Co8 5.0.0 Beta - The Big 3; A poll

If I understood them rightly, Cerulean's concerns were with how many flags etc were being stored in memory, or some such thing: too many would cause a crash, he thought.

I guess as the years go by and everyone's gear gets better with more RAM, it becomes less of an issue :shrug: Hard to tell if any given crash is caused by that, anyway. But folks are still getting through to the end without it beoming unplayable.

Speaking of C-Blue, I staggered onto the first page of this thread by mistake... :flamed:

 

Re: 5.0.x Cumulative Mods & Fixes (CMF) Bug Reports and Discussion Thread

OKAY ... of course I've got all sorts of problems with this again. :blegh:

Firstly, the testing conditions: as long as I'm not violating any rules I already identified (like the 'don't put similarly worded py files in the scr folder or the scripts won't compile' rule, which I spent another 3 hours agonizing over again today until I figured it out a second time), I've got 3 saves that work with the last py00334generic_spawner I published - one for the cave exit, one for the courtyard, and one for the back door. And I confirmed Sitra's attachee.destroy() problem. If you exit the Moathouse to any of those three areas before killing Lareth, then re-enter, kill Lareth, and re-exit, the ambush won't fire. This is probably why we've seen these mysterious "I still don't get the ambush!" reports.

Now the new problems: (cur_map == xxxx) is a deal-breaker for me. The moment I add that to an otherwise working script, the ambush is no-go. Likewise for the script modification of separating the 283 flag check as the last 'if' condition independantly and adding the 'else' condition for attachee.destroy(). That renders the ambush inoperable in my otherwise working saves.

Now the new proposed solution: I tested the original scripts with attachee.destroy() simply removed from the code, and everything works. The ambush does not repeat itself or anything like that. So ... being as I never really understood what attachee.destroy() did before anyway, is it absolutely neccesary to the script? If not, apparently removing it alone would solve the leave-come back-leave problem.

[edit]

I see one problem with this already - if you exit into the courtyard and the bandits are there, the script apparently repeats itself. As soon as there are no bandits around, the Big 3 appear out of thin air. This is kind of cool in a way actually (although it would be better if they walked in), but unfortunately they fight with the remaining bandits (those further out from the door).

[another edit]

Okay, a suitable workaround seems to be to move the attachee.destroy() command to correspond to the ambush command, as opposed to the last set of 'if' conditions. Doing so makes the ambush occur if you leave-come back-leave as described above, and it takes care of the courtyard bandits problem too. Presumably, the script does not self destruct until the ambush actually occurs.

The code looks like this:

Code:

	if (attachee.map == 5091):	# Moathouse Cave Exit
		game.areas[8] = 1
		if not (is_follower(8002) or is_follower(8004) or is_follower(8005) or is_follower(8010) or game.party_alignment == LAWFUL_EVIL):
			if ((game.global_flags[44] != 1) and (game.global_flags[45] != 1) and (game.global_flags[700] != 1) and (game.global_flags[37] == 1) and (game.global_flags[283] == 0)):
				Big_Three_Ambush_1(attachee)
				attachee.destroy()

	if (attachee.map == 5002):	# Moathouse Courtyard & Back Door
        	for leader in game.party:
            		xx, yy = location_to_axis(leader.location)
            		if (xx == 462) and (yy == 447):		# Courtyard
               			banditcheck = find_npc_near(attachee,14070)
                		if (banditcheck == OBJ_HANDLE_NULL):
                    			if not (is_follower(8002) or is_follower(8004) or is_follower(8005) or is_follower(8010) or game.party_alignment == LAWFUL_EVIL):
                        			if ((game.global_flags[44] != 1) and (game.global_flags[45] != 1) and (game.global_flags[700] != 1) and (game.global_flags[37] == 1) and (game.global_flags[283] == 0)):
                            				Big_Three_Ambush_2(attachee)
							attachee.destroy()
            		elif (xx == 437) and (yy == 445):	# Back Door
                		if not (is_follower(8002) or is_follower(8004) or is_follower(8005) or is_follower(8010) or game.party_alignment == LAWFUL_EVIL):
                    			if ((game.global_flags[44] != 1) and (game.global_flags[45] != 1) and (game.global_flags[700] != 1) and (game.global_flags[37] == 1) and (game.global_flags[283] == 0)):
                        			Big_Three_Ambush_3(attachee)
						attachee.destroy()

 

Re: 5.0.x Cumulative Mods & Fixes (CMF) Bug Reports and Discussion Thread

Without studying anything in detail, I shall randomly throw out solutions :wave:

I have never heard of cur_map, I think it is something Sitra made up to make life easier in his scripts, so if you are not importing his scipts it won't work. Stick to attachee.map like he said and it should be fine.

Attachee.destroy in this case destroys the spawner so it can never set another big 3 ambush again. It probably does nothing because before it fires it checks for a bunch of flags that presumably change during the course of the big 3 spawning firing. I would leave it there as a cleanup, to be thorough.

EDIT: Ahhh, its breaking the script. Hmmm... then there must be more conditional things in the Big_Three_Ambush_1(attachee) part since it is still deciding whether or not to fire in there? Put it at the end of each of the 3 ambush scripts intead then, after they have definitely fired.

This sounds like a lot of hassle but it should be bullet-proof when you are finished :thumbsup:

 

Forgot to mention - this has been re-added to the CMf, in the form I posted above. We'll see if it works properly in wider circulation.