14-10-18 |
dd |
giveaway |
anti_capture_sequence |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,10.00]
Total: 931 W: 288 L: 354 D: 289
|
sprt @ 10+0.1 th 1 |
Prefer giveaway captures which put the capturing piece en prise |
07-08-18 |
ia |
giveaway |
anti_shallow |
diff
|
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 16582 W: 8319 L: 7979 D: 284
|
sprt @ 10+0.1 th 1 |
conditionally skip shallow depth pruning |
14-09-18 |
ia |
giveaway |
razoring |
diff
|
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 6849 W: 2738 L: 2561 D: 1550
|
sprt @ 10+0.1 th 1 |
skip razoring |
26-09-18 |
ia |
giveaway |
master |
diff
|
ELO: -534.80 +-52.2 (95%) LOS: 0.0%
Total: 1000 W: 39 L: 951 D: 10
|
1000 @ 10+0.1 th 1 |
elo measurement with respect to MV-SF |
20-09-18 |
ia |
giveaway |
futility2 |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,10.00]
Total: 2147 W: 809 L: 864 D: 474
|
sprt @ 10+0.1 th 1 |
skip futility pruning |
22-09-18 |
ia |
giveaway |
capture_see |
diff
|
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 291 W: 147 L: 64 D: 80
|
sprt @ 10+0.1 th 1 |
skip SEE pruning for captures |
20-09-18 |
ia |
giveaway |
anti_threats2 |
diff
|
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 462 W: 226 L: 136 D: 100
|
sprt @ 10+0.1 th 1 |
tweak threat evaluation |
20-09-18 |
ia |
giveaway |
quiet_see |
diff
|
LLR: 2.94 (-2.94,2.94) [0.00,10.00]
Total: 108 W: 86 L: 7 D: 15
|
sprt @ 10+0.1 th 1 |
skip SEE pruning |
14-09-18 |
ia |
giveaway |
futility |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,10.00]
Total: 423 W: 132 L: 212 D: 79
|
sprt @ 10+0.1 th 1 |
skip futility pruning |
06-09-18 |
ia |
giveaway |
master |
diff
|
ELO: -680.57 +-81.4 (95%) LOS: 0.0%
Total: 1000 W: 18 L: 979 D: 3
|
1000 @ 10+0.1 th 1 |
measure elo difference between fairy and multivariant fork |
02-09-18 |
ia |
giveaway |
anti_no_initiative |
diff
|
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 1451 W: 646 L: 539 D: 266
|
sprt @ 10+0.1 th 1 |
skip initiative bonus |
01-09-18 |
ia |
giveaway |
anti_lmr |
diff
|
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 799 W: 359 L: 264 D: 176
|
sprt @ 10+0.1 th 1 |
less LMR for mandatory captures |
01-09-18 |
ia |
giveaway |
lazy |
diff
|
ELO: 4.52 +-18.4 (95%) LOS: 68.5%
Total: 1000 W: 371 L: 358 D: 271
|
1000 @ 10+0.1 th 1 |
remove lazy eval |
25-08-18 |
ia |
giveaway |
mobility |
diff
|
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 1946 W: 816 L: 705 D: 425
|
sprt @ 10+0.1 th 1 |
tweak mobility evaluation |
19-08-18 |
dd |
giveaway |
anti_capture_count |
diff
|
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 14456 W: 4924 L: 4663 D: 4869
|
sprt @ 30+0.3 th 1 |
Extend giveaway search based upon more accurate capture count (take 2; LTC) |
18-08-18 |
dd |
giveaway |
anti_capture_count |
diff
|
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 6450 W: 2386 L: 2221 D: 1843
|
sprt @ 10+0.1 th 1 |
Extend giveaway search based upon more accurate capture count (take 2) |
12-08-18 |
ia |
giveaway |
master |
diff
|
ELO: -682.85 +-55.8 (95%) LOS: 0.0%
Total: 2000 W: 35 L: 1958 D: 7
|
2000 @ 10+0.1 th 1 |
elo measurement |
15-08-18 |
dd |
giveaway |
anti_capture_count |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,10.00]
Total: 4164 W: 1359 L: 1383 D: 1422
|
sprt @ 30+0.3 th 1 |
Extend giveaway search based upon more accurate capture count (LTC) |
14-08-18 |
dd |
giveaway |
anti_capture_count |
diff
|
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 7598 W: 2767 L: 2588 D: 2243
|
sprt @ 10+0.1 th 1 |
Extend giveaway search based upon more accurate capture count |
13-08-18 |
ia |
giveaway |
probcut |
diff
|
ELO: -9.73 +-13.3 (95%) LOS: 7.6%
Total: 2000 W: 740 L: 796 D: 464
|
2000 @ 10+0.1 th 1 |
tweak probcut margin |
13-08-18 |
dd |
giveaway |
anti_gives_check |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,10.00]
Total: 519 W: 156 L: 231 D: 132
|
sprt @ 10+0.1 th 1 |
Overload gives_check for antichess |
12-08-18 |
dd |
giveaway |
anti_search_reduce |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,10.00]
Total: 3231 W: 1129 L: 1166 D: 936
|
sprt @ 10+0.1 th 1 |
Do not consider TT captures and promotions for giveaway LMR |
12-08-18 |
dd |
giveaway |
anti_search_prune |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,10.00]
Total: 1859 W: 658 L: 715 D: 486
|
sprt @ 10+0.1 th 1 |
Increase antichess search pruning |
10-08-18 |
ia |
giveaway |
anti_pv |
diff
|
LLR: 7.17 (-2.94,2.94) [0.00,10.00]
Total: 234 W: 231 L: 2 D: 1
|
sprt @ 10+0.1 th 1 |
negative piece values for antichess |
09-08-18 |
dd |
giveaway |
anti_capture_count |
diff
|
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 5100 W: 1815 L: 1671 D: 1614
|
sprt @ 30+0.3 th 1 |
Extend giveaway search based upon capture count (LTC) |
09-08-18 |
ia |
giveaway |
anti_threats |
diff
|
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 704 W: 401 L: 290 D: 13
|
sprt @ 10+0.1 th 1 |
threat evaluation for antichess, losers |
08-08-18 |
dd |
giveaway |
anti_capture_count |
diff
|
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 2227 W: 864 L: 754 D: 609
|
sprt @ 10+0.1 th 1 |
Extend giveaway search based upon capture count |
07-08-18 |
ia |
giveaway |
anti_extension |
diff
|
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 485 W: 290 L: 183 D: 12
|
sprt @ 10+0.1 th 1 |
extend forced captures |
04-08-18 |
dd |
giveaway |
move_horizon |
diff
|
LLR: 2.95 (-2.94,2.94) [0.00,10.00]
Total: 7882 W: 2791 L: 2611 D: 2480
|
sprt @ 30+0.3 th 1 |
Tweak antichess time management move horizon (LTC) |
04-08-18 |
dd |
giveaway |
move_horizon |
diff
|
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 3197 W: 1160 L: 1040 D: 997
|
sprt @ 10+0.1 th 1 |
Tweak antichess time management move horizon |
04-08-18 |
dd |
giveaway |
tune_move_horizon |
diff
|
1000/1000 iterations
2000/2000 games played
|
2000 @ 10+0.1 th 1 |
Tune antichess move horizon starting at 30 |
03-08-18 |
dd |
giveaway |
tune_move_horizon |
diff
|
3102/5000 iterations
6211/10000 games played
|
10000 @ 10+0.1 th 1 |
Tune antichess move horizon |
02-08-18 |
dd |
giveaway |
anti_move_horizon |
diff
|
LLR: 2.96 (-2.94,2.94) [0.00,10.00]
Total: 2317 W: 860 L: 751 D: 706
|
sprt @ 10+0.1 th 1 |
Tweak antichess move horizon |
01-08-18 |
ia |
giveaway |
anti_futility |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,10.00]
Total: 2114 W: 1002 L: 1073 D: 39
|
sprt @ 10+0.1 th 1 |
increase futility margin |
30-07-18 |
ia |
giveaway |
see |
diff
|
ELO: -5.73 +-14.7 (95%) LOS: 22.3%
Total: 2000 W: 923 L: 956 D: 121
|
2000 @ 10+0.1 th 1 |
consider winning moves in SEE |
28-07-18 |
ia |
giveaway |
anti_early_pruning |
diff
|
LLR: 2.97 (-2.94,2.94) [0.00,10.00]
Total: 86 W: 83 L: 1 D: 2
|
sprt @ 10+0.1 th 1 |
skip early pruning |
27-07-18 |
ia |
giveaway |
master |
diff
|
ELO: -1079.24 +-389.5 (95%) LOS: 0.0%
Total: 2000 W: 4 L: 1996 D: 0
|
2000 @ 10+0.1 th 1 |
measure elo difference |
25-07-18 |
ia |
giveaway |
fishnet-250718 |
diff
|
ELO: 12.34 +-12.5 (95%) LOS: 97.4%
Total: 2000 W: 710 L: 639 D: 651
|
2000 @ 30+0.3 th 1 |
regression test |
25-06-18 |
dd |
giveaway |
has_game_cycle |
diff
|
LLR: -3.00 (-2.94,2.94) [-10.00,5.00]
Total: 1035 W: 331 L: 391 D: 313
|
sprt @ 30+0.3 th 1 |
Simplify variant tests in has_game_cycle (LTC) |
23-06-18 |
dd |
giveaway |
has_game_cycle |
diff
|
LLR: 2.96 (-2.94,2.94) [-10.00,5.00]
Total: 4328 W: 1588 L: 1563 D: 1177
|
sprt @ 10+0.1 th 1 |
Simplify variant tests in has_game_cycle |
16-06-18 |
ia |
giveaway |
anti_razor2 |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,10.00]
Total: 4071 W: 1446 L: 1472 D: 1153
|
sprt @ 10+0.1 th 1 |
tweak razor margins |
03-06-18 |
ia |
giveaway |
fishnet-050518 |
diff
|
ELO: -6.95 +-12.3 (95%) LOS: 13.5%
Total: 2000 W: 638 L: 678 D: 684
|
2000 @ 30+0.3 th 1 |
regression test |
26-05-18 |
dd |
giveaway |
anti_see_ge |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,10.00]
Total: 156 W: 26 L: 113 D: 17
|
sprt @ 10+0.1 th 1 |
Order giveaway moves by whether or not it puts the piece en prise |
26-05-18 |
dd |
giveaway |
anti_see_ge |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,10.00]
Total: 204 W: 40 L: 119 D: 45
|
sprt @ 10+0.1 th 1 |
Widen search in positions where many captures are possible |
25-05-18 |
dd |
giveaway |
anti_see_ge |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,10.00]
Total: 6603 W: 2334 L: 2327 D: 1942
|
sprt @ 10+0.1 th 1 |
Short-circuit antichess static exchange evaluation |
24-05-18 |
dd |
giveaway |
anti_capture_futility |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,10.00]
Total: 695 W: 208 L: 278 D: 209
|
sprt @ 10+0.1 th 1 |
Enable giveaway futility pruning for captures |
24-05-18 |
dd |
giveaway |
anti_capture_razor |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,10.00]
Total: 1048 W: 344 L: 410 D: 294
|
sprt @ 10+0.1 th 1 |
Enable giveaway razoring for captures |
22-05-18 |
dd |
giveaway |
anti_capture_prune |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,10.00]
Total: 5744 W: 1917 L: 1921 D: 1906
|
sprt @ 30+0.3 th 1 |
Disable antichess capture-based pruning (LTC) |
23-05-18 |
dd |
giveaway |
anti_capture |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,10.00]
Total: 339 W: 89 L: 166 D: 84
|
sprt @ 10+0.1 th 1 |
Redefine giveaway variant captureOrPromotion term |
21-05-18 |
dd |
giveaway |
anti_probcut |
diff
|
LLR: 2.95 (-2.94,2.94) [-10.00,5.00]
Total: 948 W: 322 L: 278 D: 348
|
sprt @ 30+0.3 th 1 |
Simplify away antichess probcut-skip (LTC) |