Re: [gnugo-devel] more owl revisions

[SP LEE]

Hi Paul,

There are already many patches and I have the difficulty to patch the
related source files. Can I download the latest source files from Savannah
CVS site?

SP Lee

----- Original Message -----
From: "Paul Pogonyshev" <address@hidden>
To: <address@hidden>
Sent: Tuesday, May 06, 2003 9:56 PM
Subject: [gnugo-devel] more owl revisions


> this patch retires "If eyespace includes an owl inessential string, we
must
> assume that the pessimistic min is 0" policy. it must have no regression
> breakage (didn't have a while ago) and give a slight decrease in owl
nodes.
>
> Paul
>
>
> Index: engine/owl.c
> ===================================================================
> RCS file: /cvsroot/gnugo/gnugo/engine/owl.c,v
> retrieving revision 1.156
> diff -u -p -r1.156 owl.c
> --- engine/owl.c 4 May 2003 11:39:48 -0000 1.156
> +++ engine/owl.c 6 May 2003 21:47:33 -0000
> @@ -2604,12 +2604,10 @@ owl_determine_life(struct local_owl_data
>     || (mw[pos] < 3 * mz[pos] && mz[pos] > 5))
>   pessimistic_min = 0;
>
> +      /* It appears that this policy is needed no longer. */
> +#if 0
>        /* If this eyespace includes an owl inessential string, we must
assume
> -       * assume that the pessimistic min is 0.
> -       *
> -       * See owl1:304 for an example where this policy is important.
> -       * FIXME: However, a better approach would be to improve
inessential
> -       * strings detection (S13 in owl1:304 must count as essential).
> +       * that the pessimistic min is 0.
>         */
>        if (pessimistic_min > 0) {
>   for (pos2 = BOARDMIN; pos2 < BOARDMAX; pos2++) {
> @@ -2621,6 +2619,7 @@ owl_determine_life(struct local_owl_data
>     }
>   }
>        }
> +#endif
>
>        eyes_attack_points[num_eyes] = NO_MOVE;
>        eyevalue_list[num_eyes] = eyevalue;
> @@ -3758,10 +3757,10 @@ owl_reasons(int color)
>   * dies because the victim only formed a nakade shape.
>   *
>   * FIXME: This code overlaps heavily with some code in
> -    * examine_move_safety() in move_reasons.c. The caching
> -    * scheme should minimize the performance hit, but of course
> -    * it's unfortunate to have the code duplication.
> -    */
> + *   examine_move_safety() in move_reasons.c. The caching
> + *   scheme should minimize the performance hit, but of course
> + *   it's unfortunate to have the code duplication.
> + */
>   int move = DRAGON2(pos).owl_attack_point;
>
>   /* No worries if we catch something big. */
>
>
> _______________________________________________
> gnugo-devel mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/gnugo-devel
>