The American Checker Federation

I think it's no secret that I've been working on a new checkers program for a while now. So far I solved "Perfect Play" databases for all of DB-7 (4x3, 5x2, and 6x1) some of DB-8 (4x4 and most of 5x3, should be done in a few days), DB-9 5 kings vs. 4 kings and 4 kings + 1 checker vs. 4 kin...

4 checkers vs. 4 checkers.

White to move and win in 119. There is no longer win with 4 checkers vs. 4 checkers

I see value in so-called perfect solutions when they establish draws in near hopeless situations, or for adjudicating games that would otherwise disrupt the timing of a tournament. But I see no value in prolonging a win after it becomes apparent to all reasonable parties that it is a win, or in imp...

Just friendly remind you, that good portion of advanced problems with solutions (and also many, many other nice things) you can find on the Al Lyman site - former ACF website ( http://www.acfcheckers.com ). http://www.acfcheckers.com/advanced.htm Regards, Alex Hi Alex, Since I completed the 8-piece...

I finished solving all of the 4x4 endings with Perfect Play. The longest win is 293-ply. I can now instantaneously show the quickest wins and strongest defense to any 4x4 position. I also probed the databases for things like the longest win in a bridge ending and the longest win with 4 checkers agai...

John Acker wrote:Ed, can you post the link for the latest version, please? Thanks.

If you search on "Gui checkers checkerboard" you should find the version Ed Gilbert worked on after I gave him my source code. He's on vacation until March and I forget where he uploaded it.

Well, the 4x3 set has been solved with perfect play since 2001, but the claim could not yet be made that all 7-piece endings were "strongly solved." Gil Dodgen and I decided to skip the 5x2 and 6x1 set, since, as you alluded to, it's not really that interesting. Since I'm doing db8 4x4 rig...

My computer has reported that today it finished solving all of the 6x1 slices with "perfect play," down to the last one, 6 checkers against 1 checker. With white to move with the 6 checkers, the position with the longest win requires 29-ply to complete. I seem to recall Gil Dodgen's birthd...

Richard Pask wrote:If I order a board and pieces from the ACF website, are they able to ship it to the UK (and give me the equivalent total sterling price)?

Oh no, did Donald Trump talk Theresa May into outlawing draughts?

Ed, One question: If your database plays for the shortest win and defends for the longest loss, what does it do with draws? Since draws can go for an infinite number of moves, when does it concede a draw? I guess that's two questions. One comment: A checker game is not a team sport. A team of stron...

So, my resolution after this match - I can do it better next time! One of the things I am working on is very large perfect play databases (10 pieces). These databases require 2 bytes for every position, and there are over 8 trillion positions. It will take about 20 terabytes to store everything. Bu...

A few comments I have: First, congratulations to Alex. Given the speed of checkers programs today, combined with their opening books and endgame databases, they are almost playing with a "draw in hand." The opening books navigate almost every pitfall at the start, and once out of book, the...

A special thanks to Ed Gilbert for updating CheckerBoard to allow for incremental time controls, rebuilding the GUI Checkers 1.05 book into the new version 1.12 book, and modifying the GUI Checkers timing/logic/search abort under time pressure code to make it a much more challenging opponent under f...

There is a new ending that has been solved: 4 kings + 1 checker vs. 4 kings.

Amazingly, the longest win is 291-ply, the same length as the longest known win so far for 8 pieces. Here is the position.



White to move wins in 291.