[gnugo-devel] Manual page revised

[Gunnar Farneback]

This patch revises the GNU Go manual page. Essentially it removes a
lot of obsolete options but also a couple of debugging options which
should be of marginal interest to people who read the page. There
probably is some information which would be worth adding to the page
as well, but this patch only tries to reduce the amount of
desinformation.

I have only included the diff of doc/gnugo.pod below. The patch also
affects the generated file doc/gnugo.6.

/Gunnar

Index: doc/gnugo.pod
===================================================================
RCS file: /cvsroot/gnugo/gnugo/doc/gnugo.pod,v
retrieving revision 1.3
diff -u -r1.3 gnugo.pod
--- doc/gnugo.pod       6 Sep 2001 21:34:20 -0000       1.3
+++ doc/gnugo.pod       28 Aug 2002 19:22:50 -0000
@@ -10,13 +10,10 @@
 [B<--color <color>>]
 [B<--handicap <num>>]
 [B<--komi <num>>]
-[B<-p, --playstyle <style>>]
 [B<--quiet>]
 [B<-v, --version>]
 [B<-h, --help>]
-[B<--help analyze>]
 [B<--help debug>]
-[B<--help playstyle>]
 [B<--copyright>]
 [B<--mode <mode>>]
 [B<--replay <color>>]
@@ -26,21 +23,15 @@
 [B<--printsgf <filename>>]
 [B<-D, --depth <num>>]
 [B<-B, --backfill_depth <num>>]
-[B<--score [end|last|move]> ]
+[B<--score [estimate|finish|aftermath]> ]
 [B<-a, --allpats>]
 [B<-T, --printboard>]
 [B<-d, --debug <level>>]
-[B<-H, --hash <level>>]
 [B<-w, --worms>]
 [B<-m, --moyo <level>>]
 [B<-b, --benchmark num>]
-[B<-s, --stack>]
-[B<-S, --statistics>]
 [B<-t, --trace>]
 [B<-r, --seed num>]
-[B<--decidestring <string>>]
-[B<--analyzerfile <filename>>]
-[B<--analyze <options>>]
 
 =head1 DESCRIPTION
 
@@ -55,7 +46,7 @@
 The game of Go Go is a game of strategy between two players usually played on a
 19x19 grid called B<goban>. The two players put black and white B<stones> on
 the goban to enclose B<territory>. Go was invented about 4000 years ago in
-ancient China, other names for this game are (Chinese) B<Wei Chi>, (Korean)
+ancient China. Other names for this game are (Chinese) B<Wei Chi>, (Korean)
 B<Baduk> and (Ing) B<Goe>.
 
 =head2 Playing a game in ASCII mode
@@ -74,11 +65,11 @@
 graphic display. To use GNU Go under X-windows, obtain the most recent 
 version of CGoban from Bill Shubert's web site
 
-L<http://www.inetarena.com/~wms/comp/cgoban/>
+http://www.igoweb.org/~wms/comp/cgoban/index.html
 
 Start CGoban. When the CGoban Control panel comes up, select `Go Modem.'
 You will get the Go Modem Protocol Setup. Choose one (or both) of the
-playsers to be ``Program,'' and fill out the box to the path to
+players to be ``Program,'' and fill out the box to the path to
 gnugo. After clicking OK, you get the Game Setup window. Choose
 ``Rules Set'' to be Japanese (otherwise handicaps won't work). Set the
 board size and handicap if you want. Click OK and you are ready to go.
@@ -86,7 +77,7 @@
 In the Go Modem Protocol Setup window, when you specify the path
 to GNU Go, you can give it command line options, such as --quiet
 to suppress most messages. Since the Go Modem Protocol preempts
-standard I/O other messages are sent to stderr, even if they are
+standard I/O, other messages are sent to stderr, even if they are
 not error messages. These will appear in the terminal from which
 you started CGoban. 
 
@@ -113,7 +104,7 @@
 
 =head2 Documentation
 
-The files in the gnugo-3.0/doc directory contain detailed documentation about
+The files in the F<doc> directory contain detailed documentation about
 debugging options and internal program structure. Other documentation may
 be found in comments throughout the source code.
 
@@ -140,7 +131,7 @@
 Hollosi developed the current standard, which may be found
 at 
 
-L<http://www.sbox.tu-graz.ac.at/home/h/hollosi/sgf/>
+http://www.red-bean.com/sgf/
 
 GNU Go supports the Smart Go Format.
 
@@ -168,7 +159,7 @@
 
 B<-L, --until I<move>>
 
-Stop loading just before I<move> is played (eg L10).
+Stop loading just before I<move> is played (e.g. 154 or L10).
 
 B<-o, --outfile I<file>>
 
@@ -178,7 +169,7 @@
 
 B<--boardsize I<num>>
 
-Set the board size to use (3-21). Default is 19, other common formats are
+Set the board size to use (5-21). Default is 19, other common formats are
 13 and 9.
 
 B<--color I<color>>
@@ -205,83 +196,28 @@
 
 Display help message.
 
-B<--help analyze>
-
-Display help about analyzer options.
-
 B<--help debug>
 
 Display help about debugging options. 
 
-B<--help playstyle>
-
-Display help about playstyle options. 
-
 B<--copyright>
 
 Display copyright notice.
 
 =head2 Debugging and advanced options:
 
-B<-a, --allpats>
-
-Try all patterns, even those valued less than the best move
-found so far. This causes GNU Go to run more slowly with no
-improvement in strength but is useful for developers.
-
 B<-T, --printboard>
 
 Show board each move.
 
-B<-d, --debug I<level>>
-
-Debugging output (see liberty.h for bits).
-
-B<-D, --depth I<num>>
-
-Deep reading cutoff (default 14; higher=stronger, slower).
-
-B<-B, --backfill_depth I<num>>
-
-Deep reading cutoff (default 8; higher=stronger, slower).
-
-B<-H, --hash I<level>>
+B<--level I<num>>
 
-Hash (see liberty.h for bits). 
-
-B<-w, --worms>
-
-Worm debugging.
-
-B<-m, --moyo I<level>>
-
-Moyo debugging, show moyo board. The parameter is an integer,
-which can be a sum of the following possibilities (e.g. B<-m> 9
-to get territory evaluation+delta_moyo values). See docs/MOYO
-for definitions. Use color rxvt window or linux console for 
-color diagrams.
-
-    1 = territory (5/21 algorithm)
-    2 = table of delta_terri values
-    4 = moyo evaluation (5/10 algorithm)
-    8 = table of delta_moyo values
-   16 = area (4/0 algorithm)
-   32 = list of area characteristics
-   64 = table of meta_connect values
-  128 = trace -p fearless option 
+Level of play. (default 10; smaller=faster, weaker).
 
 B<-b, --benchmark I<num>>
 
 Benchmarking mode - can be used with B<-l>.
 
-B<-s, --stack>
-
-Stack trace (for debugging purposes).
-
-B<-S, --statistics>
-
-Print statistics (for debugging purposes).
-
 B<-t, --trace>
 
 Verbose tracing (use twice or more to trace reading).
@@ -290,127 +226,36 @@
 
 Set random number seed.
 
-B<--decidestring I<string>>
-
-I<string> is the coordinate of one stone in the string.
-Determine whether the string can be captured, and if so,
-whether it can be defended. Use with B<-t> or B<-o>. See docs/READING
-for suggestions about using --decidestring.
-
-B<--score [I<end|last|move>]>
+B<--score [I<estimate|finish|aftermath>]>
 
 Count or estimate territory of the input file. Usage:
 
-B<gnugo --score last -l filename>
-
-Loads the SGF file and estimates the winner after the last stored move by
-measuring the influence.
+B<gnugo --score estimate -l filename>
 
-B<gnugo --score end -l filename>
+Loads the SGF file and estimates the score by measuring the
+influence. Use with B<-L> if you want the estimate somewhere else than
+at the end of the file.
 
-Loads the SGF file and gnugo continues to play after the last stored move by
-itself up to the very end. Then the winner is determined by counting the
-territory.
+B<gnugo --score finish -l filename>
 
-B<gnugo --score L10 -l filename>
+Loads the SGF file and gnugo continues to play by itself up to the
+very end. Then the winner is determined by counting the territory.
 
-Loads the SGF file until a stone is placed on L10. Now the winner will be
-estimated as with gnugo --score last.
+B<gnugo --score aftermath -l filename>
 
-B<gnugo --score 100 -l filename>
-
-Loads the SGF file until move number 100. Now the winner will be estimated as
-with gnugo B<--score last>.
+Similar to B<--score finish> except that a more accurate but slower
+algorithm is used to determine the final status of the groups.
 
 If the option B<-o outputfilename> is provided, 
 the results will also be written as comment at the end of the output file.
 
-If the option B<--analyzerfile outputfilename> is provided, the results
-will be written as comment at the end of the output file, the result property
-will be set and the territory will be marked.
-
 B<--printsgf I<outfile>>
 
 Load SGF file, output final position (requires B<-l>).
 
-B<--analyzerfile I<name>>
-
-Filenname for analyzer output.
-
-B<--analyze I<options>>
-
-Analyze a game stored as sgf file by using --testmode. Game tree variations
-are will be ignored. This option may also be used with --score or --benchmark.
-The analyze functions will be executed on every move in --benchmark
-and --testmode game.
-
-Analyzed board states on other modes:
-
-B<--score end:>
-
-gnugo analyzes every move it makes at the end of the file until the game is
-finished.
-
-B<--score last:>
-
-board state at the end of the file will be analyzed
-
-B<--score <movenum>:>
-
-board state just before <movenum> will be analyzed
-
-B<--score <position>:>
-
-board state just before <position> is occupied will be analyzed
-
-B<--testmode annotation:>
-
-board state just before the annotated node is reached will be analyzed.
-
-The results of the analyze functions are written to the filename provided by
---analyzerfilename.
-
-
-Current options are :
-
- areacolor     show color of influence 
- capture       show capturing moves
- considered    show considered moves
- defense       show defending moves
- dragoninfo    print info about all dragons
- dragonstatus  show dragonstatus the dragons are are marked on
-               the board with the status: aLive, Dead, Critial, Unknown
- eyeinfo       print info about all eyes
- eyes          show eyes and their vital points
- moyocolor     show color of moyos
- moyovalue     show moyo values 
- overwrite     overwrites game info from the inputfile
- recommended   show moves recommend by modules show moves recommended
-               by the various modules:
-
- A = attacker
- B = very big move
- D = defender
- E = eyefinder
- F = fuseki
- L = liberty filling
- P = pattern
- p = reconsidered pattern
- S = semeai
-
- terricolor    show color of territories (moyo)
- territory     show territory (worms)
- terrivalue    show territory values (moyo)
- worminfo      print info about all worms
- wormliberties show liberties of the worm
-
-You may use "option1 option2 ..." or option1,option2,... to specifiy more than
-one option for --analyze. The analyzer functions will work in all playing
-modes except ASCII and GMP. Testmode GAME does not test variations when
-invoked with --analyze.
-
 =head1 BUGS
 
 If you find a bug, please send the SGF output file to address@hidden
+together with a description of the bug.
 
 =cut