Will this mod ever be updated for WotC? I always loved XCOM, but the RNG aspect of it really destroys the enjoyment of the game for me.
There's a reason why so many people are abusing grenades, combat protocols, Soulfire aim PCS etc: 100% Reliable Damage. All of these things are also why the mid-to-late game is so much easier than the early game: You can analyze your turns in the best possible way with your abilities to kill a whole pod in a single round with minimal RNG factor. But in the early-to-mid game, especially the first 5~ missions? An absolute RNG-fest, just look at Gatecrasher on Legend difficulty.
And even in the end game, if your whole turn is relied upon a 90% shot and it misses, well, say bye to one of your soldiers. People always talk about how if you don't have a "backup-plan" after something doesn't go as you planned, then it's your fault, but these "backup-plans" only work in no-timer missions. In time-based missions you need to rush fast, and especially on legendary- you have so many pods to fight to get to your objective, you need to be absolutely sure that your 80%+ shots hit, cause if they don't, there's hardly any backup plan. You can't abuse overwatch ambushes for pods, you can't always have perfect cover opportunities, you can't kill every single enemy in one round: at the end of the day, it all comes back to RNG.
Not even gonna start ranting about timer-based missions and the dodge mechanic.
WotC is seriously unplayable because of that reason for me, so right now Vanilla with Long War 2 + A Better Advent will do. Great mod and i really hope this will someday get updated.
Btw there's another great mod on steam called "Reliable Damage" which also unfortunetaly only works for vanilla.
If this were updated, it would be the most downloaded mod for this game. I had three agents, on a very early mission, miss on three separate 91-94% chance shots, and one of them was flanking the enemy from with NOTHING between them, meaning he got a massive accuracy bonus on top of him being a sniper in addition to not being sighted yet. The enemy, as expected, hit with only a 67% chance shot and took off half my specialist's health. Hell, I missed on a 95% attack with a 25% sneaking accuracy bonus BACKSTAB on a pitiful sectoid!
Whoever was hired to code the agent's accuracy, they are probably related to that idiot programmer who misspelled 'tether' for Aliens: Colonial Marines.
This. This exactly describes the issue of why plain simple RNG die rolls just isn't enough. It is simply not fun to experience these kind of situations even if they are mathematically accurate; it's just not fun, it's very very unfun. A game should first and foremost be fun.
Due to (as usual) life concerns stealing all my time, I never got to really come back to this mod. I still really don't have the time, however I've been hearing good things about War of the Chosen and might pick it up to give it a play through. If that's the case, I'd really like to have the modified toHit % chances of this mod, so I might try investigating.
Modding XCOM 2 was great in that there was tons of tools/support, but it also mean there was a lot of sunk costs in terms of familiarizing yourself with both the tools/dev environment, but also the inner workings of the game. So it might take equally long to figure out what's changed with War of the Chosen modding, just to make the small changes likely required to get this mod to work.
So long story short, I might see what I can do, but don't hold your breath. If anyone knows of any similar mods that do the same thing, please share as it'd be much easier to just use some other mod then have to remake/fix this one myself
I think there is one major flaw in the RNG system that is XCOM, and it has to do with the lack of positioning awareness the game seems to have. These are trained soldiers, hit chances should be much MUCH higher than what happens. 100% chance hits should be a norm when you are point blank flanking an enemy, again, that should be the norm.
The constant rhetoric we get is that these soldiers are the best of the best, finest troops from around the world... I'd hate to think of anything worse than the rag tag group that is the XCOM squad. When you have an enemy flanked at even a moderate distance, it shouldn't be a matter of whether you hit or miss, it should be a matter of how WELL you hit them. Damage being on a roll would make more sense.
This leads to one change that would solve the error of XCOM RNG. Chance to hit should be mostly dictated by your range, if you're in your effective range you've got a 100% chance to hit, but cover can "absorb" damage and again distance and flanking calculations can determine how much damage and how much "health" the cover has. RNG to dictate just how much damage you do within a few degrees of freedom is all you really need. These soldiers shouldn't be missing their targets, not the amount you see in any XCOM game. But the effectiveness of the hits should be a deciding factor.
This would inevitably up the difficulty of the game, as any soldier you have that isn't 100% covered WOULD get hit, even if just a glancing wound you'd see soldiers piling up in the nurses office with scrapes, burns, and missing limbs. The biggest problem with humans and RNG, is that a physical dice roll is just that, a physical dice roll, humans have nobody to blame but themselves for rolling the dice the way they did, but when you add the computer rolling for you, a sense of "It used loaded dice on me" can seep in, especially when they get unlucky roll after unlucky roll (and the computer opponent is getting lucky roll after lucky roll).
That's my two cents on the situation that is RNG, randomness in videogames is indeed bad design. This includes RNG or Dice.
I hate XCOM's RNG/rolls/whatever you want to call it.
It's just doesn't feel like "true" chance, whether that is manipulated by my own feelings or not, is irrelevant really. Not that any computed roll is fair anyhow but it's definitely one of the LEAST fair in any game I've played so far. For a start, all of the outcomes are predetermined (not a good thing, in my eyes) before you even set foot in a mission and each round has a set of outcomes assigned (that never change, if you perform the same actions and tested by reloading over and over). Sure, you can alter this by executing commands in a different order. Say, you have 50% chance to hit/hack, throw everybody on Overwatch, skip turn or whatever and it can change to another value on the next round. But that's just another predetermined set of values that have been chosen and you are being forced to play with.
I just don't understand why it's done this way. Each shot should be rolled just before you take the shot, not determined before the current round. It's a pretty stupid and unfair method of dice-rolling. After all, the setup for the animation takes what 2-3 seconds so surely a small calculation can be done in that time? Last time I checked, computing happened to be pretty fast.
"I just don't understand why it's done this way. Each shot should be rolled just before you take the shot, not determined before the current round. It's a pretty stupid and unfair method of dice-rolling. After all, the setup for the animation takes what 2-3 seconds so surely a small calculation can be done in that time? Last time I checked, computing happened to be pretty fast."
Typically it is done this way to prevent save scumming. If you haven't heard the term, that is saving and reloading until the desired roll is obtained.
Ideally, what I would have liked to have is a removal of this foolish pre-calculated roll order, so that a 95% is a real 95% roll with a real 5% chance to miss. But with playing with higher difficulty and your mod, the game remained difficult and punishing without loosing a game on a single roll. Thank you a thousand time.
I read that XCOM uses a type of RNG called an LCG, which is prone to produce long streaks of similar numbers. During a full game, these numbers will add up to a correct percentage, but the distribution will be incorrect, which might be what a lot of players experience (beside confirmation bias). I mean: if you flip a coin a million times, you will experience roughly half of them landing as heads, and half as tails. What you will not experience if the first 500k landing as heads, and the last 500k landing as tails, since such a streak of similar outcomes is almost impossible (or rather, so unlikely that a human lifespan is not enough to experience it, even if we flip one coin per second). It's fine to miss one 90% shot, or even two in a row. But myself and many others have experienced missing several 90% shots during one mission, then missing in a similar way next mission, and then again on a third mission (and so on). The probability of so many streaks of similar numbers occuring is frankly very very very low. All evidence I have seen online that the RNG is fair does this: it takes all the shots made and crunches it out into probability, but it never observes the distribution of those shots taken (which is where I think the problem is). I'd think that this is what needs to be adressed. I am not a modder (nor asking you to address any of this). But if the LCG is the problem, then long streaks of similar numbers produced by the LCG is what is needed to be looked at. I'm posting my thoughts here because you seem to be the only one actually interested in modding how the RNG works.
Ty you so much for this mod I was literally about to delete the game as broken after I missed twice point blank range, that's unacceptable idc about mathematical anything, iv'e fired over a thousand rounds in my life, and i have yet in my 34 years ever missed point blank, it's never happened ever, you would literally have to be the suckiest shot on the planet, anyone that disagrees has never fired a gun and they just talking nerd talk. anyways i veer off-topic I just wanted to thank you for unbreaking the game.
Very nice mod, thanks for making it! However from the first look these 75% shots are too much powerful with your mod. For instance - sniper is likely to hit enemy in cover with 75% shot and that is not fair at all. I'd start increasing chances dramatically at 80%+ shots and leave lesser percentage with less Aim bonus. (For instance - 75% shot will have 18% chance to miss instead of 12.5%). I'll see how it works in the game but I am just afraid that I am not likely to miss a shot with this. Anyway that is my opinion. It would be cool if you made configurable file so people can make changes from there. It's a shame I already deleted XCOM2 modding tools, otherwise I could have tried to recompile scripts
It actually is configurable via config file. By default is uses the average sum method as it's a very simple calculation, and was intended more for Enemy Unknown modding, as more complex math was trickier. So I'm using it as a starting point for this mod.
However in the mod's config/ini, you can enable the experimental gaussian/normal distribution, where you can change the parameters of the curve to adjust it more. The suggested values for that right now aim to do exactly what you say, adjust it more at the high end and try to keep it mostly inline at the low end (I target a 65% aim equal to vanilla/default). So give that a try if you want to experiment
Despite that though, from the bit of testing I've done, I think the default average sum numbers end up working out quite well. Still plenty of misses, doesn't make the game feel like a cake walk or anything. I'd imagine it just goes to show you that given % chance probability, it's still really hard for us to get a handle on predicting actual outcomes and results. I remember the feeling from Enemy Unknown where it was "a 90% chance to hit, sure feels a lot higher than seems to play out in the game".
Hi again! So after playing with this mod a while my fears came true - game becomes too easy. I still believe that changes to hit% should apply for 85%+ shots. Now even in the middle of the game I feel sure about my sniper hitting enemy in full cover with about 75% chance (which in my opinion should not happen - I wold not take risk of 25% shot in vanilla but here I know it's 12.5% to miss and I make decision to take the shot lightly instead of blowing up cover/retreating/hnunkering down/using a beacon etc).
I didn't understand what you said about tweaking .ini files. Even if I do tweak them - how will I know what actual hit percentage is? In-game it'll still show the same number. I wish you'd make this mod more configurable if you can.
If the game is feeling too easy, then it may just mean your perceptions of what the shot %'s are, are actually in line with reality. This mod wouldn't need to exist if more of us were like that
That being said I'm working on an update to refine the alternative Gaussian method/ini tweaks. That'll give a chart with the % breakdown, but due to the nature of statistics, it's not something you can't calculate on the fly that well. I may see about making an excel spreadsheet for the next update that can at least let you try out different values for the INI file and see what the resulting shot percentages would be.
Thanks for the feedback. I'll take a look at that and get back to you. This mod only works on the players side though and won't affect alien shots.The Fixed Roll Value is something I used in testing, but I don't believe I did anything that should have stopped it from working. I haven't been able to get back to this in a little bit (planning a small update soon, just have to wait till I get the free time :/) so a game update might have affected the mod in some unforseen way. (Should be pretty safe from game updates though I had thought).
To get a better idea of what is going on in the meantime, you can check out the log file (it's kept in your Documents\My Games\Xcom 2\XcomGame\logs or similar, look for the combat log.
With the mod enabled you will get messages in that log detailing all the combat events (attacks), whether or not they hit miss, what types of hits, and the rolls that you were getting.
So if fixed roll value is being applied you should see the roll being reported as 1.
Also if the mod is enabled the combat log will have entries mentioning FairRNGMod. If you don't see the log or that in there, it's a good sign that the mod isn't enabled/active for some reason.
40 comments
There's a reason why so many people are abusing grenades, combat protocols, Soulfire aim PCS etc: 100% Reliable Damage. All of these things are also why the mid-to-late game is so much easier than the early game: You can analyze your turns in the best possible way with your abilities to kill a whole pod in a single round with minimal RNG factor. But in the early-to-mid game, especially the first 5~ missions? An absolute RNG-fest, just look at Gatecrasher on Legend difficulty.
And even in the end game, if your whole turn is relied upon a 90% shot and it misses, well, say bye to one of your soldiers. People always talk about how if you don't have a "backup-plan" after something doesn't go as you planned, then it's your fault, but these "backup-plans" only work in no-timer missions. In time-based missions you need to rush fast, and especially on legendary- you have so many pods to fight to get to your objective, you need to be absolutely sure that your 80%+ shots hit, cause if they don't, there's hardly any backup plan. You can't abuse overwatch ambushes for pods, you can't always have perfect cover opportunities, you can't kill every single enemy in one round: at the end of the day, it all comes back to RNG.
Not even gonna start ranting about timer-based missions and the dodge mechanic.
WotC is seriously unplayable because of that reason for me, so right now Vanilla with Long War 2 + A Better Advent will do. Great mod and i really hope this will someday get updated.
Btw there's another great mod on steam called "Reliable Damage" which also unfortunetaly only works for vanilla.
Whoever was hired to code the agent's accuracy, they are probably related to that idiot programmer who misspelled 'tether' for Aliens: Colonial Marines.
Too bad it does not work with WoTC :/
Thanks a bunch!
Modding XCOM 2 was great in that there was tons of tools/support, but it also mean there was a lot of sunk costs in terms of familiarizing yourself with both the tools/dev environment, but also the inner workings of the game. So it might take equally long to figure out what's changed with War of the Chosen modding, just to make the small changes likely required to get this mod to work.
So long story short, I might see what I can do, but don't hold your breath. If anyone knows of any similar mods that do the same thing, please share as it'd be much easier to just use some other mod then have to remake/fix this one myself
Pretty good read.
I think there is one major flaw in the RNG system that is XCOM, and it has to do with the lack of positioning awareness the game seems to have. These are trained soldiers, hit chances should be much MUCH higher than what happens. 100% chance hits should be a norm when you are point blank flanking an enemy, again, that should be the norm.
The constant rhetoric we get is that these soldiers are the best of the best, finest troops from around the world... I'd hate to think of anything worse than the rag tag group that is the XCOM squad. When you have an enemy flanked at even a moderate distance, it shouldn't be a matter of whether you hit or miss, it should be a matter of how WELL you hit them. Damage being on a roll would make more sense.
This leads to one change that would solve the error of XCOM RNG. Chance to hit should be mostly dictated by your range, if you're in your effective range you've got a 100% chance to hit, but cover can "absorb" damage and again distance and flanking calculations can determine how much damage and how much "health" the cover has. RNG to dictate just how much damage you do within a few degrees of freedom is all you really need. These soldiers shouldn't be missing their targets, not the amount you see in any XCOM game. But the effectiveness of the hits should be a deciding factor.
This would inevitably up the difficulty of the game, as any soldier you have that isn't 100% covered WOULD get hit, even if just a glancing wound you'd see soldiers piling up in the nurses office with scrapes, burns, and missing limbs. The biggest problem with humans and RNG, is that a physical dice roll is just that, a physical dice roll, humans have nobody to blame but themselves for rolling the dice the way they did, but when you add the computer rolling for you, a sense of "It used loaded dice on me" can seep in, especially when they get unlucky roll after unlucky roll (and the computer opponent is getting lucky roll after lucky roll).
That's my two cents on the situation that is RNG, randomness in videogames is indeed bad design. This includes RNG or Dice.
It's just doesn't feel like "true" chance, whether that is manipulated by my own feelings or not, is irrelevant really. Not that any computed roll is fair anyhow but it's definitely one of the LEAST fair in any game I've played so far. For a start, all of the outcomes are predetermined (not a good thing, in my eyes) before you even set foot in a mission and each round has a set of outcomes assigned (that never change, if you perform the same actions and tested by reloading over and over). Sure, you can alter this by executing commands in a different order. Say, you have 50% chance to hit/hack, throw everybody on Overwatch, skip turn or whatever and it can change to another value on the next round. But that's just another predetermined set of values that have been chosen and you are being forced to play with.
I just don't understand why it's done this way. Each shot should be rolled just before you take the shot, not determined before the current round. It's a pretty stupid and unfair method of dice-rolling. After all, the setup for the animation takes what 2-3 seconds so surely a small calculation can be done in that time? Last time I checked, computing happened to be pretty fast.
Typically it is done this way to prevent save scumming. If you haven't heard the term, that is saving and reloading until the desired roll is obtained.
Ideally, what I would have liked to have is a removal of this foolish pre-calculated roll order, so that a 95% is a real 95% roll with a real 5% chance to miss.
But with playing with higher difficulty and your mod, the game remained difficult and punishing without loosing a game on a single roll. Thank you a thousand time.
Anyway that is my opinion. It would be cool if you made configurable file so people can make changes from there.
It's a shame I already deleted XCOM2 modding tools, otherwise I could have tried to recompile scripts
However in the mod's config/ini, you can enable the experimental gaussian/normal distribution, where you can change the parameters of the curve to adjust it more. The suggested values for that right now aim to do exactly what you say, adjust it more at the high end and try to keep it mostly inline at the low end (I target a 65% aim equal to vanilla/default). So give that a try if you want to experiment
Despite that though, from the bit of testing I've done, I think the default average sum numbers end up working out quite well. Still plenty of misses, doesn't make the game feel like a cake walk or anything. I'd imagine it just goes to show you that given % chance probability, it's still really hard for us to get a handle on predicting actual outcomes and results. I remember the feeling from Enemy Unknown where it was "a 90% chance to hit, sure feels a lot higher than seems to play out in the game".
So after playing with this mod a while my fears came true - game becomes too easy. I still believe that changes to hit% should apply for 85%+ shots. Now even in the middle of the game I feel sure about my sniper hitting enemy in full cover with about 75% chance (which in my opinion should not happen - I wold not take risk of 25% shot in vanilla but here I know it's 12.5% to miss and I make decision to take the shot lightly instead of blowing up cover/retreating/hnunkering down/using a beacon etc).
I didn't understand what you said about tweaking .ini files. Even if I do tweak them - how will I know what actual hit percentage is? In-game it'll still show the same number.
I wish you'd make this mod more configurable if you can.
Cheers!
That being said I'm working on an update to refine the alternative Gaussian method/ini tweaks. That'll give a chart with the % breakdown, but due to the nature of statistics, it's not something you can't calculate on the fly that well. I may see about making an excel spreadsheet for the next update that can at least let you try out different values for the INI file and see what the resulting shot percentages would be.
My config is as follows, as I wanted to test a 100% hit game.
FIXED_ROLL_VALUE=1
useAverageSum=false
useGaussian=false
However, in missions the shots still seem to miss (even 80+% ones).
Am I doing something wrong or is my usage of the config file wrong?
To get a better idea of what is going on in the meantime, you can check out the log file (it's kept in your Documents\My Games\Xcom 2\XcomGame\logs or similar, look for the combat log.
With the mod enabled you will get messages in that log detailing all the combat events (attacks), whether or not they hit miss, what types of hits, and the rolls that you were getting.
So if fixed roll value is being applied you should see the roll being reported as 1.
Also if the mod is enabled the combat log will have entries mentioning FairRNGMod. If you don't see the log or that in there, it's a good sign that the mod isn't enabled/active for some reason.