From: Mark Wooding <mdw@distorted.org.uk>
Date: Sun, 28 May 2017 08:25:19 +0000 (+0100)
Subject: math/limlee.c: Fiddle with the large-prime heuristics.
X-Git-Tag: 2.4.2~5
X-Git-Url: http://www.chiark.greenend.org.uk/ucgi/~mdw/git/catacomb/commitdiff_plain/2427b43e44e51b1c5f5e00c95cd08ee84b788b05?hp=6a2654167a4c6ece96c96a6a235d805ca21661da

math/limlee.c: Fiddle with the large-prime heuristics.

Experimentation shows me that large primes just don't last very long
with the old heuristic.  Switch to another one that's more tolerant but
still weeds out no-hopers.
---

diff --git a/math/limlee.c b/math/limlee.c
index 13bbc212..fbed738a 100644
--- a/math/limlee.c
+++ b/math/limlee.c
@@ -225,7 +225,7 @@ static int next(int rq, pgen_event *ev, limlee_stepctx *l)
 
     if (l->qq.p) {
       dist = l->u.s.disp < 0 ? -l->u.s.disp : l->u.s.disp;
-      if (dist && dist > l->u.s.steps/dist) {
+      if (dist && dist > l->u.s.steps/3) {
 	l->pops->pfree(&l->qq, l);
 	l->qq.p = 0;
       }