Looking for modified Phalanx22 source to compile under GCC f

[Andrew Tanner]

Geschrieben von: / Posted by: Andrew Tanner at 22 June 2000 17:37:47:

[Dann Corbit]

Geschrieben von: / Posted by: Dann Corbit at 23 June 2000 09:19:14:
Als Antwort auf: / As an answer to: Looking for modified Phalanx22 source to compile under GCC for windows -nt- (no text) geschrieben von: / posted by: Andrew Tanner at 22 June 2000 17:37:47:
http://www.crosswinds.net/~dobes/phalanx/
or
ftp://38.168.214.175/pub/
are the best places I know to look. If you can't find what you want there, it probably does not exist.


my ftp site

[Pete Galati]

Geschrieben von: / Posted by: Pete Galati at 23 June 2000 17:44:26:
Als Antwort auf: / As an answer to: [lookie here] geschrieben von: / posted by: Dann Corbit at 23 June 2000 09:19:14:

http://www.crosswinds.net/~dobes/phalanx/
or
ftp://38.168.214.175/pub/
are the best places I know to look. If you can't find what you want there, it probably does not exist.

The one on Dusan's site compiles just fine with djgpp. I didn't have time to figure out the makefile, so I compiled it like this:
gcc *.c -o phal.exe
seems to be up and running:
Phalanx XXII
primary book ./pbook.phalanx, 48749 bytes
secondary book ./sbook.phalanx, 259074 bytes
learning file ./learn.phalanx, 262144 bytes
eco file ./eco.phalanx, 171176 bytes
tellics set 1 Phalanx XXII, 820 kB hashtable, 47/253 kB P/S opening book
[ white, 1 ]
e4
your move is Pe2-e4
my move is Pe7-e6
1. ... e7e6
[ white, 2 ]
d4
your move is Pd2-d4
my move is Pd7-d5
2. ... d7d5
[ white, 3 ]

[Andrew Tanner]

Geschrieben von: / Posted by: Andrew Tanner at 23 June 2000 21:38:18:
Als Antwort auf: / As an answer to: Re: [lookie here] geschrieben von: / posted by: Pete Galati at 23 June 2000 17:44:26:
I tried many times compiling the source on Dusan's site with Cygnus but I haven't tried DJGPP. Thanks for the tip.

http://www.crosswinds.net/~dobes/phalanx/
or
ftp://38.168.214.175/pub/
are the best places I know to look. If you can't find what you want there, it probably does not exist.

The one on Dusan's site compiles just fine with djgpp. I didn't have time to figure out the makefile, so I compiled it like this:
gcc *.c -o phal.exe
seems to be up and running:
Phalanx XXII
primary book ./pbook.phalanx, 48749 bytes
secondary book ./sbook.phalanx, 259074 bytes
learning file ./learn.phalanx, 262144 bytes
eco file ./eco.phalanx, 171176 bytes
tellics set 1 Phalanx XXII, 820 kB hashtable, 47/253 kB P/S opening book
[ white, 1 ]
e4
your move is Pe2-e4
my move is Pe7-e6
1. ... e7e6
[ white, 2 ]
d4
your move is Pd2-d4
my move is Pd7-d5
2. ... d7d5
[ white, 3 ]

[Andrew Tanner]

Geschrieben von: / Posted by: Andrew Tanner at 23 June 2000 22:53:35:
Als Antwort auf: / As an answer to: Re: [lookie here] geschrieben von: / posted by: Pete Galati at 23 June 2000 17:44:26:
I've now tried downloading Dusan's latest source and also the latest version of DJGPP's gcc (1-14-00) and using the same command : gcc *.c -o phalanx.exe gives me pages of errors such as: search.c:510 'ipom' undeclared
phalanx.h:37 parse error before 'int 64' etc. What version of Djgpp are you using?

[Pete Galati]

Geschrieben von: / Posted by: Pete Galati at 24 June 2000 02:58:10:
Als Antwort auf: / As an answer to: Re: [lookie here] geschrieben von: / posted by: Andrew Tanner at 23 June 2000 22:53:35:

I've now tried downloading Dusan's latest source and also the latest version of DJGPP's gcc (1-14-00) and using the same command : gcc *.c -o phalanx.exe gives me pages of errors such as: search.c:510 'ipom' undeclared
phalanx.h:37 parse error before 'int 64' etc. What version of Djgpp are you using?

Actually, it's probably the most recent, I'm not sure how to check, it's dated 9-24-99.
A second look at the Phalanx I compiled that way indicated that it wasn't really working once it got past the opening book.
So I messed with the makefile a bit, and it looked like this:

CC = gcc
####### choose the line that makes the binary faster on your machine
# CFLAGS = -O3 -Wall -fomit-frame-pointer -funroll-loops
CFLAGS = -O -Wall -fomit-frame-pointer
####### debug/tuning options for developers
# CFLAGS = -O -Wall -g3 -static
# CFLAGS = -O -Wall -pg
#######
### DEFINES
### -DSHOW_FORCED_MOVES
-DPBOOK_FILE=\"pbook.phalanx\"
-DSBOOK_FILE=\"sbook.phalanx\"
-DLEARN_FILE=\"learn.phalanx\"
### -DPBOOK_DIR=\"/usr/local/lib\"
### -DSBOOK_DIR=\"/usr/local/lib\"
### -DLEARN_DIR=\"/var/local/lib\"
### -DQCAPSONLY
(the above 3 lines were uncommented)
# DEFINES = -DGNUFUN
LDFLAGS =
OBJ = .o/phalanx.o .o/bcreate.o .o/search.o .o/io.o .o/data.o \
.o/evaluate.o .o/genmoves.o .o/moving.o .o/hash.o .o/static.o \
.o/levels.o .o/book.o .o/killers.o .o/endgame.o .o/learn.o
phalanx: .o $(OBJ)
$(CC) $(CFLAGS) $(DEFINES) $(LDFLAGS) $(OBJ) -o phalanx
.o/%.o: makefile %.c
$(CC) $(CFLAGS) $(DEFINES) -c $*.c -o .o/$*.o
.o:
mkdir .o
clean:
rm -rf .o phalanx
backup:
tar -czvf Archive/phalanx.tgz \
makefile *.c *.h pbook.phalanx sbook.phalanx test.fin
One of the things that allowed it to compile I think was commenting out the line that said:
DEFINES = -DGNUFUN
because that caused phalanx.h to include a header file that I don't have.
But I think it's still not working properly. Sorry. But probably one of the keys to making it work is messing around with the makfile and compiling it with "make"
Pete