This engine features the following:

• It's just Min-Max with care taken in state evaluation

• After each depth is completed, makes an estimate of time needed to complete the next depth. Does not start next depth if estimated completion time exceeds predetermined limit.
• Estimate assumes exponential growth.

• Store Score, Depth, Move, Node Type (exact, upper bound, lower bound)
• Stored in a large array used as a hash table
• Sized via the frown test to achieve an acceptable collision rate

• Random seed unoptimized

• Offsets generated with a greedy algorithm

• Material Evaluation with standard values
• Piece Square State Evaluation
• Draws valued 1 centipawn higher than being checkmated
• Stalemate detection
• 100 ply capture / pawn move detection
• Threefold Board State Repetition detection
• Opponent-initiated draws valued same as being checkmated. Some of y'all lack threefold detection and blunder your way into a tie. If I refuse to let myself get into a situation where you have the power to make that blunder, I get fewer ties.

• Metadata cached into 32 bit integer, accelerating move application
• Flag available for "Best Possible Move", triggering immediate move send
• Lazy inCheck filtering
• Moves stored in custom "Framed Stack" data structure to maximize locality

• PV Move first, if valid
• Move from Transposition Table next, if available
• Then Castling Moves
• Then Captures, in order of Least Valuable Attacker / Most Valuable Target
• Finally quiet moves, ordered by Piece Square differential

• Reduce depth of search of quiet moves late in the move order.

http://www.chess2u.com/t7448-komodo-variety-opening-book-komodo-polyglot-book

• Reads Polyglot format
• Fitness Proportional move selection from entries in book
• Competing with Komodo Variety book

• Output each time alpha increases at root node
• Output a variety of statistics at end of each game

• Searcher runs on seperate thread from Time Limiter
• Thread synchronization via Rendezvous construct from Plan 9

• Insufficient Material detection
• Lazy move generation
• Board::isValidMove(pvMove) to check it prior to move generation
• Endgame tablebase

• History Table, didn't help
• Quiescent Search, underperformed
• Pondering, y'all are too unpredictable for pondering to work