VLTC tuning

Tuning some parameters that scale well with longer time control:

Failed STC:
https://tests.stockfishchess.org/tests/view/6313424d8202a039920e130a
LLR: -2.94 (-2.94,2.94) <-1.75,0.25>
Total: 42680 W: 11231 L: 11540 D: 19909
Ptnml(0-2): 191, 5008, 11232, 4737, 172

Passed LTC:
https://tests.stockfishchess.org/tests/view/6311e2cd874169ca52ae7933
LLR: 2.94 (-2.94,2.94) <0.50,2.50>
Total: 53448 W: 14782 L: 14437 D: 24229
Ptnml(0-2): 101, 5214, 15740, 5577, 92

Passed VLTC:
https://tests.stockfishchess.org/tests/view/6312530cfa99a92e3002c927
LLR: 2.95 (-2.94,2.94) <0.50,2.50>
Total: 123336 W: 33465 L: 33007 D: 56864
Ptnml(0-2): 38, 11466, 38204, 11920, 40

closes #4154

Bench: 5609606

src/search.cpp

@@ -63,7 +63,7 @@ namespace {
 
   // Futility margin
   Value futility_margin(Depth d, bool improving) {
-    return Value(168 * (d - improving));
+    return Value(174 * (d - improving));
   }
 
   // Reductions lookup table, initialized at startup
@@ -362,7 +362,7 @@ void Thread::search() {
               trend = (us == WHITE ?  make_score(tr, tr / 2)
                                    : -make_score(tr, tr / 2));
 
-              int opt = sigmoid(prev, 8, 17, 144, 13966, 183);
+              int opt = sigmoid(prev, 8, 17, 144, 15012, 183);
               optimism[ us] = Value(opt);
               optimism[~us] = -optimism[us];
           }
@@ -778,7 +778,7 @@ namespace {
     // return a fail low.
     if (   !PvNode
         && depth <= 7
-        && eval < alpha - 348 - 258 * depth * depth)
+        && eval < alpha - 341 - 267 * depth * depth)
     {
         value = qsearch<NonPV>(pos, ss, alpha - 1, alpha);
         if (value < alpha)
@@ -797,7 +797,7 @@ namespace {
     // Step 9. Null move search with verification search (~22 Elo)
     if (   !PvNode
         && (ss-1)->currentMove != MOVE_NULL
-        && (ss-1)->statScore < 14695
+        && (ss-1)->statScore < 15344
         &&  eval >= beta
         &&  eval >= ss->staticEval
         &&  ss->staticEval >= beta - 15 * depth - improvement / 15 + 201 + complexity / 24
@@ -808,7 +808,7 @@ namespace {
         assert(eval - beta >= 0);
 
         // Null move dynamic reduction based on depth, eval and complexity of position
-        Depth R = std::min(int(eval - beta) / 147, 5) + depth / 3 + 4 - (complexity > 650);
+        Depth R = std::min(int(eval - beta) / 152, 5) + depth / 3 + 4 - (complexity > 650);
 
         ss->currentMove = MOVE_NULL;
         ss->continuationHistory = &thisThread->continuationHistory[0][0][NO_PIECE][0];
@@ -844,7 +844,7 @@ namespace {
         }
     }
 
-    probCutBeta = beta + 179 - 46 * improving;
+    probCutBeta = beta + 173 - 46 * improving;
 
     // Step 10. ProbCut (~4 Elo)
     // If we have a good enough capture and a reduced search returns a value
@@ -906,9 +906,9 @@ namespace {
         return qsearch<PV>(pos, ss, alpha, beta);
 
     if (    cutNode
-        &&  depth >= 8
+        &&  depth >= 9
         && !ttMove)
-        depth--;
+        depth -= 2;
 
 moves_loop: // When in check, search starts here
 
@@ -1009,7 +1009,7 @@ namespace {
                   && !PvNode
                   && lmrDepth < 6
                   && !ss->inCheck
-                  && ss->staticEval + 281 + 179 * lmrDepth + PieceValue[EG][pos.piece_on(to_sq(move))]
+                  && ss->staticEval + 277 + 187 * lmrDepth + PieceValue[EG][pos.piece_on(to_sq(move))]
                    + captureHistory[movedPiece][to_sq(move)][type_of(pos.piece_on(to_sq(move)))] / 6 < alpha)
                   continue;
 
@@ -1173,7 +1173,7 @@ namespace {
                          + (*contHist[0])[movedPiece][to_sq(move)]
                          + (*contHist[1])[movedPiece][to_sq(move)]
                          + (*contHist[3])[movedPiece][to_sq(move)]
-                         - 4334;
+                         - 4560;
 
           // Decrease/increase reduction for moves with a good/bad history (~30 Elo)
           r -= ss->statScore / 15914;
@@ -1188,7 +1188,7 @@ namespace {
           // Do full depth search when reduced LMR search fails high
           if (value > alpha && d < newDepth)
           {
-              const bool doDeeperSearch = value > (alpha + 78 + 11 * (newDepth - d));
+              const bool doDeeperSearch = value > (alpha + 73 + 12 * (newDepth - d));
               value = -search<NonPV>(pos, ss+1, -(alpha+1), -alpha, newDepth + doDeeperSearch, !cutNode);
 
               int bonus = value > alpha ?  stat_bonus(newDepth)
@@ -1337,14 +1337,14 @@ namespace {
                          quietsSearched, quietCount, capturesSearched, captureCount, depth);
 
     // Bonus for prior countermove that caused the fail low
-    else if (   (depth >= 4 || PvNode)
+    else if (   (depth >= 5 || PvNode)
              && !priorCapture)
     {
         //Assign extra bonus if current node is PvNode or cutNode
         //or fail low was really bad
         bool extraBonus =    PvNode
                           || cutNode
-                          || bestValue < alpha - 70 * depth;
+                          || bestValue < alpha - 66 * depth;
 
         update_continuation_histories(ss-1, pos.piece_on(prevSq), prevSq, stat_bonus(depth) * (1 + extraBonus));
     }