Breaking News (23.10.2023):   1st of all, let's start via Russian proverb: Vek zhivi, vek uchis = Live and learn    Yes.. under these cond. the situation is completely different than recent: Results   To be honest, I expected more from Eduard Nemeth's engine (Stockfish 220823 sl)   Same words are going for the rest.. maybe previous learned Exp.. are as reason..   Sure, other factors can be..very likely the EXP experts will explain better than Me   But there is one true, under these serious conditions: Eman is said the last word!   Really impressive points...and here I do not think that it is due to Error margin etc.   Note also this time, I picked the high-rated learning engines (exc. SF-Poly 220723)   Meanwhile, exactly with same Top players another new tour is organized: XXXX
    XXXIX Champion: Eman 9.51 - My Congratulations to Khalid Omar, SF team!
Rank Name                  Elo    +    - games score oppo. draws    1 Eman 9.51            3820    7    8   420   52%  3804   91%    2 Brainlearn 26        3810    8    8   420   51%  3806   94%    3 Tactical 041023      3808    8    8   420   50%  3806   92%    4 SpecTral 5.0 sl      3804    8    8   420   50%  3807   91%    5 SF-Poly 220723       3803    8    8   420   49%  3807   92%    6 Stockfish 220823 sl  3800    8    8   420   49%  3808   88%    7 Raid v2.73           3800    8    8   420   49%  3808   90% 
       • Calculated by Bayeselo / Raid v2.73 is fixed to 3800 Elo  / Blue Name = Last Update
More Details• All of NNUE chess engines are AVX2, tested via their default Evalfiles• All of the current engine matches are played via Unique opening suite       For more fair conditions: each of the opening line is repeated as twice• Sure where possible: all of engines are played via move overhead 100   As far as I've noticed so far (as usually): no any game to be lost on time• Current overall draw percentage (based on 1470 games) a bit high: 92%• All the played tournament games are available (in PGN) NNUE XXXIX