]> git.sesse.net Git - stockfish/history - src
Precalculate FutilityMargins
[stockfish] / src /
2010-01-28 Joona KiiskiPrecalculate FutilityMargins
2010-01-28 Joona KiiskiUse calculate_reduction() function to simplify code
2010-01-28 Joona KiiskiBugfix: reduction was not set to zero in full depth...
2010-01-28 Joona KiiskiImplement calculate_reduction function
2010-01-27 Joona KiiskiStandardize set_option function
2010-01-27 Joona KiiskiDo not initialize RootPosition at startup
2010-01-27 Marco CostalbaAspiration window rewrite
2010-01-27 Marco CostalbaBe sure we exit while loop with lock held
2010-01-27 Joona KiiskiFix capture pruning
2010-01-27 Marco CostalbaSimplify locking in sp_search and sp_search_pv
2010-01-27 Marco CostalbaTemporary revert "captures pruning" due to an assert
2010-01-27 Marco CostalbaAdded some FIXME to track needed tests
2010-01-27 Marco CostalbaIntegrate gains table in History
2010-01-27 Marco CostalbaIntroduce update_gains() and refactor some code
2010-01-27 Joona KiiskiFix some silly bugs
2010-01-27 Joona KiiskiMaxGain based futility pruning for captures
2010-01-27 Joona KiiskiMaxGain based pruning
2010-01-27 Joona KiiskiImplement post futility pruning
2010-01-27 Joona KiiskiCollect MaxGain statistics
2010-01-27 Joona KiiskiImplement MaxGain table
2010-01-27 Joona KiiskiRemove pointless gcc flag when generating dependencies
2010-01-27 Joona KiiskiRemove InfiniteSearch hack
2010-01-27 Joona KiiskiRemove last use of uip.eof()
2010-01-26 Marco CostalbaReduce lock contention in sp_search_pv()
2010-01-25 Marco CostalbaFix a possible crash in thread_is_available()
2010-01-25 Marco CostalbaSmall split() cleanup
2010-01-25 Marco CostalbaPrune evasions with negative SEE in qsearch
2010-01-25 Marco CostalbaAvoid copy a Position to get a move's san notation
2010-01-25 Marco CostalbaFix a race in idle_loop() exiting
2010-01-25 Marco CostalbaIn split() release the lock before slow search stack...
2010-01-25 Marco CostalbaDo not copy master position in split()
2010-01-25 Marco CostalbaUse fast_copy() instead of full copy in sp_search
2010-01-25 Marco CostalbaBetter document how Position c'tor works
2010-01-25 Marco CostalbaFix a couple of MSVC casting warnings
2010-01-25 Marco CostalbaCopy only the search stack tail in split()
2010-01-24 Tord RomstadMerge branch 'master' of ssh://free2.projectlocker...
2010-01-24 Tord RomstadFixes a Chess960 bug when playing with more than one...
2010-01-24 Marco CostalbaCheck for thread creation successful completion
2010-01-24 Marco CostalbaFix some races in SMP code
2010-01-22 Tord RomstadMake sure we make a move at the end of the search when...
2010-01-20 Marco CostalbaIf near beta generate checks at -OnePly
2010-01-20 Marco CostalbaRetire LMRPVMoves and LMRNonPVMoves
2010-01-19 Marco CostalbaFix enum Value issue with gcc 4.4
2010-01-19 Marco CostalbaSmall lnArray[] cleanup
2010-01-19 Marco CostalbaSilence some silly MSVC warnings
2010-01-19 Marco CostalbaOrder check moves used in qsearch
2010-01-18 Marco CostalbaAvoid an useless evaluate() call
2010-01-18 Marco CostalbaAllow SearchStack to link an EvalInfo object
2010-01-18 Joona KiiskiMake reduction search code SMP-friendly
2010-01-17 Joona KiiskiSmall cleanup of unused code in sp_search
2010-01-17 Marco CostalbaFix silly MSVC warning
2010-01-17 Marco CostalbaInitialize futilityMargin in EvalInfo c'tor
2010-01-17 Joona KiiskiRetire quick_evaluate()
2010-01-17 Joona KiiskiRazor at depth one, but do razoring only when not in...
2010-01-13 Marco CostalbaAllow negative history values
2010-01-11 Marco CostalbaStore node evaluation in SearchStack
2010-01-11 Marco CostalbaDecrease NullMoveMargin and adjust razoring
2010-01-11 Marco CostalbaDo not wait when AbortSearch is set
2010-01-10 Marco CostalbaFix sending of best move during an infinite search
2010-01-09 Marco CostalbaFix threads count setting
2010-01-08 Marco CostalbaOptimal tune for 8 cores
2010-01-08 Marco CostalbaSync qsearch with search
2010-01-08 Marco CostalbaUse full evaluation in null search
2010-01-07 Joona KiiskiSlow down reductions
2010-01-07 Joona KiiskiUse logarithmic LMR also at root
2010-01-07 Joona KiiskiLogarithmic LMR
2010-01-07 Marco CostalbaFix 'position ..... moves ' parsing bug
2010-01-06 Marco CostalbaFix en-passant parsing from fen string
2010-01-05 Marco CostalbaIntroduce refine_eval()
2010-01-05 Marco CostalbaIncrease null reduction at high depths
2010-01-04 Marco CostalbaFix a compile error under gcc
2010-01-04 Marco CostalbaLast round of search.cpp cleanup
2010-01-03 Marco CostalbaSpace inflate bottom part of search.cpp
2010-01-03 Marco CostalbaRename MaxActiveSplitPoints
2010-01-03 Marco CostalbaRetire 'finished' from MovePicker
2010-01-03 Marco CostalbaRename SingleReplyExtension in SingleEvasionExtension
2010-01-03 Marco CostalbaSmall cleanup in search.cpp
2010-01-02 Marco CostalbaRetire approximateEval field from SplitPoint
2009-12-31 Marco CostalbaDouble HistoryMax and reduce aging
2009-12-30 Marco CostalbaFix a little warning under gcc compiler
2009-12-30 Marco CostalbaOptimized bitScanReverse32()
2009-12-30 Marco CostalbaRestore development version
2009-12-30 Marco CostalbaStockfish 1.6.2
2009-12-30 Marco CostalbaBetter fix for gcc optimization issue
2009-12-27 Marco CostalbaScore enum should be at least 32 bits
2009-12-27 Marco CostalbaRevert small pop_1st_bit() optimization
2009-12-27 Marco CostalbaRemove a bogus assert
2009-12-27 Marco CostalbaUse THREAD_MAX instead of hardcoded 8
2009-12-27 Marco CostalbaRestore development version
2009-12-26 Marco CostalbaSet maximum hash table size to 2GB
2009-12-26 Marco CostalbaStockfish 1.6.1
2009-12-26 Marco CostalbaWorkaround optimization bug in gcc
2009-12-26 Marco CostalbaFix description of Score enum
2009-12-26 Marco CostalbaFix a typo in ReducedStateInfo
2009-12-22 Marco CostalbaStockfish 1.6
2009-12-21 Marco CostalbaScore definition gives a compile error under gcc
2009-12-19 Marco CostalbaSetup Release Candidate 1
2009-12-19 Marco CostalbaFix a comment in HistoryMax description
2009-12-18 Marco CostalbaFix book name is hard coded as book.bin
2009-12-15 Marco CostalbaIn non-PV IID don't call evaluate when in check