Page 6 of 6
Re: Rigged rolls!?
Posted: 14 March 2023, 06:19
by Stroom
Rufasu wrote: ↑13 March 2023, 22:58
Computer random numbers are not truly random.
I am looking for randomness that is closer to true randomness.
It makes for a more interesting game.
You won't be able to spot "true randomness" from computer generated randomness. In the case of 2xd6 rolls the randomness is indistinguishable. You just don't like randomness. Randomness like this has been OK in card games, video games, EVEN ONLINE CASINOS where money is actually on the line.
Again... give us PROOF that something is broken not just pointless rambling.
Re: Rigged rolls!?
Posted: 14 March 2023, 09:51
by Rufasu
Stroom wrote: ↑14 March 2023, 06:19
Rufasu wrote: ↑13 March 2023, 22:58
Computer random numbers are not truly random.
I am looking for randomness that is closer to true randomness.
It makes for a more interesting game.
You won't be able to spot "true randomness" from computer generated randomness. In the case of 2xd6 rolls the randomness is indistinguishable. You just don't like randomness. Randomness like this has been OK in card games, video games, EVEN ONLINE CASINOS where money is actually on the line.
Again... give us PROOF that something is broken not just pointless rambling.
There are several random number algorithms.
One of them is the "Linear congruential generators" which does not have a very good quality of random numbers.
I feel that the dice of Catan have a similar periodicity as the result of the process of that algorithm.
I have no way to prove this feeling because it is a professional intuition.
And I could be wrong.
However, since there are other people who feel that Catan's dice have a periodicity in the way they appear, I think it would be a good idea to adopt the Mersenne Twister, for example.
If I didn't like randomness, I would not do Catan.
And I wouldn't want better randomness.
I like randomness because I like to be random.
Translated with
www.DeepL.com/Translator (free version)
Re: Rigged rolls!?
Posted: 14 March 2023, 20:34
by Stroom
Rufasu wrote: ↑14 March 2023, 09:51
There are several random number algorithms.
One of them is the "Linear congruential generators" which does not have a very good quality of random numbers.
I feel that the dice of Catan have a similar periodicity as the result of the process of that algorithm.
I have no way to prove this feeling because it is a professional intuition.
And I could be wrong.
However, since there are other people who feel that Catan's dice have a periodicity in the way they appear, I think it would be a good idea to adopt the Mersenne Twister, for example.
If I didn't like randomness, I would not do Catan.
And I wouldn't want better randomness.
I like randomness because I like to be random.
Translated with
www.DeepL.com/Translator (free version)
The default RNG used by PHP is perfectly fine. You are just showing signs of human bias when it comes to randomness. No matter what you do about randomness, people WILL complain. The only way to shut people up would be to create a pseudorandomness solution which would be closer to the bell curve in less rolls. But that would not be random anymore and it causes other tactical problems.
Still, you are just basing all your claims on feelings, not actual proof. Human bias.
Re: Rigged rolls!?
Posted: 14 March 2023, 23:27
by Rufasu
>Stroom
Thanks for the information.
By PHP default, do you mean the mt_rand function?
This function is notorious for being buggy, so if you are using PHP8 or later, you should be fine.
Re: Rigged rolls!?
Posted: 15 March 2023, 13:10
by Jellby
Re: Rigged rolls!?
Posted: 15 March 2023, 16:27
by Rufasu
Thank you Jellby, I will check it out.
With the bga_rand function, I now know that my concerns are unfounded.
I guess I was just unlucky.
Re: Rigged rolls!?
Posted: 15 March 2023, 21:17
by cowboy_dan
Rufasu wrote: ↑15 March 2023, 16:27
Thank you Jellby, I will check it out.
With the bga_rand function, I now know that my concerns are unfounded.
I guess I was just unlucky.
I really respect someone who can revise their initial assumption based on new evidence.