[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

[XG] Questions re "Outplays", "error compared to XG" and the meaning of "different"

Posted By: eXtreme Gammon
Date: Friday, 10 September 2010, at 3:12 p.m.

In Response To: [XG] Questions re "Outplays", "error compared to XG" and the meaning of "different" (Maik Stiebler)

Maik Stiebler: Let’s say I analyse a match with 3-ply in the first pass, and there are play decisions where the actual move would not make it through the filter, but turns out to be best when analysed at 3-ply. Will that move be treated as “difference”, so that further analysis/rollouts will be applied? If so, will it be eligible as an “outplay”?

Yes it will be considered as different and it potentialy an outplay. "Error compared to XG" compares to 3-ply Normal search interval. Same thing apply to cube action (3-ply does not analyze cube decision where the 1-ply is more than 0.200 equity from a cube action).

Maik Stiebler: When I chose to analyse with 1-ply on first pass and rollouts for all differences, errors etc., to my surprise I found the “error compared to XG” check to be meaningless, probably because no outplays were found because XG did not have a 3-ply eval for the candidate moves. Did I get this right? Is this the intended behavior? Shouldn’t the best available eval (i.e. 1-ply) be the basis for “error compared to XG” then?

That's correct "Error compare to XG" requires that you do at least a 3-ply for the first pass. Using the 1-ply to compare to XG would defeat the purpose of this option: i.e. to compare your play to a fixed level regardless of the depth of analysis.

Messages In This Thread

 

Post Response
Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:
 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.