passed but doubled pawns

Discussion of chess software programming and technical issues.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
jdart
Posts: 3824
Joined: Fri Mar 10, 2006 4:23 am
Location: http://www.arasanchess.org

passed but doubled pawns

Post by jdart » Mon Jun 15, 2015 12:42 am

This is an example of a passed but doubled pawn:

[D] 8/r5kp/p2RB1p1/3P4/1p1P4/nP4P1/P3K2P/8 b - - 0 36

(from Fischer-Larsen, Denver (m/1) 1971)

What score does your program give to the pawn on d4? Mine recognizes it as both passed and doubled gives a small minus score but maybe White should get more credit for such pawns.

--Jon

bob
Posts: 20549
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: passed but doubled pawns

Post by bob » Mon Jun 15, 2015 1:39 am

jdart wrote:This is an example of a passed but doubled pawn:

[D] 8/r5kp/p2RB1p1/3P4/1p1P4/nP4P1/P3K2P/8 b - - 0 36

(from Fischer-Larsen, Denver (m/1) 1971)

What score does your program give to the pawn on d4? Mine recognizes it as both passed and doubled gives a small minus score but maybe White should get more credit for such pawns.

--Jon
I only count the most advanced one as passed. But even that has a problem since with the other pawn, a piece can't support the lead passer from the back so it will miss out on all sorts of bonuses.

jdart
Posts: 3824
Joined: Fri Mar 10, 2006 4:23 am
Location: http://www.arasanchess.org

Re: passed but doubled pawns

Post by jdart » Mon Jun 15, 2015 2:08 am

I think sometimes the "behind" pawn is weak, but still, there are cases where it can be beneficial because White has two chances to promote on that file, i.e.
can give up a pawn and still have a passer.

--Jon

mar
Posts: 2001
Joined: Fri Nov 26, 2010 1:00 pm
Location: Czech Republic
Full name: Martin Sedlak

Re: passed but doubled pawns

Post by mar » Mon Jun 15, 2015 8:45 am

bob wrote:I only count the most advanced one as passed.
I do the same, actually giving the same bonus to doubled passer(s) doesn't seem logical to me (doubled penalty won't compensate for that), so maybe having a special "doubled passer" bonus? - if it's worth it which I doubt...
Back when I had rook behind passer bonus, I also made sure I only gave bonus to one rook if they were doubled.

Ferdy
Posts: 4110
Joined: Sun Aug 10, 2008 1:15 pm
Location: Philippines

Re: passed but doubled pawns

Post by Ferdy » Mon Jun 15, 2015 9:53 am

jdart wrote:This is an example of a passed but doubled pawn:

[D] 8/r5kp/p2RB1p1/3P4/1p1P4/nP4P1/P3K2P/8 b - - 0 36

(from Fischer-Larsen, Denver (m/1) 1971)

What score does your program give to the pawn on d4? Mine recognizes it as both passed and doubled gives a small minus score but maybe White should get more credit for such pawns.

--Jon
In Deuterium it is a pawn with doubled-pawn penalty plus isolated-pawn penalty. But giving a bonus to it if it reaches the 6th rank is an interesting try for me.

Joost Buijs
Posts: 986
Joined: Thu Jul 16, 2009 8:47 am
Location: Almere, The Netherlands

Re: passed but doubled pawns

Post by Joost Buijs » Mon Jun 15, 2015 3:44 pm

mar wrote:
bob wrote:I only count the most advanced one as passed.
I do the same, actually giving the same bonus to doubled passer(s) doesn't seem logical to me (doubled penalty won't compensate for that), so maybe having a special "doubled passer" bonus? - if it's worth it which I doubt...
Back when I had rook behind passer bonus, I also made sure I only gave bonus to one rook if they were doubled.
I only score the front pawn as passed, but it can be a mistake, in the past I saw several games with doubled passers where the second passer secured the win.

Don't you think you have to give an extra bonus when there is a doubled rook behind the passer instead of a single one?
I have the feeling that you should although I never tried and tested it in practice.

bob
Posts: 20549
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: passed but doubled pawns

Post by bob » Mon Jun 15, 2015 5:01 pm

mar wrote:
bob wrote:I only count the most advanced one as passed.
I do the same, actually giving the same bonus to doubled passer(s) doesn't seem logical to me (doubled penalty won't compensate for that), so maybe having a special "doubled passer" bonus? - if it's worth it which I doubt...
Back when I had rook behind passer bonus, I also made sure I only gave bonus to one rook if they were doubled.
I still do that. IE rook has to directly attack the pawn. So only the closest rook gets that bonus. Although I do add this in the passed pawn scoring now, trying to organize things a bit better and not having passed pawn code scattered from can to can't.

bob
Posts: 20549
Joined: Mon Feb 27, 2006 6:30 pm
Location: Birmingham, AL

Re: passed but doubled pawns

Post by bob » Mon Jun 15, 2015 5:04 pm

Joost Buijs wrote:
mar wrote:
bob wrote:I only count the most advanced one as passed.
I do the same, actually giving the same bonus to doubled passer(s) doesn't seem logical to me (doubled penalty won't compensate for that), so maybe having a special "doubled passer" bonus? - if it's worth it which I doubt...
Back when I had rook behind passer bonus, I also made sure I only gave bonus to one rook if they were doubled.
I only score the front pawn as passed, but it can be a mistake, in the past I saw several games with doubled passers where the second passer secured the win.

Don't you think you have to give an extra bonus when there is a doubled rook behind the passer instead of a single one?
I have the feeling that you should although I never tried and tested it in practice.
I think the second rook can be used better elsewhere. IE one term I used to have was that a passed pawn became more valuable if one of the files beside it was open AND I had a rook on that file (there is a name for this, maybe in Kmoch's book, but I don't remember the term). The usual benefit for a single rook behind the passer is that if the opponent blockades the pawn, the blockader can't move or the pawn will advance (since the blockader attacks the next square, but so does the rook after the pawn moves). Putting a second rook behind the pawn doesn't add anything in most cases, and the rook is doing little since it is not on an open file where it can swing to any part of the board.

PK
Posts: 824
Joined: Mon Jan 15, 2007 10:23 am
Location: Warsza
Contact:

Re: passed but doubled pawns

Post by PK » Mon Jun 15, 2015 8:40 pm

It probably depends on the values one use, but for Rodent it worked best to score the second of a doubled passers like a candidate passer (which means midgame bonus 0, endgame bonus: passer_bonus / 3.

Joost Buijs
Posts: 986
Joined: Thu Jul 16, 2009 8:47 am
Location: Almere, The Netherlands

Re: passed but doubled pawns

Post by Joost Buijs » Mon Jun 15, 2015 9:31 pm

bob wrote:
Joost Buijs wrote:
mar wrote:
bob wrote:I only count the most advanced one as passed.
I do the same, actually giving the same bonus to doubled passer(s) doesn't seem logical to me (doubled penalty won't compensate for that), so maybe having a special "doubled passer" bonus? - if it's worth it which I doubt...
Back when I had rook behind passer bonus, I also made sure I only gave bonus to one rook if they were doubled.
I only score the front pawn as passed, but it can be a mistake, in the past I saw several games with doubled passers where the second passer secured the win.

Don't you think you have to give an extra bonus when there is a doubled rook behind the passer instead of a single one?
I have the feeling that you should although I never tried and tested it in practice.
I think the second rook can be used better elsewhere. IE one term I used to have was that a passed pawn became more valuable if one of the files beside it was open AND I had a rook on that file (there is a name for this, maybe in Kmoch's book, but I don't remember the term). The usual benefit for a single rook behind the passer is that if the opponent blockades the pawn, the blockader can't move or the pawn will advance (since the blockader attacks the next square, but so does the rook after the pawn moves). Putting a second rook behind the pawn doesn't add anything in most cases, and the rook is doing little since it is not on an open file where it can swing to any part of the board.
When the pawn is blocked it is not very useful to have 2 rooks pushing from behind.
In case the pawn is not blocked it all depends upon how many times the squares in front of the pawn are attacked by the opponent.
I can think of several situations where 2 rooks behind the pawn can be useful.
On the other hand when such a situation arises the search will also find it although it will probably be somewhat later.

Post Reply