I think (?) you are arguing at cross purposes. Whether you could tell if you were in a simulation is one question that everyone else seems to be arguing about, and you are worrying about whether it would be reasonable to make a sub-perfect simulation.

( I already argued that it is hard to prove you are not inside a simulation but maybe doable if the simulation is very close -close enough- to the real nature used as prototype)

I basically proposed a way out of thinking we are in a simulation with a theory of everything possibility that notices failure in our situation once and if such theory becomes available to us. I am also arguing of course that whoever is simulating real nature is unavoidably using something as complex as nature itself so basically its worthless effort of a faithful simulation of their world, just go ahead and create a universe instead. And it will also require them knowing all of nature's laws.


Essentially i am arguing any simulation that is not nature itself is a failed one.

Also you havent proven to me that the true randomness assumed in QM is replicated perfectly by deterministic random number generators. So essentially how will the simulation proceed without using nature as a guide already. (this is also a clue of a possibility to spot you live in a simulation if your randomness magically fits some algorithm of a pseudo random number generator-hard but maybe doable)

I'd think that you'd just get the data inside the simulation and it would be what you built your laws upon.

And I know that you've done this. The particular people who you are disagreeing with are only worrying about 1/2 of the two problems...

That theory of everything would coincide perfectly with the programming though. Also, every time we have had a theory fail, we just went on to develop a new theory.

It really would depend on what you wanted to simulate and experiment on.

No need to worry about the objects past Jupiter if all you want to know is how a masque de z acts in a world within and without a twoplustwo.com smp forum.

I am of course, assuming that we can make pretty lights up in the sky and whatever sorts of math and science things you would read available through just scanning them in.

Doesn't need to be a perfect replication. Just needs to be replicated sufficiently well that you can't tell whether it is truly random or not. Even if it were a difficult problem, trial and error would fix it.

Is it possible that some aspects of quantum mechanics exist as a means of approximation to save space? For example, if a "real" universe has electrons that behave classically, could a simulation where they behave probabilistically be represented with less matter than the original?

(Sorry masque, but I'm sure due to time zone diffences you are asleep.)

The laws of nature we derive from looking at stuff is the zip file.

Will the beer be better?

Opinions will vary. Overall, people will believe that the beer of yesteryear was better, even though it wasn't. That is the only important law one could possibly derive from the data.

have you ever actually witnessed 2 electrons with this much energy collide, or have you just interpreted the results of such a collision?

How about,

Real circles exist in nature (electrons orbiting?)

A circle cannot be perfectly simulated

This

Simualtion in no way implies what Masque seems to think is necessary.

anyway maybe the basic stuff of physical laws is some very simply stuff like cellular automa which could easily be simulated.

This seems close to maximally falsified by recent developments in physics. All the dynamics of 1 gr of matter in 3D can be faithfully simulated by 2D conformal field theories..

Yeah, or the extent to which it can exist in nature....which is the exact same.

No it int, because number one a precise safe formulation of holography is not yet available convincingly (is string theory or quantum gravity a fact of proven science yet) and even if it does it doesnt invalidate what i said because then the 2 "worlds" are properly correlated 1-1 to be identical for all practical purposes as the boundary is the effective description of the 3D. How you can choose to view 3D world doesnt suddenly make it massively economical. It is still equivalent to the real world if you choose to use the boundary instead.

Just so you know what you're signing on for, he's advocating replacing the real numbers with computable numbers in the development of mathematics, and replacing real analysis with computable analysis.

Computable numbers don't satisfy the least upper bound property that the real numbers do. There are only 2 ways to complete the rationals: the reals, and the p-adics.

But you don't need a gram of matter to simulate a gram of matter....which if what you said. (And I don't know what you mean by "precise safe formulation"....because that's exactly what I would say Maldacena duality is)

oh ok....I didn't realize (and don't think I meant ) to sign up for that. "Computable" seems to mean different things in different sentences with varying degrees of precision.

If holography or a better version of it proves true then i still am correct in what i said because clearly that 1 gram is no different than the equivalent boundary used to describe it. I do not see what you argue here. If there is a faithful correspondence/connection between the boundary and the 3D system (in the sense that the boundary fully encodes the information of the 3D world) why arent they the same thing in all that matters? Why doesnt that yield the 3D world an illusion instead and simply reformulates what we perceive as world in a better way? Also realize that i was aware of such connections when i said what i did.

someone enlighten me please, I will repeat question

perfect circles or any kind of curve- exists in reality but there is no way to perfectly simulate a true circle... is that too simple an argument?

Don't irrational numbers have some kind of function in true nature? Square roots of prime numbers and things like that would be needed somewhere surely otherwise there are going to be mistakes, the real universe would display a different outcome due to its analogue precision... It just seems that an infinitude of computing power/memory (omniscience hence god) would be required, quantum minds and quantum computers.

I wasn't advocating anything.

That you don't need matter to simulate matter.

Because they are also different. They have different dimensions, one contains gravity the other doesn't etc.

To make it even more clear...if somebody said "gravitational physics can never be simulated by non gravitational theories" and somebody else points out AdS/CFT....I don't think it's a reasonable response to say "Well, if the theories are equivalent the CFT side is a gravitational theory so I'm still right"

You are free to make that interpretation....I don't see why it is relevant though.

Ok?

I don't see how it is possible for something to "exist in reality" but there is no way to "perfectly simulate" it.

You are arguing using unproven theories and frankly equating nature with a present and not yet final mathematical description of it that may eventually prove not adequate and therefore lead to unfaithful simulation.

Furthermore you didnt understand either because of my poor choice of words or because of a consistent polemic tendency to disagree with me (probably both) that when i said you need matter to simulate perfectly matter i meant nature to simulate nature or something equivalent as complex as that (in prior posts it should be obvious i didnt take a position as rigid as only identical matter - as it appeared in one isolated post used for emphasis- but i intended to mean as complex as that matter including of course any holographic representation of it) . If that proves a boundary or not is irrelevant to the point. The point was that the simulation ends up being as complex as nature itself so its essentially no different.

Although it is for me hard to prove that claim i can happily rely on the observation that any simulation we attempt is only a partial poor representation of the real thing and the harder we try the more intense the usage of resources that eventually will become as gigantic as the nature simulated in effect. How many supercomputers do you need to simulate a nuclear weapon explosion and how far from the real thing is that simulation still??? So to simulate what happens in a 10^12 m^3 volume you need a 1 m^3 volume of computing "matter" and you still of course have failed to faithfully represent it in all details i can imagine are needed to be truly faithful (ie you do not consider full refined QED or QCD or post standard model physics when performing a nuclear weapons simulation so your volume of study wont include effects present in reality).

In any case lets stop hiding behind words or sentences and lets reframe what i said.

I said that in order to faithfully simulate a fraction of nature (such as 1 gram of matter) you probably need as complex a computing structure as that matter itself.

Furthermore i maintain a possibility that even this is not true. ie that you cannot simply simulate a fraction of nature without the whole thing. (a fraction of nature is never truly independent of all else).


All this was meant to imply that a very advanced civilization say that wishes to simulate nature faithfully will probably recognize that it is as complex as creating it for real. So far our best simulations not only use a ton of resources but typically only replicate to some degree only fraction of what is going on in some small volume of spacetime.

isnt reality quantized tho? (im asking cuz I dont know, I thought it was tho)

@Mt.FishNoob: You have to understand that even if the universe is analog in space and time, as long as energy is discrete, which quantum theory says that it is, then the analog information can be represented in discrete form without losing any information provided that it is sampled at a high enough frequency in time and space. That frequency is called the Nyquist frequency, and it is twice the highest frequency component in the analog signal. That means that I can sample an analog signal so that it is just a finite set of numbers or dots on a graph, and there is only 1 way that those dots can be connected to produce any possible analog signal. The analog information between the points is redundant. I can reconstruct the exact analog signal from those discrete samples. It isn't just an approximation. It's called ideal bandlimited interpolation. That's the basis for my field of digital signal processing. In practice we don't do a perfect job of reconstruction because we only have a limited number of bits to represent numbers, but that doesn't have to be the case if the energy is quantized and we use enough bits to represent the full dynamic range of the energy. The fact that I can turn any bandlimited (meaning frequencies can't go to infinity or energy can't change amplitudes faster than a certain rate) signal into digital data and then reconstruct the original analog data from the digitized samples means that I can create a digital state corresponding to every possible state of analog information.

No way for a partition of it to perfectly simulate it. The only possibility in my silly mind of it occurring in the first place is that the universe is god and god is omniscient.

I read about bandlimiting on wikipeida, I don't understand it.

'frequencies' can't go to infinity, so that is not true analogue then?

Isn't there an upper boundary anyway of what can be perceived? Surely it can just be limited to a number lower than infinity.