uman

To help Fall Back Son:Yea IDK if that is real C (I haven't coded in forever) but you get the idea.

__________________
Few people deserve to be compared to (Rush) Limbaugh, most of them were convicted at the Nuremburg trials.
--WilliamSChips on Slashdot

ZenMasterJG

You can also pull true random numbers from random.org's via HTTP, SOAP, or CORBA, if you don't have a hardware random number generator. One of the things on my project list is to make the most useless hardware random number generator ever... using raindrops. No rain, no random bits for you! Luckily I live in Boston where it rains more or less constantly, so that shouldn't be a problem

DaWei

It is for this approach that buffers/queues were invented. Unfortunately, the approach tends to fail when one needs it the most.

__________________
Abstraction doesn't make it impossible to write bad code; it makes it possible to write superior code.
Contributor's Corner: Grumpy on C++ Exceptions DaWei on Pointers

Fall Back Son

I understand -- I was joking. Sorry for the confusion. And sorry if I'm being a bit thick here, but being truly random just indicates a perfect distribution, as far as randomness goes, over a given set of numbers. Mathematically, that would mean any number in the set is equally likely to be 'chosen' as any other number. Correct?

lectricpharaoh

Being truly random means the system is non-deterministic. Any psuedo-random generator, on the other hand, is entirely deterministic. If you know the seed and algorithm, you can determine the next number in the sequence with perfect certainty. This is not the case with truly random numbers.

DaWei's point was that this distinction exists. In many cases, psuedo-random numbers appear sufficiently random to meet the needs of the task at hand, such as in many games, but the distinction is still there. The fact that they are deterministic can be an advantage (in that it allows identical data to be replicated in subsequent tests) or a disadvantage (where this quality can be exploited to predict numbers, which would be bad in, say, an electronic slot machine in a casino).

@OP: Algorithms to generate psuedo-random numbers will use a seed value, and apply a series of operations on it to a) derive a new seed, and b) return a new pseudo-random number. Those that generate truly random numbers typically function by listening for some kind of noise. This noise might be actual sound-type noise, electromagnetic noise, radioactive decay, or whatever. The main requirement is that it fluctuates over time (the faster the better, generally, as it can thus be sampled at a greater speed). By measuring some aspect of said noise (such as amplitude or frequency), and comparing it to some threshold value, a binary one or zero result can be returned. By repeating this process, an arbitrarily long sequence of bits (and thereby bytes) can be assembled that is truly random. Such systems often apply some normalization to ensure there is a roughly uniform distribution; often this is done by adjusting the threshold value. For example, if the average amplitude of the signal over a given time period was x, the threshold could be set at 50% of x in the hopes that a value was equally likely to be either a one or zero. This normalization needs to be done carefully, however, to prevent bias from entering the system. After all, sometimes you really do get ten 'tails' results in a row when flipping a coin.

__________________
And once again, Probability proves itself willing to sneak into a back alley and service Drama as would a copper-piece harlot.
- Vaarsuvius, Order of the Stick

Fall Back Son

Thank you.. I appreciate it . Another question (this is a question I found on my "possible midterm questions" study sheet and was unable to answer):

"Write a single C statement that will print a random number from the set 2, 4, 6, 8, 10"

If anyone can give a push in the right direction for that I would appreciate it. I couldn't find anything in my lecture notes for it (which isn't that surprising since the possible midterm questions have been compiled over the years & the lecture notes are kept fairly concise).

MicDareall

make an integer array holding those values and use rand() to get a random integer between 0 and length of the array to use as an index in the array.

jim mcnamara

For randomness good enough for encryption purposes,
consult information about /dev/random in the UNIX operating system.
It is implemented in C

Fall Back Son

Thanks - don't know why I didn't think of that

but how would it be put into one line of code?

DaWei

printf ("%d", (rand () % 5 + 1) * 2);

__________________
Abstraction doesn't make it impossible to write bad code; it makes it possible to write superior code.
Contributor's Corner: Grumpy on C++ Exceptions DaWei on Pointers