Changelog for crafty-23.4-1.173.x86_64.rpm :

* Wed Jul 24 2013 asterios.dramisAATTgmail.com- Update to 23.4:
* Bug in hash implementation fixed. It was possible to get a hash hit at ply=2, and alter the ply=1 PV when it should be left with no change. This could cause Crafty to display one move as best and play another move entirely. Usually this move was OK, but it could, on occasion, be an outright blunder. This has been fixed. The search.c code has been re-written to eliminate SearchRoot() entirely, which simplifies the code. The only duplication is in SearchParallel() which would be messier to eliminate. Ditto for quiesce.c, which now only has Quiesce() and QuiesceEvasions(). The old QuiesceChecks() has been combined with Quiesce, again to eliminate duplication and simplify the code. Stupid bug in code that handles time-out. One place checked the wrong variable and could cause a thread to attempt to output a PV after the search was in the process of timing out. That thread could back up an incomplete/bogus search result to the root in rare occasions, which would / could result in Crafty kibitzing one move but playing a different move. On occasion, the move played could be a horrible blunder. Minor bug caused HashStore() to lose a best move when overwriting an old position with a null-move search result. Minor change by Tracy to lazy evaluation cutoff to speed up evaluation somewhat. Old \"Trojan Horse\" attack detection was removed. At today\'s depths, it was no longer needed. New hash idea stores the PV for an EXACT hash entry in a separate hash table. When an EXACT hit happens, this PV can be added to the incomplete PV we have so far so that we have the exact path that leads to the backed up score. New \"phash\" (path hash) command to set the number of entries in this path hash table. For longer searches, a larger table avoids path table collisions which will produce those short paths that end in . If appears in too many PVs, phash should be increased. The \"Trojan Horse\" code has been completely removed, which also resulted in the removal of the last bit of \"pre-processing code\" in preeval.c, so it has been completely removed as well. Current search depths are such that the Trojan Horse code is simply not needed any longer. A minor bug in TimeSet() fixed where on rare occasions, near the end of a time control, time_limit could be larger than the max time allowed (absolute_time_limit). We never check against this limit until we exceed the nominal time limit. Cleanup on the draw by repetition code to greatly simplify the code as well as speed it up. From 23.3:
* Null-move search restriction changed to allow null-move searches at any node where the side on move has at least one piece of any type. Minor bug in ValidMove() fixed to catch \"backward\" pawn moves and flag them as illegal. This sometimes caused an error when annotating games and annotating for both sides. The killer move array could have \"backward\" moves due to flipping sides back and forth, which would cause some odd PV displays. Small change to \"reduce-at-root\". We no longer reduce moves that are flagged as \"do not search in parallel\". Check extension modified so that if \"SEE\" says the check is unsafe, the extension is not done. We found this to be worth about +10 Elo. We also now reduce any capture move that appears in the \"REMAINING_MOVES\" phase since they can only appear there if SEE returns a score indicating loss of material. We now reduce a bit more aggressively, reducing by` 2 plies once we have searched at least 4 moves at any ply. I tried going to 3 on very late moves, but could not find any case where this was better, even limiting it to near the root or other ideas. But reducing by 2 plies after the at least 2 moves are searched was an improvement. Minor change is that the first move is never reduced. It is possible that there is no hash move, no non-losing capture, and no killer move. That drops us into the REMAINING_MOVES phase which could reduce the first move searched, which was not intended. Very minor tweak, but a tweak all the same.- Removed crafty-misc.tar.bz2 tarball. + Custom files for compiling/using crafty were added as separate Source: files (medtosmall.txt was removed since an upstream book.bin file is provided, so there is no need to compile one). + Files that can be downloaded from the official site were added in a new crafty-data.tar.bz2 tarball (bitmaps.tar.bz2 was merged in this tarball, added sound.zip which contains wav files for crafty, book.bin was added as a separate Source: since it is usually updated with each new release). + A script was added (update_crafty-data.sh) to easily update files in crafty-data.tar.bz2 using wget.- Updated README.SuSE (from old crafty-misc.tar.bz2) to reflect the current status (mainly no automatic creation of user config/book files). Also renamed it to README.openSUSE.- Updated crafty.rc (from old crafty-misc.tar.bz2) to fix tbpath and egtb.- Updated crafty.6 (from old crafty-misc.tar.bz2) to fix paths and to reflect the current status (mainly no automatic creation of user config/book files).- Updated make_books script (from old crafty-misc.tar.bz2) to fix compilation of book
*.bin files and remove the compilation from medtosmall.txt).- Rebased and updated crafty-23.2-Makefile.patch to keep libnuma requirement, remove -pg flag (incompatible with -fomit-frame-pointer from rpm optflags) and to use rpm optflags. Renamed it also to Makefile.patch.- Rebased and updated crafty-23.2-paths.patch to apply cleanly and to fix some more paths. Also renamed it to paths.patch.- Added a patch (numacompat1.patch) to add a missing include for numacompat1.h (fixes compilation with libnuma for x86_64).- Added a patch (ix86.patch) to fix compilation in ix86 (based on patch from Archlinux).- Added libnuma-devel as build requirement for x86_64.- Recommend xboard package which can be used by xcrafty.- Suggest esound-utils which can be used by crafty-speak.
* Tue Jun 08 2010 larsAATTlinux-schulserver.de- re-take package from packman- use spec-cleaner to cleanup specfile and add tags to patches- use fdupes
* Sat Mar 06 2010 packmanAATTlinks2linux.de - 23.2- updated to 23.2
* This release includes some Xboard compatibility updates as well as a small ELO increase over version 23.1 (about +10 ELO)- patches refreshed
* Fri Nov 20 2009 AxelKoellhoferAATTweb.de - 23.1- updated to 23.1
* Sun Aug 23 2009 AxelKoellhoferAATTweb.de - 23.1- modified script \"xcrafty\" to use a smaller boardsize
* Fri Aug 14 2009 AxelKoellhoferAATTweb.de - 23.1- added patch to set some paths according to the installed package crafty will now read /etc/crafty.rc and use books, endgame tables as well as the help file from /usr/share/crafty if no user-specific $HOME/.craftyrc and $HOME/.crafty/ exist- added smallbook.bin
* Mon Aug 10 2009 AxelKoellhoferAATTweb.de - 23.0- updated to 23.0- added new (and compatible) book.bin- changed make_books to build a bookc.bin file (computer opening book)- added minimal /etc/crafty.rc example-file- updated crafty-Makefile.patch for building with openSUSE >= 11.2 (thx to gentoo upstream for the decisive hint)- added security-203541.patch, see http://bugs.debian.org/203541 for more information
* Mon Sep 22 2008 larsAATTlinux-schulserver.de- moved to Education base repository
* Tue Jun 10 2008 larsAATTlinux-schulserver.de- compile on amd64- add numactl only on ix86 machines
* Wed Jun 04 2008 larsAATTlinux-schulserver.de- remove numactl from buildrequires