Who programmed the dice?

[Nekojin]

The problems with the Mersenne Twister algorithm are acknowledged by PHP library developers.

The problems of the Mersenne Twister involve speed and security, not whether it's random enough. Try harder.

[euklid314]

Please correct me but random_int() is NOT a Mersenne Twister algorithm. As I wrote the PHP functions rand() and the improved mt_rand() - latter using Mersenne Twister - are not cryptographically secure. They are still perfectly fine for applications by BGA - but they were NEVER used!

BGA uses random_int() which IS cryptographically secure and is NOT based on the Mersenne Twister algorithm. I have no idea why people are discussing Mersenne Twister (which is good enough on its own but not as perfect as random_int()) - there must be some urban myth behind it.

Rufasu, please discuss topics that are related to BGA: What is your problem with the cryptographically secure RNG function random_int() which you can check on your own as intensive as you like since you have a programming background.

[Nekojin]

Please correct me but random_int() is NOT a Mersenne Twister algorithm. As I wrote the PHP functions rand() and the improved mt_rand() - latter using Mersenne Twister - are not cryptographically secure. They are still perfectly fine for applications by BGA - but they were NEVER used!

BGA uses random_int() which IS cryptographically secure and is NOT based on the Mersenne Twister algorithm. I have no idea why people are discussing Mersenne Twister (which is good enough on its own but not as perfect as random_int()) - there must be some urban myth behind it.

I was told in a different discussion that BGA uses the Mersenne Twister as its root RNG algorithm. If that is incorrect, I apologize for bringing it into this discussion.

[euklid314]

Please correct me but random_int() is NOT a Mersenne Twister algorithm. As I wrote the PHP functions rand() and the improved mt_rand() - latter using Mersenne Twister - are not cryptographically secure. They are still perfectly fine for applications by BGA - but they were NEVER used!

BGA uses random_int() which IS cryptographically secure and is NOT based on the Mersenne Twister algorithm. I have no idea why people are discussing Mersenne Twister (which is good enough on its own but not as perfect as random_int()) - there must be some urban myth behind it.

I was told in a different discussion that BGA uses the Mersenne Twister as its root RNG algorithm. If that is incorrect, I apologize for bringing it into this discussion.

No problem. As I wrote "please correct me", - but all information I can gather informs me that random_int() is 4-times slower than rand() and Mersenne Twister mt_rand() but in contrast it is cryptographically safe. I quote: "Guessing the next value is not possible even if an attacker collects a huge number of values."

The function random_int() is no Mersenne Twister algorithm - it has completely different properties (slower, but cryptographically secure).

[Stroom]

The problems with the Mersenne Twister algorithm are acknowledged by PHP library developers.
Experts in the field of random number algorithms have also admitted this.
Furthermore, there are a number of new random number algorithms being developed.
Not knowing this implies a lack of knowledge.
It is common knowledge for developers who work with PHP.
To intentionally cite a piece of statistics is an act of deception to those who do not know computer science.

Again you show that you know nothing about this issue. You only cite that "it has problems". But WHAT are the problems? Do these problems affect 2d6? Not really. You will very likely not notice the difference if the RNG is changed. It will still be random and it will still give you "unfair" distributions of dice rolls.

It is easy to complain without giving any concrete issue that needs to be fixed. Either prove that something else is better while still being random or stop posting.

[Jellby]

I never win the lottery.
Lottery balls are made of plastic: https://lotterymachines.co.uk/products/lottery-balls/
Plastic has many problems, that's a known fact: https://pexpo.in/blogs/news/disadvantages-of-plastic
Therefore, until they change the material of the lottery balls, the lottery will be unfair.

Oh, wait... I never play the lottery either

[Rufasu]

I am so disappointed that they are only thinking of running away when they have been pointed out the problem so many times by players.
How can you claim that there is no problem when the PHP developers and those who study computer science are constantly updating it?
Have you proven that you are better than the PHP developers and that there are no problems?

[Romain672]

I don't want to say wrong thing so: https://stackoverflow.com/questions/102 ... ion-in-php :

If you're using [php randomness] for something else [...] where the numbers produced are not being used in any cryptographical way, then you may be able to get away with using rand() or mt_rand(). But, if your applications/modules rely on good random numbers for their security, you really need to take advantage of the OS sources, as described above.

The thing is, if you use randomness to per example generate random password, and someone found an exploit on it, and that exploit let them find the password of another user, then they got all the access that user have. If that user is an administrator, they have now access to the whole site. If that's a big site (first, they should have more protection), that can be extremely bad.
But bga use it not to protect anything private but to generate a random roll on a Catan game. Even if one time out of 100k (or less), that roll isn't random, who care? Nobody would tell the difference.

[Suoivax]

I am so disappointed that they are only thinking of running away when they have been pointed out the problem so many times by players.
How can you claim that there is no problem when the PHP developers and those who study computer science are constantly updating it?
Have you proven that you are better than the PHP developers and that there are no problems?

Do us, and yourself, a favor. Take two six sided dice, and roll them 25 times, recording the results.

do this about two hundred times, recording them all.

then post them here.

[Nekojin]

I am so disappointed that they are only thinking of running away when they have been pointed out the problem so many times by players.
How can you claim that there is no problem when the PHP developers and those who study computer science are constantly updating it?
Have you proven that you are better than the PHP developers and that there are no problems?

Saying, "Something's wrong, so fix it!" a dozen times doesn't make you more right. You've shown how grossly ignorant you are on the matter. If there's actually a problem, I highly recommend that you let people who actually understand the math - and the problems - be the ones to present that information.

Because all you've been doing is attempting to flip a table that's nailed down pretty securely.

Edit:

Let me be very, VERY clear here. There's a process that you're ignoring here.

1.) Identify the problem.
2.) Analyze how the problem impacts the system, and how significant that is.
3.) Determine whether it's worth the time and effort to fix the problem, based on information from 2.
4.) If 3 determines that it needs to be fixed, then fix it.

You keep shouting about 4, but you haven't demonstrated even once that you've made any effort to get past 1.