playout.c

   1 #define DEBUG
   2 #include <assert.h>
   3 #include <math.h>
   4 #include <stdio.h>
   5 #include <stdlib.h>
   6 #include <string.h>
   7
   8 #include "board.h"
   9 #include "debug.h"
  10 #include "engine.h"
  11 #include "move.h"
  12 #include "ownermap.h"
  13 #include "playout.h"
  14
  15 /* Whether to set global debug level to the same as the playout
  16  * has, in case it is different. This can make sure e.g. tactical
  17  * reading produces proper level of debug prints during simulations.
  18  * But it is safe to enable this only in single-threaded instances! */
  19 //#define DEBUGL_BY_PLAYOUT
  20
  21 #define PLDEBUGL(n) DEBUGL_(policy->debug_level, n)
  22
  23
  24 int
  25 play_random_game(struct playout_setup *setup,
  26                  struct board *b, enum stone starting_color,
  27                  struct playout_amafmap *amafmap,
  28                  struct board_ownermap *ownermap,
  29                  struct playout_policy *policy)
  30 {
  31         assert(setup && policy);
  32
  33         int gamelen = setup->gamelen - b->moves;
  34         if (gamelen < 10)
  35                 gamelen = 10;
  36
  37         if (policy->setboard)
  38                 policy->setboard(policy, b);
  39 #ifdef DEBUGL_BY_PLAYOUT
  40         int debug_level_orig = debug_level;
  41         debug_level = policy->debug_level;
  42 #endif
  43
  44         enum stone color = starting_color;
  45
  46         int passes = is_pass(b->last_move.coord) && b->moves > 0;
  47
  48         while (gamelen-- && passes < 2) {
  49                 coord_t coord;
  50                 coord = policy->choose(policy, b, color);
  51
  52                 if (is_pass(coord)) {
  53 play_random:
  54                         /* Defer to uniformly random move choice. */
  55                         /* This must never happen if the policy is tracking
  56                          * internal board state, obviously. */
  57                         assert(!policy->setboard);
  58                         board_play_random(b, color, &coord, (ppr_permit) policy->permit, policy);
  59
  60                 } else {
  61                         struct move m;
  62                         m.coord = coord; m.color = color;
  63                         if (board_play(b, &m) < 0) {
  64                                 if (PLDEBUGL(4)) {
  65                                         fprintf(stderr, "Pre-picked move %d,%d is ILLEGAL:\n",
  66                                                 coord_x(coord, b), coord_y(coord, b));
  67                                         board_print(b, stderr);
  68                                 }
  69                                 goto play_random;
  70                         }
  71                 }
  72
  73 #if 0
  74                 /* For UCT, superko test here is downright harmful since
  75                  * in superko-likely situation we throw away literally
  76                  * 95% of our playouts; UCT will deal with this fine by
  77                  * itself. */
  78                 if (unlikely(b->superko_violation)) {
  79                         /* We ignore superko violations that are suicides. These
  80                          * are common only at the end of the game and are
  81                          * rather harmless. (They will not go through as a root
  82                          * move anyway.) */
  83                         if (group_at(b, coord)) {
  84                                 if (DEBUGL(3)) {
  85                                         fprintf(stderr, "Superko fun at %d,%d in\n", coord_x(coord, b), coord_y(coord, b));
  86                                         if (DEBUGL(4))
  87                                                 board_print(b, stderr);
  88                                 }
  89                                 return 0;
  90                         } else {
  91                                 if (DEBUGL(6)) {
  92                                         fprintf(stderr, "Ignoring superko at %d,%d in\n", coord_x(coord, b), coord_y(coord, b));
  93                                         board_print(b, stderr);
  94                                 }
  95                                 b->superko_violation = false;
  96                         }
  97                 }
  98 #endif
  99
 100                 if (PLDEBUGL(7)) {
 101                         fprintf(stderr, "%s %s\n", stone2str(color), coord2sstr(coord, b));
 102                         if (PLDEBUGL(8))
 103                                 board_print(b, stderr);
 104                 }
 105
 106                 if (unlikely(is_pass(coord))) {
 107                         passes++;
 108                 } else {
 109                         /* We don't care about nakade counters, since we want
 110                          * to avoid taking pre-nakade moves into account only
 111                          * if they happenned in the tree before nakade nodes;
 112                          * but this is always out of the tree. */
 113                         if (amafmap) {
 114                                 if (amafmap->map[coord] == S_NONE || amafmap->map[coord] == color)
 115                                         amafmap->map[coord] = color;
 116                                 else if (amafmap->record_nakade)
 117                                         amaf_op(amafmap->map[coord], +);
 118                                 amafmap->game[amafmap->gamelen].coord = coord;
 119                                 amafmap->game[amafmap->gamelen].color = color;
 120                                 amafmap->gamelen++;
 121                                 assert(amafmap->gamelen < sizeof(amafmap->game) / sizeof(amafmap->game[0]));
 122                         }
 123
 124                         passes = 0;
 125                 }
 126
 127                 if (setup->mercymin && abs(b->captures[S_BLACK] - b->captures[S_WHITE]) > setup->mercymin)
 128                         break;
 129
 130                 color = stone_other(color);
 131         }
 132
 133         float score = board_fast_score(b);
 134         int result = (starting_color == S_WHITE ? score * 2 : - (score * 2));
 135
 136         if (DEBUGL(6)) {
 137                 fprintf(stderr, "Random playout result: %d (W %f)\n", result, score);
 138                 if (DEBUGL(7))
 139                         board_print(b, stderr);
 140         }
 141
 142         if (ownermap)
 143                 board_ownermap_fill(ownermap, b);
 144
 145         if (b->ps)
 146                 free(b->ps);
 147
 148 #ifdef DEBUGL_BY_PLAYOUT
 149         debug_level = debug_level_orig;
 150 #endif
 151
 152         return result;
 153 }