Pins?!

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

swami
Posts: 6640
Joined: Thu Mar 09, 2006 4:21 am

Re: Pins?!

Post by swami »

bob wrote:If you pin a queen on a king with a rook or bishop, it is obvious and is of no real use in a test position. But some pins simply restrict the opponent's mobility, and they need to be avoided or broken as soon as possible. Those are interesting since they don't win material outright, but may well lead to a long-term winning position because of the lack of mobility for the pinned piece, or any defending pieces.
Ok, make that my next test to work on. Currently I'm working on "Re-Capturing" Test suite Ie, Which pieces to use to recapture. Once that is done, I may get on to studying games and collecting positions on Strategical variant of Pin next. It's interesting for me as well.
User avatar
Graham Banks
Posts: 41423
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Pins?!

Post by Graham Banks »

swami wrote:Currently I'm working on "Re-Capturing" Test suite Ie, Which pieces to use to recapture. Once that is done, I may get on to studying games and collecting positions on Strategical variant of Pin next. It's interesting for me as well.
You're doing a good job Swami (and Dann). 8-)
gbanksnz at gmail.com
swami
Posts: 6640
Joined: Thu Mar 09, 2006 4:21 am

Re: Pins?!

Post by swami »

Graham Banks wrote:
swami wrote:Currently I'm working on "Re-Capturing" Test suite Ie, Which pieces to use to recapture. Once that is done, I may get on to studying games and collecting positions on Strategical variant of Pin next. It's interesting for me as well.
You're doing a good job Swami (and Dann). 8-)
Thanks for the compliment, Graham. It should be noted though that without Dann's help this suite ain't possible at all. He spent time analysing, running long tests, categorizing the rights and wrongs and going over right ones again with longer analysis, and finally selecting the final 100. I study games and look for pattern, do some analysis and send him about 160-200 tests, he comes up with the perfect 100 after doing extensive research. He also does something with the SQL to edit out minor bugs with tags.
jesper_nielsen

Re: Pins?!

Post by jesper_nielsen »

bob wrote:
swami wrote:Hi Jesper, It's a really good idea. But most of the pins occur in tactical type positions. Our goal is to collect positions that are strategical/positional in nature, as best as we can.

I'd add Pins - that don't involve material gain so the idea remains strategical as opposed to tactical. :)

Strategical version of "pin" - forces opponent to take the passive role and lose the mobility of his pieces by defending the pinned piece. Doesn't gain material but limits the activity of opponents piece.

Tactical Version of "Pin" - threatens to gain material via absolute or relative pin.

Thanks for the idea. I've updated the ideas page with the added "strategical" pin here
Most books use pins as part of a tactical motif, but I have seen quite a few positions over the years where a pin is critical. In one case, a black king on g7, black pawns on f7,g6,h7, black knight on f6. White pinned the knight with a bishop, and now the game is won for white. The black king could not escape the pin, could not break the pin (due to other pieces/pawns on the board) and the final result was that it locked the black king on g7 for many moves as black had to waste time to set up things to break the pin, while white did what it wanted elsewhere.

Those kinds of positions would be useful. The result of the pin was a loss, but it was _way_ beyond any possible search horizon. What one needed was to realize that the pin effectively locked the black king on g7, while the black king really needed to centralize to prevent white's king from penetrating and winning pawns on the queen-side.

It was deep enough that the search would not show the win, so the eval had to realize that if the king was stuck around g7 it was in serious trouble. Some of those kinds of positions would be pretty useful, although one does not need to actually evaluate the pin to solve them, but you do need to know that for some unknown reason, the king can't reach a good square, until by luck you expose the plan to break/avoid the pin and move on...
It was actually a situation like this that made me think about pins.

A total patzer ... lets call him Repsej to protect the poor bugger :oops: ... once allowed a position like:

White: Pf3, Pg2, Ng3, Kf2
Black: Bh4

There was no way to move the king without losing the knight. And since the king was in desperate need to get out of the crumbling castle, the game was utterly lost. The pin simply killed all counterplay, because it locked up the defending pieces.

Kind regards,
Jesper