“strategically” oriented and purely “tactical" problems

[shajbat]

Since your last update I have noticed that the quality of “tactical” problems has improved by a large factor.  I have also noticed that the number of “strategically” oriented problems has also increased.  Prior to your last update I would encounter “strategically” oriented problems here and there.  But, now I encounter more of them.  Here, of course by “strategically” oriented I mean problems where one side gets an advantage in terms of a “won position” .  These problems have great instructional value.  However, is it possible for you to organize and separate these from the main body of purely “tactical” problems?  I am not sure whether this is even feasible.

As usual; thank you very much for this wonderful site and all you labor on our behalf!

Shajbat

[richard]

Hi Shajbat,

If you mean the recent update made earlier this month then any changes in that department are an unexpected surprise :-)  There was only one change made to the generator to fix a bug  that should have been unrelated to the balance of tactical versus strategic.  It is possible that there was some unforeseen side effects of the bugfix, but another possibility is that changes made over the last 2-4 months means that there is more of a tendency to produce more problems with a "strategic" nature and that as I add more problems to the set they are starting to overwhelm those produced by the older generators. It is not trivial at the moment to separate the problems into the two sets you mention, largely because the generator doesn't understand when it is producing one type or the other.  It might be possible to find some heuristics that characterized each problem , but to do this with 100% accuracy would probably be difficult.

Regards,
Richard.

[tama]

Shajbat i think its because maybe you have improved and are now getting more difficult problems. Because i don't really get what strategic vs tactical problems are. All problems are a mix of both. Maybe you were getting simple fork problems and thought that was a purely tactical problem and then got a problem were you had to calculate longer and thought that was a strategic problem. I don't know but the resent update i think added 2000 new problems out of the previous 15000 so i don't think it would change the entire system all that much.

[shajbat]

Tama hello,


Thanks for your feedback.  I have done around 5000 problems here.  I have been pretty steady; I shoot in the 1900 range (hi and low).  Hence, your point that I might be improving, therefore I am seeing different type of problems is not necessarily correct.

Notwithstanding the simple fork problems; by strategically oriented problems I mean problems where after a series of moves one side ends up with a “positional advantage” not necessarily material advantage.  As mentioned in my earlier post I am seeing these a bit more.

One last point; I see a lot of folks here who are concerned about rating improvement here on this site.  I do recall a saying by GM Romanishian who once said:  Chess players must practice tactics everyday as musicians do practicing scales.  Those of us here who also play a musical instrument perhaps understand where he is coming from.  Practicing scales everyday does not make one a great musician; the same goes with tactics.  However, without practicing scales there is no music; the same goes for tactics. 

Thanks

Shajbat

[drahacikfm]

shajbat, there are no problems here that are purely strategic and do not win any material.  In some rare cases the problem stops when the material is still equal.  But there is always some forced win of material coming in the next few moves which might not have been included in the solution.  There are a number of problems here that need more moves in the solution in order to show the tactical idea, but there is always some tactical idea, not just positional advantage with equal material.

The Gold membership helps a lot with this, because it shows the variation that the computer sees after the final position of the problem.  In those variations you can see the forced win of material that comes later.

[shajbat]

Drahacikfm,

I had not looked at it from the perspective you are presenting.  Yes I have seen many problems where the lines provided were not deep enough.

Next time I see one that I think falls under the above category I will have Rybka take a crack at it. 

Thanks very much

Shajbat

[tama]

I like what you said in your last paragraph about the scales. He means that tactics are just one aspect of the game.