🚀 go-pugleaf

I'd like to make some suggestions regarding Gnuchess. Some days ago I posted
a request in gnu.chess for a executable version of Gnuchess 3.0 for DOS.
(I'd don't have a C complier, though we are supposed to be getting one soon..)
It seems to me that it would be useful to include this in the distribution
of Gnuchess, since it would save time for those who wish to run it only on a
PC. Compatibility is not a problem and the 'exe' file could perhaps be encoded
to ascii.

The development of Gnuchess seems to be increasing at a rapid rate. We've gone
from 1.55 (the version I have working on a Pyramid) in January to 3.0 in March!
While this is a healthy sign of support, it is not always clear what the
significance of the changes are. Here I'm thinking of hash tables, 'memset'
options etc. It would be useful to have some detailed knowledge of these changes
without having to read the source code. (In any case, I'm strictly a FORTRAN
number-cruncher and don't know anything about C!) Incidently, version 3.0
won't compile properly 'as-is' on the Pyramid, presumably because of the
'memset' feature.

Also, there is the tricky question of the strength of the program. One
of the things I'd like to do is to test the program (on a PC) on the set of
test problems from Pierre Nolot as discussed in "Computer Chess Reports"
in the 88-89 edition. Timings for commercial programs are included, and the
comparison would be interesting. It would also be useful to compare Gnuchess
with the the commercial PC programs 'Chessmaster 2100' (which I have) and the
newer Rexchess.

Finally, to all those people who have contributed to the program - thanks!!