On Mon 26 Feb 2007 (12:16 +0100), Massimiliano Maini wrote: > > The latest build is the 26th, I made a few small changes, one of which > > was something to do with stopping rollouts. So this might improve > > things (with new rollouts). > > Just uploaded on gnubg.org new install and exes rebuilt from this > morning's code. If you have a recent (i.e. 2007) intall, just download > the updated exes (with and/or without multi-thread, but the problem was > probably visible only in the multi-thread exes).
No, it affected anyone doing rollouts cubeful or cubeless if their standard rollout options had a different choice for cubeless or cubeful. The ranking was done using cubeful or cubeless based on the user's normal rollout settings, the rollout was done cubeful or cubeless based on the settings used with the rollout. It affects single processor, non-threaded as well (just verified on a single CPU non threaded machine - extending Chase's rollout when Settings->Rollouts->cubeful was checked gave incorrect rankings, extending the same rollout when it was unchecked gave correct rankings. -- Jim Segrave [EMAIL PROTECTED] _______________________________________________ Bug-gnubg mailing list [email protected] http://lists.gnu.org/mailman/listinfo/bug-gnubg
