FIBS Board backgammon forum

Backgammon => Backgammon problems => Topic started by: PersianLord on April 27, 2008, 06:36:12 PM

Poll
Question: White to play 3-3
Option 1: 24/18(2) votes: 0
Option 2: 24/21(2)-13/10-8/5 votes: 0
Option 3: 24/15*-8/5 votes: 10
Option 4: 24/21-8/5-13/7 votes: 0
Option 5: 24/21(2)-13/7 votes: 0
Title: checker play problem
Post by: PersianLord on April 27, 2008, 06:36:12 PM
This position was taken from a match on FIBS.

(http://bp0.blogger.com/_nvjbHZxa_5Y/SBS26SrVEuI/AAAAAAAAABE/X2IoPxklz1g/s400/final.JPG)

Pip-count : White 144-137 Gray
Score      : White    0-0     Gray
Match      : 5-pointer

Spoiler
Since I don't have any software at the time being, I'll not be able to provide an analysis and a rollout. My own desired move is 24/21*-8/5, which I whispered during the match. White's actuall play was 24/21(2)
[close]
Title: Re: checker play problem
Post by: spielberg on April 27, 2008, 07:04:15 PM
I'll see what BGBlitz (yes Ive actually bought some software for once!) and GnuBg say in a bit but at first sight automatically went for the same move as you Ali -- it just felt right.

A small point to speed things up (which you'll not be able to do without GnuBG). Try posting the code for the board position as well as an image of it.



Steve


PS I've now done the analysis with GnuBG. No great surprise - you and I both correct Ali. The set up numbers are: Position ID: 4PPEBwBmreABMA Match ID:  cAmgAAAAAAAA.
I saw quite a sizeable change in MWC from anything but the "natural" move (at least -3% MWC I think). This reminded me of some fine guidance and advice dorb gave me once. He suspects many expensive move errors come from misplays of doubles and ALWAYS tries to concentrate more (yes even more) when he rolls one.

Title: Re: checker play problem
Post by: PersianLord on April 27, 2008, 07:14:45 PM
Quote from: spielberg on April 27, 2008, 07:04:15 PM
I'll see what BGBlitz (yes Ive actually bought some software for once!) and GnuBg say in a bit but at first sight automatically went for the same move as you Ali -- it just felt right.

A small point to speed things up (which you'll not be able to do without GnuBG). Try posting the code for the board position as well as an image of it.



Steve

Thnx Steve.

As said, I don't have gnubg at the time being (i'll have it this weekend). I used to post the position ID and match ID, but for this, I just loaded the watched match from 3dfibs logs and posted it.

regards
Title: Re: checker play problem
Post by: playBunny on April 29, 2008, 08:15:35 PM
Quote from: spielberg on April 27, 2008, 07:04:15 PM
This reminded me of some fine guidance and advice dorb gave me once. He suspects many expensive move errors come from misplays of doubles and ALWAYS tries to concentrate more (yes even more) when he rolls one.

Having gone through many matches of players at Vog and at DailyGammon, I can agree. From the lowest to the highest levels, doubles are very often misplayed. One aspect of the error is that there's a tendency to think in terms of moving a pair of checkers together. Although it's often the case that that's the best move, it's can become a habit to play that way. A better habit is to think of the four dice separately and let it end up as paired dice only by happenstance.
Title: Re: checker play problem
Post by: Hardy_whv on April 29, 2008, 11:41:19 PM
PL: Just clicked on the spoiler section of your comment. You write "White's actuall play was 24/21(2)". Looks like White forgot to play two 3's then  :laugh:

Btw, I fully support your move. The hit is a must here. So the last 3 seems to be cristal clear: Make the 5-point.

Hardy  B)
Title: Re: checker play problem
Post by: PersianLord on April 30, 2008, 06:34:12 AM
Quote from: PersianLord on April 27, 2008, 06:36:12 PM
This position was taken from a match on FIBS.

(http://bp0.blogger.com/_nvjbHZxa_5Y/SBS26SrVEuI/AAAAAAAAABE/X2IoPxklz1g/s400/final.JPG)

Pip-count : White 144-137 Gray
Score      : White    0-0     Gray
Match      : 5-pointer

Spoiler
Since I don't have any software at the time being, I'll not be able to provide an analysis and a rollout. My own desired move is 24/21*-8/5, which I whispered during the match. White's actuall play was 24/18(2)
[close]
Title: Re: checker play problem
Post by: PersianLord on April 30, 2008, 06:35:55 AM
Quote from: Hardy_whv on April 29, 2008, 11:41:19 PM
PL: Just clicked on the spoiler section of your comment. You write "White's actuall play was 24/21(2)". Looks like White forgot to play two 3's then  :laugh:

Btw, I fully support your move. The hit is a must here. So the last 3 seems to be cristal clear: Make the 5-point.

Hardy  B)

yep, you're right. i modified the post.
Title: Re: checker play problem
Post by: blitzxz on May 02, 2008, 12:18:19 AM
Quote from: playBunny on April 29, 2008, 08:15:35 PM
Having gone through many matches of players at Vog and at DailyGammon, I can agree. From the lowest to the highest levels, doubles are very often misplayed. One aspect of the error is that there's a tendency to think in terms of moving a pair of checkers together. Although it's often the case that that's the best move, it's can become a habit to play that way. A better habit is to think of the four dice separately and let it end up as paired dice only by happenstance.

Well said. It's easy to miss moves complitely when moving doubles simple because there is so many possibilities. I once analyzed 1-1 roll with gnu and there was over 160 different moves...