an computer and they can tell us where we went wrong. 

loved

My head is spinning after reading so many comments. Comments at TT problems often help me to understand why I struggled with the problem. I wish that comments that include chess board would load faster there. Users can provide each other in such comments with guiding analysis that users want as indicated in the comments here. In other areas of chess.com the boards load just fine.

I didn't read any comments in this forum topic about the rating of Chess Mentor and how that rating does or does not relate to TT rating. But I do agree that Chess Mentor is more for positional play, so it seems completely reasonable to me that one's ratings at Chess Mentor could be a few hundred point higher or lower than at TT.

It's hard for me to imagine how someone finds a complaint that TT doesn't allow for mouse slips. To do that, TT would have to be a mind reader. Likewise with getting the idea on the first move. If you miss the first of 3 moves you may have the idea but your execution is poor and hey look that's what TT is for - to help you improve your technique/execution.

TT has helped me to look more for mates in games - and also to find more mates. I like to think that I overlook fewer things now too, but what metric can measure that? Again I need a mind reader, no?

I don't even want to know the algorithm for calculating the rating of a TT problem because I'm sure it must be quite complex, but . . . to add my two cents, perhaps there is a way to include in the algorithm that calculates my TT rating some of the data generated by my live games, so that if my live games are more heavily weighted toward 25 or 30 minute standard games, my TT score is assessed a bit differently than if my live games are 90% 2-minute blitz games with 10 seconds added per move, etc. This could even take into account how recently these games were played and how many games with such a time limit were played. I don't think there's any benefit to having a blitz mode for TT and a standard mode or any other type of mode time difference.

Also how about sounds for getting TT problem right or wrong? If I could, I'd add the sound of Pac Man when Pac Man dies for when I miss a TT problem. etc.

alexanderthemonk

I use TT all the time, on the iphone, it's pretty incredible.

Now allow me to complain...

The idea SHOULD be that you solve the whole problem before moving the first piece, otherwise you're just rushing blindly into something that feels right, because that kind of approach seems to optimize your rating on TT... That's garbage as far as training is concerned. I like the idea of requiring the moves to be done within 10 seconds of each other after the initial move.

I know, I know, we shouldn't be concerned about the rating, but why swim against the tide that is competitive instinct when we could be using it to our advantage.

Also, while I personally feel that mate in 17 is the same as mate in 2, I respect that tactics trainer is about sharpening your tactics, so I do not believe this aspect of it should be changed.

*edit: I forgot to add my thoughts about the ticking timer. First of all, there's gonna be a ticking timer OTB, so I don't really mind it being there. The only thing I don't like about it is that I use it to cheat. "Oh, 20 second problem? What did he hang..." Contrast that puzzle with another where you have 5 minutes but his knight is clearly hanging, and you'll realize that there's probably a checkmate to find. With both these types of problems throughout the ratings, removing your timer hardly seems viable.

BigHickory

Tactical Trainer needs lots of work.