Commit | Line | Data |
---|---|---|
7eeb782e AT |
1 | loadsgf games/strategy1.sgf |
2 | 1 reg_genmove white | |
3 | #? [Q13] | |
4 | ||
5 | loadsgf games/strategy2.sgf 5 | |
6 | 2 reg_genmove black | |
7 | #? [!G2|D2] | |
8 | ||
9 | # # Current GNU Go would never play those moves. Removed. | |
10 | # loadsgf games/strategy2.sgf 9 | |
11 | # 3 reg_genmove black | |
12 | # #? [!H1|H2] | |
13 | # | |
14 | # loadsgf games/strategy2.sgf 11 | |
15 | # 4 reg_genmove black | |
16 | # #? [!F1] | |
17 | ||
18 | loadsgf games/strategy3.sgf 14 | |
19 | 5 reg_genmove white | |
20 | #? [R17|P17] | |
21 | ||
22 | # incident 104 | |
23 | # Probably requires a dynamic connection analysis to solve. | |
24 | loadsgf games/incident104.sgf 63 | |
25 | 6 reg_genmove white | |
26 | #? [E10]* | |
27 | ||
28 | # incident 106 | |
29 | loadsgf games/incident104.sgf 215 | |
30 | 7 reg_genmove white | |
31 | #? [S19] | |
32 | ||
33 | # incident 107 | |
34 | # tm - moved the test out to move 228; at 216, T17 is arguable bigger. | |
35 | loadsgf games/incident107.sgf 228 | |
36 | 8 reg_genmove black | |
37 | #? [N2|N1]* | |
38 | ||
39 | # incident 109 | |
40 | loadsgf games/incident108b.sgf 172 | |
41 | 9 reg_genmove black | |
42 | #? [P15] | |
43 | ||
44 | # incident 110 | |
45 | loadsgf games/incident108b.sgf 176 | |
46 | 10 reg_genmove black | |
47 | #? [Q7] | |
48 | ||
49 | #CATEGORY=BAD_PROBLEM | |
50 | #DESCRIPTION=Not clear enough. | |
51 | #SEVERITY=0 | |
52 | # From STRATEGY: | |
53 | # 2.7.178 makes the move at P17 which is better than the game | |
54 | # move at P18 but probably inferior to the recommended move at P15. | |
55 | # It is not totally clear to me that the move at P15 is really | |
56 | # best since the 3-3 point is still open. | |
57 | # incident 111 | |
58 | loadsgf games/incident108b.sgf 178 | |
59 | 11 reg_genmove black | |
60 | #? [P15] | |
61 | ||
62 | # incident 112 | |
63 | # Suboptimal to make eye at L4. | |
64 | loadsgf games/incident108b.sgf 270 | |
65 | 12 reg_genmove black | |
66 | #? [N4] | |
67 | ||
68 | # incident 113 | |
69 | # See also reading:194 | |
70 | loadsgf games/incident108b.sgf 292 | |
71 | 13 reg_genmove black | |
72 | #? [N18] | |
73 | ||
74 | # incident 121 | |
75 | # E9 is not necessarily the best move since upper left weak too. | |
76 | # But the game move at R7 is not urgent! | |
77 | loadsgf games/incident121.sgf 24 | |
78 | 14 reg_genmove black | |
79 | #? [E9] | |
80 | ||
81 | # incident 132 | |
82 | loadsgf games/incident121.sgf 116 | |
83 | 16 reg_genmove black | |
84 | #? [H12] | |
85 | ||
86 | ||
87 | #CATEGORY=PATTERN_TUNING | |
88 | #DESCRIPTION=N11 may not be biggest, but should be found. | |
89 | #SEVERITY=2 | |
90 | # 2.7.179: This can surely be fixed with the addition of a | |
91 | # pattern. After N11, the strings at M12 and N10 cannot both be | |
92 | # saved. This is not necessarily the biggest move on the board but | |
93 | # at least this move should be found. | |
94 | # incident 165 | |
95 | # H10 looks fine - tm | |
96 | loadsgf games/incident165.sgf 108 | |
97 | 17 reg_genmove black | |
98 | #? [N11|H10] | |
99 | ||
100 | ||
101 | # incident 166a | |
102 | # Original move was N3, which is pointless. | |
103 | # E5 is an overplay, see reading test 124. | |
104 | # F3 seems fine to protect the two cuts -arend | |
105 | # N11 may well be biggest but is already tested above. | |
106 | loadsgf games/incident165.sgf 142 | |
107 | 18 restricted_genmove black E4 E5 F3 N3 | |
108 | #? [E4|F3] | |
109 | ||
110 | # # Current GNU Go would never consider passing here. Removed. | |
111 | # loadsgf games/strategy4.sgf | |
112 | # 19 reg_genmove black | |
113 | # #? [!PASS] | |
114 | ||
115 | ||
116 | #CATEGORY=OWL_TUNING | |
117 | #DESCRIPTION=S13 is terrible - negative value. | |
118 | #SEVERITY=8 | |
119 | # 2.7.179: Owl code finds Q11 but it remains undervalued at 17 points. | |
120 | # incident 290 | |
121 | loadsgf games/incident290.sgf 30 | |
122 | 20 reg_genmove black | |
123 | #? [Q11] | |
124 | ||
125 | # incident 291 | |
126 | loadsgf games/incident291.sgf 54 | |
127 | 21 reg_genmove black | |
128 | #? [S13] | |
129 | ||
130 | # incident 292 | |
131 | loadsgf games/incident291.sgf 70 | |
132 | 22 reg_genmove black | |
133 | #? [K3] | |
134 | ||
135 | # incident 294 | |
136 | loadsgf games/incident291.sgf 88 | |
137 | 23 reg_genmove black | |
138 | #? [!K5] | |
139 | ||
140 | # incident 295 | |
141 | # endgame mistake | |
142 | loadsgf games/incident291.sgf 250 | |
143 | 24 reg_genmove black | |
144 | #? [F4] | |
145 | ||
146 | # Moved to filllib.tst | |
147 | # # incident 296 | |
148 | # # liberty filling mistake | |
149 | # loadsgf games/incident291.sgf 274 | |
150 | # 25 reg_genmove black | |
151 | # #? [PASS] | |
152 | ||
153 | loadsgf games/strategy5.sgf 12 | |
154 | 26 reg_genmove black | |
155 | #? [D3] | |
156 | ||
157 | ||
158 | #CATEGORY=PATTERN_TUNING | |
159 | #DESCRIPTION=Blocking in the corner is better. | |
160 | #SEVERITY=3 | |
161 | # This test passed before 2.7.183, but for the wrong reason: | |
162 | # the move at C3 was allegedly an owl attack/defense point for | |
163 | # the dragon at D3, which of course it is not. | |
164 | # | |
165 | # Thus this example should be regarded as having been already | |
166 | # broken. | |
167 | # | |
168 | loadsgf games/strategy5.sgf 14 | |
169 | 27 reg_genmove black | |
170 | #? [C3] | |
171 | ||
172 | loadsgf games/strategy5.sgf 20 | |
173 | 28 restricted_genmove black G3 F2 | |
174 | #? [G3] | |
175 | ||
176 | loadsgf games/strategy5.sgf 40 | |
177 | 29 reg_genmove black | |
178 | #? [R4]* | |
179 | ||
180 | loadsgf games/strategy5.sgf 44 | |
181 | 30 restricted_genmove black T7 S4 | |
182 | #? [S4] | |
183 | ||
184 | # This is a problem with the semeai analyzer and revise_semeai(). | |
185 | # But there is no semeai. | |
186 | loadsgf games/strategy6.sgf 274 | |
187 | 31 reg_genmove black | |
188 | #? [PASS] | |
189 | ||
190 | loadsgf games/strategy7.sgf 23 | |
191 | 32 reg_genmove black | |
192 | #? [A4] | |
193 | ||
194 | loadsgf games/strategy8.sgf 283 | |
195 | 33 reg_genmove white | |
196 | #? [O7] | |
197 | ||
198 | loadsgf games/nicklas/nicklas8.sgf 72 | |
199 | 34 reg_genmove black | |
200 | #? [E17] | |
201 | ||
202 | loadsgf games/nicklas/nicklas8.sgf 80 | |
203 | 35 restricted_genmove black B17 H17 | |
204 | #? [H17] | |
205 | ||
206 | loadsgf games/nicklas/nicklas8.sgf 82 | |
207 | 36 restricted_genmove black H17 D18 | |
208 | #? [H17] | |
209 | ||
210 | loadsgf games/nicklas/nicklas8.sgf 84 | |
211 | 37 restricted_genmove black A17 A18 C19 H17 | |
212 | #? [H17] | |
213 | ||
214 | # The life code solves this mistake. | |
215 | loadsgf games/nicklas/nicklas8.sgf 86 | |
216 | 38 restricted_genmove black H17 B19 | |
217 | #? [H17] | |
218 | ||
219 | loadsgf games/nicklas/nicklas8.sgf 96 | |
220 | 39 reg_genmove black | |
221 | #? [P16] | |
222 | ||
223 | ||
224 | #CATEGORY=STRATEGY | |
225 | #DESCRIPTION=L14 is not a terrible way to attack O13 group. | |
226 | #SEVERITY=2 | |
227 | #Consider retiring this problem, or adding L14. L14 is not terrible. | |
228 | #M11 is better than L4 | |
229 | # 2.7.231: Broke again. M11 is overrated at 28 points. The | |
230 | # large followup value is responsible. | |
231 | # I disagree. M11 makes territory while attacking. -arend | |
232 | loadsgf games/nicklas/nicklas8.sgf 98 | |
233 | 40 reg_genmove black | |
234 | #? [E5|M11]* | |
235 | ||
236 | ||
237 | #CATEGORY=STRATEGY | |
238 | #DESCRIPTION=N16 is better than M11. Could be a tough attack, though. | |
239 | #SEVERITY=4 | |
240 | # 2.7.179: After fixing strategy 40, the spurious defenses of D19 | |
241 | # will be removed. Now N16 will be the most valuable move. So | |
242 | # this should be fixed automatically when 40 is fixed. | |
243 | # | |
244 | # 2.7.180: this test passes. | |
245 | # | |
246 | # 2.7.220: N12 is found as an owl attack point for the S14 dragon. | |
247 | # This attack will probably fail. More seriously, M16 is preferred | |
248 | # to N16. | |
249 | # | |
250 | loadsgf games/nicklas/nicklas8.sgf 106 | |
251 | 41 reg_genmove black | |
252 | #? [N16] | |
253 | ||
254 | loadsgf games/strategy9.sgf 232 | |
255 | 42 reg_genmove black | |
256 | #? [C17] | |
257 | ||
258 | # A14 works but is inferior shape since it leaves more ko threats. | |
259 | loadsgf games/strategy10.sgf | |
260 | 43 reg_genmove white | |
261 | #? [B13|B14|A14] | |
262 | ||
263 | #CATEGORY=STRATEGY | |
264 | #DESCRIPTION=L4 is awkward. | |
265 | #SEVERITY=2 | |
266 | # 2.7.179: Whether or not the recommended move at G2 is found, the | |
267 | # move at A6 is egregiously bad. In this situation an owl critical | |
268 | # dragon adjoins an owl dead one. We need to revisit our policy for | |
269 | # such situations. The reading code thinks the worm at B4 can be | |
270 | # defended, so this is an owl lunch. | |
271 | # | |
272 | # 2.7.180: After revision of semeai.c, the move at A6 is no longer | |
273 | # found. Now GNU plays at M16 on the top. Arguably stabilizing | |
274 | # the bottom at G2 or (worse but still OK) at J2 is better but | |
275 | # I would class this result as ACCEPTABLE. | |
276 | # | |
277 | # 3.1.9: Best move here is tough. GNU Go correctly tries | |
278 | # to stabilize G3 group. | |
279 | # | |
280 | # See also reading:158 | |
281 | loadsgf games/strategy11.sgf 77 | |
282 | 44 reg_genmove black | |
283 | #? [G2]* | |
284 | ||
285 | # G17/G18 and C1/H1 are both huge. | |
286 | loadsgf games/strategy11.sgf 127 | |
287 | 45 reg_genmove black | |
288 | #? [G17|G18|C1|H1] | |
289 | ||
290 | loadsgf games/strategy11.sgf 245 | |
291 | 46 reg_genmove black | |
292 | #? [E1] | |
293 | ||
294 | loadsgf games/strategy12.sgf 9 | |
295 | 47 reg_genmove black | |
296 | #? [!N17] | |
297 | ||
298 | #CATEGORY=FUSEKI_STRATEGY | |
299 | #DESCRIPTION=B17 better than D17 | |
300 | #SEVERITY=3 | |
301 | loadsgf games/strategy12.sgf 19 | |
302 | 48 reg_genmove black | |
303 | #? [B17]* | |
304 | ||
305 | loadsgf games/strategy12.sgf 21 | |
306 | # D16 is the strongest but these are acceptable | |
307 | 49 reg_genmove black | |
308 | #? [C17|D16|B17]* | |
309 | ||
310 | ||
311 | #CATEGORY=OWL_TUNING | |
312 | #DESCRIPTION=P9 is pointless compared to Q9. | |
313 | #SEVERITY=8 | |
314 | # Q9 is clearly better than Q11 because it stops a black connection | |
315 | # along the edge. | |
316 | # So much better, that I removed Q11 option -trevor | |
317 | loadsgf games/strategy13.sgf | |
318 | 50 reg_genmove white | |
319 | #? [Q9]* |