Exponential futility margins

[stockfish] / src / search.cpp

/*
  Stockfish, a UCI chess playing engine derived from Glaurung 2.1
  Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
  Copyright (C) 2008-2009 Marco Costalba

  Stockfish is free software: you can redistribute it and/or modify
  it under the terms of the GNU General Public License as published by
  the Free Software Foundation, either version 3 of the License, or
  (at your option) any later version.

  Stockfish is distributed in the hope that it will be useful,
  but WITHOUT ANY WARRANTY; without even the implied warranty of
  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  GNU General Public License for more details.

  You should have received a copy of the GNU General Public License
  along with this program.  If not, see <http://www.gnu.org/licenses/>.
*/


////
//// Includes
////

#include <cassert>
#include <cstring>
#include <fstream>
#include <iostream>
#include <sstream>

#include "book.h"
#include "evaluate.h"
#include "history.h"
#include "misc.h"
#include "movegen.h"
#include "movepick.h"
#include "lock.h"
#include "san.h"
#include "search.h"
#include "thread.h"
#include "tt.h"
#include "ucioption.h"


////
//// Local definitions
////

namespace {

  /// Types

  // IterationInfoType stores search results for each iteration
  //
  // Because we use relatively small (dynamic) aspiration window,
  // there happens many fail highs and fail lows in root. And
  // because we don't do researches in those cases, "value" stored
  // here is not necessarily exact. Instead in case of fail high/low
  // we guess what the right value might be and store our guess
  // as a "speculated value" and then move on. Speculated values are
  // used just to calculate aspiration window width, so also if are
  // not exact is not big a problem.

  struct IterationInfoType {

    IterationInfoType(Value v = Value(0), Value sv = Value(0))
    : value(v), speculatedValue(sv) {}

    Value value, speculatedValue;
  };


  // The BetaCounterType class is used to order moves at ply one.
  // Apart for the first one that has its score, following moves
  // normally have score -VALUE_INFINITE, so are ordered according
  // to the number of beta cutoffs occurred under their subtree during
  // the last iteration. The counters are per thread variables to avoid
  // concurrent accessing under SMP case.

  struct BetaCounterType {

    BetaCounterType();
    void clear();
    void add(Color us, Depth d, int threadID);
    void read(Color us, int64_t& our, int64_t& their);
  };


  // The RootMove class is used for moves at the root at the tree.  For each
  // root move, we store a score, a node count, and a PV (really a refutation
  // in the case of moves which fail low).

  struct RootMove {

    RootMove();
    bool operator<(const RootMove&); // used to sort

    Move move;
    Value score;
    int64_t nodes, cumulativeNodes;
    Move pv[PLY_MAX_PLUS_2];
    int64_t ourBeta, theirBeta;
  };


  // The RootMoveList class is essentially an array of RootMove objects, with
  // a handful of methods for accessing the data in the individual moves.

  class RootMoveList {

  public:
    RootMoveList(Position& pos, Move searchMoves[]);
    inline Move get_move(int moveNum) const;
    inline Value get_move_score(int moveNum) const;
    inline void set_move_score(int moveNum, Value score);
    inline void set_move_nodes(int moveNum, int64_t nodes);
    inline void set_beta_counters(int moveNum, int64_t our, int64_t their);
    void set_move_pv(int moveNum, const Move pv[]);
    inline Move get_move_pv(int moveNum, int i) const;
    inline int64_t get_move_cumulative_nodes(int moveNum) const;
    inline int move_count() const;
    Move scan_for_easy_move() const;
    inline void sort();
    void sort_multipv(int n);

  private:
    static const int MaxRootMoves = 500;
    RootMove moves[MaxRootMoves];
    int count;
  };


  /// Constants

  // Search depth at iteration 1
  const Depth InitialDepth = OnePly /*+ OnePly/2*/;

  // Depth limit for selective search
  const Depth SelectiveDepth = 7 * OnePly;

  // Use internal iterative deepening?
  const bool UseIIDAtPVNodes = true;
  const bool UseIIDAtNonPVNodes = false;

  // Internal iterative deepening margin. At Non-PV moves, when
  // UseIIDAtNonPVNodes is true, we do an internal iterative deepening
  // search when the static evaluation is at most IIDMargin below beta.
  const Value IIDMargin = Value(0x100);

  // Easy move margin. An easy move candidate must be at least this much
  // better than the second best move.
  const Value EasyMoveMargin = Value(0x200);

  // Problem margin. If the score of the first move at iteration N+1 has
  // dropped by more than this since iteration N, the boolean variable
  // "Problem" is set to true, which will make the program spend some extra
  // time looking for a better move.
  const Value ProblemMargin = Value(0x28);

  // No problem margin. If the boolean "Problem" is true, and a new move
  // is found at the root which is less than NoProblemMargin worse than the
  // best move from the previous iteration, Problem is set back to false.
  const Value NoProblemMargin = Value(0x14);

  // Null move margin. A null move search will not be done if the approximate
  // evaluation of the position is more than NullMoveMargin below beta.
  const Value NullMoveMargin = Value(0x300);

  // Pruning criterions. See the code and comments in ok_to_prune() to
  // understand their precise meaning.
  const bool PruneEscapeMoves    = false;
  const bool PruneDefendingMoves = false;
  const bool PruneBlockingMoves  = false;

  // Margins for futility pruning in the quiescence search, and at frontier
  // and near frontier nodes.
  const Value FutilityMarginQS = Value(0x80);

  // Each move futility margin is decreased
  const Value IncrementalFutilityMargin = Value(0xA);

  // Remaining depth:                  1 ply         1.5 ply       2 ply         2.5 ply       3 ply         3.5 ply
  const Value FutilityMargins[12] = { Value(0x100), Value(0x120), Value(0x200), Value(0x220), Value(0x250), Value(0x270),
  //                                   4 ply         4.5 ply       5 ply         5.5 ply       6 ply         6.5 ply
                                      Value(0x2A0), Value(0x2C0), Value(0x340), Value(0x360), Value(0x3A0), Value(0x3C0) };
  // Razoring
  const Depth RazorDepth = 4*OnePly;

  // Remaining depth:                 1 ply         1.5 ply       2 ply         2.5 ply       3 ply         3.5 ply
  const Value RazorMargins[6]     = { Value(0x180), Value(0x300), Value(0x300), Value(0x3C0), Value(0x3C0), Value(0x3C0) };

  // Remaining depth:                 1 ply         1.5 ply       2 ply         2.5 ply       3 ply         3.5 ply
  const Value RazorApprMargins[6] = { Value(0x520), Value(0x300), Value(0x300), Value(0x300), Value(0x300), Value(0x300) };


  /// Variables initialized by UCI options

  // Minimum number of full depth (i.e. non-reduced) moves at PV and non-PV nodes
  int LMRPVMoves, LMRNonPVMoves; // heavy SMP read access for the latter

  // Depth limit for use of dynamic threat detection
  Depth ThreatDepth; // heavy SMP read access

  // Last seconds noise filtering (LSN)
  const bool UseLSNFiltering = true;
  const int LSNTime = 4000; // In milliseconds
  const Value LSNValue = value_from_centipawns(200);
  bool loseOnTime = false;

  // Extensions. Array index 0 is used at non-PV nodes, index 1 at PV nodes.
  // There is heavy SMP read access on these arrays
  Depth CheckExtension[2], SingleReplyExtension[2], PawnPushTo7thExtension[2];
  Depth PassedPawnExtension[2], PawnEndgameExtension[2], MateThreatExtension[2];

  // Iteration counters
  int Iteration;
  BetaCounterType BetaCounter; // has per-thread internal data

  // Scores and number of times the best move changed for each iteration
  IterationInfoType IterationInfo[PLY_MAX_PLUS_2];
  int BestMoveChangesByIteration[PLY_MAX_PLUS_2];

  // MultiPV mode
  int MultiPV;

  // Time managment variables
  int SearchStartTime;
  int MaxNodes, MaxDepth;
  int MaxSearchTime, AbsoluteMaxSearchTime, ExtraSearchTime, ExactMaxTime;
  int RootMoveNumber;
  bool InfiniteSearch;
  bool PonderSearch;
  bool StopOnPonderhit;
  bool AbortSearch; // heavy SMP read access
  bool Quit;
  bool FailHigh;
  bool FailLow;
  bool Problem;

  // Show current line?
  bool ShowCurrentLine;

  // Log file
  bool UseLogFile;
  std::ofstream LogFile;

  // MP related variables
  int ActiveThreads = 1;
  Depth MinimumSplitDepth;
  int MaxThreadsPerSplitPoint;
  Thread Threads[THREAD_MAX];
  Lock MPLock;
  Lock IOLock;
  bool AllThreadsShouldExit = false;
  const int MaxActiveSplitPoints = 8;
  SplitPoint SplitPointStack[THREAD_MAX][MaxActiveSplitPoints];
  bool Idle = true;

#if !defined(_MSC_VER)
  pthread_cond_t WaitCond;
  pthread_mutex_t WaitLock;
#else
  HANDLE SitIdleEvent[THREAD_MAX];
#endif

  // Node counters, used only by thread[0] but try to keep in different
  // cache lines (64 bytes each) from the heavy SMP read accessed variables.
  int NodesSincePoll;
  int NodesBetweenPolls = 30000;

  // History table
  History H;


  /// Functions

  Value id_loop(const Position& pos, Move searchMoves[]);
  Value root_search(Position& pos, SearchStack ss[], RootMoveList& rml, Value alpha, Value beta);
  Value search_pv(Position& pos, SearchStack ss[], Value alpha, Value beta, Depth depth, int ply, int threadID);
  Value search(Position& pos, SearchStack ss[], Value beta, Depth depth, int ply, bool allowNullmove, int threadID);
  Value qsearch(Position& pos, SearchStack ss[], Value alpha, Value beta, Depth depth, int ply, int threadID);
  void sp_search(SplitPoint* sp, int threadID);
  void sp_search_pv(SplitPoint* sp, int threadID);
  void init_node(SearchStack ss[], int ply, int threadID);
  void update_pv(SearchStack ss[], int ply);
  void sp_update_pv(SearchStack* pss, SearchStack ss[], int ply);
  bool connected_moves(const Position& pos, Move m1, Move m2);
  bool value_is_mate(Value value);
  bool move_is_killer(Move m, const SearchStack& ss);
  Depth extension(const Position& pos, Move m, bool pvNode, bool capture, bool check, bool singleReply, bool mateThreat, bool* dangerous);
  bool ok_to_do_nullmove(const Position& pos);
  bool ok_to_prune(const Position& pos, Move m, Move threat, Depth d);
  bool ok_to_use_TT(const TTEntry* tte, Depth depth, Value beta, int ply);
  void update_history(const Position& pos, Move m, Depth depth, Move movesSearched[], int moveCount);
  void update_killers(Move m, SearchStack& ss);

  bool fail_high_ply_1();
  int current_search_time();
  int nps();
  void poll();
  void ponderhit();
  void print_current_line(SearchStack ss[], int ply, int threadID);
  void wait_for_stop_or_ponderhit();
  void init_ss_array(SearchStack ss[]);

  void idle_loop(int threadID, SplitPoint* waitSp);
  void init_split_point_stack();
  void destroy_split_point_stack();
  bool thread_should_stop(int threadID);
  bool thread_is_available(int slave, int master);
  bool idle_thread_exists(int master);
  bool split(const Position& pos, SearchStack* ss, int ply,
             Value *alpha, Value *beta, Value *bestValue,
             const Value futilityValue, const Value approximateValue,
             Depth depth, int *moves,
             MovePicker *mp, int master, bool pvNode);
  void wake_sleeping_threads();

#if !defined(_MSC_VER)
  void *init_thread(void *threadID);
#else
  DWORD WINAPI init_thread(LPVOID threadID);
#endif

}


////
//// Functions
////


/// perft() is our utility to verify move generation is bug free. All the
/// legal moves up to given depth are generated and counted and the sum returned.

int perft(Position& pos, Depth depth)
{
    Move move;
    int sum = 0;
    MovePicker mp = MovePicker(pos, MOVE_NONE, depth, H);

    // If we are at the last ply we don't need to do and undo
    // the moves, just to count them.
    if (depth <= OnePly) // Replace with '<' to test also qsearch
    {
        while (mp.get_next_move()) sum++;
        return sum;
    }

    // Loop through all legal moves
    CheckInfo ci(pos);
    while ((move = mp.get_next_move()) != MOVE_NONE)
    {
        StateInfo st;
        pos.do_move(move, st, ci, pos.move_is_check(move, ci));
        sum += perft(pos, depth - OnePly);
        pos.undo_move(move);
    }
    return sum;
}


/// think() is the external interface to Stockfish's search, and is called when
/// the program receives the UCI 'go' command. It initializes various
/// search-related global variables, and calls root_search(). It returns false
/// when a quit command is received during the search.

bool think(const Position& pos, bool infinite, bool ponder, int side_to_move,
           int time[], int increment[], int movesToGo, int maxDepth,
           int maxNodes, int maxTime, Move searchMoves[]) {

  // Look for a book move
  if (!infinite && !ponder && get_option_value_bool("OwnBook"))
  {
      Move bookMove;
      if (get_option_value_string("Book File") != OpeningBook.file_name())
          OpeningBook.open("book.bin");

      bookMove = OpeningBook.get_move(pos);
      if (bookMove != MOVE_NONE)
      {
          std::cout << "bestmove " << bookMove << std::endl;
          return true;
      }
  }

  // Initialize global search variables
  Idle = false;
  SearchStartTime = get_system_time();
  for (int i = 0; i < THREAD_MAX; i++)
  {
      Threads[i].nodes = 0ULL;
      Threads[i].failHighPly1 = false;
  }
  NodesSincePoll = 0;
  InfiniteSearch = infinite;
  PonderSearch = ponder;
  StopOnPonderhit = false;
  AbortSearch = false;
  Quit = false;
  FailHigh = false;
  FailLow = false;
  Problem = false;
  ExactMaxTime = maxTime;

  // Read UCI option values
  TT.set_size(get_option_value_int("Hash"));
  if (button_was_pressed("Clear Hash"))
  {
      TT.clear();
      loseOnTime = false; // reset at the beginning of a new game
  }

  bool PonderingEnabled = get_option_value_bool("Ponder");
  MultiPV = get_option_value_int("MultiPV");

  CheckExtension[1] = Depth(get_option_value_int("Check Extension (PV nodes)"));
  CheckExtension[0] = Depth(get_option_value_int("Check Extension (non-PV nodes)"));

  SingleReplyExtension[1] = Depth(get_option_value_int("Single Reply Extension (PV nodes)"));
  SingleReplyExtension[0] = Depth(get_option_value_int("Single Reply Extension (non-PV nodes)"));

  PawnPushTo7thExtension[1] = Depth(get_option_value_int("Pawn Push to 7th Extension (PV nodes)"));
  PawnPushTo7thExtension[0] = Depth(get_option_value_int("Pawn Push to 7th Extension (non-PV nodes)"));

  PassedPawnExtension[1] = Depth(get_option_value_int("Passed Pawn Extension (PV nodes)"));
  PassedPawnExtension[0] = Depth(get_option_value_int("Passed Pawn Extension (non-PV nodes)"));

  PawnEndgameExtension[1] = Depth(get_option_value_int("Pawn Endgame Extension (PV nodes)"));
  PawnEndgameExtension[0] = Depth(get_option_value_int("Pawn Endgame Extension (non-PV nodes)"));

  MateThreatExtension[1] = Depth(get_option_value_int("Mate Threat Extension (PV nodes)"));
  MateThreatExtension[0] = Depth(get_option_value_int("Mate Threat Extension (non-PV nodes)"));

  LMRPVMoves    = get_option_value_int("Full Depth Moves (PV nodes)") + 1;
  LMRNonPVMoves = get_option_value_int("Full Depth Moves (non-PV nodes)") + 1;
  ThreatDepth   = get_option_value_int("Threat Depth") * OnePly;

  Chess960 = get_option_value_bool("UCI_Chess960");
  ShowCurrentLine = get_option_value_bool("UCI_ShowCurrLine");
  UseLogFile = get_option_value_bool("Use Search Log");
  if (UseLogFile)
      LogFile.open(get_option_value_string("Search Log Filename").c_str(), std::ios::out | std::ios::app);

  MinimumSplitDepth = get_option_value_int("Minimum Split Depth") * OnePly;
  MaxThreadsPerSplitPoint = get_option_value_int("Maximum Number of Threads per Split Point");

  read_weights(pos.side_to_move());

  // Set the number of active threads
  int newActiveThreads = get_option_value_int("Threads");
  if (newActiveThreads != ActiveThreads)
  {
      ActiveThreads = newActiveThreads;
      init_eval(ActiveThreads);
  }

  // Wake up sleeping threads
  wake_sleeping_threads();

  for (int i = 1; i < ActiveThreads; i++)
      assert(thread_is_available(i, 0));

  // Set thinking time
  int myTime = time[side_to_move];
  int myIncrement = increment[side_to_move];

  if (!movesToGo) // Sudden death time control
  {
      if (myIncrement)
      {
          MaxSearchTime = myTime / 30 + myIncrement;
          AbsoluteMaxSearchTime = Max(myTime / 4, myIncrement - 100);
      } else { // Blitz game without increment
          MaxSearchTime = myTime / 30;
          AbsoluteMaxSearchTime = myTime / 8;
      }
  }
  else // (x moves) / (y minutes)
  {
      if (movesToGo == 1)
      {
          MaxSearchTime = myTime / 2;
          AbsoluteMaxSearchTime =
             (myTime > 3000)? (myTime - 500) : ((myTime * 3) / 4);
      } else {
          MaxSearchTime = myTime / Min(movesToGo, 20);
          AbsoluteMaxSearchTime = Min((4 * myTime) / movesToGo, myTime / 3);
      }
  }

  if (PonderingEnabled)
  {
      MaxSearchTime += MaxSearchTime / 4;
      MaxSearchTime = Min(MaxSearchTime, AbsoluteMaxSearchTime);
  }

  // Fixed depth or fixed number of nodes?
  MaxDepth = maxDepth;
  if (MaxDepth)
      InfiniteSearch = true; // HACK

  MaxNodes = maxNodes;
  if (MaxNodes)
  {
      NodesBetweenPolls = Min(MaxNodes, 30000);
      InfiniteSearch = true; // HACK
  }
  else if (myTime && myTime < 1000)
      NodesBetweenPolls = 1000;
  else if (myTime && myTime < 5000)
      NodesBetweenPolls = 5000;
  else
      NodesBetweenPolls = 30000;

  // Write information to search log file
  if (UseLogFile)
      LogFile << "Searching: " << pos.to_fen() << std::endl
              << "infinite: "  << infinite
              << " ponder: "   << ponder
              << " time: "     << myTime
              << " increment: " << myIncrement
              << " moves to go: " << movesToGo << std::endl;


  // We're ready to start thinking. Call the iterative deepening loop function
  //
  // FIXME we really need to cleanup all this LSN ugliness
  if (!loseOnTime)
  {
      Value v = id_loop(pos, searchMoves);
      loseOnTime = (   UseLSNFiltering
                    && myTime < LSNTime
                    && myIncrement == 0
                    && v < -LSNValue);
  }
  else
  {
      loseOnTime = false; // reset for next match
      while (SearchStartTime + myTime + 1000 > get_system_time())
          ; // wait here
      id_loop(pos, searchMoves); // to fail gracefully
  }

  if (UseLogFile)
      LogFile.close();

  Idle = true;
  return !Quit;
}


/// init_threads() is called during startup.  It launches all helper threads,
/// and initializes the split point stack and the global locks and condition
/// objects.

void init_threads() {

  volatile int i;

#if !defined(_MSC_VER)
  pthread_t pthread[1];
#endif

  for (i = 0; i < THREAD_MAX; i++)
      Threads[i].activeSplitPoints = 0;

  // Initialize global locks
  lock_init(&MPLock, NULL);
  lock_init(&IOLock, NULL);

  init_split_point_stack();

#if !defined(_MSC_VER)
  pthread_mutex_init(&WaitLock, NULL);
  pthread_cond_init(&WaitCond, NULL);
#else
  for (i = 0; i < THREAD_MAX; i++)
      SitIdleEvent[i] = CreateEvent(0, FALSE, FALSE, 0);
#endif

  // All threads except the main thread should be initialized to idle state
  for (i = 1; i < THREAD_MAX; i++)
  {
      Threads[i].stop = false;
      Threads[i].workIsWaiting = false;
      Threads[i].idle = true;
      Threads[i].running = false;
  }

  // Launch the helper threads
  for(i = 1; i < THREAD_MAX; i++)
  {
#if !defined(_MSC_VER)
      pthread_create(pthread, NULL, init_thread, (void*)(&i));
#else
      DWORD iID[1];
      CreateThread(NULL, 0, init_thread, (LPVOID)(&i), 0, iID);
#endif

      // Wait until the thread has finished launching
      while (!Threads[i].running);
  }
}


/// stop_threads() is called when the program exits.  It makes all the
/// helper threads exit cleanly.

void stop_threads() {

  ActiveThreads = THREAD_MAX;  // HACK
  Idle = false;  // HACK
  wake_sleeping_threads();
  AllThreadsShouldExit = true;
  for (int i = 1; i < THREAD_MAX; i++)
  {
      Threads[i].stop = true;
      while(Threads[i].running);
  }
  destroy_split_point_stack();
}


/// nodes_searched() returns the total number of nodes searched so far in
/// the current search.

int64_t nodes_searched() {

  int64_t result = 0ULL;
  for (int i = 0; i < ActiveThreads; i++)
      result += Threads[i].nodes;
  return result;
}


// SearchStack::init() initializes a search stack. Used at the beginning of a
// new search from the root.
void SearchStack::init(int ply) {

  pv[ply] = pv[ply + 1] = MOVE_NONE;
  currentMove = threatMove = MOVE_NONE;
  reduction = Depth(0);
}

void SearchStack::initKillers() {

  mateKiller = MOVE_NONE;
  for (int i = 0; i < KILLER_MAX; i++)
      killers[i] = MOVE_NONE;
}

namespace {

  // id_loop() is the main iterative deepening loop.  It calls root_search
  // repeatedly with increasing depth until the allocated thinking time has
  // been consumed, the user stops the search, or the maximum search depth is
  // reached.

  Value id_loop(const Position& pos, Move searchMoves[]) {

    Position p(pos);
    SearchStack ss[PLY_MAX_PLUS_2];

    // searchMoves are verified, copied, scored and sorted
    RootMoveList rml(p, searchMoves);

    // Print RootMoveList c'tor startup scoring to the standard output,
    // so that we print information also for iteration 1.
    std::cout << "info depth " << 1 << "\ninfo depth " << 1
              << " score " << value_to_string(rml.get_move_score(0))
              << " time " << current_search_time()
              << " nodes " << nodes_searched()
              << " nps " << nps()
              << " pv " << rml.get_move(0) << "\n";

    // Initialize
    TT.new_search();
    H.clear();
    init_ss_array(ss);
    IterationInfo[1] = IterationInfoType(rml.get_move_score(0), rml.get_move_score(0));
    Iteration = 1;

    Move EasyMove = rml.scan_for_easy_move();

    // Iterative deepening loop
    while (Iteration < PLY_MAX)
    {
        // Initialize iteration
        rml.sort();
        Iteration++;
        BestMoveChangesByIteration[Iteration] = 0;
        if (Iteration <= 5)
            ExtraSearchTime = 0;

        std::cout << "info depth " << Iteration << std::endl;

        // Calculate dynamic search window based on previous iterations
        Value alpha, beta;

        if (MultiPV == 1 && Iteration >= 6 && abs(IterationInfo[Iteration - 1].value) < VALUE_KNOWN_WIN)
        {
            int prevDelta1 = IterationInfo[Iteration - 1].speculatedValue - IterationInfo[Iteration - 2].speculatedValue;
            int prevDelta2 = IterationInfo[Iteration - 2].speculatedValue - IterationInfo[Iteration - 3].speculatedValue;

            int delta = Max(2 * abs(prevDelta1) + abs(prevDelta2), ProblemMargin);

            alpha = Max(IterationInfo[Iteration - 1].value - delta, -VALUE_INFINITE);
            beta  = Min(IterationInfo[Iteration - 1].value + delta,  VALUE_INFINITE);
        }
        else
        {
            alpha = - VALUE_INFINITE;
            beta  =   VALUE_INFINITE;
        }

        // Search to the current depth
        Value value = root_search(p, ss, rml, alpha, beta);

        // Write PV to transposition table, in case the relevant entries have
        // been overwritten during the search.
        TT.insert_pv(p, ss[0].pv);

        if (AbortSearch)
            break; // Value cannot be trusted. Break out immediately!

        //Save info about search result
        Value speculatedValue;
        bool fHigh = false;
        bool fLow = false;
        Value delta = value - IterationInfo[Iteration - 1].value;

        if (value >= beta)
        {
            assert(delta > 0);

            fHigh = true;
            speculatedValue = value + delta;
            BestMoveChangesByIteration[Iteration] += 2; // Allocate more time
        }
        else if (value <= alpha)
        {
            assert(value == alpha);
            assert(delta < 0);

            fLow = true;
            speculatedValue = value + delta;
            BestMoveChangesByIteration[Iteration] += 3; // Allocate more time
        } else
            speculatedValue = value;

        speculatedValue = Min(Max(speculatedValue, -VALUE_INFINITE), VALUE_INFINITE);
        IterationInfo[Iteration] = IterationInfoType(value, speculatedValue);

        // Erase the easy move if it differs from the new best move
        if (ss[0].pv[0] != EasyMove)
            EasyMove = MOVE_NONE;

        Problem = false;

        if (!InfiniteSearch)
        {
            // Time to stop?
            bool stopSearch = false;

            // Stop search early if there is only a single legal move
            if (Iteration >= 6 && rml.move_count() == 1)
                stopSearch = true;

            // Stop search early when the last two iterations returned a mate score
            if (  Iteration >= 6
                && abs(IterationInfo[Iteration].value) >= abs(VALUE_MATE) - 100
                && abs(IterationInfo[Iteration-1].value) >= abs(VALUE_MATE) - 100)
                stopSearch = true;

            // Stop search early if one move seems to be much better than the rest
            int64_t nodes = nodes_searched();
            if (   Iteration >= 8
                && !fLow
                && !fHigh
                && EasyMove == ss[0].pv[0]
                && (  (   rml.get_move_cumulative_nodes(0) > (nodes * 85) / 100
                       && current_search_time() > MaxSearchTime / 16)
                    ||(   rml.get_move_cumulative_nodes(0) > (nodes * 98) / 100
                       && current_search_time() > MaxSearchTime / 32)))
                stopSearch = true;

            // Add some extra time if the best move has changed during the last two iterations
            if (Iteration > 5 && Iteration <= 50)
                ExtraSearchTime = BestMoveChangesByIteration[Iteration]   * (MaxSearchTime / 2)
                                + BestMoveChangesByIteration[Iteration-1] * (MaxSearchTime / 3);

            // Stop search if most of MaxSearchTime is consumed at the end of the
            // iteration.  We probably don't have enough time to search the first
            // move at the next iteration anyway.
            if (current_search_time() > ((MaxSearchTime + ExtraSearchTime)*80) / 128)
                stopSearch = true;

            if (stopSearch)
            {
                //FIXME: Implement fail-low emergency measures
                if (!PonderSearch)
                    break;
                else
                    StopOnPonderhit = true;
            }
        }

        if (MaxDepth && Iteration >= MaxDepth)
            break;
    }

    rml.sort();

    // If we are pondering, we shouldn't print the best move before we
    // are told to do so
    if (PonderSearch)
        wait_for_stop_or_ponderhit();
    else
        // Print final search statistics
        std::cout << "info nodes " << nodes_searched()
                  << " nps " << nps()
                  << " time " << current_search_time()
                  << " hashfull " << TT.full() << std::endl;

    // Print the best move and the ponder move to the standard output
    if (ss[0].pv[0] == MOVE_NONE)
    {
        ss[0].pv[0] = rml.get_move(0);
        ss[0].pv[1] = MOVE_NONE;
    }
    std::cout << "bestmove " << ss[0].pv[0];
    if (ss[0].pv[1] != MOVE_NONE)
        std::cout << " ponder " << ss[0].pv[1];

    std::cout << std::endl;

    if (UseLogFile)
    {
        if (dbg_show_mean)
            dbg_print_mean(LogFile);

        if (dbg_show_hit_rate)
            dbg_print_hit_rate(LogFile);

        StateInfo st;
        LogFile << "Nodes: " << nodes_searched() << std::endl
                << "Nodes/second: " << nps() << std::endl
                << "Best move: " << move_to_san(p, ss[0].pv[0]) << std::endl;

        p.do_move(ss[0].pv[0], st);
        LogFile << "Ponder move: " << move_to_san(p, ss[0].pv[1])
                << std::endl << std::endl;
    }
    return rml.get_move_score(0);
  }


  // root_search() is the function which searches the root node.  It is
  // similar to search_pv except that it uses a different move ordering
  // scheme (perhaps we should try to use this at internal PV nodes, too?)
  // and prints some information to the standard output.

  Value root_search(Position& pos, SearchStack ss[], RootMoveList &rml, Value alpha, Value beta) {

    Value oldAlpha = alpha;
    Value value;
    CheckInfo ci(pos);

    // Loop through all the moves in the root move list
    for (int i = 0; i <  rml.move_count() && !AbortSearch; i++)
    {
        if (alpha >= beta)
        {
            // We failed high, invalidate and skip next moves, leave node-counters
            // and beta-counters as they are and quickly return, we will try to do
            // a research at the next iteration with a bigger aspiration window.
            rml.set_move_score(i, -VALUE_INFINITE);
            continue;
        }
        int64_t nodes;
        Move move;
        StateInfo st;
        Depth ext, newDepth;

        RootMoveNumber = i + 1;
        FailHigh = false;

        // Remember the node count before the move is searched. The node counts
        // are used to sort the root moves at the next iteration.
        nodes = nodes_searched();

        // Reset beta cut-off counters
        BetaCounter.clear();

        // Pick the next root move, and print the move and the move number to
        // the standard output.
        move = ss[0].currentMove = rml.get_move(i);
        if (current_search_time() >= 1000)
            std::cout << "info currmove " << move
                      << " currmovenumber " << i + 1 << std::endl;

        // Decide search depth for this move
        bool moveIsCheck = pos.move_is_check(move);
        bool captureOrPromotion = pos.move_is_capture_or_promotion(move);
        bool dangerous;
        ext = extension(pos, move, true, captureOrPromotion, moveIsCheck, false, false, &dangerous);
        newDepth = (Iteration - 2) * OnePly + ext + InitialDepth;

        // Make the move, and search it
        pos.do_move(move, st, ci, moveIsCheck);

        if (i < MultiPV)
        {
            // Aspiration window is disabled in multi-pv case
            if (MultiPV > 1)
                alpha = -VALUE_INFINITE;

            value = -search_pv(pos, ss, -beta, -alpha, newDepth, 1, 0);
            // If the value has dropped a lot compared to the last iteration,
            // set the boolean variable Problem to true. This variable is used
            // for time managment: When Problem is true, we try to complete the
            // current iteration before playing a move.
            Problem = (Iteration >= 2 && value <= IterationInfo[Iteration-1].value - ProblemMargin);

            if (Problem && StopOnPonderhit)
                StopOnPonderhit = false;
        }
        else
        {
            if (   newDepth >= 3*OnePly
                && i >= MultiPV + LMRPVMoves
                && !dangerous
                && !captureOrPromotion
                && !move_is_castle(move))
            {
                ss[0].reduction = OnePly;
                value = -search(pos, ss, -alpha, newDepth-OnePly, 1, true, 0);
            } else
                value = alpha + 1; // Just to trigger next condition

            if (value > alpha)
            {
                value = -search(pos, ss, -alpha, newDepth, 1, true, 0);
                if (value > alpha)
                {
                    // Fail high! Set the boolean variable FailHigh to true, and
                    // re-search the move with a big window. The variable FailHigh is
                    // used for time managment: We try to avoid aborting the search
                    // prematurely during a fail high research.
                    FailHigh = true;
                    value = -search_pv(pos, ss, -beta, -alpha, newDepth, 1, 0);
                }
            }
        }

        pos.undo_move(move);

        // Finished searching the move. If AbortSearch is true, the search
        // was aborted because the user interrupted the search or because we
        // ran out of time. In this case, the return value of the search cannot
        // be trusted, and we break out of the loop without updating the best
        // move and/or PV.
        if (AbortSearch)
            break;

        // Remember the node count for this move. The node counts are used to
        // sort the root moves at the next iteration.
        rml.set_move_nodes(i, nodes_searched() - nodes);

        // Remember the beta-cutoff statistics
        int64_t our, their;
        BetaCounter.read(pos.side_to_move(), our, their);
        rml.set_beta_counters(i, our, their);

        assert(value >= -VALUE_INFINITE && value <= VALUE_INFINITE);

        if (value <= alpha && i >= MultiPV)
            rml.set_move_score(i, -VALUE_INFINITE);
        else
        {
            // PV move or new best move!

            // Update PV
            rml.set_move_score(i, value);
            update_pv(ss, 0);
            TT.extract_pv(pos, ss[0].pv, PLY_MAX);
            rml.set_move_pv(i, ss[0].pv);

            if (MultiPV == 1)
            {
                // We record how often the best move has been changed in each
                // iteration. This information is used for time managment: When
                // the best move changes frequently, we allocate some more time.
                if (i > 0)
                    BestMoveChangesByIteration[Iteration]++;

                // Print search information to the standard output
                std::cout << "info depth " << Iteration
                          << " score " << value_to_string(value)
                          << ((value >= beta)?
                              " lowerbound" : ((value <= alpha)? " upperbound" : ""))
                          << " time " << current_search_time()
                          << " nodes " << nodes_searched()
                          << " nps " << nps()
                          << " pv ";

                for (int j = 0; ss[0].pv[j] != MOVE_NONE && j < PLY_MAX; j++)
                    std::cout << ss[0].pv[j] << " ";

                std::cout << std::endl;

                if (UseLogFile)
                    LogFile << pretty_pv(pos, current_search_time(), Iteration, nodes_searched(), value,
                                         ((value >= beta)? VALUE_TYPE_LOWER
                                          : ((value <= alpha)? VALUE_TYPE_UPPER : VALUE_TYPE_EXACT)),
                                         ss[0].pv)
                            << std::endl;

                if (value > alpha)
                    alpha = value;

                // Reset the global variable Problem to false if the value isn't too
                // far below the final value from the last iteration.
                if (value > IterationInfo[Iteration - 1].value - NoProblemMargin)
                    Problem = false;
            }
            else // MultiPV > 1
            {
                rml.sort_multipv(i);
                for (int j = 0; j < Min(MultiPV, rml.move_count()); j++)
                {
                    int k;
                    std::cout << "info multipv " << j + 1
                              << " score " << value_to_string(rml.get_move_score(j))
                              << " depth " << ((j <= i)? Iteration : Iteration - 1)
                              << " time " << current_search_time()
                              << " nodes " << nodes_searched()
                              << " nps " << nps()
                              << " pv ";

                    for (k = 0; rml.get_move_pv(j, k) != MOVE_NONE && k < PLY_MAX; k++)
                        std::cout << rml.get_move_pv(j, k) << " ";

                    std::cout << std::endl;
                }
                alpha = rml.get_move_score(Min(i, MultiPV-1));
            }
        } // New best move case

        assert(alpha >= oldAlpha);

        FailLow = (alpha == oldAlpha);
    }
    return alpha;
  }


  // search_pv() is the main search function for PV nodes.

  Value search_pv(Position& pos, SearchStack ss[], Value alpha, Value beta,
                  Depth depth, int ply, int threadID) {

    assert(alpha >= -VALUE_INFINITE && alpha <= VALUE_INFINITE);
    assert(beta > alpha && beta <= VALUE_INFINITE);
    assert(ply >= 0 && ply < PLY_MAX);
    assert(threadID >= 0 && threadID < ActiveThreads);

    Move movesSearched[256];
    EvalInfo ei;
    StateInfo st;
    const TTEntry* tte;
    Move ttMove, move;
    Depth ext, newDepth;
    Value oldAlpha, value;
    bool isCheck, mateThreat, singleReply, moveIsCheck, captureOrPromotion, dangerous;
    int moveCount = 0;
    Value bestValue = -VALUE_INFINITE;

    if (depth < OnePly)
        return qsearch(pos, ss, alpha, beta, Depth(0), ply, threadID);

    // Initialize, and make an early exit in case of an aborted search,
    // an instant draw, maximum ply reached, etc.
    init_node(ss, ply, threadID);

    // After init_node() that calls poll()
    if (AbortSearch || thread_should_stop(threadID))
        return Value(0);

    if (pos.is_draw())
        return VALUE_DRAW;

    if (ply >= PLY_MAX - 1)
        return pos.is_check() ? quick_evaluate(pos) : evaluate(pos, ei, threadID);

    // Mate distance pruning
    oldAlpha = alpha;
    alpha = Max(value_mated_in(ply), alpha);
    beta = Min(value_mate_in(ply+1), beta);
    if (alpha >= beta)
        return alpha;

    // Transposition table lookup. At PV nodes, we don't use the TT for
    // pruning, but only for move ordering.
    tte = TT.retrieve(pos.get_key());
    ttMove = (tte ? tte->move() : MOVE_NONE);

    // Go with internal iterative deepening if we don't have a TT move
    if (UseIIDAtPVNodes && ttMove == MOVE_NONE && depth >= 5*OnePly)
    {
        search_pv(pos, ss, alpha, beta, depth-2*OnePly, ply, threadID);
        ttMove = ss[ply].pv[ply];
    }

    // Initialize a MovePicker object for the current position, and prepare
    // to search all moves
    isCheck = pos.is_check();
    mateThreat = pos.has_mate_threat(opposite_color(pos.side_to_move()));
    CheckInfo ci(pos);
    MovePicker mp = MovePicker(pos, ttMove, depth, H, &ss[ply]);

    // Loop through all legal moves until no moves remain or a beta cutoff
    // occurs.
    while (   alpha < beta
           && (move = mp.get_next_move()) != MOVE_NONE
           && !thread_should_stop(threadID))
    {
      assert(move_is_ok(move));

      singleReply = (isCheck && mp.number_of_evasions() == 1);
      moveIsCheck = pos.move_is_check(move, ci);
      captureOrPromotion = pos.move_is_capture_or_promotion(move);

      movesSearched[moveCount++] = ss[ply].currentMove = move;

      // Decide the new search depth
      ext = extension(pos, move, true, captureOrPromotion, moveIsCheck, singleReply, mateThreat, &dangerous);
      newDepth = depth - OnePly + ext;

      // Make and search the move
      pos.do_move(move, st, ci, moveIsCheck);

      if (moveCount == 1) // The first move in list is the PV
          value = -search_pv(pos, ss, -beta, -alpha, newDepth, ply+1, threadID);
      else
      {
        // Try to reduce non-pv search depth by one ply if move seems not problematic,
        // if the move fails high will be re-searched at full depth.
        if (    depth >= 3*OnePly
            &&  moveCount >= LMRPVMoves
            && !dangerous
            && !captureOrPromotion
            && !move_is_castle(move)
            && !move_is_killer(move, ss[ply]))
        {
            ss[ply].reduction = OnePly;
            value = -search(pos, ss, -alpha, newDepth-OnePly, ply+1, true, threadID);
        }
        else
            value = alpha + 1; // Just to trigger next condition

        if (value > alpha) // Go with full depth non-pv search
        {
            ss[ply].reduction = Depth(0);
            value = -search(pos, ss, -alpha, newDepth, ply+1, true, threadID);
            if (value > alpha && value < beta)
            {
                // When the search fails high at ply 1 while searching the first
                // move at the root, set the flag failHighPly1. This is used for
                // time managment:  We don't want to stop the search early in
                // such cases, because resolving the fail high at ply 1 could
                // result in a big drop in score at the root.
                if (ply == 1 && RootMoveNumber == 1)
                    Threads[threadID].failHighPly1 = true;

                // A fail high occurred. Re-search at full window (pv search)
                value = -search_pv(pos, ss, -beta, -alpha, newDepth, ply+1, threadID);
                Threads[threadID].failHighPly1 = false;
          }
        }
      }
      pos.undo_move(move);

      assert(value > -VALUE_INFINITE && value < VALUE_INFINITE);

      // New best move?
      if (value > bestValue)
      {
          bestValue = value;
          if (value > alpha)
          {
              alpha = value;
              update_pv(ss, ply);
              if (value == value_mate_in(ply + 1))
                  ss[ply].mateKiller = move;
          }
          // If we are at ply 1, and we are searching the first root move at
          // ply 0, set the 'Problem' variable if the score has dropped a lot
          // (from the computer's point of view) since the previous iteration.
          if (   ply == 1
              && Iteration >= 2
              && -value <= IterationInfo[Iteration-1].value - ProblemMargin)
              Problem = true;
      }

      // Split?
      if (   ActiveThreads > 1
          && bestValue < beta
          && depth >= MinimumSplitDepth
          && Iteration <= 99
          && idle_thread_exists(threadID)
          && !AbortSearch
          && !thread_should_stop(threadID)
          && split(pos, ss, ply, &alpha, &beta, &bestValue, VALUE_NONE, VALUE_NONE,
                   depth, &moveCount, &mp, threadID, true))
          break;
    }

    // All legal moves have been searched.  A special case: If there were
    // no legal moves, it must be mate or stalemate.
    if (moveCount == 0)
        return (isCheck ? value_mated_in(ply) : VALUE_DRAW);

    // If the search is not aborted, update the transposition table,
    // history counters, and killer moves.
    if (AbortSearch || thread_should_stop(threadID))
        return bestValue;

    if (bestValue <= oldAlpha)
        TT.store(pos.get_key(), value_to_tt(bestValue, ply), VALUE_TYPE_UPPER, depth, MOVE_NONE);

    else if (bestValue >= beta)
    {
        BetaCounter.add(pos.side_to_move(), depth, threadID);
        move = ss[ply].pv[ply];
        if (!pos.move_is_capture_or_promotion(move))
        {
            update_history(pos, move, depth, movesSearched, moveCount);
            update_killers(move, ss[ply]);
        }
        TT.store(pos.get_key(), value_to_tt(bestValue, ply), VALUE_TYPE_LOWER, depth, move);
    }
    else
        TT.store(pos.get_key(), value_to_tt(bestValue, ply), VALUE_TYPE_EXACT, depth, ss[ply].pv[ply]);

    return bestValue;
  }


  // search() is the search function for zero-width nodes.

  Value search(Position& pos, SearchStack ss[], Value beta, Depth depth,
               int ply, bool allowNullmove, int threadID) {

    assert(beta >= -VALUE_INFINITE && beta <= VALUE_INFINITE);
    assert(ply >= 0 && ply < PLY_MAX);
    assert(threadID >= 0 && threadID < ActiveThreads);

    Move movesSearched[256];
    EvalInfo ei;
    StateInfo st;
    const TTEntry* tte;
    Move ttMove, move;
    Depth ext, newDepth;
    Value approximateEval, nullValue, value, futilityValue, futilityValueScaled;
    bool isCheck, useFutilityPruning, singleReply, moveIsCheck, captureOrPromotion, dangerous;
    bool mateThreat = false;
    int moveCount = 0;
    Value bestValue = -VALUE_INFINITE;

    if (depth < OnePly)
        return qsearch(pos, ss, beta-1, beta, Depth(0), ply, threadID);

    // Initialize, and make an early exit in case of an aborted search,
    // an instant draw, maximum ply reached, etc.
    init_node(ss, ply, threadID);

    // After init_node() that calls poll()
    if (AbortSearch || thread_should_stop(threadID))
        return Value(0);

    if (pos.is_draw())
        return VALUE_DRAW;

    if (ply >= PLY_MAX - 1)
        return pos.is_check() ? quick_evaluate(pos) : evaluate(pos, ei, threadID);

    // Mate distance pruning
    if (value_mated_in(ply) >= beta)
        return beta;

    if (value_mate_in(ply + 1) < beta)
        return beta - 1;

    // Transposition table lookup
    tte = TT.retrieve(pos.get_key());
    ttMove = (tte ? tte->move() : MOVE_NONE);

    if (tte && ok_to_use_TT(tte, depth, beta, ply))
    {
        ss[ply].currentMove = ttMove; // can be MOVE_NONE
        return value_from_tt(tte->value(), ply);
    }

    approximateEval = quick_evaluate(pos);
    isCheck = pos.is_check();

    // Null move search
    if (    allowNullmove
        &&  depth > OnePly
        && !isCheck
        && !value_is_mate(beta)
        &&  ok_to_do_nullmove(pos)
        &&  approximateEval >= beta - NullMoveMargin)
    {
        ss[ply].currentMove = MOVE_NULL;

        pos.do_null_move(st);

        // Null move dynamic reduction based on depth
        int R = (depth >= 5 * OnePly ? 4 : 3);

        // Null move dynamic reduction based on value
        if (approximateEval - beta > PawnValueMidgame)
            R++;

        nullValue = -search(pos, ss, -(beta-1), depth-R*OnePly, ply+1, false, threadID);

        pos.undo_null_move();

        if (nullValue >= beta)
        {
            if (depth < 6 * OnePly)
                return beta;

            // Do zugzwang verification search
            Value v = search(pos, ss, beta, depth-5*OnePly, ply, false, threadID);
            if (v >= beta)
                return beta;
        } else {
            // The null move failed low, which means that we may be faced with
            // some kind of threat. If the previous move was reduced, check if
            // the move that refuted the null move was somehow connected to the
            // move which was reduced. If a connection is found, return a fail
            // low score (which will cause the reduced move to fail high in the
            // parent node, which will trigger a re-search with full depth).
            if (nullValue == value_mated_in(ply + 2))
                mateThreat = true;

            ss[ply].threatMove = ss[ply + 1].currentMove;
            if (   depth < ThreatDepth
                && ss[ply - 1].reduction
                && connected_moves(pos, ss[ply - 1].currentMove, ss[ply].threatMove))
                return beta - 1;
        }
    }
    // Null move search not allowed, try razoring
    else if (   !value_is_mate(beta)
             && depth < RazorDepth
             && approximateEval < beta - RazorApprMargins[int(depth) - 2]
             && ss[ply - 1].currentMove != MOVE_NULL
             && ttMove == MOVE_NONE
             && !pos.has_pawn_on_7th(pos.side_to_move()))
    {
        Value rbeta = beta - RazorMargins[int(depth) - 2];
        Value v = qsearch(pos, ss, rbeta-1, rbeta, Depth(0), ply, threadID);
        if (v < rbeta)
          return v;
    }

    // Go with internal iterative deepening if we don't have a TT move
    if (UseIIDAtNonPVNodes && ttMove == MOVE_NONE && depth >= 8*OnePly &&
        evaluate(pos, ei, threadID) >= beta - IIDMargin)
    {
        search(pos, ss, beta, Min(depth/2, depth-2*OnePly), ply, false, threadID);
        ttMove = ss[ply].pv[ply];
    }

    // Initialize a MovePicker object for the current position, and prepare
    // to search all moves.
    MovePicker mp = MovePicker(pos, ttMove, depth, H, &ss[ply]);
    CheckInfo ci(pos);
    futilityValue = VALUE_NONE;
    useFutilityPruning = depth < SelectiveDepth && !isCheck;

    // Avoid calling evaluate() if we already have the score in TT
    if (tte && (tte->type() & VALUE_TYPE_EVAL))
        futilityValue = value_from_tt(tte->value(), ply) + FutilityMargins[int(depth) - 2];

    // Move count pruning limit
    const int MCLimit = 3 + (1 << (3*int(depth)/8));

    // Loop through all legal moves until no moves remain or a beta cutoff
    // occurs.
    while (   bestValue < beta
           && (move = mp.get_next_move()) != MOVE_NONE
           && !thread_should_stop(threadID))
    {
      assert(move_is_ok(move));

      singleReply = (isCheck && mp.number_of_evasions() == 1);
      moveIsCheck = pos.move_is_check(move, ci);
      captureOrPromotion = pos.move_is_capture_or_promotion(move);

      movesSearched[moveCount++] = ss[ply].currentMove = move;

      // Decide the new search depth
      ext = extension(pos, move, false, captureOrPromotion, moveIsCheck, singleReply, mateThreat, &dangerous);
      newDepth = depth - OnePly + ext;

      // Futility pruning
      if (    useFutilityPruning
          && !dangerous
          && !captureOrPromotion
          &&  move != ttMove)
      {
          //std::cout << std::endl;
          //for (int d = 2; d < 14; d++)
          //    std::cout << d << ", " << 300 + 2*(1 << (3*d/4)) << std::endl;

          //std::cout << std::endl;
/*
            300 + 2*(1 << (3*d/4))

            2 -> 256 -  304
            3 -> 288 -  308
            4 -> 512 -  316
            5 -> 544 -  316
            6 -> 592 -  332
            7 -> 624 -  364
            8 -> 672 -  428
            9 -> 704 -  428
           10 -> 832 -  556
           11 -> 864 -  812
           12 -> 928 -  1324
           13 -> 960 -  1324


            3 + (1 << (3*int(depth)/8))

            1 * onePly - > moveCount >= 4
            2 * onePly - > moveCount >= 5
            3 * onePly - > moveCount >= 7
            4 * onePly - > moveCount >= 11
            5 * onePly - > moveCount >= 11
            6 * onePly - > moveCount >= 19
            7 * onePly - > moveCount >= 35
*/
          // History pruning. See ok_to_prune() definition
          if (   moveCount >= MCLimit
              && ok_to_prune(pos, move, ss[ply].threatMove, depth)
              && bestValue > value_mated_in(PLY_MAX))
              continue;

          // Value based pruning
          if (approximateEval < beta)
          {
              if (futilityValue == VALUE_NONE)
                  futilityValue =  evaluate(pos, ei, threadID)
                                 + (300 + 2 * (1 << (3 * int(depth) /4)))
                                 + 4*IncrementalFutilityMargin;

              futilityValueScaled = futilityValue - moveCount * IncrementalFutilityMargin;

              if (futilityValueScaled < beta)
              {
                  if (futilityValueScaled > bestValue)
                      bestValue = futilityValueScaled;
                  continue;
              }
          }
      }

      // Make and search the move
      pos.do_move(move, st, ci, moveIsCheck);

      // Try to reduce non-pv search depth by one ply if move seems not problematic,
      // if the move fails high will be re-searched at full depth.
      if (    depth >= 3*OnePly
          &&  moveCount >= LMRNonPVMoves
          && !dangerous
          && !captureOrPromotion
          && !move_is_castle(move)
          && !move_is_killer(move, ss[ply]))
      {
          ss[ply].reduction = OnePly;
          value = -search(pos, ss, -(beta-1), newDepth-OnePly, ply+1, true, threadID);
      }
      else
        value = beta; // Just to trigger next condition

      if (value >= beta) // Go with full depth non-pv search
      {
          ss[ply].reduction = Depth(0);
          value = -search(pos, ss, -(beta-1), newDepth, ply+1, true, threadID);
      }
      pos.undo_move(move);

      assert(value > -VALUE_INFINITE && value < VALUE_INFINITE);

      // New best move?
      if (value > bestValue)
      {
        bestValue = value;
        if (value >= beta)
            update_pv(ss, ply);

        if (value == value_mate_in(ply + 1))
            ss[ply].mateKiller = move;
      }

      // Split?
      if (   ActiveThreads > 1
          && bestValue < beta
          && depth >= MinimumSplitDepth
          && Iteration <= 99
          && idle_thread_exists(threadID)
          && !AbortSearch
          && !thread_should_stop(threadID)
          && split(pos, ss, ply, &beta, &beta, &bestValue, futilityValue, approximateEval,
                   depth, &moveCount, &mp, threadID, false))
        break;
    }

    // All legal moves have been searched.  A special case: If there were
    // no legal moves, it must be mate or stalemate.
    if (moveCount == 0)
        return (pos.is_check() ? value_mated_in(ply) : VALUE_DRAW);

    // If the search is not aborted, update the transposition table,
    // history counters, and killer moves.
    if (AbortSearch || thread_should_stop(threadID))
        return bestValue;

    if (bestValue < beta)
        TT.store(pos.get_key(), value_to_tt(bestValue, ply), VALUE_TYPE_UPPER, depth, MOVE_NONE);
    else
    {
        BetaCounter.add(pos.side_to_move(), depth, threadID);
        move = ss[ply].pv[ply];
        if (!pos.move_is_capture_or_promotion(move))
        {
            update_history(pos, move, depth, movesSearched, moveCount);
            update_killers(move, ss[ply]);
        }
        TT.store(pos.get_key(), value_to_tt(bestValue, ply), VALUE_TYPE_LOWER, depth, move);
    }

    assert(bestValue > -VALUE_INFINITE && bestValue < VALUE_INFINITE);

    return bestValue;
  }


  // qsearch() is the quiescence search function, which is called by the main
  // search function when the remaining depth is zero (or, to be more precise,
  // less than OnePly).

  Value qsearch(Position& pos, SearchStack ss[], Value alpha, Value beta,
                Depth depth, int ply, int threadID) {

    assert(alpha >= -VALUE_INFINITE && alpha <= VALUE_INFINITE);
    assert(beta >= -VALUE_INFINITE && beta <= VALUE_INFINITE);
    assert(depth <= 0);
    assert(ply >= 0 && ply < PLY_MAX);
    assert(threadID >= 0 && threadID < ActiveThreads);

    EvalInfo ei;
    StateInfo st;
    Move ttMove, move;
    Value staticValue, bestValue, value, futilityValue;
    bool isCheck, enoughMaterial, moveIsCheck;
    const TTEntry* tte = NULL;
    int moveCount = 0;
    bool pvNode = (beta - alpha != 1);

    // Initialize, and make an early exit in case of an aborted search,
    // an instant draw, maximum ply reached, etc.
    init_node(ss, ply, threadID);

    // After init_node() that calls poll()
    if (AbortSearch || thread_should_stop(threadID))
        return Value(0);

    if (pos.is_draw())
        return VALUE_DRAW;

    // Transposition table lookup, only when not in PV
    if (!pvNode)
    {
        tte = TT.retrieve(pos.get_key());
        if (tte && ok_to_use_TT(tte, depth, beta, ply))
        {
            assert(tte->type() != VALUE_TYPE_EVAL);

            return value_from_tt(tte->value(), ply);
        }
    }
    ttMove = (tte ? tte->move() : MOVE_NONE);

    // Evaluate the position statically
    isCheck = pos.is_check();
    ei.futilityMargin = Value(0); // Manually initialize futilityMargin

    if (isCheck)
        staticValue = -VALUE_INFINITE;

    else if (tte && (tte->type() & VALUE_TYPE_EVAL))
    {
        // Use the cached evaluation score if possible
        assert(ei.futilityMargin == Value(0));

        staticValue = tte->value();
    }
    else
        staticValue = evaluate(pos, ei, threadID);

    if (ply >= PLY_MAX - 1)
        return pos.is_check() ? quick_evaluate(pos) : evaluate(pos, ei, threadID);

    // Initialize "stand pat score", and return it immediately if it is
    // at least beta.
    bestValue = staticValue;

    if (bestValue >= beta)
    {
        // Store the score to avoid a future costly evaluation() call
        if (!isCheck && !tte && ei.futilityMargin == 0)
            TT.store(pos.get_key(), value_to_tt(bestValue, ply), VALUE_TYPE_EV_LO, Depth(-127*OnePly), MOVE_NONE);

        return bestValue;
    }

    if (bestValue > alpha)
        alpha = bestValue;

    // Initialize a MovePicker object for the current position, and prepare
    // to search the moves.  Because the depth is <= 0 here, only captures,
    // queen promotions and checks (only if depth == 0) will be generated.
    MovePicker mp = MovePicker(pos, ttMove, depth, H);
    CheckInfo ci(pos);
    enoughMaterial = pos.non_pawn_material(pos.side_to_move()) > RookValueMidgame;

    // Loop through the moves until no moves remain or a beta cutoff
    // occurs.
    while (   alpha < beta
           && (move = mp.get_next_move()) != MOVE_NONE)
    {
      assert(move_is_ok(move));

      moveCount++;
      ss[ply].currentMove = move;

      moveIsCheck = pos.move_is_check(move, ci);

      // Futility pruning
      if (   enoughMaterial
          && !isCheck
          && !pvNode
          && !moveIsCheck
          &&  move != ttMove
          && !move_is_promotion(move)
          && !pos.move_is_passed_pawn_push(move))
      {
          futilityValue =  staticValue
                         + Max(pos.midgame_value_of_piece_on(move_to(move)),
                               pos.endgame_value_of_piece_on(move_to(move)))
                         + (move_is_ep(move) ? PawnValueEndgame : Value(0))
                         + FutilityMarginQS
                         + ei.futilityMargin;

          if (futilityValue < alpha)
          {
              if (futilityValue > bestValue)
                  bestValue = futilityValue;
              continue;
          }
      }

      // Don't search captures and checks with negative SEE values
      if (   !isCheck
          &&  move != ttMove
          && !move_is_promotion(move)
          &&  pos.see_sign(move) < 0)
          continue;

      // Make and search the move
      pos.do_move(move, st, ci, moveIsCheck);
      value = -qsearch(pos, ss, -beta, -alpha, depth-OnePly, ply+1, threadID);
      pos.undo_move(move);

      assert(value > -VALUE_INFINITE && value < VALUE_INFINITE);

      // New best move?
      if (value > bestValue)
      {
          bestValue = value;
          if (value > alpha)
          {
              alpha = value;
              update_pv(ss, ply);
          }
       }
    }

    // All legal moves have been searched.  A special case: If we're in check
    // and no legal moves were found, it is checkmate.
    if (!moveCount && pos.is_check()) // Mate!
        return value_mated_in(ply);

    assert(bestValue > -VALUE_INFINITE && bestValue < VALUE_INFINITE);

    // Update transposition table
    move = ss[ply].pv[ply];
    if (!pvNode)
    {
        // If bestValue isn't changed it means it is still the static evaluation of
        // the node, so keep this info to avoid a future costly evaluation() call.
        ValueType type = (bestValue == staticValue && !ei.futilityMargin ? VALUE_TYPE_EV_UP : VALUE_TYPE_UPPER);
        Depth d = (depth == Depth(0) ? Depth(0) : Depth(-1));

        if (bestValue < beta)
            TT.store(pos.get_key(), value_to_tt(bestValue, ply), type, d, MOVE_NONE);
        else
            TT.store(pos.get_key(), value_to_tt(bestValue, ply), VALUE_TYPE_LOWER, d, move);
    }

    // Update killers only for good check moves
    if (alpha >= beta && !pos.move_is_capture_or_promotion(move))
        update_killers(move, ss[ply]);

    return bestValue;
  }


  // sp_search() is used to search from a split point.  This function is called
  // by each thread working at the split point.  It is similar to the normal
  // search() function, but simpler.  Because we have already probed the hash
  // table, done a null move search, and searched the first move before
  // splitting, we don't have to repeat all this work in sp_search().  We
  // also don't need to store anything to the hash table here:  This is taken
  // care of after we return from the split point.

  void sp_search(SplitPoint* sp, int threadID) {

    assert(threadID >= 0 && threadID < ActiveThreads);
    assert(ActiveThreads > 1);

    Position pos = Position(sp->pos);
    CheckInfo ci(pos);
    SearchStack* ss = sp->sstack[threadID];
    Value value;
    Move move;
    bool isCheck = pos.is_check();
    bool useFutilityPruning =     sp->depth < SelectiveDepth
                              && !isCheck;

    while (    sp->bestValue < sp->beta
           && !thread_should_stop(threadID)
           && (move = sp->mp->get_next_move(sp->lock)) != MOVE_NONE)
    {
      assert(move_is_ok(move));

      bool moveIsCheck = pos.move_is_check(move, ci);
      bool captureOrPromotion = pos.move_is_capture_or_promotion(move);

      lock_grab(&(sp->lock));
      int moveCount = ++sp->moves;
      lock_release(&(sp->lock));

      ss[sp->ply].currentMove = move;

      // Decide the new search depth.
      bool dangerous;
      Depth ext = extension(pos, move, false, captureOrPromotion, moveIsCheck, false, false, &dangerous);
      Depth newDepth = sp->depth - OnePly + ext;

      // Prune?
      if (    useFutilityPruning
          && !dangerous
          && !captureOrPromotion)
      {
          // History pruning. See ok_to_prune() definition
          if (   moveCount >= 2 + int(sp->depth)
              && ok_to_prune(pos, move, ss[sp->ply].threatMove, sp->depth)
              && sp->bestValue > value_mated_in(PLY_MAX))
              continue;

          // Value based pruning
          if (sp->approximateEval < sp->beta)
          {
              if (sp->futilityValue == VALUE_NONE)
              {
                  EvalInfo ei;
                  sp->futilityValue =  evaluate(pos, ei, threadID)
                                    + FutilityMargins[int(sp->depth) - 2];
              }

              if (sp->futilityValue < sp->beta)
              {
                  if (sp->futilityValue > sp->bestValue) // Less then 1% of cases
                  {
                      lock_grab(&(sp->lock));
                      if (sp->futilityValue > sp->bestValue)
                          sp->bestValue = sp->futilityValue;
                      lock_release(&(sp->lock));
                  }
                  continue;
              }
          }
      }

      // Make and search the move.
      StateInfo st;
      pos.do_move(move, st, ci, moveIsCheck);

      // Try to reduce non-pv search depth by one ply if move seems not problematic,
      // if the move fails high will be re-searched at full depth.
      if (   !dangerous
          &&  moveCount >= LMRNonPVMoves
          && !captureOrPromotion
          && !move_is_castle(move)
          && !move_is_killer(move, ss[sp->ply]))
      {
          ss[sp->ply].reduction = OnePly;
          value = -search(pos, ss, -(sp->beta-1), newDepth - OnePly, sp->ply+1, true, threadID);
      }
      else
          value = sp->beta; // Just to trigger next condition

      if (value >= sp->beta) // Go with full depth non-pv search
      {
          ss[sp->ply].reduction = Depth(0);
          value = -search(pos, ss, -(sp->beta - 1), newDepth, sp->ply+1, true, threadID);
      }
      pos.undo_move(move);

      assert(value > -VALUE_INFINITE && value < VALUE_INFINITE);

      if (thread_should_stop(threadID))
          break;

      // New best move?
      if (value > sp->bestValue) // Less then 2% of cases
      {
          lock_grab(&(sp->lock));
          if (value > sp->bestValue && !thread_should_stop(threadID))
          {
              sp->bestValue = value;
              if (sp->bestValue >= sp->beta)
              {
                  sp_update_pv(sp->parentSstack, ss, sp->ply);
                  for (int i = 0; i < ActiveThreads; i++)
                      if (i != threadID && (i == sp->master || sp->slaves[i]))
                          Threads[i].stop = true;

                  sp->finished = true;
              }
          }
          lock_release(&(sp->lock));
      }
    }

    lock_grab(&(sp->lock));

    // If this is the master thread and we have been asked to stop because of
    // a beta cutoff higher up in the tree, stop all slave threads.
    if (sp->master == threadID && thread_should_stop(threadID))
        for (int i = 0; i < ActiveThreads; i++)
            if (sp->slaves[i])
                Threads[i].stop = true;

    sp->cpus--;
    sp->slaves[threadID] = 0;

    lock_release(&(sp->lock));
  }


  // sp_search_pv() is used to search from a PV split point.  This function
  // is called by each thread working at the split point.  It is similar to
  // the normal search_pv() function, but simpler.  Because we have already
  // probed the hash table and searched the first move before splitting, we
  // don't have to repeat all this work in sp_search_pv().  We also don't
  // need to store anything to the hash table here: This is taken care of
  // after we return from the split point.

  void sp_search_pv(SplitPoint* sp, int threadID) {

    assert(threadID >= 0 && threadID < ActiveThreads);
    assert(ActiveThreads > 1);

    Position pos = Position(sp->pos);
    CheckInfo ci(pos);
    SearchStack* ss = sp->sstack[threadID];
    Value value;
    Move move;

    while (    sp->alpha < sp->beta
           && !thread_should_stop(threadID)
           && (move = sp->mp->get_next_move(sp->lock)) != MOVE_NONE)
    {
      bool moveIsCheck = pos.move_is_check(move, ci);
      bool captureOrPromotion = pos.move_is_capture_or_promotion(move);

      assert(move_is_ok(move));

      lock_grab(&(sp->lock));
      int moveCount = ++sp->moves;
      lock_release(&(sp->lock));

      ss[sp->ply].currentMove = move;

      // Decide the new search depth.
      bool dangerous;
      Depth ext = extension(pos, move, true, captureOrPromotion, moveIsCheck, false, false, &dangerous);
      Depth newDepth = sp->depth - OnePly + ext;

      // Make and search the move.
      StateInfo st;
      pos.do_move(move, st, ci, moveIsCheck);

      // Try to reduce non-pv search depth by one ply if move seems not problematic,
      // if the move fails high will be re-searched at full depth.
      if (   !dangerous
          &&  moveCount >= LMRPVMoves
          && !captureOrPromotion
          && !move_is_castle(move)
          && !move_is_killer(move, ss[sp->ply]))
      {
          ss[sp->ply].reduction = OnePly;
          value = -search(pos, ss, -sp->alpha, newDepth - OnePly, sp->ply+1, true, threadID);
      }
      else
          value = sp->alpha + 1; // Just to trigger next condition

      if (value > sp->alpha) // Go with full depth non-pv search
      {
          ss[sp->ply].reduction = Depth(0);
          value = -search(pos, ss, -sp->alpha, newDepth, sp->ply+1, true, threadID);

          if (value > sp->alpha && value < sp->beta)
          {
              // When the search fails high at ply 1 while searching the first
              // move at the root, set the flag failHighPly1.  This is used for
              // time managment: We don't want to stop the search early in
              // such cases, because resolving the fail high at ply 1 could
              // result in a big drop in score at the root.
              if (sp->ply == 1 && RootMoveNumber == 1)
                  Threads[threadID].failHighPly1 = true;

              value = -search_pv(pos, ss, -sp->beta, -sp->alpha, newDepth, sp->ply+1, threadID);
              Threads[threadID].failHighPly1 = false;
        }
      }
      pos.undo_move(move);

      assert(value > -VALUE_INFINITE && value < VALUE_INFINITE);

      if (thread_should_stop(threadID))
          break;

      // New best move?
      lock_grab(&(sp->lock));
      if (value > sp->bestValue && !thread_should_stop(threadID))
      {
          sp->bestValue = value;
          if (value > sp->alpha)
          {
              sp->alpha = value;
              sp_update_pv(sp->parentSstack, ss, sp->ply);
              if (value == value_mate_in(sp->ply + 1))
                  ss[sp->ply].mateKiller = move;

              if (value >= sp->beta)
              {
                  for (int i = 0; i < ActiveThreads; i++)
                      if (i != threadID && (i == sp->master || sp->slaves[i]))
                          Threads[i].stop = true;

                  sp->finished = true;
              }
        }
        // If we are at ply 1, and we are searching the first root move at
        // ply 0, set the 'Problem' variable if the score has dropped a lot
        // (from the computer's point of view) since the previous iteration.
        if (   sp->ply == 1
            && Iteration >= 2
            && -value <= IterationInfo[Iteration-1].value - ProblemMargin)
            Problem = true;
      }
      lock_release(&(sp->lock));
    }

    lock_grab(&(sp->lock));

    // If this is the master thread and we have been asked to stop because of
    // a beta cutoff higher up in the tree, stop all slave threads.
    if (sp->master == threadID && thread_should_stop(threadID))
        for (int i = 0; i < ActiveThreads; i++)
            if (sp->slaves[i])
                Threads[i].stop = true;

    sp->cpus--;
    sp->slaves[threadID] = 0;

    lock_release(&(sp->lock));
  }

  /// The BetaCounterType class

  BetaCounterType::BetaCounterType() { clear(); }

  void BetaCounterType::clear() {

    for (int i = 0; i < THREAD_MAX; i++)
        Threads[i].betaCutOffs[WHITE] = Threads[i].betaCutOffs[BLACK] = 0ULL;
  }

  void BetaCounterType::add(Color us, Depth d, int threadID) {

    // Weighted count based on depth
    Threads[threadID].betaCutOffs[us] += unsigned(d);
  }

  void BetaCounterType::read(Color us, int64_t& our, int64_t& their) {

    our = their = 0UL;
    for (int i = 0; i < THREAD_MAX; i++)
    {
        our += Threads[i].betaCutOffs[us];
        their += Threads[i].betaCutOffs[opposite_color(us)];
    }
  }


  /// The RootMove class

  // Constructor

  RootMove::RootMove() {
    nodes = cumulativeNodes = ourBeta = theirBeta = 0ULL;
  }

  // RootMove::operator<() is the comparison function used when
  // sorting the moves.  A move m1 is considered to be better
  // than a move m2 if it has a higher score, or if the moves
  // have equal score but m1 has the higher node count.

  bool RootMove::operator<(const RootMove& m) {

    if (score != m.score)
        return (score < m.score);

    return theirBeta <= m.theirBeta;
  }

  /// The RootMoveList class

  // Constructor

  RootMoveList::RootMoveList(Position& pos, Move searchMoves[]) : count(0) {

    MoveStack mlist[MaxRootMoves];
    bool includeAllMoves = (searchMoves[0] == MOVE_NONE);

    // Generate all legal moves
    MoveStack* last = generate_moves(pos, mlist);

    // Add each move to the moves[] array
    for (MoveStack* cur = mlist; cur != last; cur++)
    {
        bool includeMove = includeAllMoves;

        for (int k = 0; !includeMove && searchMoves[k] != MOVE_NONE; k++)
            includeMove = (searchMoves[k] == cur->move);

        if (!includeMove)
            continue;

        // Find a quick score for the move
        StateInfo st;
        SearchStack ss[PLY_MAX_PLUS_2];
        init_ss_array(ss);

        moves[count].move = cur->move;
        pos.do_move(moves[count].move, st);
        moves[count].score = -qsearch(pos, ss, -VALUE_INFINITE, VALUE_INFINITE, Depth(0), 1, 0);
        pos.undo_move(moves[count].move);
        moves[count].pv[0] = moves[count].move;
        moves[count].pv[1] = MOVE_NONE; // FIXME
        count++;
    }
    sort();
  }


  // Simple accessor methods for the RootMoveList class

  inline Move RootMoveList::get_move(int moveNum) const {
    return moves[moveNum].move;
  }

  inline Value RootMoveList::get_move_score(int moveNum) const {
    return moves[moveNum].score;
  }

  inline void RootMoveList::set_move_score(int moveNum, Value score) {
    moves[moveNum].score = score;
  }

  inline void RootMoveList::set_move_nodes(int moveNum, int64_t nodes) {
    moves[moveNum].nodes = nodes;
    moves[moveNum].cumulativeNodes += nodes;
  }

  inline void RootMoveList::set_beta_counters(int moveNum, int64_t our, int64_t their) {
    moves[moveNum].ourBeta = our;
    moves[moveNum].theirBeta = their;
  }

  void RootMoveList::set_move_pv(int moveNum, const Move pv[]) {
    int j;
    for(j = 0; pv[j] != MOVE_NONE; j++)
      moves[moveNum].pv[j] = pv[j];
    moves[moveNum].pv[j] = MOVE_NONE;
  }

  inline Move RootMoveList::get_move_pv(int moveNum, int i) const {
    return moves[moveNum].pv[i];
  }

  inline int64_t RootMoveList::get_move_cumulative_nodes(int moveNum) const {
    return moves[moveNum].cumulativeNodes;
  }

  inline int RootMoveList::move_count() const {
    return count;
  }


  // RootMoveList::scan_for_easy_move() is called at the end of the first
  // iteration, and is used to detect an "easy move", i.e. a move which appears
  // to be much bester than all the rest.  If an easy move is found, the move
  // is returned, otherwise the function returns MOVE_NONE.  It is very
  // important that this function is called at the right moment:  The code
  // assumes that the first iteration has been completed and the moves have
  // been sorted. This is done in RootMoveList c'tor.

  Move RootMoveList::scan_for_easy_move() const {

    assert(count);

    if (count == 1)
        return get_move(0);

    // moves are sorted so just consider the best and the second one
    if (get_move_score(0) > get_move_score(1) + EasyMoveMargin)
        return get_move(0);

    return MOVE_NONE;
  }

  // RootMoveList::sort() sorts the root move list at the beginning of a new
  // iteration.

  inline void RootMoveList::sort() {

    sort_multipv(count - 1); // all items
  }


  // RootMoveList::sort_multipv() sorts the first few moves in the root move
  // list by their scores and depths. It is used to order the different PVs
  // correctly in MultiPV mode.

  void RootMoveList::sort_multipv(int n) {

    for (int i = 1; i <= n; i++)
    {
      RootMove rm = moves[i];
      int j;
      for (j = i; j > 0 && moves[j-1] < rm; j--)
          moves[j] = moves[j-1];
      moves[j] = rm;
    }
  }


  // init_node() is called at the beginning of all the search functions
  // (search(), search_pv(), qsearch(), and so on) and initializes the search
  // stack object corresponding to the current node.  Once every
  // NodesBetweenPolls nodes, init_node() also calls poll(), which polls
  // for user input and checks whether it is time to stop the search.

  void init_node(SearchStack ss[], int ply, int threadID) {

    assert(ply >= 0 && ply < PLY_MAX);
    assert(threadID >= 0 && threadID < ActiveThreads);

    Threads[threadID].nodes++;

    if (threadID == 0)
    {
        NodesSincePoll++;
        if (NodesSincePoll >= NodesBetweenPolls)
        {
            poll();
            NodesSincePoll = 0;
        }
    }
    ss[ply].init(ply);
    ss[ply+2].initKillers();

    if (Threads[threadID].printCurrentLine)
        print_current_line(ss, ply, threadID);
  }


  // update_pv() is called whenever a search returns a value > alpha.  It
  // updates the PV in the SearchStack object corresponding to the current
  // node.

  void update_pv(SearchStack ss[], int ply) {
    assert(ply >= 0 && ply < PLY_MAX);

    ss[ply].pv[ply] = ss[ply].currentMove;
    int p;
    for(p = ply + 1; ss[ply+1].pv[p] != MOVE_NONE; p++)
      ss[ply].pv[p] = ss[ply+1].pv[p];
    ss[ply].pv[p] = MOVE_NONE;
  }


  // sp_update_pv() is a variant of update_pv for use at split points.  The
  // difference between the two functions is that sp_update_pv also updates
  // the PV at the parent node.

  void sp_update_pv(SearchStack* pss, SearchStack ss[], int ply) {
    assert(ply >= 0 && ply < PLY_MAX);

    ss[ply].pv[ply] = pss[ply].pv[ply] = ss[ply].currentMove;
    int p;
    for(p = ply + 1; ss[ply+1].pv[p] != MOVE_NONE; p++)
      ss[ply].pv[p] = pss[ply].pv[p] = ss[ply+1].pv[p];
    ss[ply].pv[p] = pss[ply].pv[p] = MOVE_NONE;
  }


  // connected_moves() tests whether two moves are 'connected' in the sense
  // that the first move somehow made the second move possible (for instance
  // if the moving piece is the same in both moves).  The first move is
  // assumed to be the move that was made to reach the current position, while
  // the second move is assumed to be a move from the current position.

  bool connected_moves(const Position& pos, Move m1, Move m2) {

    Square f1, t1, f2, t2;
    Piece p;

    assert(move_is_ok(m1));
    assert(move_is_ok(m2));

    if (m2 == MOVE_NONE)
        return false;

    // Case 1: The moving piece is the same in both moves
    f2 = move_from(m2);
    t1 = move_to(m1);
    if (f2 == t1)
        return true;

    // Case 2: The destination square for m2 was vacated by m1
    t2 = move_to(m2);
    f1 = move_from(m1);
    if (t2 == f1)
        return true;

    // Case 3: Moving through the vacated square
    if (   piece_is_slider(pos.piece_on(f2))
        && bit_is_set(squares_between(f2, t2), f1))
      return true;

    // Case 4: The destination square for m2 is attacked by the moving piece in m1
    p = pos.piece_on(t1);
    if (bit_is_set(pos.attacks_from(p, t1), t2))
        return true;

    // Case 5: Discovered check, checking piece is the piece moved in m1
    if (   piece_is_slider(p)
        && bit_is_set(squares_between(t1, pos.king_square(pos.side_to_move())), f2)
        && !bit_is_set(squares_between(t1, pos.king_square(pos.side_to_move())), t2))
    {
        Bitboard occ = pos.occupied_squares();
        Color us = pos.side_to_move();
        Square ksq = pos.king_square(us);
        clear_bit(&occ, f2);
        if (type_of_piece(p) == BISHOP)
        {
            if (bit_is_set(bishop_attacks_bb(ksq, occ), t1))
                return true;
        }
        else if (type_of_piece(p) == ROOK)
        {
            if (bit_is_set(rook_attacks_bb(ksq, occ), t1))
                return true;
        }
        else
        {
            assert(type_of_piece(p) == QUEEN);
            if (bit_is_set(queen_attacks_bb(ksq, occ), t1))
                return true;
        }
    }
    return false;
  }


  // value_is_mate() checks if the given value is a mate one
  // eventually compensated for the ply.

  bool value_is_mate(Value value) {

    assert(abs(value) <= VALUE_INFINITE);

    return   value <= value_mated_in(PLY_MAX)
          || value >= value_mate_in(PLY_MAX);
  }


  // move_is_killer() checks if the given move is among the
  // killer moves of that ply.

  bool move_is_killer(Move m, const SearchStack& ss) {

      const Move* k = ss.killers;
      for (int i = 0; i < KILLER_MAX; i++, k++)
          if (*k == m)
              return true;

      return false;
  }


  // extension() decides whether a move should be searched with normal depth,
  // or with extended depth.  Certain classes of moves (checking moves, in
  // particular) are searched with bigger depth than ordinary moves and in
  // any case are marked as 'dangerous'. Note that also if a move is not
  // extended, as example because the corresponding UCI option is set to zero,
  // the move is marked as 'dangerous' so, at least, we avoid to prune it.

  Depth extension(const Position& pos, Move m, bool pvNode, bool captureOrPromotion,
                  bool check, bool singleReply, bool mateThreat, bool* dangerous) {

    assert(m != MOVE_NONE);

    Depth result = Depth(0);
    *dangerous = check | singleReply | mateThreat;

    if (*dangerous)
    {
        if (check)
            result += CheckExtension[pvNode];

        if (singleReply)
            result += SingleReplyExtension[pvNode];

        if (mateThreat)
            result += MateThreatExtension[pvNode];
    }

    if (pos.type_of_piece_on(move_from(m)) == PAWN)
    {
        Color c = pos.side_to_move();
        if (relative_rank(c, move_to(m)) == RANK_7)
        {
            result += PawnPushTo7thExtension[pvNode];
            *dangerous = true;
        }
        if (pos.pawn_is_passed(c, move_to(m)))
        {
            result += PassedPawnExtension[pvNode];
            *dangerous = true;
        }
    }

    if (   captureOrPromotion
        && pos.type_of_piece_on(move_to(m)) != PAWN
        && (  pos.non_pawn_material(WHITE) + pos.non_pawn_material(BLACK)
            - pos.midgame_value_of_piece_on(move_to(m)) == Value(0))
        && !move_is_promotion(m)
        && !move_is_ep(m))
    {
        result += PawnEndgameExtension[pvNode];
        *dangerous = true;
    }

    if (   pvNode
        && captureOrPromotion
        && pos.type_of_piece_on(move_to(m)) != PAWN
        && pos.see_sign(m) >= 0)
    {
        result += OnePly/2;
        *dangerous = true;
    }

    return Min(result, OnePly);
  }


  // ok_to_do_nullmove() looks at the current position and decides whether
  // doing a 'null move' should be allowed.  In order to avoid zugzwang
  // problems, null moves are not allowed when the side to move has very
  // little material left.  Currently, the test is a bit too simple:  Null
  // moves are avoided only when the side to move has only pawns left.  It's
  // probably a good idea to avoid null moves in at least some more
  // complicated endgames, e.g. KQ vs KR.  FIXME

  bool ok_to_do_nullmove(const Position& pos) {

    return pos.non_pawn_material(pos.side_to_move()) != Value(0);
  }


  // ok_to_prune() tests whether it is safe to forward prune a move.  Only
  // non-tactical moves late in the move list close to the leaves are
  // candidates for pruning.

  bool ok_to_prune(const Position& pos, Move m, Move threat, Depth d) {

    assert(move_is_ok(m));
    assert(threat == MOVE_NONE || move_is_ok(threat));
    assert(!pos.move_is_check(m));
    assert(!pos.move_is_capture_or_promotion(m));
    assert(!pos.move_is_passed_pawn_push(m));
    assert(d >= OnePly);

    Square mfrom, mto, tfrom, tto;

    mfrom = move_from(m);
    mto = move_to(m);
    tfrom = move_from(threat);
    tto = move_to(threat);

    // Case 1: Castling moves are never pruned
    if (move_is_castle(m))
        return false;

    // Case 2: Don't prune moves which move the threatened piece
    if (!PruneEscapeMoves && threat != MOVE_NONE && mfrom == tto)
        return false;

    // Case 3: If the threatened piece has value less than or equal to the
    // value of the threatening piece, don't prune move which defend it.
    if (   !PruneDefendingMoves
        && threat != MOVE_NONE
        && pos.move_is_capture(threat)
        && (   pos.midgame_value_of_piece_on(tfrom) >= pos.midgame_value_of_piece_on(tto)
            || pos.type_of_piece_on(tfrom) == KING)
        && pos.move_attacks_square(m, tto))
        return false;

    // Case 4: Don't prune moves with good history
    if (!H.ok_to_prune(pos.piece_on(mfrom), mto, d))
        return false;

    // Case 5: If the moving piece in the threatened move is a slider, don't
    // prune safe moves which block its ray.
    if (  !PruneBlockingMoves
        && threat != MOVE_NONE
        && piece_is_slider(pos.piece_on(tfrom))
        && bit_is_set(squares_between(tfrom, tto), mto)
        && pos.see_sign(m) >= 0)
        return false;

    return true;
  }


  // ok_to_use_TT() returns true if a transposition table score
  // can be used at a given point in search.

  bool ok_to_use_TT(const TTEntry* tte, Depth depth, Value beta, int ply) {

    Value v = value_from_tt(tte->value(), ply);

    return   (   tte->depth() >= depth
              || v >= Max(value_mate_in(100), beta)
              || v < Min(value_mated_in(100), beta))

          && (   (is_lower_bound(tte->type()) && v >= beta)
              || (is_upper_bound(tte->type()) && v < beta));
  }


  // update_history() registers a good move that produced a beta-cutoff
  // in history and marks as failures all the other moves of that ply.

  void update_history(const Position& pos, Move m, Depth depth,
                      Move movesSearched[], int moveCount) {

    H.success(pos.piece_on(move_from(m)), move_to(m), depth);

    for (int i = 0; i < moveCount - 1; i++)
    {
        assert(m != movesSearched[i]);
        if (!pos.move_is_capture_or_promotion(movesSearched[i]))
            H.failure(pos.piece_on(move_from(movesSearched[i])), move_to(movesSearched[i]));
    }
  }


  // update_killers() add a good move that produced a beta-cutoff
  // among the killer moves of that ply.

  void update_killers(Move m, SearchStack& ss) {

    if (m == ss.killers[0])
        return;

    for (int i = KILLER_MAX - 1; i > 0; i--)
        ss.killers[i] = ss.killers[i - 1];

    ss.killers[0] = m;
  }


  // fail_high_ply_1() checks if some thread is currently resolving a fail
  // high at ply 1 at the node below the first root node.  This information
  // is used for time managment.

  bool fail_high_ply_1() {

    for(int i = 0; i < ActiveThreads; i++)
        if (Threads[i].failHighPly1)
            return true;

    return false;
  }


  // current_search_time() returns the number of milliseconds which have passed
  // since the beginning of the current search.

  int current_search_time() {
    return get_system_time() - SearchStartTime;
  }


  // nps() computes the current nodes/second count.

  int nps() {
    int t = current_search_time();
    return (t > 0)? int((nodes_searched() * 1000) / t) : 0;
  }


  // poll() performs two different functions:  It polls for user input, and it
  // looks at the time consumed so far and decides if it's time to abort the
  // search.

  void poll() {

    static int lastInfoTime;
    int t = current_search_time();

    //  Poll for input
    if (Bioskey())
    {
        // We are line oriented, don't read single chars
        std::string command;
        if (!std::getline(std::cin, command))
            command = "quit";

        if (command == "quit")
        {
            AbortSearch = true;
            PonderSearch = false;
            Quit = true;
            return;
        }
        else if (command == "stop")
        {
            AbortSearch = true;
            PonderSearch = false;
        }
        else if (command == "ponderhit")
            ponderhit();
    }
    // Print search information
    if (t < 1000)
        lastInfoTime = 0;

    else if (lastInfoTime > t)
        // HACK: Must be a new search where we searched less than
        // NodesBetweenPolls nodes during the first second of search.
        lastInfoTime = 0;

    else if (t - lastInfoTime >= 1000)
    {
        lastInfoTime = t;
        lock_grab(&IOLock);
        if (dbg_show_mean)
            dbg_print_mean();

        if (dbg_show_hit_rate)
            dbg_print_hit_rate();

        std::cout << "info nodes " << nodes_searched() << " nps " << nps()
                  << " time " << t << " hashfull " << TT.full() << std::endl;
        lock_release(&IOLock);
        if (ShowCurrentLine)
            Threads[0].printCurrentLine = true;
    }
    // Should we stop the search?
    if (PonderSearch)
        return;

    bool overTime =     t > AbsoluteMaxSearchTime
                     || (RootMoveNumber == 1 && t > MaxSearchTime + ExtraSearchTime && !FailLow) //FIXME: We are not checking any problem flags, BUG?
                     || (  !FailHigh && !FailLow && !fail_high_ply_1() && !Problem
                         && t > 6*(MaxSearchTime + ExtraSearchTime));

    if (   (Iteration >= 3 && (!InfiniteSearch && overTime))
        || (ExactMaxTime && t >= ExactMaxTime)
        || (Iteration >= 3 && MaxNodes && nodes_searched() >= MaxNodes))
        AbortSearch = true;
  }


  // ponderhit() is called when the program is pondering (i.e. thinking while
  // it's the opponent's turn to move) in order to let the engine know that
  // it correctly predicted the opponent's move.

  void ponderhit() {

    int t = current_search_time();
    PonderSearch = false;
    if (Iteration >= 3 &&
       (!InfiniteSearch && (StopOnPonderhit ||
                            t > AbsoluteMaxSearchTime ||
                            (RootMoveNumber == 1 &&
                             t > MaxSearchTime + ExtraSearchTime && !FailLow) ||
                            (!FailHigh && !FailLow && !fail_high_ply_1() && !Problem &&
                             t > 6*(MaxSearchTime + ExtraSearchTime)))))
      AbortSearch = true;
  }


  // print_current_line() prints the current line of search for a given
  // thread.  Called when the UCI option UCI_ShowCurrLine is 'true'.

  void print_current_line(SearchStack ss[], int ply, int threadID) {

    assert(ply >= 0 && ply < PLY_MAX);
    assert(threadID >= 0 && threadID < ActiveThreads);

    if (!Threads[threadID].idle)
    {
        lock_grab(&IOLock);
        std::cout << "info currline " << (threadID + 1);
        for (int p = 0; p < ply; p++)
            std::cout << " " << ss[p].currentMove;

        std::cout << std::endl;
        lock_release(&IOLock);
    }
    Threads[threadID].printCurrentLine = false;
    if (threadID + 1 < ActiveThreads)
        Threads[threadID + 1].printCurrentLine = true;
  }


  // init_ss_array() does a fast reset of the first entries of a SearchStack array

  void init_ss_array(SearchStack ss[]) {

    for (int i = 0; i < 3; i++)
    {
        ss[i].init(i);
        ss[i].initKillers();
    }
  }


  // wait_for_stop_or_ponderhit() is called when the maximum depth is reached
  // while the program is pondering.  The point is to work around a wrinkle in
  // the UCI protocol:  When pondering, the engine is not allowed to give a
  // "bestmove" before the GUI sends it a "stop" or "ponderhit" command.
  // We simply wait here until one of these commands is sent, and return,
  // after which the bestmove and pondermove will be printed (in id_loop()).

  void wait_for_stop_or_ponderhit() {

    std::string command;

    while (true)
    {
        if (!std::getline(std::cin, command))
            command = "quit";

        if (command == "quit")
        {
            Quit = true;
            break;
        }
        else if (command == "ponderhit" || command == "stop")
            break;
    }
  }


  // idle_loop() is where the threads are parked when they have no work to do.
  // The parameter "waitSp", if non-NULL, is a pointer to an active SplitPoint
  // object for which the current thread is the master.

  void idle_loop(int threadID, SplitPoint* waitSp) {
    assert(threadID >= 0 && threadID < THREAD_MAX);

    Threads[threadID].running = true;

    while(true) {
      if(AllThreadsShouldExit && threadID != 0)
        break;

      // If we are not thinking, wait for a condition to be signaled instead
      // of wasting CPU time polling for work:
      while(threadID != 0 && (Idle || threadID >= ActiveThreads)) {
#if !defined(_MSC_VER)
        pthread_mutex_lock(&WaitLock);
        if(Idle || threadID >= ActiveThreads)
          pthread_cond_wait(&WaitCond, &WaitLock);
        pthread_mutex_unlock(&WaitLock);
#else
        WaitForSingleObject(SitIdleEvent[threadID], INFINITE);
#endif
      }

      // If this thread has been assigned work, launch a search
      if(Threads[threadID].workIsWaiting) {
        Threads[threadID].workIsWaiting = false;
        if(Threads[threadID].splitPoint->pvNode)
          sp_search_pv(Threads[threadID].splitPoint, threadID);
        else
          sp_search(Threads[threadID].splitPoint, threadID);
        Threads[threadID].idle = true;
      }

      // If this thread is the master of a split point and all threads have
      // finished their work at this split point, return from the idle loop.
      if(waitSp != NULL && waitSp->cpus == 0)
        return;
    }

    Threads[threadID].running = false;
  }


  // init_split_point_stack() is called during program initialization, and
  // initializes all split point objects.

  void init_split_point_stack() {
    for(int i = 0; i < THREAD_MAX; i++)
      for(int j = 0; j < MaxActiveSplitPoints; j++) {
        SplitPointStack[i][j].parent = NULL;
        lock_init(&(SplitPointStack[i][j].lock), NULL);
      }
  }


  // destroy_split_point_stack() is called when the program exits, and
  // destroys all locks in the precomputed split point objects.

  void destroy_split_point_stack() {
    for(int i = 0; i < THREAD_MAX; i++)
      for(int j = 0; j < MaxActiveSplitPoints; j++)
        lock_destroy(&(SplitPointStack[i][j].lock));
  }


  // thread_should_stop() checks whether the thread with a given threadID has
  // been asked to stop, directly or indirectly.  This can happen if a beta
  // cutoff has occured in thre thread's currently active split point, or in
  // some ancestor of the current split point.

  bool thread_should_stop(int threadID) {
    assert(threadID >= 0 && threadID < ActiveThreads);

    SplitPoint* sp;

    if(Threads[threadID].stop)
      return true;
    if(ActiveThreads <= 2)
      return false;
    for(sp = Threads[threadID].splitPoint; sp != NULL; sp = sp->parent)
      if(sp->finished) {
        Threads[threadID].stop = true;
        return true;
      }
    return false;
  }


  // thread_is_available() checks whether the thread with threadID "slave" is
  // available to help the thread with threadID "master" at a split point.  An
  // obvious requirement is that "slave" must be idle.  With more than two
  // threads, this is not by itself sufficient:  If "slave" is the master of
  // some active split point, it is only available as a slave to the other
  // threads which are busy searching the split point at the top of "slave"'s
  // split point stack (the "helpful master concept" in YBWC terminology).

  bool thread_is_available(int slave, int master) {
    assert(slave >= 0 && slave < ActiveThreads);
    assert(master >= 0 && master < ActiveThreads);
    assert(ActiveThreads > 1);

    if(!Threads[slave].idle || slave == master)
      return false;

    if(Threads[slave].activeSplitPoints == 0)
      // No active split points means that the thread is available as a slave
      // for any other thread.
      return true;

    if(ActiveThreads == 2)
      return true;

    // Apply the "helpful master" concept if possible.
    if(SplitPointStack[slave][Threads[slave].activeSplitPoints-1].slaves[master])
      return true;

    return false;
  }


  // idle_thread_exists() tries to find an idle thread which is available as
  // a slave for the thread with threadID "master".

  bool idle_thread_exists(int master) {
    assert(master >= 0 && master < ActiveThreads);
    assert(ActiveThreads > 1);

    for(int i = 0; i < ActiveThreads; i++)
      if(thread_is_available(i, master))
        return true;
    return false;
  }


  // split() does the actual work of distributing the work at a node between
  // several threads at PV nodes.  If it does not succeed in splitting the
  // node (because no idle threads are available, or because we have no unused
  // split point objects), the function immediately returns false.  If
  // splitting is possible, a SplitPoint object is initialized with all the
  // data that must be copied to the helper threads (the current position and
  // search stack, alpha, beta, the search depth, etc.), and we tell our
  // helper threads that they have been assigned work.  This will cause them
  // to instantly leave their idle loops and call sp_search_pv().  When all
  // threads have returned from sp_search_pv (or, equivalently, when
  // splitPoint->cpus becomes 0), split() returns true.

  bool split(const Position& p, SearchStack* sstck, int ply,
             Value* alpha, Value* beta, Value* bestValue, const Value futilityValue,
             const Value approximateEval, Depth depth, int* moves,
             MovePicker* mp, int master, bool pvNode) {

    assert(p.is_ok());
    assert(sstck != NULL);
    assert(ply >= 0 && ply < PLY_MAX);
    assert(*bestValue >= -VALUE_INFINITE && *bestValue <= *alpha);
    assert(!pvNode || *alpha < *beta);
    assert(*beta <= VALUE_INFINITE);
    assert(depth > Depth(0));
    assert(master >= 0 && master < ActiveThreads);
    assert(ActiveThreads > 1);

    SplitPoint* splitPoint;
    int i;

    lock_grab(&MPLock);

    // If no other thread is available to help us, or if we have too many
    // active split points, don't split.
    if(!idle_thread_exists(master) ||
       Threads[master].activeSplitPoints >= MaxActiveSplitPoints) {
      lock_release(&MPLock);
      return false;
    }

    // Pick the next available split point object from the split point stack
    splitPoint = SplitPointStack[master] + Threads[master].activeSplitPoints;
    Threads[master].activeSplitPoints++;

    // Initialize the split point object
    splitPoint->parent = Threads[master].splitPoint;
    splitPoint->finished = false;
    splitPoint->ply = ply;
    splitPoint->depth = depth;
    splitPoint->alpha = pvNode? *alpha : (*beta - 1);
    splitPoint->beta = *beta;
    splitPoint->pvNode = pvNode;
    splitPoint->bestValue = *bestValue;
    splitPoint->futilityValue = futilityValue;
    splitPoint->approximateEval = approximateEval;
    splitPoint->master = master;
    splitPoint->mp = mp;
    splitPoint->moves = *moves;
    splitPoint->cpus = 1;
    splitPoint->pos.copy(p);
    splitPoint->parentSstack = sstck;
    for(i = 0; i < ActiveThreads; i++)
      splitPoint->slaves[i] = 0;

    // Copy the current position and the search stack to the master thread
    memcpy(splitPoint->sstack[master], sstck, (ply+1)*sizeof(SearchStack));
    Threads[master].splitPoint = splitPoint;

    // Make copies of the current position and search stack for each thread
    for(i = 0; i < ActiveThreads && splitPoint->cpus < MaxThreadsPerSplitPoint;
        i++)
      if(thread_is_available(i, master)) {
        memcpy(splitPoint->sstack[i], sstck, (ply+1)*sizeof(SearchStack));
        Threads[i].splitPoint = splitPoint;
        splitPoint->slaves[i] = 1;
        splitPoint->cpus++;
      }

    // Tell the threads that they have work to do.  This will make them leave
    // their idle loop.
    for(i = 0; i < ActiveThreads; i++)
      if(i == master || splitPoint->slaves[i]) {
        Threads[i].workIsWaiting = true;
        Threads[i].idle = false;
        Threads[i].stop = false;
      }

    lock_release(&MPLock);

    // Everything is set up.  The master thread enters the idle loop, from
    // which it will instantly launch a search, because its workIsWaiting
    // slot is 'true'.  We send the split point as a second parameter to the
    // idle loop, which means that the main thread will return from the idle
    // loop when all threads have finished their work at this split point
    // (i.e. when // splitPoint->cpus == 0).
    idle_loop(master, splitPoint);

    // We have returned from the idle loop, which means that all threads are
    // finished. Update alpha, beta and bestvalue, and return.
    lock_grab(&MPLock);
    if(pvNode) *alpha = splitPoint->alpha;
    *beta = splitPoint->beta;
    *bestValue = splitPoint->bestValue;
    Threads[master].stop = false;
    Threads[master].idle = false;
    Threads[master].activeSplitPoints--;
    Threads[master].splitPoint = splitPoint->parent;
    lock_release(&MPLock);

    return true;
  }


  // wake_sleeping_threads() wakes up all sleeping threads when it is time
  // to start a new search from the root.

  void wake_sleeping_threads() {
    if(ActiveThreads > 1) {
      for(int i = 1; i < ActiveThreads; i++) {
        Threads[i].idle = true;
        Threads[i].workIsWaiting = false;
      }
#if !defined(_MSC_VER)
      pthread_mutex_lock(&WaitLock);
      pthread_cond_broadcast(&WaitCond);
      pthread_mutex_unlock(&WaitLock);
#else
      for(int i = 1; i < THREAD_MAX; i++)
        SetEvent(SitIdleEvent[i]);
#endif
    }
  }


  // init_thread() is the function which is called when a new thread is
  // launched.  It simply calls the idle_loop() function with the supplied
  // threadID.  There are two versions of this function; one for POSIX threads
  // and one for Windows threads.

#if !defined(_MSC_VER)

  void *init_thread(void *threadID) {
    idle_loop(*(int *)threadID, NULL);
    return NULL;
  }

#else

  DWORD WINAPI init_thread(LPVOID threadID) {
    idle_loop(*(int *)threadID, NULL);
    return NULL;
  }

#endif

}