                         | |||
| 61P-43Z-33 | |||
| Position ID: 4HPhASLg2+ABMA Match ID: cIkNAAAAAAAE | |||
Alert: very bad move ( -0.730)
| # | Ply | Move | Equity | |
|---|---|---|---|---|
| 1 | 3 | 8/5(2) 7/4*(2) | +0.682853 | |
| 0.635069 0.271643 0.016624 - 0.364931 0.086373 0.003412 | ||||
| 3-ply cubeful prune [grandmaster] | ||||
| • | 2 | 3 | 24/21 7/4*/1* 7/4 | -0.046975 ( -0.729828) |
| 0.484576 0.149840 0.006555 - 0.515424 0.149947 0.008387 | ||||
| 3-ply cubeful prune [grandmaster] | ||||
This is Diagram #22 in the Nactation Tutorial. Nack uses it to discuss the top play, which is making both the 4pt and 5pt. What I noticed, however, is that GnuBg uses an odd notation for the franchise-buying move listed second that is quite inelegant. GnuBg gives us:
24/21, 7/4*/1*, 7/4
Surely it should be listed as:
24/21, 7/4*, 7/1*
XG gets this right. In fact, the play is so bad that I had to add it to XG’s list of candidate moves. XG had ignored it in its initial evaluation. At first, I tried adding it using the same notation as GnuBg: 24/21 7/4*/1* 7/4. Now, I claim this is inelegant. Imagine my surprise, when XG took it to the next level. It chastised me with an error, saying, “Moves added to the move list must be valid moves.” When I used 24/21 7/4* 7/1*, XG suddenly found that the move was valid.
Programmers take note. This may be a position you would like to add to your test suite.
Mike