# Further thoughts on Peltier-Purvis

2 views

### Josh Purinton

Dec 16, 2008, 5:39:42 AM12/16/08
Inspired by Roman's note, Aunt Beast finds a mistake on move 38 of this game of the 2007-2008 WGOSA tournament:

21- (Purvis*-Peltier+SmallBeast) 1(22)2 1(23)1[2,3] 4(24)23 7(25)7[5,6,24] 7(26)25[8-13] 4(27)24 2(28)22 8(29)26 8(30)8[9-13] 9(31)9[10-13] 9(32@30)31 30(33)32 10(34)10[11] 14(35)14[15-17] 14(36)35[15] 16(37)36 12(38)13 16(39)16 II

Apparently 10(38)34[12] would have won, yielding *2 + *2 + 4 + 2L0, which is a misere P-position.

### danny purvis

Dec 16, 2008, 8:26:31 AM12/16/08
Thanks very much! I will look forward to playing through this, and I will publish it if you don't object.

From: Josh Purinton <josh.p...@gmail.com>
Sent: Tuesday, December 16, 2008 5:39:42 AM
Subject: Further thoughts on Peltier-Purvis

### Jeff Peltier

Dec 16, 2008, 9:39:54 AM12/16/08
Yes this is true and really amazing as 4spots has genus 1°²·· and 0L2 has 2¹4.., how is that sum of the two has genus 3°²°·· ?
I don' remember if I was aware when I played move 37. Can you pursue the analysis one move earlier?

--
Jeff
--
--
Jean-François

### Dan Hoey

Dec 16, 2008, 10:16:54 AM12/16/08
Jeff Peltier wrote:
> Yes this is true and really amazing as 4spots has genus 1°²·· and 0L2 has
> 2¹4.., how is that sum of the two has genus 3°²°·· ?

2^14 is not tame. You can't tell the genus of the sum from the genus of
the addends unless all are tame.

Dan

### Josh Purinton

Dec 16, 2008, 12:01:07 PM12/16/08
On Tue, Dec 16, 2008 at 9:39 AM, Jeff Peltier wrote:
I don' remember if I was aware when I played move 37. Can you pursue the analysis one move earlier?

Great idea! As it happens, move 37 was also a mistake.

27(37)4[5] would have won, yielding *3 + 4 + 1P2 + 2L1.

I asked Aunt Beast to look at move 26, but steam began coming out of her ears, and I could tell it was going to take too long, so I told her not to bother.