gnugo-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[gnugo-devel] GNU Go 3.1.33


From: bump
Subject: [gnugo-devel] GNU Go 3.1.33
Date: Fri, 19 Apr 2002 17:03:24 -0700

GNU Go 3.1.33 is now at http://www.gnu.org/software/gnugo/devel.html
and in the CVS. The regressions should be identical to 3.1.32,
so it is unfortunate that this isn't true for VC, as Trevor reports.

The tarball contains the regressions this time, in preparation
for the 3.2 release.

experimental_reading_1_33.2. Corrects an error in experimental_reading_1_33.1 
arend_1_33.6. TODO redone. 
replacement_1_33.1. Turns off replacement checking for the stable release. 
gunnar_1_33.2. Fixes some warnings on Alpha. 
experimental_reading_1_33.1. Don't compile the experimental code by default. 
arend_1_33.5. Revised normalize float. 
arend_1_33.4. (in part) Improved traces and revised hash_init().
gunnar_1_33.1. Float normalization in mkpat.c. 
score_1_33.1. Fixes --score estimate crash. 
teun_1_33.1a. Shadowed variables fixed in reading.c.
twogtp_1_33.1. Twogtp gets the engine name and version numbers.
arend_1_33.2. Record passes in move list. 
arend_1_33.3. Bugfix in readconnect.c. 

- regression/READING, STRATEGY and NICKLAS removed
- makevcdist.pl, doc/gnugo-32.[jpg|eps]added to EXTRADIST
- eval3.sh and semeai2.tst added to automake control
- TODO list in texinfo updated
- class field in struct pattern and class_mask in made unsigned int
- input_FILE and output_FILE in mkpat.c explicitly initialized
- tree based reading not compiled by default
- consistency checking of replacement patterns disabled for the stable release
- more Handtalk games in CVS
- Jessie Annala THANKed
- new function gg_normalize_float2int to address optics.c-platform dependency 
- hash_init behaviour made independent of boardsize
- output cleanup: --statistics, and EITHER_MOVE-trace
- floats normalized in finish_pattern() in mkpat.c
- put GNU Go's passes in permanent move history
- assertion in compute_connection_distances corrected

Dan



reply via email to

[Prev in Thread] Current Thread [Next in Thread]