Who programmed the dice?

[Rufasu]

I just want to get away, that's the attitude of the people here,
I'm really fed up.
Everything is illogical and fraudulent.
Many people are frustrated with the BGA's random number process.
Despite this, they are trying to push it through using only the paradox of probability.
Please prove first that the paradox of probability can be applied to this program.
Mathematically speaking, for the paradox of probability to occur, the probability space must be implemented correctly.
However, no algorithm has been invented that can perfectly reproduce the probability space.
You implement an unknown algorithm and claim that the random number process is complete.
Prove it.
If you can prove it, it will be a major breakthrough worldwide.

If you deny the proposition that the random number process is unsatisfactory, then you should prove it.

[Nekojin]

Now you've devolved into insults and gibberish.

Your argument is that all RNGs are faulty, therefore... what? What solution is possible when your accusation is that the current system, like every other system that's ever been invented, isn't absolutely perfect?

What final result are you seeking? It doesn't seem like you have one, you're just flinging poo at the walls and hoping something sticks.

[Jellby]

Please explain how you can distinguish the BGA "random" dice from rolling real physical dice. If I give you two series of rolls, how would you decide which one comes from BGA and which one comes from real dice?

[Romain672]

Personnally, I explained that even if people are researching things about the randomness of a dice, that doesn't mean that a random dice used in bga is faulty and should be changed.

"Many people are frustrated with the BGA's random number process." => That's the case of a lot of random generation on games. In among us, people were complaining that a player was too much impostor. In league of legends (even if that's more complicated for some subjects) people are complaining their teammates are worst (which is non sense since that would mean that others people would have better teammate on average) OR that a smurf queue exist OR show a single game where they got no chance to win and forget the others ones. When you ask people to estimate the amount of % of anything (like their part on conflict, their part of household chores...) and add up all the actors i think it will be about 140%. If you ask people the probability of an event with probability of <20%, or >80% it will be extremely difficult to be right, events of 1% chance happens more frequently than people think to believe.

I'm unsure what kind of answers you want to receive? Do you want some about informatics only and theory about how to generate a random number (then I think you should post on 'discuss about board game' and not on catan)? Do you want some popularization/vulgarization (what i try to do) to understand why having something good enough is fine? Do you want to understand why people complains about randomness all accross the web? Is there a specific problem about dice which need to be look up (which one?) because people complaining about randomness gives various things and when that can be verified in 80+% of cases, they don't understand probability/they missremember/they lie in purpose/they cherry pick a specific event.

It's unclear to me.

I think we can all agree that bga can't give true randomness results. But that's not a problem if the rolls are enough good to not be dissociable from some randoms rolls in real life (and it's possible that the roll in real dice would not be random because of more weight in the 6 face compared to the 1 face).
At that point... Let me troll slightly: would you want bga to code physics to make us able to roll dices, and make it ridiculously more likely to roll 1s because of weight?


But yeah if you want a true answer, I will talk for myself at least, I don't know if the rolls are enough random or not. I have just seen from my games that they seem to be (per example on the crew, i noted the rare hands i got (like only 2colors/4), and they seem to be about what i should have with the number of my total games). Then I have seen that rare maths subjects done here look to not find anything wrong (there was one exception which was looked by one admin, it wasn't revealed what it was, so yeah that can be bug if you collect data, but that wasn't about the randomness).
But on the other side of the spectrum, we got plenty of people saying wrong things, not understanding probability...
At one point you need to choose a side, and I choosed mine.

And there is something else. If something is wrong about the randomness like let's said something big, every 10 rolls the result is 3. Then by just collecting data of 100k rolls, you should be able to see easily not only that the 3 has been rolled too much, but that big series of 3s will occur. When you do the average the serie will be slightly lower than expected, the series of 2 you expect will be much lower (same for others numbers) and so on...
You can check many differents things with a lot of data, even weird claims.
What is the chance to have something annoying about the randomness and which happen to be extremely rare to not be find by the radar of many differents things you could check above. It's way more likely that the bug could be seen by a lot of differents ways.
But like I said earlier, if you wanna talk about the php randomness, and how you create randomness with php, that shouldn't be on a catan subject which isn't linked to php at all. You will even got more specific answers on another sub-forum (or even a specialised one, why don't ask that to a php forum?) than on that Catan one. Just arrive clean and said things like 'I'm curious how the php work' and you will got way more peaceful discussions.

[frogstar_A]

Many people are frustrated with the BGA's random number process.

Yes. Because many people are very very bad at understanding probability.

With so many millions of games played on BGA even million to one chance events could be expected to happen quite a few times. Laughably common events like playing a game of Catan and the number 6 not coming up much would be expected to happen daily.

There is a thread on random numbers every week, because random number sequences would be expected to produce highly unusual patterns every week when used as often as BGAs, and whoever gets the most unusual rolls starts a thread.

[Rufasu]

It is ludicrous to state to a disgruntled player that he or she has faith in probability or in the BGA.
It is wrong from a mathematical, computer science, and business standpoint.
As I have said many times, the BGA should do what every developer does.
Just improve the quality of the game, constantly upgrade the PHP version and use new libraries.
If you have time to write long excuses, write programs.

[Romain672]

Let me be more direct: rare people have done some analysis on randomness and find nothing. I will not be able to find it back but it's present on the forum, I remember reading it at least two times.
At that point, since yourself don't even state what is wrong with it, I have no reason to be on your side, while I have multiple reason to be on bga's one.

Just improve the quality of the game, constantly upgrade the PHP version and use new libraries.

I don't know it, but isn't that already done? Or they are in an old version of php?
But even that old version should be fine for randomness (ie what I or others said earlier)?

If you have time to write long excuses, write programs.

You can do it too you know.
Nobody here is payed or work for bga.

Note sure I answer more here, I advised you to goes on other subforum with what you seem to want. But if you really want 'just' your last post, that's just non sense to me, few point changing something who work really well.

[Stroom]

It is ludicrous to state to a disgruntled player that he or she has faith in probability or in the BGA.
It is wrong from a mathematical, computer science, and business standpoint.
As I have said many times, the BGA should do what every developer does.
Just improve the quality of the game, constantly upgrade the PHP version and use new libraries.
If you have time to write long excuses, write programs.

You are wrong.
You have human biases when it comes to the "fairness" of RNG. Weird dice rolls happen most of the time. You will rarely get a perfectly normal distribution with 70-80 dice rolls. Your arguments have no basis whatsoever. You just say "something is wrong". No proof. No understanding about how any changes would affect the outcome.

Demanding that something very special and time consuming must be done because "people are not happy" is stupid. The people who don't like RNG just complain because they can not stand that RNG is not ALWAYS in their favor. Their complaints are flawed. So your argument is bad.

If you want the RNG to be more human-friendly then it stops being random. Giving it rules that one number must not be rolled more than 3 times in a row" etc breaks the randomness aspect. If you would manipulate the RNG to actually help one player based on sme metrics, the other players will complain. For Catan, basically 2/3 or 3/4 of the players will not be happy with the dice rolls when they lose. So it's pointless to make the dice more unfair.

You are not going to improve any "quality of the game" by changing the RNG system. It is still going to be as random as the current one. You don't even say what is specifically broken so there is nothing to fix, really.
You are wrong.

[Nekojin]

It is ludicrous to state to a disgruntled player that he or she has faith in probability or in the BGA.
It is wrong from a mathematical, computer science, and business standpoint.

This is the same mentality that feeds into mentality like, "The customer is always right," which, in the moment that it was first used, was entirely fine. But that mentality, extended to infinity, results in companies being ruined by bad actors, people looking to take advantage of the goodwill of the store.

Similarly, gaming companies that have bent over backward to add every idea that the fans demand results in a worthless sludge of a game that barely resembles the original product. Quite a few games have gone from being rising stars to failed duds because of listening to the customers too much. Not every person with a suggestion has a complete comprehension of what's going on under the hood. Not every customer is right.

As I have said many times, the BGA should do what every developer does.
Just improve the quality of the game, constantly upgrade the PHP version and use new libraries.

I'm not a BGA programmer. Neither is romain672. Neither is Jellby (As far as I know). They might be programmers. They might even be programmers for game representations here on BGA... but that doesn't make them BGA programmers (at least, in the way that you mean it).

That said... the idea of "continually upgrading" is another path to failure. Real programmers for game engines don't do that. It's far more common for devs to use a method called "leapfrogging." They pick a version of an engine that works for them, and stick with that version for a while. They only make interim upgrades when there's a discovery of a significant flaw or vulnerability, and continue to develop on a stable platform that everyone understands. When there's a significant change to the underlying engine that benefits the game, the developers will upgrade to the newer version (but often, even then, only after that new version has proven itself stable). Some developers don't even do that - they start with a set example of an engine, and stick with it from start to finish. For most games, there are few engine changes that will actually matter enough to upgrade for.

Now obviously, there's a significant difference between a single stand-alone game and a live framework like BGA. BGA is naturally going to have upgrades, updates, and so on. But that doesn't mean that it has to change to the new version of PHP as soon as it comes out. They may not even want to use the new version, if there's no major changes from the one they're currently using. Upgrading takes programming time - time that would be taken away from other purposes.

The current version of PHP is 8.3.3. That was released yesterday. 8.3.0 was nine months ago. 8.2.0 was 20 months ago, and is still fully supported. 8.1.0 was 33 months ago, and still has security support. Do you believe that BGA should be rocking 8.3.3 right now, even though it only came out yesterday?

For the record, BGA requires that new game modules that third parties create be compatible with 8.2. That gives third-party programmers a level of stability to work with, instead of having to be using the most current version, which might change tomorrow...

If you have time to write long excuses, write programs.

If you have the time to complain for pages and pages and pages without ever once taking any opportunity to demonstrate what the actual problem is... perhaps you should find a different path. The one you're on is mulishly stubborn and utterly valueless. Just because you are making an accusation, does not automatically make that accusation correct.

Instead of actually backing up your demands with evidence, you keep stomping your foot and shouting that because you're upset, we (and BGA) should just do what you wish. That's not how reality works. Trying to do that will snap your neck when you run into people with opposing demands.

[Jellby]

Besides, if you want BGA to do anything, use the "Contact us / Support" link at the bottom. This forum is for user-to-user discussion.