[gnugo-devel] Regression results for the large scale patch

[Stéphane Nicolet]

Here are the regression results (against 3.4) for the large
scale capture patch posted in
<http://mail.gnu.org/archive/html/gnugo-devel/2003-07/msg00209.html>



In total, there are 13 PASS (two more if we include the
positions shown in the original message) and 19 FAIL. The
fails and some of the passes are discussed below.

I think that 6 of the fails are real, 8 can be interpreted
as reliable alternate answers, while 5 are too difficult
for me.

Comments are most welcome, of course.

Stephane.


**********


39 unexpected FAIL: Correct 'R7', got 'R6'
   http://www.public32.com/regress/?nngs1:39

   Maybe not a fail. The engine believes that
   R6 kills R3 while protecting the territory


21 unexpected FAIL: Correct 'S16', got 'S14'
   http://www.public32.com/regress/?neurogo:21

   Real fail. S16 has more effect on the corner


2 unexpected FAIL: Correct 'Q18|S16', got 'P16'
   http://www.public32.com/regress/?arion:2

   Don't know. If P16 indeed kills R17,
   then it's OK. I'm not strong enough to tell.


13 unexpected FAIL: Correct 'R13|M8|L9', got 'J3'
   http://www.public32.com/regress/?lazarus:13

   Maybe not a fail. J3 threatens to kill the
   group at the bottom.


64 unexpected FAIL: Correct 'B13|C13|C12|E14|D14', got 'B12'
   http://www.public32.com/regress/?strategy2:64

   Maybe not a fail. Doesn't seem different from B13.


89 unexpected FAIL: Correct '(D16|B7|G13)', got 'H11'
   http://www.public32.com/regress/?strategy2:89

   Maybe not a fail. H11 has more effect in the center, but
   leaves more freedom for the white stones.


440 unexpected FAIL: Correct 'O2', got 'N3'
   http://www.public32.com/regress/?nngs:440

   Don't know. O2 captures the corner more tightly,
   but N3 sort of protects the cut in O4.


700 unexpected FAIL: Correct 'K17', got 'K16'
   http://www.public32.com/regress/?nngs:700

   Real fail. K17 is better than K16.


22 unexpected FAIL: Correct 'F2', got 'A5'
   http://www.public32.com/regress/?global:22

   Maybe not a fail. A5 seems to kill the bottom
   group while preparing a gote capture of four
   stones on the left.


870 unexpected FAIL: Correct 'R6', got 'O6'
   http://www.public32.com/regress/?trevord:870

   Don't know.


167 unexpected FAIL: Correct 'D4', got 'E5'
   http://www.public32.com/regress/?strategy4:167

   Don't know. Maybe E5 is playable, but not best.


200 unexpected FAIL: Correct 'P6|P7|Q7|S18', got 'S8'
   http://www.public32.com/regress/?strategy4:200

   Real fail.


110 unexpected FAIL: Correct 'B14|B13', got 'M16'
   http://www.public32.com/regress/?nngs2:110

   Maybe not a fail. I would say that 3.4 had
   this one as an accidental pass since the
   evaluations of L18 and B14 were so close.


240 unexpected FAIL: Correct 'F17', got 'G10'
   http://www.public32.com/regress/?nngs2:240

   Real fail. Protecting in F17 is bigger than
   capturing J9.


240 unexpected FAIL: Correct 'G13|G12', got 'J12'
   http://www.public32.com/regress/?nngs3:240

   Maybe not a fail. J12 seems OK to me.


330 unexpected FAIL: Correct 'P12|Q11', got 'O10'
   http://www.public32.com/regress/?nngs3:330

   Maybe not a fail. O10 is a good move too,
   to capture on a large scale.


490 unexpected FAIL: Correct 'P15', got 'T13'
   http://www.public32.com/regress/?nngs3:490

   Don't know. T13 threatens to connect to the Q10
   stone but seems a little bit strange.
   Btw, why are the S14-T14 stones dead ? Shouldn't
   they be aggregated to the Black critical dragon ?


33 unexpected FAIL: Correct 'S17|R3', got 'O4'
   http://www.public32.com/regress/?manyfaces1:33

   Real fail. If Black is to play in that corner,
   R3 is indeed better.


4 unexpected FAIL: Correct 'F4', got 'K1'
   http://www.public32.com/regress/?13x13b:4

   Real fail.



15 unexpected PASS!
   http://www.public32.com/regress/?nngs1:15


42 unexpected PASS!
   http://www.public32.com/regress/?nngs1:42

   The patch finds that O9 also kills M6.


1406 unexpected PASS!
   http://www.public32.com/regress/?nicklas1:1406

   When was this test broken?


290 unexpected PASS!
   http://www.public32.com/regress/?nngs:290

114 unexpected PASS!
   http://www.public32.com/regress/?strategy3:114

   D7 kills the corner while protecting the
   territory


202 unexpected PASS!
   http://www.public32.com/regress/?strategy4:202

140 unexpected PASS!
   http://www.public32.com/regress/?nngs2:140

   Finds the right connection to prevent
   White from entering the territory


600 unexpected PASS!
   http://www.public32.com/regress/?nngs2:600

   When was this test broken?


470 unexpected PASS!
   http://www.public32.com/regress/?nngs3:470

   It is passed, but why is J17 a bad move,
   by the way ?


950 unexpected PASS!
   http://www.public32.com/regress/?nngs3:950

   F15 is the proper way to capture H14.


1030 unexpected PASS!
   http://www.public32.com/regress/?nngs3:1030

   Proper way to capture S11.


160 unexpected PASS!
  http://www.public32.com/regress/?nngs4:160

640 unexpected PASS!
  http://www.public32.com/regress/?nngs4:640
  Not on Trevor's site yet
































... /// ... \\\ ...