Re: [gnugo-devel] slowest moves statistic from regress.pike

[bump]

> When measuring performance improvements/losses, we effectivel only measure
> average time spent per move. However, both for the "gnugo user
> experience", and for reasonable time management in games with time
> limits, the worst case behaviour is very important.

On the subject of slow moves, I'd like to revive discussion of
move 214 in the game ShaiHulud-gnugo-3.5.3-200401231947.sgf
which I just added to the CVS.

There is probably a bug here.

http://mail.gnu.org/archive/html/gnugo-devel/2004-01/msg00179.html

Dan