Well, Only two of those matches involved CRef bots (1619017 and 1619014) and both those matches were lost by the CRef engine on time. What could have happened is that the engines thought they were still in the match, i.e., they did not finish their matches in time for the next round. The script that runs the CRef bots in parallel waits until all six engines finish their games, than restarts them collectively for the next match. There is also a timeout that kills all engines if the time is obviously too long, explaining why they were able to recover after missing a round (the script is basically the one from the 'lite' standard engine pack). The other two matches did not have any of my engines involved. So, if those are related, and my engines recovered on their own, that points to a server-side issue.
René PS. Glad to see a strong bot giving Magnus some competition (and perhaps finally attempt to scale the 3,000 Elo mark). On Mon, Jul 4, 2011 at 12:37 PM, Rémi Coulom <[email protected]> wrote: > Hi, > > I have just observed a strange phenomenon on CGOS: > - games 1619017, 1619014, 1619012, and 1619011 were lost on time, almost > simultaneously > - all the CRef bots disappeared from the server at the next round. > > Rémi > > On 2 juil. 2011, at 04:29, Brian Sheppard wrote: > > > Is the 9x9 server down? Or is it that no one is connected? > > > > _______________________________________________ > > Computer-go mailing list > > [email protected] > > http://dvandva.org/cgi-bin/mailman/listinfo/computer-go > > _______________________________________________ > Computer-go mailing list > [email protected] > http://dvandva.org/cgi-bin/mailman/listinfo/computer-go >
_______________________________________________ Computer-go mailing list [email protected] http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
