| 17-08-18 |
Elb |
npm_r |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 4345 W: 747 L: 854 D: 2744
|
sprt
@ 10+0.1 th 1 |
More nmp reduction in middlegame, less in endgame |
| 17-08-15 |
Elb |
lmr_variant` |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 18510 W: 3248 L: 3295 D: 11967
|
sprt
@ 10+0.1 th 1 |
Take 2 |
| 17-08-14 |
Elb |
lmr_variant |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 64336 W: 10654 L: 10522 D: 43160
|
sprt
@ 10+0.1 th 1 |
Increase/decrease reduction for big differences in stat values. Inspired by lmrTweak3b patch of VoyagerOne. |
| 17-08-03 |
Elb |
reg_test |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 26491 W: 3348 L: 3381 D: 19762
|
sprt
@ 60+0.6 th 1 |
Do SPRT test to test for regression. (LTC) |
| 17-08-03 |
Elb |
reg_test |
diff
|
LLR: 2.95 (-2.94,2.94) [0.00,5.00]
Total: 36987 W: 6792 L: 6510 D: 23685
|
sprt
@ 10+0.1 th 1 |
Do SPRT test to test for regression. |
| 17-08-03 |
Elb |
reg_test |
diff
|
ELO: 1.11 +-2.8 (95%) LOS: 77.8%
Total: 20000 W: 3532 L: 3468 D: 13000
|
20000
@ 10+0.1 th 1 |
Do a quick regression test of the last 2 commits. Some reports suggest a regression: https://nextchessmove.com/dev-builds |
| 17-07-18 |
Elb |
sct |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 16468 W: 2939 L: 2994 D: 10535
|
sprt
@ 10+0.1 th 1 |
Endgame scaling: use tuned values |
| 17-07-17 |
Elb |
tune_scaling |
diff
|
29302/30000 iterations
59819/60000 games played
|
60000
@ 20+0.2 th 1 |
Tune endgame scaling by type and number of pawns |
| 17-07-14 |
Elb |
oneflank |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 61888 W: 11186 L: 11049 D: 39653
|
sprt
@ 10+0.1 th 1 |
Take 3: don't use scale factor for endgames with little material |
| 17-07-12 |
Elb |
oneflank |
diff
|
LLR: -3.81 (-2.94,2.94) [0.00,5.00]
Total: 66102 W: 11943 L: 11824 D: 42335
|
sprt
@ 10+0.1 th 1 |
Take 2: scale up or down depending on the type of endgame |
| 17-07-12 |
Elb |
oneflank |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 25883 W: 4733 L: 4747 D: 16403
|
sprt
@ 10+0.1 th 1 |
More drawish if all pawns of strong side on one flank |
| 17-06-30 |
Elb |
seep |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 133022 W: 24108 L: 23812 D: 85102
|
sprt
@ 10+0.1 th 1 |
Negative see value pruning tweak. Don't just exclude promotions but all advanced pawn pushes. |
| 17-06-29 |
Elb |
nmptr |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 8180 W: 1449 L: 1540 D: 5191
|
sprt
@ 10+0.1 th 1 |
NMP tweak |
| 17-06-28 |
Elb |
clmr |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 7469 W: 1290 L: 1384 D: 4795
|
sprt
@ 10+0.1 th 1 |
Take 2 |
| 17-06-28 |
Elb |
clmr |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 9487 W: 1613 L: 1698 D: 6176
|
sprt
@ 10+0.1 th 1 |
Less reduction in endgames |
| 17-06-28 |
Elb |
cext |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 14920 W: 2664 L: 2726 D: 9530
|
sprt
@ 10+0.1 th 1 |
No gives check extension for cut nodes |
| 17-06-27 |
Elb |
trappedr |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 25022 W: 4480 L: 4499 D: 16043
|
sprt
@ 10+0.1 th 1 |
Take 2: expand idea to more pieces |
| 17-06-27 |
Elb |
trappedr |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 10182 W: 1740 L: 1822 D: 6620
|
sprt
@ 10+0.1 th 1 |
Take 3: fixed version (put prio of take 2 to -1) |
| 17-06-26 |
Elb |
trappedr |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 10714 W: 1839 L: 1919 D: 6956
|
sprt
@ 10+0.1 th 1 |
Trapped rook in endgame (large penalty) |
| 17-06-22 |
Elb |
ttlmr |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 17552 W: 3095 L: 3192 D: 11265
|
sprt
@ 10+0.1 th 1 |
ttCapture lmr tweak |
| 17-06-20 |
Elb |
fvt |
diff
|
LLR: -2.96 (-2.94,2.94) [-3.00,1.00]
Total: 7704 W: 1341 L: 1512 D: 4851
|
sprt
@ 10+0.1 th 1 |
Remove futilityValue pruning |
| 17-06-20 |
Elb |
nmprep |
diff
|
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 7473 W: 1288 L: 1381 D: 4804
|
sprt
@ 10+0.1 th 1 |
A rather wild idea based on https://groups.google.com/forum/?fromgroups=#!topic/fishcooking/gWEPhdgVIhY |
| 17-06-16 |
Elb |
qfp |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 32135 W: 5816 L: 5863 D: 20456
|
sprt
@ 10+0.1 th 1 |
Take 3: change ordering of pruning. Probably slower, but maybe more precise. |
| 17-06-16 |
Elb |
qfp |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 12720 W: 2307 L: 2378 D: 8035
|
sprt
@ 10+0.1 th 1 |
Take 2 |
| 17-06-16 |
Elb |
qfp |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 29811 W: 5402 L: 5400 D: 19009
|
sprt
@ 10+0.1 th 1 |
Futility pruning tweak |
| 17-06-15 |
Elb |
isolated |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7907 W: 1407 L: 1499 D: 5001
|
sprt
@ 10+0.1 th 1 |
Do the opposite: count isolated pawns double |
| 17-06-15 |
Elb |
isolated |
diff
|
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 8185 W: 1452 L: 1542 D: 5191
|
sprt
@ 10+0.1 th 1 |
Don't count isolated pawns in the pawnsOnSquares, thereby reducing the bishop penalty. |
| 17-06-15 |
Elb |
isolated |
diff
|
LLR: -0.25 (-2.94,2.94) [0.00,5.00]
Total: 800 W: 146 L: 153 D: 501
|
sprt
@ 10+0.1 th 1 |
Isolated penalty tweak |
| 17-06-13 |
Elb |
nmp_mvc |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 23614 W: 3072 L: 3113 D: 17429
|
sprt
@ 60+0.6 th 1 |
Take 2 (LTC) |
| 17-06-13 |
Elb |
nmp_mvc |
diff
|
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 6164 W: 1174 L: 1023 D: 3967
|
sprt
@ 10+0.1 th 1 |
Take 2 |
| 17-06-13 |
Elb |
nmp_mvc |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 3682 W: 648 L: 759 D: 2275
|
sprt
@ 10+0.1 th 1 |
stc |
| 17-06-12 |
Elb |
king_end |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 32284 W: 5842 L: 5830 D: 20612
|
sprt
@ 10+0.1 th 1 |
Final try: (0,70) penalty |
| 17-06-11 |
Elb |
king_end |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 28565 W: 5267 L: 5269 D: 18029
|
sprt
@ 10+0.1 th 1 |
Use penalty (0,100) |
| 17-06-10 |
Elb |
king_end |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 68715 W: 12586 L: 12417 D: 43712
|
sprt
@ 10+0.1 th 1 |
Penalty for king immobility in endgame (increased penalty) |
| 17-06-09 |
Elb |
king_end |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 20648 W: 3709 L: 3746 D: 13193
|
sprt
@ 10+0.1 th 1 |
Penalty for king immobility in endgame |
| 17-06-09 |
Elb |
king_end |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 12077 W: 2192 L: 2266 D: 7619
|
sprt
@ 10+0.1 th 1 |
Take 3: give penalty if king is on edge of the board and can't move towards center |
| 17-06-09 |
Elb |
king_end |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 15166 W: 2777 L: 2837 D: 9552
|
sprt
@ 10+0.1 th 1 |
Take 2: Penalty when our king is on the first rank in endgame and can't move to 2nd rank. |
| 17-06-09 |
Elb |
king_end |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 9512 W: 1709 L: 1794 D: 6009
|
sprt
@ 10+0.1 th 1 |
Penalty for king on first rank in endgame |
| 17-06-08 |
Elb |
tempo |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 32990 W: 5946 L: 5990 D: 21054
|
sprt
@ 10+0.1 th 1 |
Take 2 |
| 17-06-08 |
Elb |
tempo |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 16534 W: 2992 L: 3092 D: 10450
|
sprt
@ 10+0.1 th 1 |
Tempo value tweak |
| 17-06-06 |
Elb |
kde |
diff
|
LLR: -3.01 (-2.94,2.94) [0.00,5.00]
Total: 31211 W: 5677 L: 5671 D: 19863
|
sprt
@ 10+0.1 th 1 |
Take 2 |
| 17-06-02 |
Elb |
kde |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 51352 W: 9376 L: 9282 D: 32694
|
sprt
@ 10+0.1 th 1 |
King danger in endgame tweak |
| 17-06-01 |
Elb |
closedness |
diff
|
LLR: -3.15 (-2.94,2.94) [0.00,5.00]
Total: 13447 W: 2413 L: 2489 D: 8545
|
sprt
@ 10+0.1 th 1 |
Bonus for a knight when there are many immobile pawns (closed position) |
| 17-06-01 |
Elb |
closedness |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7671 W: 1332 L: 1425 D: 4914
|
sprt
@ 10+0.1 th 1 |
Take 2: give penalty to bishops in case of immobile pawns, instead of pawns on same color squares. |
| 17-05-31 |
Elb |
swt |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,4.00]
Total: 8279 W: 1451 L: 1580 D: 5248
|
sprt
@ 10+0.1 th 1 |
ShelterWeakness tweak |
| 17-05-29 |
Elb |
kft2 |
diff
|
LLR: -3.20 (-2.94,2.94) [0.00,5.00]
Total: 73052 W: 13318 L: 13142 D: 46592
|
sprt
@ 10+0.1 th 1 |
Penalty when king on semiopen file (by rank) |
| 17-05-30 |
Elb |
kft |
diff
|
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 18052 W: 2302 L: 2362 D: 13388
|
sprt
@ 60+0.6 th 1 |
Try this strong yellow on LTC (low throughput) |
| 17-05-29 |
Elb |
tune_kft |
diff
|
24237/30000 iterations
49952/60000 games played
|
60000
@ 20+0.2 th 1 |
Tune penalty when king on semiopen file (by rank) |
| 17-05-24 |
Elb |
kft |
diff
|
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 140739 W: 25726 L: 25251 D: 89762
|
sprt
@ 10+0.1 th 1 |
Penalty when king on semiopen file and not on back rank |
| 17-05-24 |
Elb |
kft |
diff
|
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 40026 W: 7356 L: 7309 D: 25361
|
sprt
@ 10+0.1 th 1 |
Penalty when king on semiopen file |