Ludii Forum
Tumbleweed - Printable Version

+- Ludii Forum (https://ludii.games/forums)
+-- Forum: Suggestions (https://ludii.games/forums/forumdisplay.php?fid=10)
+--- Forum: Submit Your Games (https://ludii.games/forums/forumdisplay.php?fid=23)
+--- Thread: Tumbleweed (/showthread.php?tid=197)

Pages: 1 2


Tumbleweed - MikeZapawa - 09-30-2020

I'm thrilled to post an implementation of my game Tumbleweed. As credited in the metadata, all the work was done by Michael Amundsen and Alek Erickson. Thank you guys once again!


RE: Tumbleweed - cambolbro - 09-30-2020

Hi,

Excellent! Thank you for the submission, and for the work you all put into it.

Regards,
Cameron


RE: Tumbleweed - Michael - 09-30-2020

I just realized that my name is misspelled in the credits. It says "Admundsen" in stead of "Amundsen". No biggie, of course.


RE: Tumbleweed - MikeZapawa - 09-30-2020

(09-30-2020, 07:31 PM)Michael Wrote: I just realized that my name is misspelled in the credits. It says "Admundsen" in stead of "Amundsen". No biggie, of course.

My apologies, I've just updated the document, you go update the version in Tumble forum. 

BTW it's probably worth noted that even though Tumbleweed has very simple rules, AI plays it poorly even on small boards.


RE: Tumbleweed - Eric Piette - 10-02-2020

Hi,

I just add it to our games and it will be included in the next release.

Regards,
Eric


RE: Tumbleweed - AlekErickson - 10-03-2020

I think the current version is great, but there is a minor bug. The rules state that the game ends after 2 successive passes OR if there are no more legal moves for one player.

Right now the .lud is not ending the game even after one player runs out of legal moves. Instead, you are forced to pass, which might appear confusing to people if they don't realize they need to pass.

It's not a big deal if you know that you should pass, but maybe somebody can put this alternate game ending trigger in .


RE: Tumbleweed - Michael - 10-03-2020

My first thought is that I don’t know of an easy way to fix that. The way it is implemented now, you never run out of moves because you may always pass.


RE: Tumbleweed - AlekErickson - 10-04-2020

OK, I want to ask the developers if this may affect the AI strength.


RE: Tumbleweed - DennisSoemers - 10-04-2020

(10-04-2020, 08:41 AM)AlekErickson Wrote: OK, I want to ask the developers if this may affect the AI strength.

In theory ANY change in the game's rules (or even just changes in how exactly the same rules may be expressed in multiple different ways, because some ludemes are more computationally efficient than others) can affect AI strength.

In practice I don't expect a tiny difference like just having one single extra pass move at the end of every game (with a branching factor of just 1) would make a meaningful difference though.


RE: Tumbleweed - AlekErickson - 10-04-2020

Normally , I would drop this, but I want to squeeze every drop of AI strength out of this implementation. 
So here goes: 

Actually, the difference is not necessarily just one pass. 

First of all, in this game, under the current ruleset, when one player runs out of moves, the other player might have hundreds of moves left, each with a large branching factor. So, even though no action taken by either player can change the game result, the game tree still has a significant complexity remaining. 

In the case when the AI is assessing the position at the end of a MCTS play-out, and its play-out has reached a position where it has run out of moves, will it always understand that such a "certainly lost game", is actually certainly lost? 


If not, then we would help the AI strength by reinforcing the point that the game is over when a player has run out of moves.