Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
Absolute waste of time.
FYI: Assuming you are analyzing combatLog.txt, be warned that this data doesn't exactly match the data from the in game log. In particular, some attacks are duplicated and some are omitted. I couldn't find any pattern as to why records were omitted ot duplicated, but I don't think it invalidates the analysis.
As long the code is proprietary there's nothing we can do about this than "deal with it and move on".
The only thing we can tell is that roll generation is a deterministic process, because computers are deterministic machines.
The effort require to actually give strong evidence that the dice rolls are either fair or rigged is so much that it's not worth it. And even if you went all the way, if you proved that the dices are rigged or fair, what would you gain at the end? A pat in the back by strangers online?
This game at least gives you RNG manipulation tools to get the best out of 2 or 3 rolls, to make enemies get the worst out of 2 or 3 rolls, to prevent rolls below a certain number, to turn 1s into 20s and so on. Use them to your advantage, as you said at the beginning of the post, you are at the end of your unfair run and characters are overpowered and the game is no longer a challenge, therefore, even if the rolls were rigged it wouldn't be relevant at the end, since the rolls become less and less relevant as the game progresses.
Dice rolls are problematic only in the early game, but that's a problem with the ruleset itself, not with how this game generates the rolls.
Unity programs that are moddable can generally (but not always) be disassembled all the way back to compilable C# code. Wrath falls into this category, and you can download and use "dnSpy" to convert the DLL files into shockingly readable C# code.
It is reasonably easy to locate the code that generates random numbers for use in combat and verify that no hanky-panky is going on.
You are correct to say that Wrath uses a deterministic process to generate its random numbers.
You are Incorrect to say that computers are necessarily deterministic. All computers that are Windows 11 compatible include a "Trusted Playform Module" (TPM) which, in part, includes a hardware RNG. The numbers produced by this device are required to be generated in a non-deterministic way.
Well... Yes? 😀
More seriously, it is an interesting programming project if you are of that bent. I learned an enormous amount about regular expressions and discovered the BigInteger class. I even considered rewriting my application as an actual mod (which is how I discovered dnSpy), so I learned quite a bit.
If you are trying to convince people who are sure the RNG is biased against them, then... Yeah, that's a waste of time. But it is amusing to watch people engage in ever more extreme forms of cognitive dissonance as their beliefs are challenged, so that's something, at least. 😀
Only with really bad implementations.
I suspect you didn't mean "atrophy". Again, what does "true random" mean. You cannot calculate the next roll of a PRNG based on the information available to you. You cannot calculate the next roll of a die based on the information available to you. In both cases, given enough external information you can make that calculation. It is a lot more information needed for the die case but is the same. The only time this isn't the case os with quantum effects.
"The code is just using the standard UnityEngine.Random.Range, which is a pseudorandom number generator based on the Xorshift 128 algorithm. Nothing special here."
Then i´d say:
Xorshift:
"To be clear, there is probably no harm whatsoever at using xorshift128+, but it does systematically fail reasonable statistical tests." (which probably means not so true random...)
So "true random" would mean every roll of a d20 dice has a 1:20 chance to roll any number. In a big sample every number should be represented almost equally. If that doesn´t happen it´s not "true random", if we stress the "true" - and it´s not really that random if it´s a calculated number, based on a seed. Then it depends how the seed is created.
Random is not about knowing the circumstances, but it´s that the result is unpredictable, but would statistically equal out. Here it´s about a calculation based on a seed. If we would only take the sample the provided earlier in thee thread, we could say that it´s unlikely to roll a 14 for whatever reason, or maybe also a 15. And if it wasn´t about 5 and 9, we could also say that it´s more likely to roll above 10 than below. That´s not really random - but of course the sample is small.
Let´s say "true random" means "chaotic" in laymen terms. While "chaotic" means random, if we say that random means something unpredictable, and chaotic something which can be predicted, but appears random, because we don´t have all the information. Which is how the random number generators work. That´s not true random. It´s chaotic, so the name "RNG" is misleading. It should be "CNG". :o)
If it's reasonably easy I would have expected you to provide the answer to the actual question instead of mentioning a theoretical path to finding it.
So, let me ask you, do enemies and party members use the same seed for roll generation? (I suspect they do because it would be the easiest thing to implement and I doubt they can be arsed to do otherwise, even if they wanted to rig the rolls), but is it? as undeniable proof. Not as evidence, not as hint of something, as irrefutable proof. Whatever it is, then provide proof, after all it's reasonably easy.
Not everyone has windows 11 and wotr doesn't use said property of the module anyway, therefore this is irrelevant to the discussion. The generation of dice rolls in WOTR (or any crpg out there for that matter) is a deterministic process. And the computers where we play these games are deterministic machines as well.
There are certainly devices out there that... attempt to emulate true randomness via physics, but the fact is that we accept their "randomness" more as a convention (because of their unpredictability) than as a matter of fact. Remember that Classical physics are not truly random, quantum mechanics is considered non-deterministic by most theories, but those are theories, not proven facts, and there are also theories out there that consider quantum mechanics as deterministic, therefore even "true" rngs -could- be deterministic in the end. This is something that we will leave for the scientists of the future to answer.
This without taking into account the additional cost of such devices and if their implementation will ever be worth in substitution of pseudo RNGs for CRPGS.
That's jut another waste of time rephrased in a more elaborate form. Wasting both their time and yours. Then again, we could argue that everything we do is a waste of time, since the end result is the same. But this pessimistic/nihilistic approach appears to be out of the scope of this thread. :P
I just wanted to tell the OP that he should move on and not worry about something that loses relevance (if you know what you're doing) as you progress through the game.
I started to comment on the probability of such things. I remember this frequently. Things get so heated but in the end it was nothing. Also it is never satisfying to prove something to someone on the internet. They just ignore you. Or they gaslight you. Or later if something you said really was profound everyone just adopts it later as common wisdom without credit to you.
I have found that a lot of people online are experts at making you feel bad. They are experts at saying nothing to look profound. I don't know this goes into other areas. The worst people for this are whenever something is kind of woke(I don't want to explain it) like there is some political reasoning to something. Then the people that show up are professional gaslighters and instigators. I guess those people do everything in bad faith and are fighting a culture war so it is something different.
Though to be fair I do believe that the dice are unfair. I was just talking about my experience more and more on dealing with internet conversations.
I guess the silver lining of everyone taking what you said without credit is that no one would be able to trace it back to you. Leaving references to other papers is hard because you have to know other places that it is referenced. Despite what I said I found myself lacking somewhat in that department.
yup
I already did this -- download dnspy and analyze the code for yourself.
The only other option would be to post the entire disassembled code, which wouldn't fit in a forum post and would be so voluminous as to be difficult to analyze (Dnspy provides dependency links that makes such analysis easier). This would also be somewhat dubious from a copyright standpoint, especially on a forum monitored by the games developers.
Then, in the D20 system, if you want to up the difficulty, you don't want to touch the dice. That's not how the maths work there.
If I want to make a check impossible, the only thing I need to do is add a +30 to the original DC and call it a day. That way the player has to roll a natural 20 to win the check.
Unfair is just that, inflated difficulty checks to the point where you need improbable rolls to win unless you have a perfect strategy and build planning.
People are just paranoid over rolls because of confirmation bias.
The belief that nothing is rigged is just another form of confirmation bias.
Rigged or not, the dice in Wrath are unpleasant and sap enjoyment from the game.
Playing Owlcat games is somewhat akin to an abusive relationship, in that there's precious little out there in the new RTWP party-based cRPGs, so you play them because that's what there is, and yet you know there's something not quite right about it, something unsettlingly nagging, but you go on, just because there's no great alternative.
Roll 1s every time you get an important skill check? ah well, bad luck mate, hey, have you tried reloading?
Enemy rolls 19, 20, 19 in their first 4 attacks & kills your power-tank before you've even dented it? ah well, bad luck mate, hey, have you tried reloading?
Reload, buff a bit, reload, buff a bit more this time, reload, right, that's got it.
I seem to have over 40 hours and I'm still in act 1. Oh, is it because I've actually played act 1 approximately 4 times already, just in lots of small chunks of reloading. LMAO.
I mean, it's great that x amount of dice rolls proves y to be the mean where z is representative of a communal average, but none of that bullocks improves moment to moment gameplay in an RPG.
Something you could do as a developer: Icewind Dale is renown for being a tough challenge game with regards to combat. Experts can steamroll it with min/maxing, but most average players will have hefty challenges in that game, and it has nothing to do with lucky or unlucky dice rolls.
In order to mitigate 'unlucky' crit streaks , they made Helmets, even basic non-magical ones, prevent critical hits - so if the enemy rolls the 20, they hit, but it doesn't crit. This enables tanks to actually tank regardless of the rolls.
Amazing isn't it, a game reknown for combat difficulty actually gives a gamewide buff to the player right from the start of the game. I guess they figured out how to make combat interesting beyond just the numbers. Shocking, I know.
But Wrath seems to have a reputation for difficulty because of spikes in RNG numbers, which isn't difficulty, it's just luck. Because the combat isn't really very interesting. It's, like, the same 3 or 4 enemy types over and over with the odd boss-like sprinkled among them, and the boss is 'difficult' because it has either big numbers or an ability you are not yet levelled enough to deal with.
There wouldn't be so much discussion of dice rolls if the game wasn't so massively dependent on them for it's encounter design to have any challenge to it.
Tomato / tomato, right?
This is a totally different question, of course. There is a reason why X-COM and BG3 have "unfair" (weighted in the players favor) RNG -- in a single player game, random dice have no particular value from a fun perspective. Biasing the dice in the players favor is good, as long as the player is having fun.
Yes... You do realize that this is a blatant cheat in favor of the player and that there is a difficulty option in Wrath to do the same thing (minus the helmets), right?
Also, back in the day there were similar discussions of biased RNG for infinity engine games. Perhaps the helmet cheat reduced the frequency of complaints, but they did exist.
Alternatively, Owlcat felt that players who were not having fun would turn turn down the difficulty until they started having fun.
Before you laugh at this, keep in mind that this is the same developer who felt hidden time limits would make games more fun. Owlcat has a.... unique... perspective on CRPGs.
Your milage may vary, but my opinion is that the reason has a reputation for uneven difficulty is... Because it has uneven difficulty. Combat difficulty is set to "Would be easy, but not trivial, for a party that has completed the current act," Thus, you start each act suffering, painfully manage to win, and finally actually feel strong at the end of the act, only for the cycle to repeat in the next act. This is terrible game design, admittedly, but this is how the game is designed.
And imho being able to kill or be killed in one or two hits is not about difficultly. It´s about how long, tactical and RNG depended these fights are, even more so when tehre ae critical misses, and auto-hits, while they somehow abandoned it for chests, as there´s a 100% chance, instead of maxing out on 95%. For reasons i guess, as people would reload or complain about it much more, as it´s about single rolls, while in a short fight You have a bunch of them anyways.