This is the mail archive of the
xconq7@sources.redhat.com
mailing list for the Xconq project.
Re: Exploration patch
- To: Jim Kingdon <kingdon at panix dot com>
- Subject: Re: Exploration patch
- From: Hans Ronne <hronne at pp dot sbbs dot se>
- Date: Wed, 1 Aug 2001 11:45:36 +0200
- Cc: xconq7 at sources dot redhat dot com
>Without the enclosed patch, exploration in the standard game ends
>around turn 40 or 60 or something. I guess another solution to
>see-terrain-if-captured would be to just set it to zero (if one gets
>info on a unit, one gets info on the terrain it sits on). I've played
>a few games with this patch and it seems to work well for me.
>
>Comments?
The exploration bug should be fixed, I agree, but I'm not sure this is the
best way. If I capture an enemy unit with this patch, what I get to see is
a patchwork of seen and unseen hexes that are scattered all over the map.
Not very pretty, and doesn't make much sense either. A better fix along
these lines would be to have a make-terrain like algorithm that generates
one single random sized area of visible hexes that nucleates at the
captured unit.
Or to make it really simple, make all hexes within a fixed or randomly
determined radius from the captured unit visible!
However, I suspect the real problem here is need_explorers:
if (numcontacted == 0) {
/* If we've haven't found anybody, always explore. */
return TRUE;
} else if (numfound == 0) {
/* If we've made contact but haven't found their home base,
we still need to explore, but not so much. */
return probability(50);
} else if (numfound < numsides - 1) {
/* If we haven't found everybody's home base, we still need
to have a few units continuing to explore. */
return probability(10);
} else {
/* If everybody has been found, then we likely have more
pressing concerns; don't do more exploration. */
return FALSE;
}
I think exploration should continue also after all enemy bases have been
found. And dropping it to just 10% after the first enemy base has been
found seems a little drastic, particularly in game with many sides.
Perhaps you could test some other percentages in need_explorers and see if
they solve the problem.
Hans
Hans Ronne
hronne@pp.sbbs.se