Stockfish Testing Queue

Finished - 31839 tests

18-07-19 noo biased_ttdraw5 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 16490 W: 3604 L: 3654 D: 9232
sprt @ 10+0.1 th 1 Biased draw score from TT Take 5
18-07-19 pro ps_statupdate diff
LLR: -1.84 (-2.94,2.94) [-3.00,1.00]
Total: 10246 W: 1684 L: 1797 D: 6765
sprt @ 60+0.6 th 1 LTC: We suspect this will struggle. I will end early if so. depth cap 15, value = 46*d*d.
18-07-19 SC playOut diff
LLR: 2.94 (-2.94,2.94) [0.00,5.00]
Total: 14947 W: 3430 L: 3224 D: 8293
sprt @ 10+0.1 th 1 Restore PV bounds.
18-07-19 IIv eval diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 15022 W: 3259 L: 3364 D: 8399
sprt @ 10+0.1 th 1 CloseEnemies #1
18-07-19 IIv eval diff
LLR: -2.95 (-2.94,2.94) [0.00,4.00]
Total: 12327 W: 2703 L: 2819 D: 6805
sprt @ 10+0.1 th 1 CloseEnemies #2
18-07-19 noo biased_ttdraw4 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8576 W: 1884 L: 1973 D: 4719
sprt @ 10+0.1 th 1 Biased draw score from TT Take 4
18-07-19 SC playOut diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 8525 W: 1799 L: 1888 D: 4838
sprt @ 10+0.1 th 1 Clean up mess.
18-07-19 SC playOut2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 10188 W: 2259 L: 2340 D: 5589
sprt @ 10+0.1 th 1 Guard against messing up PV.
18-07-19 noo biased_ttdraw2 diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 25563 W: 5718 L: 5722 D: 14123
sprt @ 10+0.1 th 1 Biased draw score from TT Take 2
18-07-19 noo biased_ttdraw diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 5803 W: 1262 L: 1365 D: 3176
sprt @ 10+0.1 th 1 Biased draw score from TT First experiment to see if I translate VALUE_DRAW from TT into a biased score scaled down by depth.
18-07-19 SC playOut diff
LLR: -1.58 (-2.94,2.94) [0.00,5.00]
Total: 10572 W: 2362 L: 2380 D: 5830
sprt @ 10+0.1 th 1 Stop when safe value is reached.
18-07-19 sg stat_bonus_root_depth_n diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 10847 W: 2380 L: 2458 D: 6009
sprt @ 10+0.1 th 1 Decrease break even: -2.
18-07-19 sg stat_bonus_root_depth_n diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 42506 W: 9319 L: 9242 D: 23945
sprt @ 10+0.1 th 1 Decrease break even: -1.
18-07-19 sg stat_bonus_root_depth_n diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 37568 W: 8284 L: 8230 D: 21054
sprt @ 10+0.1 th 1 Increase break even: +1.
18-07-19 SC playOut diff
LLR: -0.13 (-2.94,2.94) [0.00,5.00]
Total: 10836 W: 2415 L: 2367 D: 6054
sprt @ 10+0.1 th 1 Check for move list size.
18-07-19 pro ps_statupdate diff
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 34162 W: 7605 L: 7508 D: 19049
sprt @ 10+0.1 th 1 depth cap 15, value = 46*d*d.
18-07-19 sg evasion_cmh_new3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 26463 W: 5849 L: 5850 D: 14764
sprt @ 10+0.1 th 1 Use contHistory[1] and constHistory[3] with factor 5/8.
18-07-19 noo master diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 28079 W: 6187 L: 6180 D: 15712
sprt @ 10+0.1 th 1 Stability & consistency check of my RMAed Ryzens...
18-07-19 sg evasion_cmh_new3 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 43904 W: 9760 L: 9674 D: 24470
sprt @ 10+0.1 th 1 Combine yellow attempts and use contHistory[1] and constHistory[3] with factor 1/2.
18-07-19 sg evasion_cmh_new3 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 39190 W: 8684 L: 8622 D: 21884
sprt @ 10+0.1 th 1 Combine yellow attempts and use contHistory[1] and constHistory[3] with factor 3/8.
18-07-19 sg evasion_cmh_new diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 64611 W: 14344 L: 14156 D: 36111
sprt @ 10+0.1 th 1 Factor 1/2 failed yellow so try now contHistory[1] * 3 / 8. Test against my passed evasion_cmh patch.
18-07-19 sg evasion_cmh_new2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 35295 W: 7754 L: 7712 D: 19829
sprt @ 10+0.1 th 1 Factor 1/2 failed yellow so try now contHistory[3] * 3 / 8. Test against my passed evasion_cmh patch.
18-07-19 xor noConnect diff
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 27734 W: 6174 L: 6064 D: 15496
sprt @ 10+0.1 th 1 Remove connectivity.
18-07-19 sg evasion_cmh_new2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 4531 W: 959 L: 1069 D: 2503
sprt @ 10+0.1 th 1 Fix bench. Factor 1/2 failed yellow so try now contHistory[3] / 3. Test against my passed evasion_cmh patch.
18-07-19 sg evasion_cmh_new2^ diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 9544 W: 2063 L: 2148 D: 5333
sprt @ 10+0.1 th 1 Factor 1/2 failed yellow so try now contHistory[3] * 3 / 4. Test against my passed evasion_cmh patch.
18-07-19 sg evasion_cmh_new diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 20162 W: 4435 L: 4467 D: 11260
sprt @ 10+0.1 th 1 Factor 1/2 failed yellow so try now contHistory[1] * 3 / 4. Test against my passed evasion_cmh patch.
18-07-19 sg evasion_cmh_new^ diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 12024 W: 2656 L: 2728 D: 6640
sprt @ 10+0.1 th 1 Factor 1/2 failed yellow so try now contHistory[1] / 3. Test against my passed evasion_cmh patch.
18-07-19 sg evasion_cmh_new2^ diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 44043 W: 9821 L: 9734 D: 24488
sprt @ 10+0.1 th 1 Use also contHistory[3] / 2. Test against my passed evasion_cmh patch.
18-07-19 xor connectivity2 diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 35652 W: 7847 L: 7803 D: 20002
sprt @ 10+0.1 th 1 Attempt 3, only when attacked.
18-07-19 pb0 hanging1 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 4978 W: 1059 L: 1166 D: 2753
sprt @ 10+0.1 th 1 Idea by Brian: count as hanging pieces attacked by more than one, and defended only by the king.
18-07-19 xor connectivity2^ diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 7648 W: 1632 L: 1726 D: 4290
sprt @ 10+0.1 th 1 With adjustment to bonus.
18-07-19 xor connectivity2 diff
LLR: -0.62 (-2.94,2.94) [0.00,5.00]
Total: 3366 W: 730 L: 741 D: 1895
sprt @ 10+0.1 th 1 Attempt 3, only when attacked.
18-07-19 sg evasion_cmh_new2 diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7921 W: 1667 L: 1759 D: 4495
sprt @ 10+0.1 th 1 Use also contHistory[3] / 4. Test against my passed evasion_cmh patch.
18-07-19 sg evasion_cmh_new2^^ diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 7763 W: 1662 L: 1755 D: 4346
sprt @ 10+0.1 th 1 Use also contHistory[3]. Test against my passed evasion_cmh patch.
18-07-19 sg evasion_cmh_new^ diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 45569 W: 10238 L: 10143 D: 25188
sprt @ 10+0.1 th 1 Use also contHistory[1] / 2. Test against my passed evasion_cmh patch.
18-07-19 sg evasion_cmh_new diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 22620 W: 5006 L: 5026 D: 12588
sprt @ 10+0.1 th 1 Use also contHistory[1] / 4. Test against my passed evasion_cmh patch.
18-07-19 sg evasion_cmh_new^^ diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 15722 W: 3466 L: 3520 D: 8736
sprt @ 10+0.1 th 1 Use also contHistory[1]. Test against my passed evasion_cmh patch.
18-07-19 pb0 pinOnBackRank diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 23830 W: 5301 L: 5314 D: 13215
sprt @ 10+0.1 th 1 Penalty for minor piece pinned at back rank inspired by Brians analisys of this game https://groups.google.com/d/msg/lczero/7Ffntxej6gc/eE3PDlYXAgAJ
18-07-19 pb0 pinOnBackRank diff
LLR: -2.95 (-2.94,2.94) [0.00,5.00]
Total: 22618 W: 5056 L: 5075 D: 12487
sprt @ 10+0.1 th 1 Take 2: only endgame penalty
18-07-19 sg lmr_checks_new diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 27777 W: 6110 L: 6105 D: 15562
sprt @ 10+0.1 th 1 My first lmr_checks test failed yellow with over 100K games. Retry it now against my passed evasion_cmh patch because this effects this for sure.
18-07-19 Viz connectivity2^ diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 12566 W: 2106 L: 2179 D: 8281
sprt @ 60+0.6 th 1 LTC for xoroshiro since framework is completely empty I tried this idea before, but want to see how it does (on mostly empty framework) after recent simplification to Overload.
18-07-19 pro ps_statupdate diff
LLR: -2.96 (-2.94,2.94) [-3.00,1.00]
Total: 65708 W: 14452 L: 14748 D: 36508
sprt @ 10+0.1 th 1 closer to master.
18-07-19 pb0 afterNull diff
LLR: -2.97 (-2.94,2.94) [0.00,5.00]
Total: 16612 W: 3649 L: 3699 D: 9264
sprt @ 10+0.1 th 1 No capture stat bonus if previous move was null-move
18-07-19 xor connectivity2^ diff
LLR: 2.96 (-2.94,2.94) [0.00,5.00]
Total: 24136 W: 5529 L: 5276 D: 13331
sprt @ 10+0.1 th 1 I tried this idea before, but want to see how it does (on mostly empty framework) after recent simplification to Overload.
18-07-19 xor connectivity2 diff
LLR: -2.94 (-2.94,2.94) [0.00,5.00]
Total: 20409 W: 4522 L: 4552 D: 11335
sprt @ 10+0.1 th 1 Attempt 2: another variant.
18-07-19 pro ps_statupdate diff
LLR: -1.57 (-2.94,2.94) [0.00,5.00]
Total: 268 W: 28 L: 95 D: 145
sprt @ 10+0.1 th 1 instead of setting bonus to 0 at depths over 17, just cap depth at 17.
18-07-19 pro ps_statupdate diff
LLR: -2.96 (-2.94,2.94) [-3.00,1.00]
Total: 19847 W: 4321 L: 4526 D: 11000
sprt @ 10+0.1 th 1 try to simplify stat_bonus (40 * d * d);
18-07-19 pro ps_statupdate diff
LLR: -0.15 (-2.94,2.94) [-3.00,1.00]
Total: 9248 W: 2038 L: 2065 D: 5145
sprt @ 10+0.1 th 1 test value of depth 17 limit in stat_bonus.
18-07-19 31m SoloQueenMob diff
LLR: -2.96 (-2.94,2.94) [0.00,5.00]
Total: 15806 W: 3428 L: 3482 D: 8896
sprt @ 10+0.1 th 1 Even though imbalanced queen positions are uncommon, I'm surprised that such large changes are yielding Elo effects of very nearly 0. Keep increasing the size of the changes until we see an effect. Decrease mob by 3.
18-07-19 pro ps_statupdate diff
LLR: -1.75 (-2.94,2.94) [-3.00,1.00]
Total: 82602 W: 18186 L: 18448 D: 45968
sprt @ 10+0.1 th 1 try a simplified stat update equation.