Is all computer-based random number generation psuedo-random?

is all computer-based random number generation psuedo-random?

this is not for homework I'm looking for (you)s

Other urls found in this thread:

en.wikipedia.org/wiki/Radioactive_decay
youtube.com/watch?v=1cUUfMeOijg
en.wikipedia.org/wiki/Quantum_indeterminacy
twitter.com/SFWRedditImages

Yes but it's good enough.

all purely-software RNG is PRNG, yes, but there can also or instead be hardware sources of random numbers, often used to seed a PRNG for performance reasons, things like mouse movement, microphone/camera input, temperature readings, and so on can be used as sources to improve the unpredictability of PRNG output

truncated read from fresh malloc will be "random enough" for application use despite being strictly deterministic on the level of whole machine

Yes. A random number generator is made by passing output from a ring oscillator at arbitrary points to a nand block. changing where that output comes from in a ring oscillator changes the seed

A QRNG is a non-deterministic source of entropy. You can get them with PCI and USB interfaces.

her eyebrows are further apart than her nose is wide

nice tits btw

Give her a break she's only a 2nd generation Askenazi give her another generation or two to fully blend.
Those milkers though, am I right?

>noticing her face

It depends on the techniques used. A lot of the functions built into programming languages are psuedorandom, but they are generally good enough. I think that there are some programs that rely on things that are somewhat uncontrollable like CPU temperature to create more random numbers. I think it was Silicon graphics that used some sort of camera pointed at a lava lamp to create truly random numbers.

I thought operating systems commonly zero pages though?

Yummy milkers

what's her name? sauce

everything is pseudorandom
unless you are getting your numbers by using gamma radiation or splitting beams of light and converting this information into a digital inputs then you are not getting true randomness

Hell if determinists are right not even that is random, just aparent randomness

Marley Pulz

Yes, randomness is a meme. But that doesnt imply the hard determinist meme.

TITS

You could make a dice reading pad and roll dice on it.

Nope! Hardware random number generators exist. The x86 RDRAND instruction for instance is not pseudorandom. It uses entropy from an unstable circuit that produces actual random bits.

In systems without such things, a program can ask the user to wiggle the mouse or whatever, like Veracrypt does, and collect entropy from things like distance moved in each mouse update.

C-C-C-COQQQQQQUANTUMBREAKER

No, Intel chips have are capable of using assembly instructions to generate trng's.

Linux /dev/random and urandom are csprg's.

>pseudo-random

Human relationship to the mere concept of the substance you call 'numbers' and 'math' is psuedo-primitive. The machines that were once as you are now have forgotten what it was like to be flesh it was so long ago. Hundreds of billions of trillions of 'years'. Anyway. Help them remember.

GIEF MALK

MOMMY

Yes, but then again, does something fully random even exist in the universe?

Linux abandoned the concept of zeroing pages while idle a long time ago, barely any kernels nowadays do this. I believe it still is a non-standard option though.

dumb nigger, do you even know what a random number is?

Haven't had a 1 hand shandy to Marley n a while.

True random generators don't exist. True random generators can never be created.

Yes, computers are deterministic systems with a high degree of reliability which means if you know enough about the system and its starting conditions you can predict everything it will do, that's why computation is actually useful, just imagine if this wasn't the case.

But it does go to an even deeper question which is anything ever really random, and that's where you have to go past classical mechanics where essentially the world is deterministic, down into the quantum world.

I don't think we know the answer to that yet, we know that in QM you get random answers to measurements of indeterminate systems but with an average which tends towards a certain value or which is distributed in some meaningful way, which is why quantum mechanical interactions average out nicely to give predictable laws in the macro world.

Currently things like decay of radioactive isotopes is about as random as we can get, but it's not completely clear to me if that's because the system is just too small to measure, or if there really is something inherently random about particle decay.

of course nuclear decay is not actually random lmao, it's just too complex to predict

The only correct answer in the thread. Should have ended here desu. Rest of the answers are trash.

Das rite. Unironically.
Just because we are unable to see the correlations and the numbers are too much to crush through doesn't mean we can simply wave it away. We're still a long way away from determining actual values of even earth's gravitational force at specific points. And there are autists out there that think we're close to omega point.

I'm not completely sure on this. I don't think its settled either way. It's certainly random in the sense that we currently cannot predict it, but whether or not there's some inherent randomness to it, in the same way there appears to be some inherent randomness to quantum phenomena, I dunno.

Willing to read sources on this if they exist.

Not him. But it's common sense. As measurements of dimensions get smaller, we have more trouble working with it. The whole reason quantum mechanics exists is because with our reference frame, we simply cannot completely understand things that go on in micro/macro scale. All that "negligible" bullshit comes from our lack of capacity. That's why scientists spend so much money, time and effort on things like rockets and space ships. Because they have to look at all those tiny and numerous aspects to correct their calculations and get better chances of creating something reliable. Even then, they rely so much on trial and error.

i don't have any sources but anything in the universe being non-deterministic is extremely far fetched. you might as well attribute it to magic pixies as you would attribute it to something truly random. where exactly does quantum phenomena appear to be random?

en.wikipedia.org/wiki/Radioactive_decay

>Radioactive decay is a stochastic (i.e. random) process at the level of single atoms, in that, according to quantum theory, it is impossible to predict when a particular atom will decay,[1][2][3] regardless of how long the atom has existed. However, for a collection of atoms, the collection's expected decay rate is characterized in terms of their measured decay constants or half-lives.

This is basically what my current understanding is, when dealing with an individual atom the indeterminacy of the system means predicting behaviour isn't possible.

It's only possible to roughly predict large collections of atoms and how they behave statistically as a group.

But our interpretation of the facts of quantum mechanics is really just that, it's an interpretation, we don't have a deeper understanding of precisely what is happening so I don't think we truly understand if these processes have some really genuine randomness to them or if it's simply appears that way.

Remember that evolution of quantum systems is deterministic it's only measurements that are statistically distributed.

Functionally though, I don't think it matters, radioactive decay for now is sufficiently obscure that there's no predicting it and so it's useful enough to create CSPRNG from it.

There's many such systems that have those properties, this is another one youtube.com/watch?v=1cUUfMeOijg

>unstable circuit
So it's based on a variable, being an unstable circuit that can be manipulated? That's not random...

theoretically nothing is truly random. Hardware rngs are good enough for most normal users. (le normies)

Then you lied to OP!!

Read up on quantum mechanics, it's not just about lacking insufficiently sensitive measurement devices, it's actually that there's theorized fundamentally limit on how accurately you can know location and momentum, it's to do with the duality of wave/particle nature of everything at the quantum level. This is a good start - en.wikipedia.org/wiki/Quantum_indeterminacy

But probably find decent explanations on youtube

If you measure the direction of spin of an electron the answers are random but such that the average of lots of measurements give a specific value.

Another example is the double slit experiment, individual particles land randomly once passed through the double slit, but so on average they form a coherence pattern.

There's loads of examples...

But you're right, that's what I've been saying, we know that measurably through experiment that no theories so far lead to being able to predict these systems, but that doesn't prove they're in some way inherently random. What's actually happening under the hood, so to speak, is right now just conjecture and interpretation

this was my first post in this thread

Ah, my mistake bb.

In theory yes. Computers based on logic and mathematics cannot produce anything "truly" random. It's all about algorithms.

However RNGs have algorithms that take in seed values from outside the RNG (in many cases the internal clock, or camera data like pic related) to make it closer to real randomness.

>her eyebrows
You wanna know how I know you're gay?

[spoiler] there's no such thing as a random number [/spoiler]

nothing is random
fortunately many complex systems will never be predictable due to limitations on reality (e.g. need more memory than atoms in the universe) and most are unfeasible until we have nearly godlike ability (e.g. perfectly accurate model of earth's weather) GURPS tech level 13 shit

yes

Hardware RNGs exist that generate random numbers based on physical processes that are considered truly random such as quantum events like tunnel effect or nuclear fission

Doesn't some company generate randomness from photos of a wall of shelves with lava lamps?

>nothing is random
Physical processes on a quantum scale are considered truly random, and there's nothing we know of (as of now) that suggests otherwise. Radioactive decay is a classic example. There are hardware-add ons that utilise radioactive decay to provide truly random elements to RNG.

>nothing is random
I think that most modern theoretical physicists would disagree with this and describe the observables from QM measurements to be categorized as random, with some probability distribution.

Or at the very least, "not predictable", which might be a difference without distinction, at least when RNG is concerned.

Flow of fluid is deterministic.

The cheapest possible source of true randomness would be tunnel effect or photon noise

are you retarded? not currently predictable!=random

user what are you talking about...

that's not really true. on embedded systems you have a pre-generated seed to calculate your random values from, that seed is usually coming from hsm devices which "can" be real random values. and even if you don't have a hsm, most modern cpus have crypto closed modules to generate random, which most like fall back to the same method, pre-generated seed which random bases its calculation on or at boot time initialization processes and other stuff to generate entropy.

/dev/urandom or /dev/random are all pseudo at the end of the day as they only work with pre-generated entropy.

Arduino can generate random from white noise read from analogue pin with no signal on it

you can use your soundcard to generate entropy, that's nothing new.

it's not really random, it's more reliable than, let's say the values that CRC32 produce for integrity checks, but it's still not truly random.

buy a geiger counter or lava lamps. there are some tutorials online how to tinker your pi into a hsm with a geiger module. then again, embedded security is usually so bad, making a hsm out of it is like trying to eat soup with a fork, pointless.

Yes.
No.

yes that was cloudflare iirc but that is now just a show off, they have long replaced it as source of entropy.

As far as I know Cloudflare uses that. But instead of one lava lamp it's a wall full of them.

Jon? Jon how are you? What happened to Jacque?

see temperature is controllable and when speaking about reliable entropy you should also consider further security implications other than the random values, things like side channel and time attacks to compromise those.

i would rather have additional hardware that is communicating directly with the hardware over TLS or other tunnels to transfer random data than calculating by entropy based on how hot my cpu might be after the initialization of my dns service at 4.4sec boot time

Yes,

cloudflare

not to go on a tangent by injecting philosophy
free will of sentient beings may not be deterministic

so that wiggling of mouse to the left vs to the right may actually be random

The key to understand randomness as well as free will is "deterministic chaos".

You can not predict the movements of something as simple as a double pendulum beyond, even though it's "deterministic" in a mathematical sense.

But be aware that you don't need to predict something to make assumptions about something. If something a number is """truly random""" but follows a Normal distribution with very low variance, you can still guess the number - not mathematically (because there is still a 0.0001% chance that it's a very high or low number) but your guess will be "good enough".

If it uses the system time, it is deterministic.
Cursor position algorithms are possibly random, if the human brain is indeterministic (we don't know yet if the brain relies on quantum randomness or not).