1  |   Order by index

Apr 13, 2003, 3:37pm

I have had some odd times in some Bingo worlds before that has to do with
the randomization of the bot and what boards are more likely to win than
others. I have had times where I have won nearly every game I was in the
world for and other times when I only had 3 or 4 numbers called throughout
each game. The thing is, each board seems to be rather consistant in its
"luck".

I think a nice PHP script to randomize each board would be a much better
system than to have 100 boards with random numbers on them. The PHP script
could come up with the random numbers and even if what I am feeling is total
bullshit, it'd be nice to know that I'm wrong. :-P Yeah, forgive me for
nitpicking here, but from the moment I heard about Bingo, I assumed that
that's how the boards were generated - it just seemed like the logical way
to do it.

At this point, it may just be a waste of time to do, especially if the
current system works better than I feel it does, but if any new bingo worlds
were to be made, this sounds like the logical way to do things so perhaps
this will at least give future owners of bingo a good idea to work with. :-)

-Dion

Apr 13, 2003, 5:23pm

Actually changing to a PHP script wouldn't help, in an of itself, because it would still be programatic and therefore pseuorandom. What's really needed is simply to add a few lines of code to make the pseudorandom process less predictable, such as using the exact time a person choosing a board clicks on it (down to the millisecond the click is recieved) as a randomizarion seed, or skipping a generated number for each line of chat examined, or seeding the generator with the average of the locations of all avatars present.

TechnoZeus

[View Quote]

Apr 13, 2003, 5:30pm

I kinda understand why computers can't generate true random numbers, but I
don't get the point of why the numbers need to be truely random. Anyone care
to explain all of this?

Ryan

[View Quote]

Apr 13, 2003, 6:31pm

Because when they aren't truly random, some numbers tend to be more common
than others.

-Dion

[View Quote] Ryan

[View Quote]

Apr 14, 2003, 8:12pm

Remember, though, that random doesn't mean uniform. With enough samples, random numbers should cover most of the possible values, but it's also possible (although unlikely) that a truely-random source could generate 5 zeros in a row.

-Agent1


[View Quote]

Apr 14, 2003, 10:23pm

Yeah, and eventually the 0's would even out. Like I said, I'm probably just
being a little weird here. Numbers like 37 and 65 seem to pop up more often
than others but that's probably just me.

-Dion

[View Quote] -Agent1


[View Quote]

Apr 15, 2003, 12:09am

[View Quote] Well, theoretically the turely random numbers are impossible is silly. If you're
using something like seconds from the epoch and a formula that's constant you'd be
getting something truely random based on it. Sortof? I just don't understand the
theory behind non-true random since it's not a brain trying to go "Well I picked a
low one last time, let's go high this time."

--Bowen--

Apr 16, 2003, 9:08am

If your using a regular scale as a base to generate a random number; it is
not random... because it is based on something static on nature...

[View Quote]

Apr 16, 2003, 7:21pm

The answer is... the numbers don't need to be truely random. What would be good though, is for the predictability to not be noticeably higher than what you would get with complete randomness.

TechnoZeus

[View Quote]

1  |   Order by index