Finished
StyxdoRecklesss1_wdl_lineardiffN=25000
LLR: -2.26 (-2.25, 2.89) [0.00, 4.00]
Games: 19158 W: 6351 L: 6413 D: 6394
Ptnml(0-2): 726, 2123, 3902, 2143, 685
Fixed Node: Linear WDL in S1, increasing a little bit since v27 over the Stage.
WorldRecklessfine-tuning-take2-s3diffN=25000
LLR: -2.41 (-2.25, 2.89) [0.00, 4.00]
Games: 6418 W: 2058 L: 2181 D: 2179
Ptnml(0-2): 242, 743, 1338, 668, 218
take 2
WorldRecklessfine-tuning-s3diffN=25000
LLR: -2.97 (-2.94, 2.94) [0.00, 4.00]
Games: 1408 W: 416 L: 604 D: 388
Ptnml(0-2): 91, 178, 299, 100, 36
fine tuning in S3
PeregrRecklesscut-in-checkdiff8.0+0.08
LLR: -2.29 (-2.25, 2.89) [0.00, 4.00]
Games: 22682 W: 5512 L: 5547 D: 11623
Ptnml(0-2): 75, 2721, 5774, 2706, 65
accept the TT cutoff for in check positions instead of letting qsearch noisy only evasions and no ttmove decide the fate
WorldRecklesscontcorrhistdiff8.0+0.08
LLR: -2.27 (-2.25, 2.89) [0.00, 4.00]
Games: 9006 W: 2196 L: 2270 D: 4540
Ptnml(0-2): 25, 1106, 2317, 1028, 27
WorldRecklessstep-wdl-s2diff8.0+0.08
LLR: -0.16 (-2.25, 2.89) [0.00, 4.00]
Games: 10898 W: 2742 L: 2717 D: 5439
Ptnml(0-2): 31, 1292, 2780, 1313, 33
start 0.65, gamma 1.12, step 47
WorldRecklessstep-wdl-s2diffN=25000
LLR: -2.67 (-2.94, 2.94) [0.00, 4.00]
Games: 38486 W: 12509 L: 12516 D: 13461
Ptnml(0-2): 1240, 4391, 8056, 4248, 1308
start 0.65, gamma 1.12, step 47
ShawnRecklesstempo_nnuediff8.0+0.08
LLR: -2.27 (-2.25, 2.89) [0.00, 4.00]
Games: 23818 W: 5891 L: 5922 D: 12005
Ptnml(0-2): 64, 2937, 5957, 2868, 83
nnue threats tempo
ShawnRecklessfds_rdiff8.0+0.08
LLR: -2.26 (-2.25, 2.89) [0.00, 4.00]
Games: 10988 W: 2661 L: 2729 D: 5598
Ptnml(0-2): 35, 1329, 2835, 1259, 36
fds r
PeregrRecklessrm-insuff-drawdiff8.0+0.08
LLR: 2.90 (-2.25, 2.89) [-3.00, 0.00]
Games: 33004 W: 8108 L: 8015 D: 16881
Ptnml(0-2): 90, 3431, 9364, 3530, 87
Remove useless insufficient material detection
WorldReckless[Integral] maindiff40.0+0.40
Elo: -7.62 +- 4.83 (95%) [N=40000]
Games: 5472 W: 1361 L: 1481 D: 2630
Ptnml(0-2): 5, 767, 1338, 595, 31
[LTC] Reckless vs. Integral
StyxdoRecklessadv_02_s1diff8.0+0.08
LLR: 3.00 (-2.25, 2.89) [0.00, 4.00]
Games: 16594 W: 4266 L: 4081 D: 8247
Ptnml(0-2): 57, 1920, 4162, 2097, 61
STC: Replace 1536 Data in S1 with Adv-01+Adv-02.
StyxdoRecklessadv_02_s1diffN=25000
LLR: -0.90 (-2.25, 2.89) [0.00, 4.00]
Games: 17708 W: 5916 L: 5899 D: 5893
Ptnml(0-2): 664, 1904, 3675, 1973, 638
Fixed Node: Replace 1536 Data in S1 with Adv-01+Adv-02.
WorldReckless[Integral] maindiff8.0+0.08
Elo: 8.77 +- 14.06 (95%) [N=40000]
Games: 832 W: 256 L: 235 D: 341
Ptnml(0-2): 4, 103, 198, 90, 21
[STC] Reckless vs. Integral
WorldRecklessv27-ae50420ddiff8.0+0.08
LLR: 2.95 (-2.25, 2.89) [0.00, 4.00]
Games: 4326 W: 1149 L: 1001 D: 2176
Ptnml(0-2): 10, 475, 1060, 593, 25
WorldRecklessv27-ae50420ddiffN=25000
LLR: 2.96 (-2.25, 2.89) [0.00, 4.00]
Games: 6226 W: 2196 L: 1985 D: 2045
Ptnml(0-2): 194, 667, 1254, 730, 268
PeregrRecklessconthist-capt-againdiff40.0+0.40
LLR: 0.46 (-2.25, 2.89) [0.00, 4.00]
Games: 25900 W: 6091 L: 5998 D: 13811
Ptnml(0-2): 20, 3009, 6794, 3112, 15
test LTC
WorldRecklesss2-no-640diff8.0+0.08
LLR: -0.82 (-2.25, 2.89) [0.00, 4.00]
Games: 23810 W: 5853 L: 5819 D: 12138
Ptnml(0-2): 81, 2843, 6030, 2863, 88
remove data-640 from S2
WorldReckless[Stockfish] openbenchdiffN=5000
Generated 20001160/20000000 Games
Elo: -74.20 +- 0.13 (95%) [N=20000000]
Games: 20001160 W: 5427646 L: 9635582 D: 4937932
[2] Adversarial 5k SN vs. Stockfish 5k
WorldRecklesss2-no-640diffN=25000
LLR: -3.24 (-2.94, 2.94) [0.00, 4.00]
Games: 32372 W: 10551 L: 10618 D: 11203
Ptnml(0-2): 1097, 3631, 6809, 3540, 1109
remove data-640 from S2
WorldRecklessscale-380diff8.0+0.08
LLR: 0.46 (-2.25, 2.89) [0.00, 4.00]
Games: 14286 W: 3600 L: 3536 D: 7150
Ptnml(0-2): 51, 1716, 3555, 1760, 61
WorldRecklessscale-380diffN=25000
LLR: -2.49 (-2.25, 2.89) [0.00, 4.00]
Games: 22774 W: 7541 L: 7602 D: 7631
Ptnml(0-2): 839, 2529, 4678, 2536, 805
StyxdoRecklessob_fixed_testdiffN=25000
LLR: -3.35 (-2.25, 2.89) [0.00, 4.00]
Games: 12256 W: 3916 L: 4070 D: 4270
Ptnml(0-2): 423, 1412, 2582, 1318, 393
OB dataset, test of before and after of fix applied by World.
StyxdoRecklessob_fixed_testdiff8.0+0.08
LLR: -2.27 (-2.25, 2.89) [0.00, 4.00]
Games: 21380 W: 5232 L: 5270 D: 10878
Ptnml(0-2): 56, 2588, 5457, 2516, 73
STC OB dataset, test of before and after of fix applied by World.
WorldRecklesswdl-0.85-s2diff8.0+0.08
LLR: -2.31 (-2.25, 2.89) [0.00, 4.00]
Games: 7522 W: 1806 L: 1887 D: 3829
Ptnml(0-2): 24, 939, 1919, 852, 27
0.75 -> 0.85