[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [computer-go] Computer Go tournament at EGF




On Wed, 9 Feb 2005, David G Doshay wrote:

On 9, Feb 2005, at 12:41 PM, Nick Wedd wrote:

My provisional plans are:
  5-round Swiss
  Time limits one hour each, sudden death
  Probably Chinese (area) rules

are you planning a 19*19 tournament only, or is there a chance of a 9*9 'side' tournament?

My program SlugGo is strong but slow. I have dropped the time for the longest games from over 5 hours down to 3 hours with no drop in playing strength, with the average game now taking about 2 hours. I do see a rare game (perhaps one in 30) that finishes in under an hour. While I can trim time directly at the expense of playing strength, I think that people would like to see SlugGo playing at full strength.

I hope to eventually be able to compete in a tournament because I think that the computer Go world does not benefit if strong programs are excluded due to a forced forfeit on time. Just trying to plant a seed for a future tournament.

i disagree with this a little bit. it is real easy for me to write a program that plays optimal Go, I'll just be forced to forfeit on time if there are _any_ time limits (or memory limits :-( ).
which is not a good argument for taking time limits away. same with slugGo - the whole challenge of computer Go is making it play reasonably good in reasonable time. It it can't play in reasonable time than it might not be all that much different from generating the whole game tree...

regards,

michal bazynski

_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/