Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!wuarchive!usc!randvax!ucla-an!stb!michael
From: michael@stb.uucp (Michael Gersten)
Newsgroups: comp.sources.games.bugs
Subject: Re: Conquer bug or just annoying change?
Keywords: conquer, movement, nomads, zone of control
Message-ID: <1989Dec12.045829.5237@stb.uucp>
Date: 12 Dec 89 04:58:29 GMT
References: <844@wet.UUCP>
Reply-To: michael@stb.uucp (Michael Gersten)
Organization: The Serial Tree BBS, +1 213 397 3137
Lines: 102

There is an ****ANNOYING**** "feature" of conquer version 4 that when
an NPC nation -- including nomads, savages, and lizards -- goes to run
its turn, it gets the present state of the board to work from, not the
PREVIOUS state of the board which a user gets. This makes it impossible
to pin anybody because no matter what you did, they can always move out.

Put another way, if you want to move to place A but an army in B is in
front of A, you will stop at B. Whether or not army B is there to stop
you when your move executes, you will stop at B. That NPC army B may
move out before you get there.

Now turn it around. If you move your army B out of the way, the NPC's
can now move straight to your army A.

I made a simple solution that solves PART of this for nomads, savages,
and lizards. Since most of those armies are running around in attack
mode looking for something to attack, and since the combat code insures
that a combat will occur if one army is in attack mode, I had the
nomad/lizard/savage movement code test to see if a combat would occur
first before they move. As distributed there is a 0% -- yes, 0%-- chance
of a battle occuring in the sector that they start in because they
are guaranteed to leave to a different sector.

Here's the code I changed. Its pretty simple. Put it in npc.c

/* Determine if a fight will occur in sector x,y. Approximate--must be
	an attacking army (of a real country) in the sector.
	Currently used for savages etc. FOR NPCs DISPLOMACY MUST BE
	ACCOUNTED FOR. Check for ispc not isactive -- although npc's
	should stop us as well, nomads and savages do not.
	This is NOT used for pirates -- they are navies, not armies.
*/
fight_scene (x,y, stat)
{
	int country, armynum;
	for (country=0; country<NTOTAL; country++)
		if (ispc(ntn[country].active))
			/* Taken from combat() -- can combat occur? */
			/* for(j=0;j<MAXARM;j++) {
			/* 	aptr = &nptr->arm[j];
			/* 	if((aptr->sold>0)
			/* 	&&(aptr->stat>=ATTACK)
			/* 	&&(aptr->stat<=SORTIE||aptr->stat>=NUMSTATUS)
			/* 	&&(!fought[aptr->xloc][aptr->yloc])){ */
			/*	}} */
			/* Also check if this army is attacking */

			for (armynum=0; armynum<MAXARM; armynum++)
/* Same location */		if (AXLOC == x && AYLOC == y)
				{
/**
printf ("Country %d, army %d in sector with someone\n", country, armynum);
printf ("Country's status is %d, others are %d\n", ASTAT, stat);
printf ("(Attack is %d, sortie is %d)\n", ATTACK, SORTIE);
**/
  /* They are attacking */	if (
				       (  ASTAT>=ATTACK && ASOLD>0
					  && (ASTAT <= SORTIE || ASTAT>=NUMSTATUS)
				       )
  /* Or we are attacking */	   || (stat >= ATTACK && stat <= SORTIE)
				   )
				{
/**
printf ("Fight occurs\n");
**/
					return 1;
				} else {
/**
printf ("No fight ");
**/
					return 0;
				} /* if in attack mode */
				} /* if x and y */
	return 0;
}

The "No fight" section occurs if there are no armies in attack mode in
the sector -- it turns out this only occurs with lizard armies in garison
mode (some of which may choose to go out and attack neighboring sectors)

Then, in do_savage, do_nomad, and do_lizard, the following change

void
do_savage()
{
	short armynum;
	int x, y;

	printf("updating savage (nation %d)\n",country);
	for(armynum=0;armynum<MAXARM;armynum++) if(P_ASOLD>0){
		P_ASTAT=ATTACK;
		if(P_ATYPE<MINLEADER) {
			P_ASOLD *= 102;
			P_ASOLD /= 100;
		}
-->>	/* If combat they do not move */
-->>		if (fight_scene (P_AXLOC, P_AYLOC, P_ASTAT))
-->>			continue;
		P_AMOVE=(curntn->maxmove * *(unitmove+P_ATYPE%UTYPE))/10;


	Michael