Summary
1. Correction-history tuning is still using the older multipliers and zero fallback.
Upstream 4e3c185 retuned the correction-history blend (including the default continuation term when no previous move is available) to improve static-eval calibration, but Revolution still multiplies the four components by the previous coefficients and skips updates when bonus == 0. That mismatch skews to_corrected_static_eval, hurting move ordering and pruning compared to Stockfish’s current baselin.
2. Null-move pruning still uses the older gate and shallow reduction.
The current code only enables null moves when a rook’s worth of material remains (or three pawns) and reduces depth by 3 + depth/4. Stockfish’s latest tuning relaxed the gate (any non-pawn material) and deepened the reduction (6 + depth/3) to cut branches more aggressively while keeping verification safeguards. Keeping the old thresholds leaves Revolution exploring many branches Stockfish now prunes
Root aspiration / low-ply history remain on the pre-patch defaults.
Revolution still zeroes rootDepth, seeds lowPlyHistory with 89, and uses the earlier aspiration re-search maths. The Stockfish patch raises the history seed, tracks best-move flips, and re-centers the aspiration window to reduce fail-high oscillations—important for keeping pace against stronger opponents.
Elo: 8.37 +/- 15.06, nElo: 16.29 +/- 29.30 LOS: 86.20 %, DrawRatio: 52.59 %, PairsRatio: 1.25 Games: 540, Wins: 137, Losses: 124, Draws: 279, Points: 276.5 (51.20 %) Ptnml(0-2): [3, 54, 142, 69, 2], WL/DD Ratio: 0.82 LLR: 0.50 (17.0%) (-2.94, 2.94) [0.00, 10.00]Elo gain
+8-+15 elo gain versus baseline version.
Download revolution 2.80 300925
Jorge Ruiz
connoisseur of both chess and anthropology, a combination that reflects his deep intellectual curiosity and passion for understanding both the art of strategic chess books