this game is super rigged

[Romain672]

More seriously, even in pro poker website, you find people which think the game is rigged, while there is direct money on the line, and players who plays it the whole day so way much.

More:
https://en.wikipedia.org/wiki/Confirmation_bias
https://en.wikipedia.org/wiki/Loss_aversion

[SluggerBaloney]

Deleted

[SluggerBaloney]

Deleted

[Giordano Bruno 1974]

I'm very far from being a young millenial and though I don't program PHP, I used to program computers at a fairly high level.

I'm aware that computers can't generate random numbers, they only appear to. Computer languages have built in functions that will generate pseudo-random numbers given an initial seed and it's child's play to create that seed.

You would know very quickly if the seed was the same each time because you would get exactly the same set of numbers each time you played the game as you said. What you would not get is a skewed set of numbers that produced odd results that people typically complain about.

And you most certainly would not get a system that actually cheats to achieve some kind of desired outcome e.g. allow beginners to win more often. As I said, you can't do that by simply rigging the initial seed. It's actually a fairly complex task because you have to program in what is a "good roll" of the dice. In many games it's not obvious what rolls are good and which are bad because it's so context sensitive. It could be done but it's not simple task and you couldn't do it accidently.

[Giordano Bruno 1974]

Or more seriously, people who see repeated direct evidence of broken PRNG's and refuse to believe what they see because they think computers are magic

https://en.wikipedia.org/wiki/Cognitive_dissonance

What do you think is repeated direct evidence of broken PRNGs? Because anecdotes are really not evidence.

And I've yet to experience anything that I think is rigged myself. Yahtzee is a game with a big dose of luck, it's supposed to be that way. Naturally some games you will get a lot of luck and some games you will have very little. That's normal.

[RicardoRix]

They use the RNG code of the underlying operating system.

Most of the games on BGA are open source, and easily viewable via the BGA Studio.

You can also create and run your own php programs that will run on the same BGA servers.

If you want to generate random number to analyse then there is this project (xtwooneone) on BGA Studio which will give you 5000 random numbers.
https://en.studio.boardgamearena.com/lobby?game=2645
Feel free to do your own analysis.

Put those 5000 numbers through statistical analysis like has been done here.
https://boardgamearena.com/forum/viewto ... 11&p=59122

We are still waiting for your 1200 dice roll analysis data and source.

[SluggerBaloney]

Deleted

[RicardoRix]

I'm saying that looking at BGA's code wouldn't tell you anything, because BGA's code does not generate a random number. You would need to know the underlying OS BGA runs on and I've never seen that info anywhere.

It's immaterial. What do you want to see the assembly code that is runs?? What exactly is it going to tell you? Will there be something specific there that means that triples occur more often in Can' Stop?

It's the results that matter, and whether or not they past statistical analysis. Isn't it your 1200 Can't Stop dice roll analysis that's lead you to believe that it's a problem?

[Giordano Bruno 1974]

I'm aware that computers can't generate random numbers, they only appear to. Computer languages have built in functions that will generate pseudo-random numbers given an initial seed and it's child's play to create that seed.

While true that's not what I was referring to. I'm saying that looking at BGA's code wouldn't tell you anything, because BGA's code does not generate a random number. You would need to know the underlying OS BGA runs on and I've never seen that info anywhere.

Ricardo has it spot on.

You can't look at the source code and expect to see anything incriminating. Random number generating is a pretty important thing for any operating system or any computer language. A huge amount of money goes into making sure these things work or games will not run properly on the machine. Are there any known issues with any particular operating system? I mean what are you looking for anyway.

The idea that a non-specialist could just glance at the code and know that these functions will produce too many triples in specific board game is ludicrous. Proving that complicated mathematics functions are skewed is not a small task. This is why intelligent people focus on statistical analysis of output data.

Now I'm told that these tests have been run in the past and that's good enough for me. I haven't noticed any anomalies myself and I believe that those that do claim these kinds of things are reporting things that wouldn't be caused by a seeding problem. The seed doesn't care if you are good player or a beginner.

[frogstar_A]

The way pseudo random numbers work is they use a seed to create a list of seemingly random numbers. These numbers are NOT random, they only appear to be. If you feed the PRNG the same seed, you'll get the same sequence of "random" numbers.
...
As always any time this topic comes up you have the usual group of zealots who believe anything done by computers is infallible, so it's pointless debating the topic with them since all they do is just get nasty and start calling you names. I was posting in the Can't Stop forums around this and it quickly devolved into people stalking my BGA account and slinging accusations and name calling. It's pretty disgusting behavior. The modern young millennial isn't great at understanding that they don't know everything.

SluggerBaloney - we all know that computers can't generate truly random numbers, the best they can do is sequences of non-random numbers that behave like the desired random distribution. Such a sequence is more than adequate for playing games like Can't Stop or Yahtzee as long as it mirrors the correct distribution. Your contention on the Can't Stop forum is that triples come up "too often" relative to the likelihood if each of the four dice really were a uniform distribution from 1 to 6.

Unfortunately you have based your conclusion on a single data set that you have now deleted. It's not really surprising that people aren't happy to just "take it on trust" that you're right based upon data they haven't seen - whether that's millennials that you've for some reason had a dig at here, or indeed people of older generations such as myself and other posters.

No-one thinks computers are "infallible". What a silly thing for anyone to think. But most of us start from the position that their pseduo-random numbers are good enough to play some trivial internet games with to pass the time.

You would know very quickly if the seed was the same each time because you would get exactly the same set of numbers each time you played the game as you said. What you would not get is a skewed set of numbers that produced odd results that people typically complain about.

And you most certainly would not get a system that actually cheats to achieve some kind of desired outcome e.g. allow beginners to win more often. As I said, you can't do that by simply rigging the initial seed. It's actually a fairly complex task because you have to program in what is a "good roll" of the dice. In many games it's not obvious what rolls are good and which are bad because it's so context sensitive. It could be done but it's not simple task and you couldn't do it accidently.

+1

We KNOW the numbers aren't random. That alone doesn't prove that triples come up too often at Can't Stop.