factor = 0 not actually 0...??

[pigglet82]

To keep track of how the AI choose in this event I set two variables in a previous event -- optionone & optiontwo -- and assigned them the value of zero. Since Denmark is catholic, the AI should never choose the first option. But when looking at the variables in the save file, optionone had the value 2 while optiontwo had the value 24. How come the AI choose the first option twice? I thought factor = 0 meant an option never triggered.

country_event = {

	id = 828282

	trigger = {
		tag = DAN
		has_global_flag = inittest
	}

	mean_time_to_happen = { days = 1 }

	title = "titletest"
	desc = "desctest"

	option = {
		name = "nametest"
		ai_chance = {
			factor = 90
			days = 1
			modifier = {
				factor = 0
				religion = catholic
			}
		}
		change_variable = { which = optionone value = 1 }
	}
	option = {
		name = "nametest"
		ai_chance = {
			factor = 10
			days = 1
			modifier = {
				factor = 10
				religion = catholic
			}
		}
		change_variable = { which = optiontwo value = 1 }
	}
}

[Darken]

well, IMO, option one should never be chosen by a catholic nation so this must either be a bug, or perhaps denmark recently became catholic and the data has not updated yet?

this should be posted as a bug report IMO because I realy doubt its WAD.

[pigglet82]

well, IMO, option one should never be chosen by a catholic nation so this must either be a bug, or perhaps denmark recently became catholic and the data has not updated yet?

Well this occured in 1399 and Denmark is catholic from the start so that shouldn't be it.

this should be posted as a bug report IMO because I realy doubt its WAD.

It certainly looks like a bug, but before I report anything I'll do a few more test-runs. Just in case it depends on some mistake of mine (not that unlikely )

[Helius]

Zero has never meant zero.

Before ai=yes/no triggers, Magna Mundi used an event to set a specific flag to identify countries as player-controlled.
Sometimes, the ai would choose that option.

[pigglet82]

Zero has never meant zero.

Before ai=yes/no triggers, Magna Mundi used an event to set a specific flag to identify countries as player-controlled.
Sometimes, the ai would choose that option.

Ok, thanks for clarifying that (sort of... since I'm still confused as to why zero don't mean zero )

[Helius]

What we tell the AI are merely suggestions. Zero is a very strong suggestion (24/26 in your case), but still just a suggestion.

[Haftetavenscrap]

Is having a modifier inside an ai_chance even legal syntax? It never worked for me.

[TheLand]

What we tell the AI are merely suggestions. Zero is a very strong suggestion (24/26 in your case), but still just a suggestion.

24/26 is not that different to 9/10... so perhaps factor=0 is not much different from factor=0.1?

[pigglet82]

What we tell the AI are merely suggestions. Zero is a very strong suggestion (24/26 in your case), but still just a suggestion.

I thought the suggestion approach applied to any other value -- that's logic -- but that zero actually ment zero. I'm no mathematical genius but I know what happens when multiplying with zero I just can't see the logic behind PI having it work this way.

[pigglet82]

Is having a modifier inside an ai_chance even legal syntax? It never worked for me.

It seems to work fine for me in the tests I've done. I can also tell you after a lot of testing that it doesn't work within random_list (which would have made my life so much easier...)

[pigglet82]

24/26 is not that different to 9/10... so perhaps factor=0 is not much different from factor=0.1?

Maybe. If I find the time, I'll try that test event again for a longer run, once with modifier zero and once with 0.1 and see if I get any significant difference.

[Helius]

Is having a modifier inside an ai_chance even legal syntax? It never worked for me.

Absolutely! SRI and MMP events use it, all the time (and have been using it, since before NA).

[Helius]

I just looked at your code again. Here's something: You no longer need to use "days = 1"! Maybe leaving that out will make a difference?

[pigglet82]

I just looked at your code again. Here's something: You no longer need to use "days = 1"! Maybe leaving that out will make a difference?

Haha... I have never used it before and didn't even know about its existence until dharper enlightened me the other day:

I do it purely because Johan said to. I have no evidence that it matters, but that's how I learned it, so that's what I do.

If days = 1 was used as a ground value which wasn't altered by the modifier, this actually makes sense. Will try without it and see if I get zero for one of the results.

[pigglet82]

If days = 1 was used as a ground value which wasn't altered by the modifier, this actually makes sense. Will try without it and see if I get zero for one of the results.

Nope, no change from what I can see. I did two test runs and the result was 2/168 and 3/201 (they were supposed to stop at a nice and even 100 but I guess my stop catch didn't really work as planned...)

[ADP101]

i know nothing about modding but heres my 2 cents. What if you put in -1?

[Brent15]

I'm going a little off topic here lads, but can anyone tell me why factors for MTTH events use decimal fractions as positive modifiers and anything at or above 1 for negative modifiers?

Does it directly modify the MTTH? The chance that it will fire at the MTTH? Or something else?

Thanks!

[orwell]

50 months x factor .8 = 40
50 months x factor 1.2 = 60
Here's a useful beginner guide to events, and I'm working on a project that will theoretically make it even easier.

[Brent15]

50 months x factor .8 = 40
50 months x factor 1.2 = 60
Here's a useful beginner guide to events, and I'm working on a project that will theoretically make it even easier.

Ah, I see, thats what I thought. A useful guide indeed, thanks man!