Hi Sup Forums, this is a different thread to most of those usually posted
I'm philosophically interested in the concept of random. I've done a fair bit of research so far but I've mostly found information on pseudo random number generators
So can someone explain to me how most programs create properly random numbers
Also is it theoretically possible to create a program that is "pure random" ie something that is completely independent from a logical process
pic sorta related
Other urls found in this thread:
en.wikipedia.org
github.com
youtube.com
en.wikipedia.org
youtube.com
youtube.com
twitter.com
Might want to ask on the math & science board.
This.
We're idiots.
pure random numbers would be fucking huge
>I'm philosophically interested in the concept of random
kys
>Also is it theoretically possible to create a program that is "pure random" ie something that is completely independent from a logical process
en.wikipedia.org
>So can someone explain to me how most programs create properly random numbers
There is no consensus/proof that properly random is/isn't a thing at all (and probably no good of evidence of either).
The current practice for programs to generate "properly" random numbers is to use the specialized hardware that, for example, measures thermal noise. Whether numbers generated by such devices are truly random remains unknown, but it's true that we humans are not capable of predicting them.
>Also is it theoretically possible to create a program that is "pure random" ie something that is completely independent from a logical process
that's a pretty hard concept to grasp
Determinism says that nothing is random, though.
Yes, there exists a theory that nothing is random. Well done.
>pretty hard concept to grasp
that's a pretty interesting way to write "makes no sense"
This is false, uncertainty principle limits how well things can be determined. There are many things that are, as we currently understand them, truly random.
So are we ded? Did you have your question answered OP, or was this a post and run?
I wrote this thing to get quantum-generated random numbers that could be verified and used by plebs:
When ever you use a random number generator all you are really doing is connecting to some shop in india where some pajeet is getting paid 3 cents an hour to play 'eenie meenie miney moe' on a numpad while another pajeet gets paid an additional 2 cents to make sure his fellow pajeets are playing correctly, this number then gets sent back to you.
Impressive.
there is several ways, on is somehow using white noise from random signals received from space converted into a waveform and calculated versus another snippet of white noise.
this is the as i understand it because im am a retard and cant explain it well answer.
sorry i cant be of any real help.
On the one hand we have things that we cannot determine, on the other we have things that are not deterministic. But if you need to know the entire state of the universe in order to predict the next number then it's functionally random and that's close enough, the distinction is philosophical for most purposes.
Though from what I understand much of the world of the very small is believed to be actually non-deterministic by nature, hence using a smoke alarm to generate coin flips.
Thanks
This, using random numbers supports sweatshops, and also in turn supports indians stealing our jobs. Immediately desist using any crypto.
I've already read this but it doesn't seem to be something your ordinary computer could use
If i asked my computer to give me a random number, how would it do it?
mathematically speaking a random number is a number situated within a certain interval that is given completely independently to any other factors
applying this to real life means the number would have to also be completely independent from anything else, meaning a computer wouldn't be able to "logically" reach it (because logic implies you start with something and end with something else) hence why I'm sceptical about the concept of anything truly random
last time i checked the thread i thought it was dead
Seems interesting, I don't have linux though so I won't be able to use this
If I've got it right though, you get numbers from experimental data and the data itself is supposed random, yes?
Have some more fractals cos I love these
Very interesting. You could XOR the output with /dev/random to produce a stream with enhanced entropy. Information theory states it would be no less predictable than any stream on its own. This is how they get around potentially untrusted hardware rngs
>meaning a computer wouldn't be able to "logically" reach it (because logic implies you start with something and end with something else) hence why I'm sceptical about the concept of anything truly random
The computer obtains random numbers not by logically reaching them but by making observations about outside world - observing things that we perceive to be random.
What I meant by not making sense is that if you make a program that is "pure random" it will be purely useless, which is an extremely undesirable quality of a program (unless you frequent desktop generals, of course).
>mathematically speaking a random number is a number situated within a certain interval
I'd like to see the definition that mentions this specific interval.
>If i asked my computer to give me a random number, how would it do it?
Crypto. That's the difference between random and pseudorandom. It samples inputs with high entropy (CPU timings, mouse movements, memory access speeds, etc) and hashes them iteratively until you have something hard to predict. Look up hashing algorithms and entropy for details. Hardware RNGs use any number of methods, and on-die CPU RNGs use logic gates that flop wildly. There is no pure randomness in the universe, except for in some properties of quantum particles.
>Crypto. That's the difference between random and pseudorandom. It samples inputs with high entropy (CPU timings, mouse movements, memory access speeds, etc) and hashes them iteratively until you have something hard to predict. Look up hashing algorithms and entropy for details.
That's generally labeled as pseudo-random.
>Hardware RNGs
That's generally labeled as properly random.
>I've already read this but it doesn't seem to be something your ordinary computer could use
If you have a somewhat modern intel CPU, it has a hardware random number generator built into it already, although its probably backdoored by the NSA.
Plenty of high security servers use PCI based random number generators though.
>If i asked my computer to give me a random number, how would it do it?
Without getting "true" random numbers from environmental noise, its done by using a cryptographically secure psuedorandom number generator (CSPRNG) which is deterministic, but unpredictable if you don't have the initial random seed.
>hence why I'm sceptical about the concept of anything truly random
If you subscribe to the current consensus on quantum mechanics, then the universe entirely non-deterministic and quantum uncertainty is a great way to get random numbers.
a random number has to be real
therefore it fits within the interval of all real numbers
i haven't learnt maths in english but all numbers belong to an realm, and when you seek a random number it usually belongs to R, so the interval is at least R (real)
after all in maths you always have to define on what interval you're working
do we know for a fact these quantum particles are absolutely random or is it because we haven't fully understood them?
to elaborate on the random number generators of CPUs, is the final number created starting from an initial input?
is it the input that varies or the transformation of the number?
I'm very curious as to what "chooses" the process and the numbers
forgot my signature image
don't know if I'm the only one but I just love these fractals
>to elaborate on the random number generators of CPUs, is the final number created starting from an initial input?
>is it the input that varies or the transformation of the number?
You've been already told numerous times that the input comes from observations of outside world. The algorithm that takes the observation and produces a nice random number from it is the same within the hardware unit.
>when you seek a random number it usually belongs to R
It's quite more often an integer if you ask me.>do we know for a fact these quantum particles are absolutely random or is it because we haven't fully understood them?
Not a direct answer answer to your question, but watch regardless, because it should be very relevant to your interests.
youtube.com
how does a cpu gather data from the outside world I thought it could only do processing?
>how does a cpu gather data from the outside world I thought it could only do processing
Does your computer have the keyboard?
don't the keyboard inputs have to be modified for the CPU to understand
besides my computer doesn't have anything that can detect quantum particles
Ask somewhere where people are actually educated, like on diy or sci.
This is the mainstream electronics board, where we make ourselves think we know shit by ricing UI's and discussing mainstream electronics on a superficial level.
>don't the keyboard inputs have to be modified for the CPU to understand
What if they do? What if they don't? If they do you can also apply similar modifications to your observations to let CPU work with them. If they don't, you can just use observations without modifying them! Well done. What is the actual answer to your question? The answer is neither, your question makes no sense; I honestly can't tell if you're trolling or th
>besides my computer doesn't have anything that can detect quantum particles
If you don't have the hardware units that can generate properly random numbers, then you can't generate properly random numbers on your machines. Great observation.
Intel claims their latest processors do have that hardware.
en.wikipedia.org
Eat a dick.
OP youtube.com
highly related
>Eat a dick.
I would, but you beat me to it.
youtube.com
chaos is how you get randomness in practice, no quantum bullshit needed.
Deterministic chaos is still deterministic, especially when you compare it to quantum effects that we theorize to be actually random, as opposed to unpredictable by us, but still deterministic chaos. It does not matter in practices because we generally only want unpredictability, but this discussion is somewhat theoretical.
bampu
...
Fractals are pretty cool. Anyway, if you wanted "true" randomness on the cheap (and you don't trust modern intel's hardware RNG), you could get a $25 USB software-defined radio (SDR). Tune it to some white noise non-station, and use the input as your randomness. Ofc, it's subject to all sorts of physical attacks, but it's a very similar principle to the "environmental noise" used by intel's hardware RNGs.
is there truly a difference between unpredictability and randomness? Do you think true randomness actually exists? makes as much sense as free will!
Nature is random, human behavior is random, our thought processes are random, our conversations are random.
Assign an integer to each of these variables, and then observe the randomness of life itself. Your computer will be overwhelmed with data.
>How Can Mirrors Be Real If Our Eyes Aren't Real
Say you get the number 4 from a random number generator, and the CPU generates a new random number 8 with that as a seed, and then a subsequent 2. Assuming its a CSPRNG, there is no way to predict the next or previous numbers if you don't have the initial seed. Seeing as the initial seed came from truely random environmental obsevations, we can have safe random numbers.