Jump to content
Sign in to follow this  
freddern

How does a computer do a random choice?

Recommended Posts

@ NurEinMensch and bn880 -- Do you realize who and how much work you are arguing with when you say there are no truly random events!?  The theorys are so strong that it is not a matter of opinion. Your argument would have some weight if we just looked at the results of things such as radioactive decay and declared that it is random just because it looks random, but that is not the case. We declare it random because we understand the process and can see that randomness is a part of it. In other words we know it is random because of the theory behind it not because of observation.

I'm aware of this. Sorry for lack of respect. As soon as I really understand QM I'll let you know.

Until then I'll just keep thinking "how can anything be random?"

And now I need my sleep. biggrin_o.gif

Share this post


Link to post
Share on other sites

The amount of work done has little to do with it. I could not care less, I am not responsible for this work. smile_o.gif You can say things like this for just about any theory or method "do you dare not agree with it, we spent so and so many hours on it".

Not only that, but it's the interpretation that is probably wrong, and if not that then it's just an incomplete theory, not wrong.

Share this post


Link to post
Share on other sites

It's threads like these that make me want to drop my combined finance/economics degree and go straight into physics.

I'm sure that in the long run however, pragmatism will overcome my scepticism at what is essentially a degree in sociology and human behaviour rock.gif

Share this post


Link to post
Share on other sites

It is interesting to see, how some people declare the end of physical development after the QT. It almost sounds like 'The Earth is flat.'

Just because we have not the means to explain a phenomen today, does not necessary mean, that we should just accept it as is. The QT is a model, that is supposed to describe how a certain part of our nature works. The advantage of this model is, that it works in situations, where previous models fail. However, this doesn't make previous models completly wrong - I don't stress Maxwell everytime, when Ohm is still sufficient to solve my problem...

Before Copernicus, the earth was center of the world and everything else was moving somehow around it. Observations forced the astronomers, to create complicated mathematic equotations and tables to predict the move of the planets according to that model. And some of those formulas and tables would still work today. And if their formulas and tables failed, it was because the formulas and tables were wrong/incomplete and not because the model was wrong.

And now that I completly lost my point biggrin_o.gif, I want to end with a quote of Niels Bohr, that pretty much sums it up (or not):

Quote[/b] ]The opposite of a correct statement is a false statement. But the opposite of a profound truth may well be another profound truth.

Edit: btw., I like it, how an almost simple technical questions slips into a philosophical discussion smile_o.gif

Share this post


Link to post
Share on other sites
Before Copernicus, the earth was center of the world and everything else was moving somehow around it. Observations forced the astronomers, to create complicated mathematic equotations and tables to predict the move of the planets according to that model. And some of those formulas and tables would still work today. And if their formulas and tables failed, it was because the formulas and tables were wrong/incomplete and not because the model was wrong.

It is a matter of perception, really. If you lock the coordinate axes of space to the center of Earth, then everything really IS orbiting Earth, albeit the orbits look really funky. I once coded a small program, which simulated the behaviour of planets in the gravity fields of our solar system. That program had on option to lock the perception to Earth, providing a set of really goofy orbits of other bodies of our solar system.

It is just that the calculations become much easier, if you lock the coordinate axes to the center of the sun.

Share this post


Link to post
Share on other sites
It is a matter of perception, really. If you lock the coordinate axes of space to the center of Earth, then everything really IS orbiting Earth, albeit the orbits look really funky. I once coded a small program, which simulated the behaviour of planets in the gravity fields of our solar system. That program had on option to lock the perception to Earth, providing a set of really goofy orbits of other bodies of our solar system.

It is just that the calculations become much easier, if you lock the coordinate axes to the center of the sun.

So you just proved me right, it was just the lack of the mathematical means to properly describe, what they were not able describe biggrin_o.gifbiggrin_o.gifbiggrin_o.gif

Share this post


Link to post
Share on other sites
Basically, unpredictable is not random.

Why not?

I'm not necessarily disputing this statement.  I'd just like to know what other properties a random number has besides unpredictability.

Anyone? rock.gif

Share this post


Link to post
Share on other sites

Interesting Brain-Knocking match we have going here.

I think the whole idea of "psuedo-random" va "random" is rather silly.

Just because we don't undertstand every possible "seed" that goes into which side a coin will land on doesn't mean it's 'randomness' is any more random than a computers output.

After all, a computer is built by people. What if Bob, who designed the first computer with a random number generator, had had a piano fall on his head the day before he submitted his plans? With Bob dead, someone else would have to design the chip. This chip would have different specifications from the one Bob was going to design, and the chip would have generated it's first random numbers under different circumstances. As a result every random number generated from that chip would be different from the chip that "Bob" would have designed.

My point is, just because we don't fully understand all of nature's "random seed(s)" in regards to natural/physical events, doesn't mean machines we've created aren't creating real 'random' numbers.

After writing this and thinking a bit (it's getting smokey in here), the whole concept of "random" seems strange to me - like an abstract definition of something we can't predict or don't understand.

Is radioactive decay *really* random, or are we giving ourselves too much credit?

Share this post


Link to post
Share on other sites
Just because we don't undertstand every possible "seed" that goes into which side a coin will land on doesn't mean it's 'randomness' is any more random than a computers output.

Well, the big difference comes when you really need something random, to be used in strong cryptography for example. A computer generated pseudo-random value is no good, since the "attacker" only needs to know the seed you used (and the random algorithm) and he gets the exactly same random numbers you used to do the encryption.

If you do this by flipping a coin 1000 times for example and using the results as the key, even if the attacker would know the exact way you flipped your hand to throw the coin its very unlikely that he would get the same results you did.

Share this post


Link to post
Share on other sites
Quote[/b] ]Well, the big difference comes when you really need something random, to be used in strong cryptography for example. A computer generated pseudo-random value is no good, since the "attacker" only needs to know the seed you used (and the random algorithm) and he gets the exactly same random numbers you used to do the encryption.

Understood...All the attacker woukld need then, is a computer that can "emulate the universe" to figure it out then. :P

How many Pentium4's do you think that would take? biggrin_o.gif

Share this post


Link to post
Share on other sites
[...]If you do this by flipping a coin 1000 times for example and using the results as the key, even if the attacker would know the exact way you flipped your hand to throw the coin its very unlikely that he would get the same results you did.

And here we go. If the attacker knew the way we flipped it with infinite precision he would definatly get the same results.

Quote[/b] ]Understood...All the attacker woukld need then, is a computer that can "emulate the universe" to figure it out then. :P

How many Pentium4's do you think that would take?

On the other hand that computer would need to emulate itself (emulating itself emulating itself...) since it is part of the universe, too.

This makes it impossible to do that in practice.

Share this post


Link to post
Share on other sites
And here we go. If the attacker knew the way we flipped it with infinite precision he would definatly get the same results.

Or not, since the coin would be not only affected by the way the coin is flipped, but also by the air it flies through (currents, pressure, etc.) and the surface it hits and from how high and dozens of other factors.

With an universe simulator you might be able to simulate that yes, but no such things exist (and never will, its not a matter of processing power but storage space and lack of knowledge of how the universe was born or what exatcly every atom in the world contains) so its not realistic to keep it as a threat to your cryptography or whatever else you are doing.

Share this post


Link to post
Share on other sites
Guest
And here we go. If the attacker knew the way we flipped it with infinite precision he would definatly get the same results.

You can't have infinite precision. The laws of physics are such as there is no such thing as infinite precision. Quantum Mechanics again.

Quote[/b] ]Just because we don't undertstand every possible "seed" that goes into which side a coin will land on doesn't mean it's 'randomness' is any more random than a computers output.

Yes it does because I can do a correlation analysis on the computer's numbers and get a direct correlation. I can give you an example - I made once a neural network that predicted the random numbers of the Java 1.1 random number generator.

While this might seem irrelevant, it's not in engineering applications. For instance many forms of adaptive filters requires that you can generate 100% uncorrelated noise. This is done by using tables of radioactive decay times. No computer generated random numbers are uncorrelated enough.

Share this post


Link to post
Share on other sites

That's where we wont agree. Some say there isn't infinite precision and point to the therory of quantum mechanics and other's (like me) say there is infinite precision, probably not in practice but in theory.

Basically what I say is you could describe the universe as a big computer doing nothing but emulating itself and as such it would be predictable if there was anything outside the universe. (which i doubt but thats another story wink_o.gif )

Share this post


Link to post
Share on other sites
Guest
That's where we wont agree. Some say there isn't infinite precision and point to the therory of quantum mechanics and other's (like me) say there is infinite precision, probably not in practice but in theory.

This is not a practical limit it is a theoretical one. As I mentioned, the computer in front of you is evident that it is correct. None of the transistors in your computer would work if the QM models were wrong. Disputing the uncertainty principle in QM is like disputing that gravity exists or that the earth isn't flat.

The only difference is that the knowledge of the average person about quantum mechanics is limited and hence the origin of many crackpot theories.

It's a physical fact staniding on an a very solid mathematical and empirical ground. You could just as well be claiming that the earth is flat - it would be equally wrong.

Just because you don't know and understand the physics behind it does not mean that it doesn't exist smile_o.gif

Share this post


Link to post
Share on other sites

Okay,

basically I don't agree with Denoirs' explanation of Quantum Uncertainty and the terminology not conept used by Kegetys.

I think Denoir inappropriately deduces that Quantum uncertainty automatically establishes there is such a thing as a 'random' in nature. He is not alone. While it is accepted we can not predict between quantum states, it does not mean the rudimentary process behind it is truly random.

I agree with Denoirs' theory if the word 'unpredictable' is used, not 'random'. I find 'Random' comes from a lack of understanding throughout our history of unpredictable occurances, and it's nothing to be ashamed of. It has no place in science for me. Yes it is now impossible to predict precisely how a coin will fall and it may never be possible, that makes it unpredictable not truly random. It serves the purpose of encryption.

Quantum Entanglement can shed some light on the acceptance of 'random' vs 'unpredictable'. If particles can be 'entangled' it means they can not behave randomly, just unpredictably to us.

It is about interpretation and semantics. Not actually questioning Quantum Mechanical theories.

Share this post


Link to post
Share on other sites
Guest
Quantum Entanglement can shed some light on the acceptance of 'random' vs 'unpredictable'.  If particles can be 'entangled' it means they can not behave randomly, just unpredictably to us.

QE describtes the relationship between particles (ie conservation of spin, momentum etc) and it is in no way relevant to the randomness of nature.

Analogy: If x is a random numer then sin(x) is a random value too.

Quote[/b] ]It is about interpretation and semantics.

No it's not. It's about people not having the first clue about physics.

Quote[/b] ]I agree with Denoirs' theory if the word 'unpredictable' is used, not 'random'. I find 'Random' comes from a lack of understanding throughout our history of unpredictable occurances, and it's nothing to be ashamed of.

Agian the source of this is people not having a clue about mathematics. The definiton fo random is:

Quote[/b] ]

random, adj (Statistics): Having a value that cannot be determined before the value is taken , but only described probabilistically.

Taken from Unwin-Hyman Dictionary of Mathematics.

Share this post


Link to post
Share on other sites

I love the way I can ask anything and get a fulfilling answer around here! (I started this tread)

(btw: did not understand cr..p) but I still love it! biggrin_o.gif

Share this post


Link to post
Share on other sites

Ok Denoir, I simply have a different understanding of the word random than you do, and the definition should be

Quote[/b] ]random, adj (Statistics): Having a value that cannot be determined before the value is taken by current or any methods known to date, but only described probabilistically.

Enough BS about the un-disputed right and wrong, it is about semantics.

EDIT: So anyway, I do NOT dispute QM in any way, perhaps just some peoples deduction that it verifies the word 'random' as per standard definition.

Share this post


Link to post
Share on other sites
Quote[/b] ]random, adj (Statistics): Having a value that cannot be determined before the value is taken , but only described probabilistically.

After reading a *tiny* bit about Heisenberg's theory, I have to say it's a pretty interesting read.

Quote[/b] ]In the sharp formulation of the law of causality-- "if we know the present exactly, we can calculate the future"-it is not the conclusion that is wrong but the premise.

--Heisenberg, in uncertainty principle paper, 1927

...and the premise is that we cannot know the present exactly, which I would have to agree with 100%.

Quote[/b] ]I can give you an example - I made once a neural network that predicted the random numbers of the Java 1.1 random number generator.

What if design flaw or some other 'random' biggrin_o.gif event (power surge/Sunstorm/Previously Unknown hardware Errata/Manufacturing defect/ect. ect.) in the computer hardware your program was running on caused one bit to be changed, and a different number to be generated?

Despite your cool little program's ability to "know the present exactly" in regards to the laws of the computer it operates in, thre will allways allways those outside/other influences which we cannot and will not be able to predict.

This bring up another abstract term to ponder: The concept of 'Perfection'

Share this post


Link to post
Share on other sites
Quote[/b] ]In the sharp formulation of the law of causality-- "if we know the present exactly, we can calculate the future"-it is not the conclusion that is wrong but the premise.

--Heisenberg, in uncertainty principle paper, 1927

Good old Heisenberg got it right. That's what I am saying.

If we knew it (which we never will) we could calculate the future. And probably the past.

That's equivalent to saying universe is a computer emulating itself.

tounge_o.gif

Share this post


Link to post
Share on other sites
Quote[/b] ]"if we know the present exactly, we can calculate the future"

We could only calculate the (possibly alternate) future which would happen without the calculation (a computer simulation of the universe for example), not the absolute future since a simulator could not simulate itself, thus in the simulated universe the unvierse simulator they build there would not work, and they couldnt see the true future that we can influence to by knowing the future the simulator shows.

Share this post


Link to post
Share on other sites
Quote[/b] ]"if we know the present exactly, we can calculate the future"

We could only calculate the (possibly alternate) future which would happen without the calculation (a computer simulation of the universe for example), not the absolute future since a simulator could not simulate itself, thus in the simulated universe the unvierse simulator they build there would not work, and they couldnt see the true future that we can influence to by knowing the future the simulator shows.

Then agian, is OUR future the true future?

Share this post


Link to post
Share on other sites
Guest
Ok Denoir, I simply have a different understanding of the word random than you do, and the definition should be
Quote[/b] ]random, adj (Statistics): Having a value that cannot be determined before the value is taken by current or any methods known to date, but only described probabilistically.

Enough BS about the un-disputed right and wrong, it is about semantics.

EDIT: So anyway, I do NOT dispute QM in any way, perhaps just some peoples deduction that it verifies the word 'random' as per standard definition.

It is defined throught the mathematical definition of 'random' not through some mysterious philosophical point of view.

I could cite the math/logic equations that define it but it won't make any difference to you.

These are not things open to interpretation they are a mathematically deduced based on the fundament of heavily verified physical facts.

I'm going to drop this discussion now since you obviously won't stop FSPiloting this issue.

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
Sign in to follow this  

×