[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/sci/ - Science & Math

View post   
File: 4 KB, 190x223, g.jpg [View same] [iqdb] [saucenao] [google]
3426258 No.3426258 [Reply] [Original]

So how long until we can produce truly random numbers from programming?

>> No.3426265

It's impossible

>> No.3426271

My own random number generator has been generating nines since 2003...
But that is the thing with randomness, I can never be sure if it's really random.

>> No.3426274

>>3426265
This, random events cannot be created - otherwise they are not truly random

>> No.3426279
File: 65 KB, 450x510, 1282172614767.jpg [View same] [iqdb] [saucenao] [google]
3426279

>>3426271

>> No.3426281

>>3426274
Situations that create conditions that are so close to random can indeed be created. However the results of this randomness still come from a number of factors that cause the 'randomness' which result in it not being random.

>> No.3426283

We could use quantum mechanics to get the "truest" randomness possible. Each potential state represents a value, then you just measure the system and interpret that as your value.

>> No.3426321

Even something genuinely random would have rough guidelines it should follow.

If we had a truly random process that could roll a die (pick between 1, 2, 3, 4, 5, or 6), after 120,000 rolls, we could expect approximately 20,000 of each number to appear.

At what point of deviation from this expected value would we question the authenticity of our random generator?

>> No.3426334

Just use circuit that suppresses external noise/signals and amplifies thermal noise. Thermal noise = pure randomness I mean you can't get entropy any higher than random motion of atoms (fluctuations of current). Depending on how much independence from the environment do you need you can just pick up either simple circuit (amplifier with a hot resistor as "signal" generator) or more complicated circuits. You can also just use casing for EM shielding - as an additional/main suppression.
How to get it digital? - Just convert the output by A/D converter.

Done! The fluctuations are random. Obviously we assume the transfer function of your amplifier is of an appropriate shape (white noise = constant so it doesn't amplify some frequencies more then others).

If anyone disagrees about the random nature of the mentioned random signal generator 2'nd law of thermodynamics would like to be revised by you.

>> No.3426342

Nope.

Certain types of radioactive decay are really, truly random and can be used to make real random number generators.

This is how REAL mathematicians and statisticians do it.

>> No.3426346
File: 20 KB, 334x359, 1177583313138..jpg [View same] [iqdb] [saucenao] [google]
3426346

>>3426258
We can't, because random doesn't exist. Its an illusion.

We live in a world of causality. Randomness cannot exist in a world of causality because it creates a paradox.

Oh, and don't listen to the idiots saying that randomness exists in quantum mechanics, because is doesn't. These idiots read this jargon out of theoretical textbooks, and have no real understanding of how quantum mechanics works.

>> No.3426350

>>3426342
Do you disagree with thermal noise being random even if modified by transfer function of amplifier?

Radioactive decay = practical.
lol man how radioactive your component is supposed to be in order to generate 10^6 - 10^9 random numbers from cubic centimeter of your matter and still keep it safe?

>> No.3426354

>>3426265
Regular computer hardware is designed to be completely deterministic, so it can only produce pseudorandom numbers. If you want a random number you've got to add some special hardware that reads in some random physical phenomenon like the noise created in a reverse biased zener diode by quantum electrodynamic effects.

>> No.3426359

>>3426350
http://www.fourmilab.ch/hotbits/

Anyone can get the numbers. Don't hear the words "radioactive decay" and leap to the "OMG WE'RE GONNA DIE" conclusion.

>> No.3426364

>>3426354
Take something deterministic, make it more complex, and reasonably incalculable, and guess what? It's still deterministic.

>> No.3426369

>>3426346
Go away, learn some things, then come back. We'll wait.

>> No.3426374

>>3426364
The noise created in electrical circuits by quantum effects is not deterministic.

>> No.3426380

>>3426374
No, it's not feasibly calculable. It's still deterministic.

>> No.3426382

randomness exists if values can be generated, and used, faster than than they can be predicted.

white noise is the easiest way to implement a random figure generator.
not really possible from programming alone though

>> No.3426386

>>3426364
how the fuck is quantum electrodynamic effects deterministic?

and you do get a qubit generator thing. (or whatever its called) its a box that you plug into your USB and it has a quantum computer chip that can generate random numbers. I dont think its available at the moment because you have to cool it but if you have the money you can get it.

>> No.3426392

>>3426380
The outcomes of quantum events are not deterministic. To determine where a particle will be in the future you must know what its momentum and position are in the present. This is not possible on the quantum scale due to Heisenberg's uncertainty principle.

>> No.3426398

>>3426359
"produces data at a modest rate (about 100 bytes per second),"

OMG so much! I think it will take just about 10 minutes to generate crappy Civ 2 map (on old computer it was under a second).

Seriously 1MB of random data will take ~2 to 3 hours to generate. Old shitty hardware was able to make it in ~10 seconds. I'm not saying radioactive is bad method but for practical purposes it's useless.

>> No.3426399

>>3426380
you've never done QM before, have you?

>> No.3426410

>>3426392
This doesn't show it's not deterministic. This shows we currently have no means of calculating it with the tools available.

>> No.3426420

>>3426410
We have means of calculating the future positions of particles. What we don't have is instruments capable of providing the numbers needed to plug into the equations. We will never have such instruments for the same reason we will never have a ship capable of travelling faster than the speed of light or a perpetual motion machine. It is a physical impossibility, not just a lack of technology.

>> No.3426423
File: 49 KB, 450x412, 1297395280997..jpg [View same] [iqdb] [saucenao] [google]
3426423

>>3426392
>To determine where a particle will be in the future you must know what its momentum and position are in the present.

No, you need to know a particle's relationship with its surrounding particles. Then you can predict what will happen. Simply calculating a single particle's position/momentum doesn't work. The Heisenberg principle is bullshit.

>> No.3426428

>>3426410
For fucks sake it's random enough to be used in ANY software application. Yes you are right it is deterministic IF you know the properties of the sub-Planck scale string-whatever quantum structures which are by definition immeasurable then you can determine the exact result of the quantum process.
So if we assume we can get information that we can't get then it is deterministic. Otherwise in practical world it's random.

>> No.3426432

FIRST OF ALL BEFORE THIS SHIT STARTS...
Let us define deterministic. A deterministic system is one in which the configuration of its previous states determines (that is it causes it to happen, with no other outcome being possible given that system) the progression of that system. Given the same initial conditions, the outcome with be the same.
With that knowledge, we can conclude that we cannot know whether quantum mechanical systems are deterministic or not.

>> No.3426436

>>3426423
There is a soccer ball in motion somewhere on the planet. I will either tell you where on the planet, or I will tell you how fast it is moving and in what direction. Are you capable of telling me where the soccer ball will land?

>> No.3426437

Let's say I have a function F(x) that is differentiable everywhere.

Now, for any real x, you can request one of the follow, but not both:
(1) the value of F(x)
(2) the value of F'(x)

Given a finite amount of selections, you could not accurately describe F to me. F would still exist and as long as I know what F is, F would not change. The value of F for any value of x is already determined.

>> No.3426449

>>3426432
it already started.

>> No.3426450

>>3426420
not the one you are replying.
In general I agree but with the perpetual stuff you have a bit over shot it. Relativity at cosmic scales is in contradiction with conservation of energy as such there is possibility to destroy or generate energy on cosmic scales as such it might be possible to create something that could work forever. - I mean it's still an open problem for more exotic parts of science (not for common once like mechanics and such obviously).

Please correct me if I'm wrong I don't want to mislead anyone.

>> No.3426453

>>3426423
That's the thing: particles aren't really particles, they're waves. They don't have fixed "exact" momenta (is that the correct plural? I don't really know Latin), they have a "smeared out" potential to interact as if they had several momenta. They don't really occupy one point, they have the potential to interact in an (actually infinite) range of space.
Your intuition means nothing at scales you have never seen.

>> No.3426454

>>3426436
Nope, but you could. Deterministic.

>> No.3426459

>>3426450 Relativity at cosmic scales is in contradiction with conservation of energy

You're going to need to back that up. It sounds like something you made up.

>> No.3426460
File: 1.41 MB, 312x161, 1275147095247.gif [View same] [iqdb] [saucenao] [google]
3426460

>But I come to you now humbly, with help for a conundrum

>> No.3426463

>>3426454
Its an analogy. There isn't anyone or anything in the universe that is capable of knowing the position and momentum of a quantum particle simultaneously.

>> No.3426469

>>3426436
No I can not. But a powerful enough computer to do the calculations can.

>> No.3426470

>>3426463
Except the universe...

>> No.3426475

>>3426459
One sec. I'll have a look for it. If you can show that it's not true I'll be glad.

>> No.3426476

>>3426469
So you don't understand physics or computation. Here's a set of linear equations.

x + 2y + 3z = 14
3x-y-8z = 12

No computer no matter how powerful can solve for all three variables.

>> No.3426484

>>3426463
>>3426469
Actually I think these both are kind of misleading. It's not really that you can't *know* the momentum and position simultaneously, but they still exist out there somewhere. It's more like the particle has a range of potentials, but spontaneously interacts as though it has only one value.

>> No.3426487

>>3426463
You certainly have high hopes for intelligence my friend.

>There isn't anyone or anything in the universe that is capable of knowing the position and momentum of a quantum particle simultaneously.

We can conclude:
The particle itself isn't aware of its position and its momentum simultaneously.
No intelligent system can ever be defined to measure both values simultaneously.

So what do you believe causes a particle to experience a change in position or momentum then?

>> No.3426489

>>3426475
http://www.physicsforums.com/showthread.php?t=506985
First sentence.


and one example was shown with photons being red shifted and hence lost part of their energy if I remember well - but not sure about the example.

>> No.3426491

>>3426476

Probably because there are inifnite solutions

>> No.3426495

>>3426476
There are infinite solutions to that system. n is any real number.
<div class="math">(n, \frac{148-17n}{13}, \frac{7n-38}{13})</div>

>> No.3426496

>>3426476
That's actually an excellent analogy. x, y, and z all exist and and can assume any feasible value, but if you are told the value of any one of the three, suddenly the other two are bound to a single value.

>> No.3426507

>>3426487
"Awareness" has nothing to do with it.

>> No.3426510

>>3426258
Krakengineer !!5XY+x7grkpt
Please can you have a quick glimpse on the :

http://www.physicsforums.com/showthread.php?t=506985
First sentence.

And explain what assumptions were maid to get that result (violation of 1'st law of thermodynamics). Since I find it as decently difficult to believe as well.

Many thanks.

>> No.3426515
File: 43 KB, 560x380, 8292-1[1].jpg [View same] [iqdb] [saucenao] [google]
3426515

>>3426507

>> No.3426533

>>3426510
Energy is a tricky concept.
Here: http://en.wikipedia.org/wiki/Conservation_of_energy#Relativity
It seems that you're actually correct.