Four consecutive fails to imprison someone at 98% chance of success is highly improbable but at least possible. Failing at 100% is just plain wrong though. WTF?
There's something very wrong with percentages in this game
- Thread starter killcreak
- Start date
-
We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
Nope, I doubt anything's wrong at all.
Rounding errors (100% likely technically isn't) and a very unlucky random seed, which isn't going to change until the game ticks. Wait a few days to change the seed.
Rounding errors (100% likely technically isn't) and a very unlucky random seed, which isn't going to change until the game ticks. Wait a few days to change the seed.
I don't know, clockwork, the only two chaplains I've ever had convert to a heresy did it over and over again with me chaining the options to imprison and demand conversion. They had a high Learning score so the mean time to happen was quite long, it's just that the game decided that this would happen numerous times in a not very long time span.
The failure to imprison is almost certainly the stored seed, but how do you explain the Heretic Chaplains? I don't play all that much compared to some people: less than 200 hours.
The failure to imprison is almost certainly the stored seed, but how do you explain the Heretic Chaplains? I don't play all that much compared to some people: less than 200 hours.
Percentages in this game are pre-generated from what it seems... like for instance I have had events pop and it says I have an 80% chance to get my ward a certain trait. I ended up not getting the trait. So I reloaded my save over and over and over, and always got that trait that I didn't want that only had 20%.
Same thing with children... I modded my game so that I can see children's adult portraits soon as there born. The way they look is almost basically predetermined and only differs base on their gender. If you save a game right before the day they are born your child will almost always look exactly the same no matter how many times you reload. If you load from a save 1 month away then you get maybe 1-3 different presets, but it will always be the same exact presets each time.
Also I was scratching my head yesterday when I gave my wife a county... cause I had my feudal taxes on low and so I should be receiving 10% from barons and counts... I gave my wife a county I had just conquered through a holy war so it had -100% taxes and levies, etc... She was giving me 5.4 gold per month... when she was only collecting .45 from being my wife... so that 10% taxes definitely isn't based on what you take in... and it definitely was based on how much gold she had cause had 94 gold... and 10% of that is 9 gold...
Same thing with children... I modded my game so that I can see children's adult portraits soon as there born. The way they look is almost basically predetermined and only differs base on their gender. If you save a game right before the day they are born your child will almost always look exactly the same no matter how many times you reload. If you load from a save 1 month away then you get maybe 1-3 different presets, but it will always be the same exact presets each time.
Also I was scratching my head yesterday when I gave my wife a county... cause I had my feudal taxes on low and so I should be receiving 10% from barons and counts... I gave my wife a county I had just conquered through a holy war so it had -100% taxes and levies, etc... She was giving me 5.4 gold per month... when she was only collecting .45 from being my wife... so that 10% taxes definitely isn't based on what you take in... and it definitely was based on how much gold she had cause had 94 gold... and 10% of that is 9 gold...
Last edited:
From what you've said, it still could very likely be the stored seed. It doesn't update every game day - depending on how long that "not very long time span" was, it could very well have still been the same seed.
What do you mean you doubt it? It's either wrong display or it's wrong calculation.
The seed should be used up as soon as it's called upon to trigger the conversion event, right? Because I did it probably about five times (give or take) to both chaplains: they give me the heretic chaplain event, I imprison then, demand they convert, they do so and are released, then I make them my chaplain again. It's been a while so I can't remember, but it was probably in the week-to-month range before the next heretic chaplain event fired and I repeated the process.
Did you not read anything I said? Everything is perfectly explainable by the fact that computers aren't magic (and the number displayed is explainable by it just not being worth it to clog up the UI with "99.99%" all the time when "100%" is good enough for all but the most improbable scenarios. They round to nearest whole number in the display - that's it).
There is nothing inherent about seeds that requires them to be "used up" when any event fires. It all depends on the implementation of the pseudo-RNG for when the seed is recalculated. And although it could be implemented to recalculate every event, that would be absolutely awful for a game like CK2 performance-wise, because that's an extra, potentially expensive operation that would need to be performed for every single event for every single character. Updating on a given tick is really the only feasible way to handle that for this game.
Keep in mind, I don't actually know when stuff is updated or how the random calculations are actually performed because I didn't make the game. But implementing a fair pseudo-RNG is not hard, and there is absolutely no reason to assume there's anything wrong with it. It's just that pseudo-RNGs have their own constraints and limitations that have to be kept in mind.
No, did you? That something is explainable doesn't make it correct. Do you know the difference between "correct" and "explainable"? I never wrote it cannot be explained, I wrote it's wrong.
Except that nothing's wrong, because it's entirely explainable from inherent limitations in computers that aren't worth programming around.
But its not wrong. 99.50% is 100% with rounding. There is therefore nothing wrong with failing even if there's <= 0.5% chance of it happening.
You finally wrote something that makes sense. Yes, someone thought it's not worth programming around it.
No, you moron, what's wrong is displaying one number while actually using another. Is it so difficult to display "99.50%" as "~100%" giving an indication that there is a failure chance?
Please stop talking, just because people are disagreeing with you doesn't mean you should be patronising them or calling them morons. Your original post and title are a little over the top to start with, and people are bound to come in with competing arguments. Even if you're not wrong, can you not entertain the possibility of it for even a second?
Maybe you should calm down, because your complaint here boils down to you wishing that computers were magic, and now you're just resorting to insulting people (people who clearly understand the issue better than you do). You're also misunderstanding the point of what I said if you think I was saying the developers were simply lazy.
They round to the nearest whole number for the percentages, and you happened to be unlucky enough to get a failure when the rounding said there was 100%. But the thing is, you could potentially run into this exact same problem even if you display far more decimal places. Eventually it will round to 100%, and there could still be a failure chance (and the only way to ensure accuracy - if it's even possible - would likely require a crazy amount of decimal places that nobody wants to see all the time).
So they decided to round to nearest whole numbers when displaying this information because the case where even that low precision gives a noticeably different outcome is so rare that most players will never experience it over hundreds or even thousands of hours of play. It is not important enough to add in extra decimal places that give the player no noticeably useful information for the vast majority of situations just to have slightly fewer rounding errors in order to handle an extremely rare case.
Deal with it.
I'm not talking, I'm writing. How much over the top are my original post and title? 0.5% or 0%?
You have to let the game progress a tick for a random seed thing to re-randomize a result.
Which fixes the OP's issue but doesn't explain the heretic chaplain at all.
Why not? Have you not been reading what I've written either?
It's the seed not updating. When you're dealing with a pseudo-RNG (the "pseudo-" there is important; it's most often left off when talking about RNGs for simplicity's sake, but you really do have to keep in mind that computers aren't capable of generating truly random numbers), the hands down most likely explanation for returning the same, improbable result is that the seed didn't change between when the two results were generated.
You realize that's still a problem though, right? It keeping a seed with the save I understand: it prevents people from simply save-scumming and turning a 20% imprison chance into a 100% chance quite so easily.
It holding onto a seed for months so the same event fires off over and over is just comical. In a game where the RNG determines so much, it shouldn't be so slapdash.