Alexandria 9.0 surprises in test suites

[Jouni]

I have not tested previous versions. But version 9 scores better than Reckless 0.90 or SF 18 in test positions. And much better in my mate suite.

[peter]

I have not tested previous versions. But version 9 scores better than Reckless 0.90 or SF 18 in test positions.

Yes, not bad, even if performance in anti engine puzzles misses MultiPV- mode. Here compared to other MultiPV=1- runs (6 threads, 30"/pos.):

Code: Select all

    Program                                    Elo   +/-  Matches  Score   Av.Op.   S.Pos.   MST1    MST2   RIndex
  
 29 Lc0v0.32.1-6147500PT-MuPV1               : 3502    3  12411    50.2 %   3500   224/325    3.8s   11.9s   0.57
 
 33 Lc0v0.33.0-BT5-3700M-MuPV1               : 3485    3  12264    47.8 %   3501   213/325    4.6s   13.3s   0.52
 34 Reckless0.9.0.3-6t-MuPV1                 : 3482    3  12065    47.3 %   3501   213/325    4.8s   13.5s   0.49
 
 36 Stockfish18-6t-MuPV1                     : 3470    3  11739    45.4 %   3501   204/325    5.4s   14.5s   0.49
 
 39 Stockfish-260101-6t-MuPV1                : 3467    3  11702    45.1 %   3502   201/325    5.4s   14.8s   0.50
 40 Alexandria-9.0.0-6t-MuPV1                : 3452    3  11741    42.8 %   3502   185/325    5.3s   15.9s   0.42
 41 Stockfish17.1-6t-MuPV1                   : 3450    3  11526    42.6 %   3502   186/325    5.9s   16.2s   0.45
 42 PlentyChess7.0.37-6t-MuPV1               : 3446    3  11603    41.9 %   3503   174/325    5.0s   16.6s   0.44

MST1  : Mean solution time (solved positions only)
MST2  : Mean solution time (solved and unsolved positions)
RIndex: Score according to solution time ranking for each position

BTW in the 325 from here

viewtopic.php?p=988828#p988828

these 2 positions have to be corrected as for their solutions, nr. 159:
2kr3r/qp1n1p2/p2p2p1/N2Pn2p/1P3Q1P/7R/P1P1B1P1/2KR4 w - - bm Ra3 Rb3; hmvc 0; c0 "Ra3=5, Rb3=5"; id "ENET-2026 070";
and nr.207:
r1bqkb1r/ppp3pp/n3pn2/1N1p1p2/3P1B2/4P3/PPP2PPP/R2QKBNR w KQkq - bm Nf3; c0 "Nf3=5"; id "Hiarcs_Buch";

In nr.159 Rb3 has to be added as second one solution and in nr. 207 Nf3 has to replace c4, regards