ShashChess

[amchess]

In general, for every position, no less than one minute.
After, I apply this formule (old but always good):
M = 2nft
where
M is your HTs size in kilobytes (alwyas a 2 power and not greater than half RAM)
n is the number of threads: physical or logical if you enable multivp
f is the clock frequence in Mhz
t is the average time in seconds per move.

Best regards,
Andrea

[peter]

In general, for every position, no less than one minute.
After, I apply this formule (old but always good):
M = 2nft
where
M is your HTs size in kilobytes (alwyas a 2 power and not greater than half RAM)
n is the number of threads: physical or logical if you enable multivp
f is the clock frequence in Mhz
t is the average time in seconds per move.

I know this old but to me also still good formula and so normally take 8G hash for 100" x 12cores x 3GHz x 2=7,2G.
And the first run with Tal, Capablanca and Petrosian all 3 checked is about half through after the time since my last posting. Till now 99/160 solved (62%)

I'll be back regards

[amchess]

In general, for every position, no less than one minute.
After, I apply this formule (old but always good):
M = 2nft
where
M is your HTs size in kilobytes (alwyas a 2 power and not greater than half RAM)
n is the number of threads: physical or logical if you enable multivp
f is the clock frequence in Mhz
t is the average time in seconds per move.

I know this old but to me also still good formula and so normally take 8G hash for 100" x 12cores x 3GHz x 2=7,2G.
And the first run with Tal, Capablanca and Petrosian all 3 checked is about half through after the time since my last posting. Till now 99/160 solved (62%)

I'll be back regards

Ok.
No problem.
Best regards,
Andrea

[peter]

In general, for every position, no less than one minute.
After, I apply this formule (old but always good):
M = 2nft
where
M is your HTs size in kilobytes (alwyas a 2 power and not greater than half RAM)
n is the number of threads: physical or logical if you enable multivp
f is the clock frequence in Mhz
t is the average time in seconds per move.

I know this old but to me also still good formula and so normally take 8G hash for 100" x 12cores x 3GHz x 2=7,2G.
And the first run with Tal, Capablanca and Petrosian all 3 checked is about half through after the time since my last posting. Till now 99/160 solved (62%)

I'll be back regards

Ok.
No problem.
Best regards,
Andrea

Pity I forgot to set Syzygy ProbeLimit to 6 instead of 0, which was preset.

Not good for positions like 313 of course

[d]5k2/R7/p4p2/r7/5KPP/8/8/8 b - - 0 1

which on the other hand is too easy with 6men yet but wasn't solved in 100" with ProbeLimit 0.

Maybe it should be replaced for runs with tbs, as well as 301, Dann showed already, being a 6men tbs- position at all and therefore should be "solved" by the GUI as it was set in my run, but wasn't, because the given best move Rh8 is only rated 5th one from 5 with 0.00- eval by the GUI loading 6men Syzygys at root positions too.

http://talkchess.com/forum3/viewtopic.p ... 80#p858480

[d]1R6/8/8/2p5/p1k5/6r1/1K6/8 w - - 0 1

Anyhow, that's the Fritz- log after the run:

Code: Select all

Solving: C:\...\Testsets\326.cbh
Maximum time = 100s.

1. ,  326   Solved in 0.55s/30; Solved: 1
2. HTC,6,  326   Solved in 1.84s/19; Solved: 2
3. ,  326   Solved in 4.84s/22; Solved: 3
4. ,  326   > 100s.
5. Eret,3,  326   > 100s.
6. HTC4 - Halloween,P,  Akobia+Garcia 2013   > 100s.
7. ,  326   Solved in 0.84s/14; Solved: 4
8. HTC,9,  326   Solved in 5.29s/22; Solved: 5
9. Olafsson,H - Unzicker,1,  Lugano 1970   Solved in 7.15s/29; Solved: 6
10. ,  326   Solved in 1.61s/18; Solved: 7
11. ,  326   Solved in 0.42s/7; Solved: 8
12. ,  326   > 100s.
13. ,  326   Solved in 0.45s/10; Solved: 9
14. ,  326   Solved in 4.84s/20; Solved: 10
15. ,  326   Solved in 41.98s/30; Solved: 11
16. ,  326   > 100s.
17. ,  326   > 100s.
18. ,  326   Solved in 0.50s/7; Solved: 12
19. ,  326   > 100s.
20. ,  326   Solved in 22.28s/27; Solved: 13
21. ,  326   Solved in 0.44s/7; Solved: 14
22. ,  326   Solved in 8.89s/40; Solved: 15
23. ,  326   > 100s.
24. ,  326   > 100s.
25. ,  326   Solved in 0.39s/8; Solved: 16
26. ,  326   Solved in 19.36s/26; Solved: 17
27. ,  326   > 100s.
28. ,  326   Solved in 9.70s/25; Solved: 18
29. ,  326   > 100s.
30. ,  326   > 100s.
31. ,  326   Solved in 36.75s/34; Solved: 19
32. ,  326   > 100s.
33. ,  326   Solved in 2.78s/19; Solved: 20
34. ,  326   Solved in 6.90s/37; Solved: 21
35. ,  326   > 100s.
36. ,  326   Solved in 5.35s/24; Solved: 22
37. ,  326   Solved in 4.13s/28; Solved: 23
38. ,  326   Solved in 19.61s/29; Solved: 24
39. ,  326   > 100s.
40. ,  326   Solved in 2.39s/38; Solved: 25
41. ,  326   > 100s.
42. ,  326   > 100s.
43. ,  326   Solved in 9.16s/24; Solved: 26
44. ,  326   Solved in 6.08s/26; Solved: 27
45. ,  326   > 100s.
46. ,  326   Solved in 8.72s/22; Solved: 28
47. ,  326   > 100s.
48. ,  326   > 100s.
49. ,  326   Solved in 0.59s/14; Solved: 29
50. ,  326   > 100s.
51. ,  326   Solved in 53.13s/30; Solved: 30
52. ,  326   Solved in 0.62s/13; Solved: 31
53. ,  326   Solved in 6.33s/23; Solved: 32
54. ,  326   > 100s.
55. ,  326   Solved in 12.53s/21; Solved: 33
56. ,  326   Solved in 1.76s/16; Solved: 34
57. ,  326   Solved in 0.64s/15; Solved: 35
58. ,  326   > 100s.
59. ,  326   Solved in 7.02s/23; Solved: 36
60. ,  326   Solved in 1.53s/17; Solved: 37
61. ,  326   Solved in 13.34s/25; Solved: 38
62. ,  326   Solved in 2.96s/23; Solved: 39
63. ,  326   > 100s.
64. ,  326   > 100s.
65. ,  326   Solved in 1.59s/16; Solved: 40
66. ,  326   Solved in 0.48s/20; Solved: 41
67. ,  326   > 100s.
68. ,  326   > 100s.
69. ,  326   Solved in 0.41s/7; Solved: 42
70. ,  326   > 100s.
71. ,  326   > 100s.
72. ,  326   Solved in 0.38s/7; Solved: 43
73. ,  326   > 100s.
74. ,  326   Solved in 0.47s/16; Solved: 44
75. 33,  326   > 100s.
76. ,  326   Solved in 0.64s/16; Solved: 45
77. ,  326   Solved in 5.76s/54; Solved: 46
78. ,  326   Solved in 1.65s/17; Solved: 47
79. ,  326   Solved in 4.56s/23; Solved: 48
80. ,  326   > 100s.
81. ,  326   Solved in 2.68s/23; Solved: 49
82. ,  326   > 100s.
83. ,  326   Solved in 48.86s/32; Solved: 50
84. ,  326   > 100s.
85. ,  326   Solved in 0.41s/7; Solved: 51
86. ,  326   Solved in 2.59s/19; Solved: 52
87. ,  326   Solved in 3.57s/26; Solved: 53
88. ,  326   > 100s.
89. ,  326   Solved in 1.33s/24; Solved: 54
90. ,  326   Solved in 77.89s/51; Solved: 55
91. ,  326   Solved in 52.95s/38; Solved: 56
92. ,  326   Solved in 7.52s/26; Solved: 57
93. ,  326   > 100s.
94. ,  326   Solved in 0.42s/9; Solved: 58
95. ,  326   Solved in 1.76s/21; Solved: 59
96. ,  326   > 100s.
97. ,  326   Solved in 11.64s/37; Solved: 60
98. ,  326   > 100s.
99. ,  326   Solved in 15.23s/40; Solved: 61
100. ,  326   Solved in 2.36s/20; Solved: 62
101. ,  326   > 100s.
102. ,  326   Solved in 0.48s/11; Solved: 63
103. ,  326   Solved in 17.38s/44; Solved: 64
104. ,  326   > 100s.
105. ,  326   Solved in 1.65s/35; Solved: 65
106. ,  326   Solved in 3.14s/22; Solved: 66
107. ,  326   > 100s.
108. ,  326   > 100s.
109. ,  326   Solved in 2.20s/24; Solved: 67
110. ,  326   Solved in 0.58s/19; Solved: 68
111. ,  326   Solved in 86.44s/59; Solved: 69
112. ,  326   Solved in 7.82s/38; Solved: 70
113. ,  326   Solved in 20.27s/33; Solved: 71
114. ,  326   > 100s.
115. ,  326   > 100s.
116. ,  326   > 100s.
117. ,  326   Solved in 13.29s/30; Solved: 72
118. ,  326   Solved in 0.51s/13; Solved: 73
119. ,  326   Solved in 3.59s/21; Solved: 74
120. ,  326   Solved in 9.24s/28; Solved: 75
121. ,  326   Solved in 47.10s/30; Solved: 76
122. ,  326   Solved in 0.41s/7; Solved: 77
123. ,  326   Solved in 12.48s/24; Solved: 78
124. ,  326   Solved in 0.66s/14; Solved: 79
125. ,  326   Solved in 85.21s/32; Solved: 80
126. ,  326   Solved in 12.36s/21; Solved: 81
127. ,  326   Solved in 7.60s/23; Solved: 82
128. ,  326   Solved in 32.82s/26; Solved: 83
129. ,  326   > 100s.
130. ,  326   Solved in 1.03s/17; Solved: 84
131. ,  326   > 100s.
132. ,  326   > 100s.
133. ,  326   > 100s.
134. ,  326   > 100s.
135. ,  326   Solved in 1.61s/17; Solved: 85
136. ,  326   Solved in 97.11s/30; Solved: 86
137. ,  326   Solved in 4.12s/18; Solved: 87
138. ,  326   > 100s.
139. ,  326   Solved in 1.17s/20; Solved: 88
140. ,  326   Solved in 0.81s/17; Solved: 89
141. ,  326   Solved in 90.42s/27; Solved: 90
142. ,  326   Solved in 0.30s/7; Solved: 91
143. ,  326   Solved in 1.78s/19; Solved: 92
144. ,  326   > 100s.
145. ,  326   > 100s.
146. ,  326   Solved in 15.32s/23; Solved: 93
147. ,  326   > 100s.
148. ,  326   > 100s.
149. ,  326   Solved in 1.42s/18; Solved: 94
150. ,  326   Solved in 0.45s/12; Solved: 95
151. ,  326   Solved in 0.72s/13; Solved: 96
152. ,  326   > 100s.
153. ,  326   > 100s.
154. ,  326   > 100s.
155. ,  326   > 100s.
156. ,  326   > 100s.
157. ,  326   Solved in 6.30s/19; Solved: 97
158. ,  326   Solved in 0.41s/7; Solved: 98
159. ,  326   Solved in 0.62s/14; Solved: 99
160. ,  326   > 100s.
161. ,  326   Solved in 8.10s/24; Solved: 100
162. ,  326   Solved in 6.16s/23; Solved: 101
163. ,  326   Solved in 31s/26; Solved: 102
164. ,  326   > 100s.
165. ,  326   Solved in 87.33s/31; Solved: 103
166. ,  326   Solved in 0.41s/7; Solved: 104
167. ,  326   Solved in 8.32s/17; Solved: 105
168. ,  326   Solved in 7.77s/21; Solved: 106
169. ,  326   Solved in 0.55s/14; Solved: 107
170. ,  326   Solved in 10.64s/18; Solved: 108
171. ,  326   > 100s.
172. ,  326   Solved in 6.88s/21; Solved: 109
173. ,  326   Solved in 0.55s/12; Solved: 110
174. ,  326   > 100s.
175. ,  326   Solved in 74.23s/36; Solved: 111
176. ,  326   > 100s.
177. ,  326   > 100s.
178. ,  326   Solved in 6.43s/19; Solved: 112
179. ,  326   Solved in 11.12s/31; Solved: 113
180. ,  326   > 100s.
181. ,  326   > 100s.
182. ,  326   Solved in 58.75s/28; Solved: 114
183. ,  326   Solved in 26.01s/28; Solved: 115
184. ,  326   Solved in 2.25s/19; Solved: 116
185. ,  326   > 100s.
186. ,  326   Solved in 44.43s/32; Solved: 117
187. ,  326   > 100s.
188. ,  326   Solved in 0.42s/10; Solved: 118
189. ,  326   Solved in 0.37s/7; Solved: 119
190. ,  326   Solved in 0.37s/8; Solved: 120
191. ,  326   Solved in 81.06s/35; Solved: 121
192. ,  326   > 100s.
193. ,  326   > 100s.
194. ,  326   > 100s.
195. ,  326   Solved in 0.81s/17; Solved: 122
196. ,  326   Solved in 0.42s/7; Solved: 123
197. ,  326   Solved in 117.70s/31; Solved: 124
198. ,  326   Solved in 5.40s/21; Solved: 125
199. ,  326   > 100s.
200. ,  326   Solved in 0.92s/28; Solved: 126
201. ,  326   Solved in 49.80s/41; Solved: 127
202. ,  326   Solved in 4.20s/22; Solved: 128
203. ,  326   Solved in 2.75s/20; Solved: 129
204. ,  326   Solved in 0.87s/16; Solved: 130
205. ,  326   > 100s.
206. ,  326   Solved in 4.23s/20; Solved: 131
207. ,  326   Solved in 2.75s/21; Solved: 132
208. ,  326   Solved in 0.48s/15; Solved: 133
209. ,  326   Solved in 0.41s/9; Solved: 134
210. ,  326   Solved in 22.01s/41; Solved: 135
211. ,  326   Solved in 11.03s/26; Solved: 136
212. ,  326   Solved in 4.95s/28; Solved: 137
213. ,  326   Solved in 5.94s/25; Solved: 138
214. ,  326   > 100s.
215. ,  326   Solved in 25.65s/41; Solved: 139
216. ,  326   Solved in 0.41s/7; Solved: 140
217. ,  326   > 100s.
218. ,  326   Solved in 0.48s/12; Solved: 141
219. ,  326   Solved in 0.97s/14; Solved: 142
220. ,  326   > 100s.
221. ,  326   Solved in 2.54s/27; Solved: 143
222. ,  326   > 100s.
223. ,  326   Solved in 12.54s/37; Solved: 144
224. ,  326   Solved in 0.62s/14; Solved: 145
225. ,  326   Solved in 5.32s/22; Solved: 146
226. ,  326   > 100s.
227. ,  326   Solved in 0.59s/14; Solved: 147
228. ,  326   > 100s.
229. ,  326   > 100s.
230. ,  326   Solved in 52.54s/52; Solved: 148
231. ,  326   > 100s.
232. ,  326   > 100s.
233. ,  326   Solved in 4.13s/31; Solved: 149
234. ,  326   > 100s.
235. ,  326   Solved in 1.17s/23; Solved: 150
236. ,  326   > 100s.
237. ,  326   Solved in 0.70s/16; Solved: 151
238. ,  326   Solved in 4.96s/33; Solved: 152
239. ,  326   Solved in 0.55s/17; Solved: 153
240. ,  326   Solved in 12.51s/39; Solved: 154
241. ,  326   Solved in 15.23s/24; Solved: 155
242. ,  326   Solved in 9.61s/29; Solved: 156
243. ,  326   Solved in 0.41s/9; Solved: 157
244. ,  326   Solved in 8.86s/24; Solved: 158
245. ,  326   Solved in 4.31s/27; Solved: 159
246. ,  326   Solved in 0.42s/11; Solved: 160
247. ,  326   Solved in 8.53s/23; Solved: 161
248. ,  326   Solved in 1.17s/18; Solved: 162
249. ,  326   Solved in 7.96s/29; Solved: 163
250. ,  326   Solved in 0.56s/17; Solved: 164
251. ,  326   Solved in 7.19s/23; Solved: 165
252. ,  326   Solved in 0.67s/14; Solved: 166
253. ,  326   Solved in 32.07s/28; Solved: 167
254. ,  326   Solved in 77.02s/28; Solved: 168
255. ,  326   Solved in 0.45s/11; Solved: 169
256. ,  326   Solved in 0.67s/15; Solved: 170
257. ,  326   > 100s.
258. ,  326   > 100s.
259. ,  326   Solved in 0.86s/17; Solved: 171
260. ,  326   > 100s.
261. ,  326   > 100s.
262. ,  326   Solved in 3.62s/29; Solved: 172
263. ,  326   > 100s.
264. ,  326   > 100s.
265. ,  326   Solved in 0.39s/7; Solved: 173
266. ,  326   Solved in 0.42s/11; Solved: 174
267. ,  326   > 100s.
268. ,  326   > 100s.
269. ,  326   Solved in 0.37s/7; Solved: 175
270. ,  326   Solved in 63.57s/29; Solved: 176
271. ,  326   Solved in 2.25s/20; Solved: 177
272. ,  326   Solved in 1.90s/19; Solved: 178
273. ,  326   > 100s.
274. ,  326   > 100s.
275. ,  326   Solved in 4.49s/23; Solved: 179
276. ,  326   > 100s.
277. ,  326   Solved in 0.47s/12; Solved: 180
278. ,  326   Solved in 33.98s/23; Solved: 181
279. ,  326   Solved in 0.50s/13; Solved: 182
280. ,  326   Solved in 0.42s/7; Solved: 183
281. ,  326   Solved in 30.16s/25; Solved: 184
282. ,  326   Solved in 30.72s/26; Solved: 185
283. ,  326   > 100s.
284. ,  326   Solved in 2.17s/20; Solved: 186
285. ,  326   > 100s.
286. ,  326   > 100s.
287. ,  326   > 100s.
288. ,  326   > 100s.
289. ,  326   Solved in 11.05s/23; Solved: 187
290. ,  326   Solved in 3.51s/21; Solved: 188
291. ,  326   > 100s.
292. ,  326   > 100s.
293. ,  326   > 100s.
294. ,  326   Solved in 0.42s/10; Solved: 189
295. ,  326   > 100s.
296. ,  326   Solved in 2.68s/18; Solved: 190
297. ,  326   > 100s.
298. ,  326   > 100s.
299. ,  326   Solved in 3.14s/26; Solved: 191
300. ,  326   Solved in 15.57s/24; Solved: 192
301. 8,  326   > 100s.
302. ,  326   > 100s.
303. ,  326   > 100s.
304. ,  326   > 100s.
305. ,  326   Solved in 1.62s/18; Solved: 193
306. ,  326   Solved in 1.20s/17; Solved: 194
307. ,  326   Solved in 5.37s/18; Solved: 195
308. ,  326   Solved in 0.48s/12; Solved: 196
309. ,  326   Solved in 85.10s/32; Solved: 197
310. ,  326   Solved in 2.34s/17; Solved: 198
311. ,  326   Solved in 57.92s/29; Solved: 199
312. ,  326   Solved in 0.42s/7; Solved: 200
313. ,  326   > 100s.
314. ,  326   Solved in 12.01s/27; Solved: 201
315. ,  326   Solved in 0.80s/17; Solved: 202
316. ,  326   > 100s.
317. ,  326   > 100s.
318. ,  326   Solved in 4.54s/24; Solved: 203
319. ,  326   Solved in 8.66s/26; Solved: 204
320. ,  326   Solved in 0.42s/14; Solved: 205
321. ,  326   Solved in 6.51s/31; Solved: 206
322. ,  326   Solved in 27.60s/34; Solved: 207
323. ,  326   Solved in 0.37s/7; Solved: 208
324. ,  326   Solved in 0.94s/20; Solved: 209
325. ,  326   Solved in 10.30s/24; Solved: 210

Result: 210 out of 325 = 64.6%. Average time = 12.45s / 21.95

Guess I will for the first comparison cut off 301 and give Houdini Tactical and Bluefish a try without tbs too, maybe at some hard midgame- positions there will be less time to solution without tbs anyhow.
As for Bluefish problem is, the Presearch (Pro Analysis) helps, so does (alternatively) the internal MultiPV (Tactical) setting, so the one and only setting comparable isn't easy to find.
As for ShashChess MultiPV would probably help too, in Shredder GUI's .eng- file it can be set as an engine- option, but first of all I'll have to have solution- lines of all of the positions too.

Pondering regards

[Dann Corbit]

What proofs do you have we were wrong? I can only judge based on numbers and code.
If not I can't do more and this discussion will end here.
All positions are well know in chess forums and even their solutions.
The suite has HARD positions, not only tactical:
1.198 Tal
2. 78 Capablanca
3. 42 Petrosian
We use them justly to improve and test Shashin theory.
The tactical positions are 1 (attacking) and 3 (defensive), so the majority.
If there are "bugs", I'm not responsible of them, they are very small and potentially there isn't a "perfect" suite.
The proof is they're continuously improved in chess forums like this one.
I can also show you the result in all of those 3 types of positions...
Anyway, all this doesn't justify the great difference in score.
Moreover, I'm a computer science engineer and I'm improving my skills even at work.
So, I've taken a look at BlueFish code:
except multipv (a very old idea),it's a small subset of ShashChess code (essentially, Crystal's kingDanger, a very good function by John Ellis, imho a very talented programmer).
After all, everyone, even not so skilled, can see it.
So, I'm not surprised.
In general, even if our code is free, we never release without testing and show the tests results, because it's not professional/scientific to pass sentence without proofs. We can surely make some errors because nobody is perfect, but like this we hope to reduce them.
Surely, we made a very hard and long work (even more than one year) to overtake the Houdini tactical version, imho a very good and original work by Robert Houdart, even now at more than two years of distance!
Indeed, in our previous release, we simply told we were near this engine, not we were the best, always based on the results and not on generic sensations".
We, and a lot of people, are happy about this release, but, just because free, clearly you can use/like it or not.
Very humbly, we always speak based on facts, very clearly and not to toot our own horn:
the only goal is to show Shashin theory is flexible enough to ensure this tactical strength without loosing in strength: this isn't true for BlueFish, Crystal or other existing engines. We are convinced that this power is very useful.
I hope I was clear.
Andrea

I did not mean to sound derogatory (which is one problem with internet forums). I like ShashChess very much. However, there are a number of defects in your test set and I guess that fixing them will probably raise the score of ShashChess, not lower it.

I would say that most of the problems have some sort of defect. Many are minor (e.g. missing a check sign on the key move). Others are varied from missing alternate moves to wrong answers.
Consider this analysis done by Shashchess:

Code: Select all

r5k1/5p2/4p1n1/6Q1/1p2b3/1P1p3B/2p5/5RK1 b - - acd 38/55+; acs 190; acn 1020583k; ce -386; bm Ra7; pv Ra7;
2B5/8/3kp3/1P3pp1/3b4/7P/6K1/8 w - - acd 65/71; acs 148; acn 823732k; ce -45; bm Kg3; pv Kg3 Ke7 h4 Be5+ Kg2 gxh4 b6 Kf6 Kh3 Bd6 Kxh4 e5 b7 e4 Kh3 Bc7 Kg2 Bb8 Bd7 f4 Kf1 Ke5 Ke2 f3+ Kf1 Ba7 Bc6 Kf4 Bb5 Ke3 Bc6 Bb8 Bd7 Bd6 Bc6 Kf4 Bd7 Bc7 Bb5 e3 Ke1 Bd6 Kf1 Kg3 Bd3 Bc7 Bc4 Bb8 Bb5 Ba7 Bd3 Kf4 Bb5 Kf5 Bc6 Kg4 Bd7+ Kg3 Bb5 Bb8 Bd3 Kf4 Bb5 Bd6 Kg1 Kg5 Kf1 Be5 Ba6 Bb8 Bb5 Kf5;
2r3k1/3qppb1/2Np1np1/1p1Pn2p/1P6/4BP2/P2QB1PP/3R2K1 b - - acd 44/66; acs 182; acn 829478k; ce -72; bm Nxc6; pv Nxc6 Bxb5 Nxd5 Qxd5 e6 Qd2 d5 Rc1 Qb7 a4 d4 Bf2 Kh7 Rc5 Bf8 Bxc6 Rxc6 Rxc6 Qxc6 b5 Qc4 Bxd4 Qxa4 b6 Qb5 Qc1 Bd6 Kh1 h4 h3 g5 Qb2 Qf1+ Bg1 Kg6 b7 Bb8 Qc2+ Kf6 Qa2 Qc1 Qa8 Qf4 Qa1+ Kg6 Qb2 Qg3 Qc2+ Kf6 Qc1 Qd6 Qc8 Qg3 Qh8+ Kg6 Qd8 Kf5 Qd3+ Kf6 Qd4+ Kg6 Qe3 Qc7 Qd3+ Kh6 Qb1;
r2qr1k1/ppp2pbp/2np2pn/8/2PPN1b1/5PP1/PP2N1BP/R1BQ1RK1 b - - acd 33/52+; acs 174; acn 701404k; ce -48; bm Bc8; pv Bc8;
8/1p2b3/6k1/5p2/p2NbP1p/P1B1P1p1/1P5P/6K1 w - - acd 58/68+; acs 195; acn 1133144k; ce -27; bm Be1; pv Be1;
8/3k4/2p5/8/p1p5/PpP5/1P2B3/r1NK4 w - - acd 68/26; acs 187; acn 1213757k; ce 0; bm Bxc4; pv Bxc4 Rb1 Bd3 Rxb2 Ne2 Ra2 Kc1 Kd6 Ng3 Ke5 Nf1 Kf4 Bb1 Rg2 Nd2 b2+ Kd1 c5 Nc4 Rf2 Nd2 Rh2 Nc4;
5r2/4brkp/6p1/nBq1p3/P1P1P3/R4PNb/2NR1P1P/3Q2K1 b - - acd 35/61+; acs 185; acn 834501k; ce -55; bm Nxc4; pv Nxc4;
2r2bk1/1R3p2/3p1np1/p1rPp2p/P3P3/1N2N2P/R4PP1/6K1 b - - acd 45/63; acs 172; acn 758478k; ce 0; bm Rc3; pv Rc3 Nxa5 Nxe4 Nc6 Rc1+ Kh2 Ra8 Rc7 Nc3 Ra3 Nb5 axb5 Rxa3 Ne7+ Bxe7 Rxc1 Rb3 Rc7 Kf8 Rb7 e4 g3 h4 b6 hxg3+ Kxg3 Bd8 Rb8 Ke7 Rb7+ Kf6 Kg4 Rb4 Rd7 Bxb6 Rxd6+ Ke7 Rc6 Rb5 Kf4 Bxe3+ Kxe3 Rxd5 Kxe4 Ra5 Rb6 Ra3 h4 Kf8 h5 gxh5 Rh6 Ra2 Rxh5;
1R6/8/8/2p5/p1k5/6r1/1K6/8 w - - acd 59/70; acs 185; acn 2812701k; ce 0; bm Ra8; pv Ra8 Rg2+ Ka3 Rg7 Rc8 Rg1 Rh8 Rd1 Kb2 Rd3 Ra8 Rd2+ Ka3 Rd6 Rh8 Ra6 Rg8 Ra7 Rf8 Re7 Rg8 Re1 Rg7 Ra1+ Kb2 Rd1 Ka3 Rd2 Rg8 Rh2 Ra8 Rh6 Rg8 Ra6 Rh8 Rd6 Rb8 Rd1 Rg8 Rf1 Rg7 Rf2 Rh7 Rd2 Rh8 Re2 Rb8 Rf2 Rc8 Rf1 Rh8 Rg1 Rh7 Ra1+ Kb2 Rf1 Rh8 Rd1 Rb8 Rd5 Ka3 Kc3 Kxa4 Kc2 Rc8;
1rnr2k1/p4pp1/2qp1n1p/4pb1P/2p1P3/B1P2BP1/P1QR1PN1/3R2K1 w - - acd 32/53-; acs 163; acn 655832k; ce -361; bm Re2; pv Re2 Bh7;
r2r2k1/2p2p1p/1p4p1/pBnPq3/P3p3/2R1P2P/1P1Q1PP1/2R3K1 w - - acd 46/71; acs 175; acn 699518k; ce 0; bm Bc6; pv Bc6 Rab8 Rc4 Nd3 Qc3 Qg7 Rc2 Nb4 Rd2 f5 Bb5 Qxc3 Rxc3 Rbc8 Bc6 Rb8 Bb5;
8/5pk1/p5p1/4r2p/3R4/5P1P/6N1/6K1 b - - acd 43/61; acs 193; acn 1112096k; ce -152; bm Rf5; pv Rf5 Kf2 g5 f4 a5 Kf3 gxf4 Nxf4 Kf6 Ra4 Kg5 Rc4 Kh4 Rd4 Kg5 Kg3 Rb5 Nd3 Rf5 Ra4 Rd5 h4+ Kf6 Nf4 Re5 Kf3 Rb5 Ke3 Kf5 Nd3 f6 Rf4+ Ke6 Rd4 Kf5 Kd2 Rb6 Rf4+ Ke6 Ra4 Kf5 Ke2 Rb5 Rf4+ Ke6 Re4+ Kd6 Ke3 Rf5 Rd4+ Kc6 Rc4+ Kd6 Ke4 Rb5 Rc8 Kd7 Rf8 Rb1 Rf7+ Kc8;
r1bqr1k1/pp3pbp/2n2np1/2Pp4/3N4/2N1BP2/PP1QB1PP/R4RK1 b - - acd 37/59-; acs 193; acn 783848k; ce 5; bm Bf8; pv Bf8 Nxc6;
r2q4/p4pkp/bp4p1/2nP1r2/8/2N2B2/PP4PP/R2Q1R1K w - - acd 46/45; acs 185; acn 776326k; ce 0; bm Be2; pv Be2 Rxf1+ Qxf1 Bxe2 Qxe2 Qf6 h3 Rd8 Rf1 Re8 Qc4 Qe5 d6 Re6 Nb5 Nd7 Rd1 Rf6 Qc6 Qe2 Nc3 Rf1+ Rxf1 Qxf1+ Kh2 Qf4+ Kh1 Qc1+ Kh2;
r3rbk1/5bp1/p2p3p/1pp2P2/PB6/R4N1P/1q3PP1/1B1QR1K1 w - - acd 45/50; acs 181; acn 820598k; ce 0; bm axb5; pv axb5 Qxb4 Rxa6 Rxe1+ Qxe1 Qxe1+ Nxe1 Rb8 b6 Bc4 Ba2 Bxa2 Rxa2 Rxb6 Kf1 Rb8 Ke2 Re8+ Kd1 Kf7 g4 Re4 Ra7+ Be7 f3 Rc4 Ke2 Ke8 Ra8+ Kd7 Ra7+ Ke8;
2rr1bk1/ppq2pp1/2b1p2p/8/2P2PQ1/2BB4/PP4PP/3R1RK1 b - - acd 38/60+; acs 180; acn 763182k; ce 60; bm Rd7; pv Rd7;
2kq1b1r/pp4pp/2b1p3/2p2p2/2P1P3/P1B2P2/1PQ4P/2K2BR1 b - - acd 38/66; acs 191; acn 846180k; ce 38; bm fxe4; pv fxe4 fxe4 Qh4 Kb1 Rg8 Ka2 Bd6 Qd3 Bxh2 Rh1 Bxe4 Qe2 Bxh1 Qxe6+ Kc7 Qxg8 Bd6 Bd3 g6 Bf6 Qxf6 Qxh7+ Kb8 Qxh1 g5 Qh7 Bf8 Qh2+ Qd6 Qxd6+ Bxd6 Kb1 Kc7 Kc2 Be7 Bf5 Kd6 Bc8 Ke5 Kd3 Kf4 Bxb7 g4 Ke2 Kg3 Kf1 Bd6 Bc8 Kh3 b3 Be7;
r2q1rk1/pp1b1ppp/4p1n1/2p5/2P5/1QP1B1P1/P3PPBP/R3K2R b - - acd 36/57+; acs 185; acn 732302k; ce 92; bm Qa5; pv Qa5;

If you check the answers in your test suite, it will say that most of them are wrong, but actually most of them are right.

[amchess]

https://github.com/amchess/ShashChess/releases/tag/14

[Cornfed]

https://github.com/amchess/ShashChess/releases/tag/14

Thanks for this!

"Improved search"?

[amchess]

Yes.
The goal is always the same: to increase the solved positions number without loosing in playing strength.
For example, latest StockfishNNUE is a lot stronger than Classical One, but this last solves more hard positions.

Andrea

[Cornfed]

Yes.
The goal is always the same: to increase the solved positions number without loosing in playing strength.
For example, latest StockfishNNUE is a lot stronger than Classical One, but this last solves more hard positions.

Andrea

So "doubling the probability of using classic eval", as you note in the link, makes for a better engine?

[amchess]

Yes