|
BGonline.org Forums
Wild rollout results (bug?)
Posted By: Chase In Response To: Wild rollout results (bug?) (Nack Ballard)
Date: Thursday, 22 April 2010, at 2:02 a.m.
Just to eliminate some obvious causes of my anomalous result, I doubled checked the XG position ID against Neil's. His position ID is identical to that of my second rollout, and identical to that of my first rollout except for the digit that represents the Jacoby setting.
I can also confirm that I ran the predefined rollout setting called "3 ply with Variance Reduction" for both of my rollouts. Since this is a canned setting, the rollout parameters cannot be changed as far as I'm aware. (The size of the search filter can be changed, but I did not do so. If I had without realizing it, the rollout output would report the change -- indicating "Huge," for example.)
As far as seeds being reproducible, both of my rollouts report that the initial seed was set to 2, which was as intended. (I believe the default seed for the program is 1, and I thought it not good for everyone to be doing rollouts using the same sequence of rolls.) That my rollouts are consistent with each other is evidence that the initial seed was set properly (or at least the same) for each rollout.
I don't have any ideas beyond this. I'll be happy to do another rollout with a different seed or other changes if it will help sort things out. I'll wait to hear from others, especially Neil and Xavier and MCG before doing so.
|
BGonline.org Forums is maintained by Stick with WebBBS 5.12.