SPCC: Testrun of Stockfish 12 avx2 finished

[pohl4711]

Testrun of Stockfish 12 avx2 finished

https://www.sp-cc.de

(Perhaps you have to clear your browsercache or reload the website)

[pohl4711]

Here the scores of Stockfish 12 and 11 vs. the same opponents:

Code: Select all

Stockfish 12 200902 

Stockfish 11 200118       : 1000 (+416,=554,- 30), 69.3 %
Slow Chess 2.2 popc       : 1000 (+776,=222,-  2), 88.7 %
Ethereal 12.25 pext       : 1000 (+756,=242,-  2), 87.7 %
Komodo 14 bmi2            : 1000 (+640,=355,-  5), 81.8 %
Xiphos 0.6 bmi2           : 1000 (+768,=227,-  5), 88.2 %
Fire 7.1 popc             : 1000 (+800,=198,-  2), 89.9 %
Houdini 6 pext            : 1000 (+667,=322,- 11), 82.8 %

Code: Select all

Stockfish 11 200118 

Stockfish 12 200902       : 1000 (+ 30,=554,-416), 30.7 %
Slow Chess 2.2 popc       : 1000 (+635,=348,- 17), 80.9 %
Ethereal 12.25 pext       : 1000 (+575,=409,- 16), 78.0 %
Komodo 14 bmi2            : 1000 (+365,=584,- 51), 65.7 %
Xiphos 0.6 bmi2           : 1000 (+639,=356,-  5), 81.7 %
Fire 7.1 popc             : 1000 (+625,=368,-  7), 80.9 %
Houdini 6 pext            : 1000 (+450,=509,- 41), 70.5 %

Because I use an AMD Ryzen CPU for testing since some weeks, all engine-binaries, Stockfish 12 played against, are popcount/avx2, of course, because bmi2/pext-compiles are extremly slow on AMD. To keep the rating-list engine-names consistent, the "bmi2"- or "pext"-extension in the engine-name is still in use for older engines (Houdini 6, Komodo 14, Ethereal 12.25 and Xiphos 0.6) - otherwise ORDO will not calculate all played games by this engine as one engine...