the next Computer Olympiad has been dated.
It will take place in Leiden (NL), from June 27, 2016 to July 03, 2016.

I want to propose a new Go variant for 9x9 board: 
"Frisbee Go simulation"
Normal go rules apply. However, when a player wants to place a stone on
"cell" (i,j), the stone will land there only with probability (1- 4*eps).
With probability eps each it will land on (i-1,j) or (i+1,j) or (i-1,j-1)
or (i-1,j+1). If one or two of these cells are outside the board the
move will count as a pass. If the landing cell is occupied by another
stone the move is also counted as a pass. Illegal moves are also counted
as pass moves. 

eps is the "uncertainty" level of the game.
Only values eps between 0 and 0.25 make sense. 
0 gives the standard go game. 

The relation tio "normal" Frisbee Go should be clear: The player wants
to throw the disk on cell (i,j) but with certain probability the
disk lands on one of the neighboring cells. 

Background of the proposal:
In the long run I want to see robots playing "robot frisbee go".
As a first step, the simulation shall help to develop good
game-theoretic programs for this discipline.

Of course, Frisbee Go Simulation will be played in the 2016 Olympiad
only if at least two programs are registered. So, may the programmers please
let me know if they are interested? Also all sorts of questions are welcome.

Cheers, Ingo.

PS. You may have a look at a picture, painted by Tanja Esser.
Computer-go mailing list

Reply via email to