Interesting. So we're a simulation that can simulate 15 (femto)meters of cosmos that is indistinguishable from reality. This isn't much but it's still absolutely mind-bending. Reminds me of the movie The 13th Floor [1]
The tech review summary seems to be saying this super intelligent universe-simulating computer builder doesn't know about Bresenham's line drawing algorithm [0] and as a result it can only walk high energy particles along a limited number of paths.
Possibly? I can see why one would want to assume that an entity that can simulate a universe would be able to know more than the collective ability of that simulation, but on the other hand, we simulate things and learn unexpected results from our simulations. I think it's entirely feasible that a simulation can generate more knowledge, or different knowledge, than exists outside the simulation.
Certainly, but the idea that some hyper intelligent entity would develop the tech to build a universe-scale finite element simulator without groking anti-aliasing along the way is simply absurd. Bresenham's algorithm is a big deal because it's essentially the first thing everyone needs to invent when they start trying to render continuous processes on a discrete grid.
Where would you introduce the trajectory-correcting algorithm?
In General Relativity, fundamental particles are simply taking shortest paths through spacetime, and where there are multiple short paths, the one that extremizes the timelike distance is favoured.
In a perturbative theory of gravitation, you introduce corrections on the particles by way of graviton interactions.
Worse, whether a local patch of energy-density amounts to a particle is observer-dependent; in both approaches above an accelerated observer sees more particles than an unaccelerated one does. That includes more gravitons in the latter approach.
It would be a lot of work to write down a theory in which particles (whose count isn't global) decide on corrections to their own trajectories that return themselves to a path similar to one along a smooth manifold even though they are "really" travelling along a non-smooth one, so long as you are trying to match well-tested results of General Relativity. (A theory which doesn't reproduce those results is maybe an interesting toy; "cheating" by using a nearly-but-not-quite minimal length that has no direct observables is an uninteresting toy, and is not really a good argument for the existence of Simulators (or Creator Gods or whatever).)
In particular, the non-linear contributions of gravity will likely lead you into an explosion in the amount of knowledge that a particle would have to acquire in order to assess its trajectory; you'd probably also want to propose an explanation of how it goes about adjusting its trajectory when it discovers it's not on the "right" one. Don't forget to make this correct for all possible observers, including the ones that will see no particle and those that will see two or more particles.
In practice, attempts to do away with infinitesimals of length that do not clearly fail to reproduce known physics also tend to produce infinities of state in matter and/or additional gauge fields, and the infinities don't succumb to information-reduction techniques like renormalization by power-set counting.
If you're looking to encode that knowledge somewhere, like in a spacetime-permeating field, the gravitational field of General Relativity is right there and already serving that purpose in a non-toy theory that (in explorable parts of the known universe) behaves completely correctly with quantum field theories of matter, including The Standard Model. :-)
Best of all, in a block universe model, the whole thing is "pre-rendered", and there is no difference between the block universe model and an initial-values-surface formalism in the presence of determinism. So you only have to fully render one frame, and you get all the others, past and future. (Quantum mechanics as we know it is deterministic -- that's unitarity.)
It kind of irks me that they basically assume a cubic lattice. Nature does not form cubic lattices, when nature forms a lattice (for example, in a foam) it tends to form roughly dodecahedral cells (with coordination number 12). Given this property, a space-time lattice would be very nearly isotropic.
Yep, it's expanding symmetrically in all directions in that the distances may increase between bodies but the masses such as suns and planets do not grow.
Introducing more edges with your dodecahedral (or an icosahedral or some large-n polytope) cell worsens the edge-vs-face problem below, because a "cell shape" discretizes rotation and we have excellent evidence that rotation is smooth in our spacetime.
Small regions of our spacetime are locally Lorentz-invariant, meaning that a number of observable quantities (magnitude of angular momentum and mass are two) do not change for a particle held at the origin of a system of coordinates as the particle is arbitrarily rotated or boosted.
Arbitrariness is important. If we are observing a distant and predictable multifrequency radiator and define spherical coordinates with the radiator at the origin, then any movement our observation gear makes that isn't exclusively radial is equivalent to a rotation of the distant radiator. You can do this by holding the radiator at the (spacelike) origin, and the observer at its fixed (spatial) coordinates; in order to keep these coordinates constant, you have to rotate the system of coordinates to counter wholly non-radial relative movement. At large distances, the rotation at the origin becomes extremely small. Small or large, Lorentz invariance means the radiator has the same mass (it's by definition rest mass since it's always at the coordinates [0,0,0,t]) and momentum.
So arbitrarily small rotation goes hand-in-hand with arbitrarily distant observers, and also with nearer observers who can displace themselves tiny amounts.
So is there a minimum rotation?
The arbitrariness of rotation is a in conflict with "cell shape", as when your "cell" distinguishes between edges and faces, the discrete nature of rotated mass-energy-momentum exposed through the (corner-filled, discrete) cell structure leads to different observables under rotation compared to that of the (smooth, continuous) spacetime of (either theory of) relativity. This deviation is larger for objects of higher momentum; and remembering Einstein's relation for rest-massless particles, E = pc = \hbar\omega = h / \lambda, that means that for different-wavelength photons emitted from the same source, the higher-frequency photons will arrive later than the lower-frequency ones. We have good observational evidence against frequency-dependent arrival times from bright distant objects (supernovas, gamma-ray bursts, even millisecond pulsars).
We can conceive of a "cell shape" which is uniform under infinitesimal rotations, but at that point you have shifted one set of infinitesimals to another, and in the context of this topic (a simulation that among other things saves on state by abolishing real numbers in dimensions of length, rotation, boost and/or translation) is pretty much a non-winner.
(Additionaly, ignoring the simulation context, you would also almost always run into difficulties if your "cell size" -- a minimum length in space or a minimal interval in spacetime -- is large enough to produce observables.)
That's pretty natural since it is extremely hard to recover the inverse square law (for gravitation and electromagnetism) from any configuration of spacetime, assuming large dimensions (large compared to the Planck length).
Ehrenfest and Weyl showed this as early as 1920 and 1922, respectively.
3+1 is also baked into the Minkowski metric by the latter's definition, which leads to the Poincaré group being the isometry group on local patches of spacetime; the Poincaré group is a subgroup of the Standard Model, and is extremely well tested in controlled, laboratory settings (and supported by an abundance of observational evidence too).
Extra dimensions of spacetime have to be small compared to the Planck length in order not to be obvious today. For all practical purposes, if such tiny dimensions exist, we can safely omit them from effective theoretical descriptions of everyday physics, just like we can ignore things like very large extra dimensions where the smallest step one can take is much bigger than the Hubble diameter.
Agreed. We can only get to an estimation of the reality for the real thing is too complex... Until we break through the simulation and end up at the event horizon!
I'm no expert on philosophy, but I feel like consciousness is a subjective experience. So externally there would be no difference, but I'm pretty sure I'm conscious.
Of course, I've also heard of the illusion of consciousness, but that's where my brain explodes.
Still, human intelligence would be simulated within that, presumably. So the platform might not be the obvious one, but the software would still be theoretically possible.
That, or there's some interface to that platform that provides a source of intelligence. Maybe we're nothing more than Docker containers running on God's hardware.
But what if consciousness itself is simulated. The problem here is that a simulation would have to adhere to a set of programmatic rules meaning that the only way consciousness could arise as we like to think we know it, is via some weird unintended side effect.
What sort of moral issues result from creating a simulation universe that gives rise to sentient beings? I feel creeped out and reminded of Source Code with Jake Gyllenhaal for some reason.
That's assuming sentience is an emergent property of the simulation rather than something else, such as being the ground from which the simulation arises from.
The author has commented that he regrets having overlooked the moral implications of simulated life in his previous works (see e.g. http://gregegan.customer.netspace.net.au/PERMUTATION/FAQ/FAQ..., which contains spoilers for Permutation City) so I wouldn't be surprised if that was the inspiration for this story.
It touches thoughtfully and humorously on: AI, simulations, robotics, ethical and moral questions regarding artificial life, among many other interesting topics. It's one of the funniest and greatest science fiction books I've ever read.
Keep in mind that if you wanted to create a simulated universe you can ensure nobody ever finds out.
If someone does find out, you can simply fix the problem and revert to a point in time just before the data leading to the discovery was captured.
Speaking academically, you might be playing with fire if you probe the outer limits of the universe and we are in a simulation run by some entity concerned with keeping that a secret.
Who's to say "real" computing, or "reality" at all, has to adhere to the laws we observe within the simulation? That's like Luigi in Super Mario Brothers saying you can't kick a turtle into a brick in the our world and not expect a gold coin to pop up in front of our faces.
I'm not sure you can. Yes you can make a chest with 2 apples contain 5 apples if you put 2 more, but it's impossible (IMHO) to make it consistent. How many apples do you have if you place 2 chests besides 2 other chests and put 1 apple in each? What if you repeat the process 2 times and then another 2 times?
Also - when some things in physics aren't intuitive (QM for example, but even adding velocities in relativity) we don't think math is wrong - we think math doesn't apply in that way to this particular phenomena.
Sure he can. He can overload the "=" or "+" operator and make them do whatever he wants in that domain.
Taking it further, what if rules and logic, and by extension, math as we know it, is just a construct that is not applicable in the "real" universe. What if, in that "real" space, 7+7=5, or mathematical operations are not possible. What if causality and time dont exist?
You can change symbols all you want, but when life appears in your simulation they won't care. If they put 2 items and then another 2 items, and 4 items are there - for all they care 2+2=4.
And if sometimes its 5 - they will do what we did with relativity when it turned out velocities doesn't add up like they "should". We haven't changed addition - we changed the equations used in physics.
I guess you could hand-made an intelligent life that is designed never to discover the true math.
The point I was trying to raise, though, is what if the concept of "adding two things" has no basis in the "real reality". I know its hard to grasp...but imagine a reality where there is no "+" or "-" operation.
If you put 2 male mice in a box with 2 female mice and wait a while, you may find 5 or more mice in the box. Or if you put 2 clouds in a box and later put 2 clouds in the box, you probably won't find 4 or 5 clouds in the box.
The act of putting things in boxes (after first perceiving and classifying them as discrete entities) and the abstract notion of addition are only as related as you need them to be for a given purpose.
> The act of putting things in boxes (after first perceiving and classifying them as discrete entities) and the abstract notion of addition are only as related as you need them to be for a given purpose.
Yes. That was my point. How would you change the abstract notion invented by creatures in your simulation if not by changing reality to shape their abstractions? And how do you change enough instances that their math develops in such way that 2+2=5?
I think our basic math (logic, set theory, integer math) with some exceptions (choice theorem) is the only useful consistent basic math there is.
I do not believe that to be true. While physics is heavily dependent on math, the reverse is not true.
> I can make 2 + 2 = 5 in a simulation.
No, I am fairly confident that you absolutely can not. You can make a simulation where putting 2 oranges in a box and adding two more oranges results in 5 oranges in the box (rather like putting together certain nucleons together into a nucleus results in a mass different than the sum of their individual masses). But in that simulation it would be a peculiar property of oranges, not a property of mathematics -- just as we don't claim addition is invalidated by the behavior of nucleon masses.
There's no direct observation of addition, it's just a symbolic construct to model something we observe. I suppose a reality where such a model never made sense is possible, but it would be a very foreign environment that is completely outside the scope of our intuitive understanding. If it made sense only for a subset of the cases it does in our current understanding, then that would be the context in which it was discovered/understood.
Addition is defined in such a way that the universe it's performed in doesn't matter (in fact, no universe is required at all), which is true for most if not all of formal mathematics. See set theory 101 for the very beginning on how's that possible.
Right. The grandparent comment assumes the world outside the simulation is like what humans think they know about what they think they observe inside the simulation.
But what's outside the simulation could be quite different. And humans could always be mistaken about both what they think they observe, their observations themselves, and the conclusions they make from these.
At best its a matter of estimating probabilities (ie. educated guesses), and even there they're probabilities regarding appearances. It has been argued as far back as Kant (over two hundred years ago) that one can never get to the "thing-in-itself" that is behind those appearances.
I wish most of the people who talk about the Simulation Argument understood this.
There is not even a reason to believe that the "outer universe" has such things as space and time or information as we know it, and no way to know what a "computation" might comprise in such a situation.
Maybe the situation is not that pessimal, and an outer universe is much like ours, but to prefer that believe one would need evidence, of which we have none.
The range of possible universes is much more broad than we can imagine. Even so, some people have imagined much further than most, through practice or some natural tendency. Our intuitions developed to fit the universe we are in. Getting beyond that is difficult and not many even try.
Various eastern philosophies have a somewhat different solution to this whole problem.
Consciousness -- as in the sheer fact of experience -- is itself the base "reality," out of which everything is made. If you examine everything you're calling "the universe," all you will ever discover is sights, sounds, thoughts about sights and sounds, belief (or certainty) that those sights and sounds are caused by something external, etc. What are all of those made of? Consciousness, of course.
In this reality, "brains" are the entities that consciousness has dreamed up to demonstrate correlations with conscious properties. In our reality, we therefore say that "the brain causes consciousness."
If you investigate your experience closely, you'll discover that the epiphanies "I exist," "I am conscious," "I am alive", "a world exists" all point to the same ineffable "miracle" of existence, the one "substance" of which all realities are made.
Upon hearing this, perhaps the mind thinks "just as there are realities that exist without time and space, why can't there be some without consciousness?" There's a very interesting answer to that question, but it won't come in the form of thoughts.
Of course, there's no reason to believe any of that. On the other hand, as the Buddha said, don't believe me....
Assuming of course that the world outside our simulation is even remotely similar to our reality, which is not necessarily the case. Our reality could be a gross simplification of the physical systems outside of it, or maybe even one of many separate simulations that has had quantum mechanics added as a sort of "What would happen if..." test.
Quantum mechanics are only non-deterministic in a single timeline. Going back in time doesn't mean it's any easier to determine the future given the current state, so you can freely "replay" without leaking information.
"There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarre and inexplicable.
There is another theory which states that this has already happened."
― Douglas Adams, The Restaurant at the End of the Universe
That seems like a major roadblock, but hackers could imagine ways around it. Let's make it easier by assuming the creator can only store so many backups before running out of resources. That seems fair if simulations are anything like what we have now and restarting from scratch is unfeasible, perhaps due to lost input data or deadlines.
Then one approach would be to silently and irreversibly corrupt the simulation in a way that will become noticeable. Perhaps you fill every document with secret messages that require AI as smart as the sims to detect and is only revealed years in the future.
By the time it is noticed it's too late and removing all those messages will require more resources than the simulation itself. Trying to catch them on demand will require understanding how the brains of the sims work internally, which may be unfeasible.
That idea is probably preventable, but I'm sure there are exploitable and non-reversible security holes in any imperfect simulation.
Who says efficiency was not a concern when developing the universe? Otherwise the creators would not have included limitations like finite light speed or Heisenberg indeterminacy principle to reduce computation time. :-)
Go with one of the more complex configurations in conway's game of life. Imagine the little squares gaining sentinence and discovering they are merely parts of a game of life simulation. A bit absurd.
Yes, but in practice no one wants to go to a backup. So the simulators will try just deleting or messing with the aware person, creating Matrix-like glitches.
What is interesting is if we are the result of an unfathomable number of transistor gates, the consciousness that arises from it would never be able to understand the 'platform' (hardware) that it sits on.. would it? The idea being that you are creating a new dimension that cannot see the dimension that gave rise to it.
I feel like no matter what we are in a 'simulation' where another dimension gave rise to ours whether it was intentional or not.
Nobody said the simulation is being done with transistors. Even if it was why wouldn't it be possible? I don't see any constraints. One day humans will fully understand neurons. One day we will make an AI using transistors who will be able to "understand" transistors
Is this really true? What about a self-aware AI? I can imagine an AI that resides on a chip and developed consciousness could conceivably control a camera/robot in the real world and literally inspect and even reverse engineer or repair/improve itself or the hardware it runs on.
True, but in the end their ability to perceive will be limited by the wires to/from those cameras to their hardware, and thus it is still possible that even the external world they thing they're observing, including themselves and their place in it, is also simulated. As in, someone could plug a "fake" camera into the AI and fool it.
I agree that an AI could be tricked into believing it's environment is the real world when it's only a simulation, just as we could be AIs in a simulation. But this doesn't preclude the possibility of an AI being able to completely inspect/understand the hardware it's running on in the real world, if given the fact that it actually is running on hardware in the real world.
I would be surprised if we are not living in simulation. Look at superposition and function collapse in quantum physics, it's equivalent to what we are doing in games today = do not render if player is not watching, save resources. Think about it, look for example at GTA V, what we can achieve today and think about how world looked like 100 years ago.
I imagine that in 1000 years from now, simulating world similar to ours will not be a problem at all. Maybe not as complex because every level you go deeper you have less resources at your disposal. Who knows which level deep we are? Maybe we are simulation in simulation in simulation...
This is a common misconception, but superposition has nothing to do with "do not render if the player is not watching." As far as we know, every particle in the universe is interacting with every other (with a light speed delay) and thus "measuring" it (modifying and responding to its wavefunction) at all times. In the two-slit experiment, for example, the wavefunction of a particle moving through a slit interacts with the wavefunction of the particle used to detect it, causing both wavefunctions to be more narrowly distributed in the space around the slit. But before or after this interaction, the whole wavefunction still needs to be simulated to produce the effects we see experimentally. The wavefunction itself is the real thing being simulated, and it's always "running."
You may be right that it will become easier to simulate worlds like ours, but it will require innovations that make a quantum computer look like an etch-a-sketch. The universe is really big and really detailed.
I see the distinction you're making (I think), but I still don't see how that invalidates GP's analogy. Surely "running" a wavefunction is computationally cheaper than re-evaluating the particle's characteristics at every universal frame. Perhaps it's not so much about our own "watching" but rather "do not calculate when irrelevant". My own analogy might be an online map frame like google maps, where tiles outside the viewing box are not loaded. This is because we can tell what needs to be calculated by the bounding box of the viewer without needing all the other finer details. I see it more as a lazy eval.
The particle's characteristics ARE the wavefunction - or to say it another way, the wavefunction is the only real characteristic a particle has.
When I run a quantum mechanics simulation at work, I might afterwards calculate the particle positions or some other classical characteristic of the system. But I need those characteristics only because I'm trying to get a classical model of the system. The system itself, when it's running, doesn't do any of that. It interacts wavefunction-to-wavefunction. As far as we know, that's how the universe works.
Yeah you're right, but in the De Broglie-Bohm theory the wavefunction is calculated, and then the movements of the particles are calculated on top. So from a computational perspective it's strictly harder.
> As far as I know not in the De-Broglie-Bohm theory
As a physicist I am telling you should not conclude that the De-Broglie-Bohm theory is tell you anything different than what comicjk said.
> The particle's characteristics ARE the wavefunction
This kind of statement is normally said to stress the fact that the best evidence and experiments indicate quantum mechanics accurately describe the physical world and even parts like superposition(single particle double slit experiment, tunneling etc) that defy more classical intuitions.
De-Broglie-Bohm theory includes the same set of non-intuitive behaviors that all valid interpretations of quantum mechanics have, in other words it is not testable different from other interpretations.
> De-Broglie-Bohm theory includes the same set of non-intuitive behaviors that all valid interpretations of quantum mechanics have, in other words it is not testable different from other interpretations.
"In 2016, Pisin Chen and Hagen Kleinert argued that the Copenhagen interpretation and the De Broglie–Bohm theory yield different results for the ratio of peak intensities in the double-slit experiment. They concluded that they are thus not mathematically equivalent."
I would want to do some additional research before agreeing with Pisin Chen and Hagen Kleinert that the De-Broglie-Bohm theory has been experimentally disproven.
Other then combining the existing research double checking the numerical simulation and algorithms in [15] would be my approach.
Pisin Chen and Hagen Kleinert paper seems to lack such an examination which would have made for a more conclusive argument.
Is it possible that the calculations you are performing afterwards are what is causing the waveform interactions to have definitive results? In other words, could the waveform interactions cause the waveforms to become coupled and cause the entire system to exist in a superposition until such time as a classical measurement is made?
The computational problem of simulating a wavefunction is suspected by computer scientists to be exponentially harder than simulating classical physics (this is the question of whether BPP=BQP). So quantum physics is the opposite of a computational optimisation!
Yes, I'm referring to big and detailed relative to what we can simulate with foreseeable technology. Sure, we can see how to maybe simulate a universe the size of a cell, and a much much larger universe could maybe simulate everything in ours. But in either case, the larger universe is going to view the smaller one as a simplification.
If I'm running Linux in a VM, I don't think of it as a simplification-- but it is way easier to just shut it off and reinstall if something goes wrong.
The players are those lucky few that have an internal voice and their own agency. They can do almost whatever they want, as long as they can figure out how.
NPCs are more like walking zombies and have to fill the role they were born into.
Sigh. Why would you dedicate your life to proving this?
You will either find you cannot prove it, in which case you have wasted your entire life on falsehood, or you prove it true in which you undo your dilute your existence and everyone you have ever known and loved.
though, if we are a simulation... maybe we can exploit bugs in it and give ourselves advantages! After all, the only bug-free program is one that was never started.
It's not entirely clear what, exactly, it would mean for this to be true. It's also not clear that defining what we mean by "Is the universe a simulation?" has any value if it is not testable.
It's like religion in that people like to anthropomorphize wildly. The Cosmological argument ("Define God to be the first cause") becomes "... and that's why God smote a fig tree."
Here, "Huh, the universe might have some funny symmetry" frequently seems to become "the Gods running the simulation we live in will turn you off if you ask the wrong questions", or some other similar wild leap.
Response 1: because ignoring things with potentially existentially negative results is the functional equivalent of sticking your fingers in your ears and shouting
'LA LA LA LA LA'. Response 2: many (most?) people have a tiny little voice in the back of their head that whispers "Jump! Jump!" whenever they're close to the edge of a roof, cliff top, etc.
You could argue this about so many scientific endeavors: the heliocentric model, the search for extraterrestrial life, climate change, the theory of evolution, the big bang theory, etc. These all had serious existential effects on society but society has also benefited greatly in other ways. I don't see why clarifying the nature of reality with regards to simulations would be any different.
I dont understand why 1 more llayer of abstraction makes a difference. You are a software running on neurons interacting with simulation of outer world.
This does indeed looks like a modern version of theosophic explanation of why the world exists. The only difference is, previously people were convinced that gods were everywhere and so there must have been one who created our world; and now we, surrounded by computers in all their shapes and forms, are getting suspicious as to what if... You know.
The title of the arxiv article is: Constraints on the Universe as a Numerical Simulation
The only time they use the word evidence is in this quote:
__
Therefore, there is a sense in which lattice QCD may be viewed as the nascent science of universe simulation, and, as will be argued in the next paragraph, very basic extrapolation of current lattice QCD resource trends into the future suggest that experimental searches for evidence that our universe is, in fact, a simulation are both interesting and logical.
We've updated the submission title from “Physicists May Have Evidence Universe Is a Computer Simulation”, which breaks the guidelines by being editorialized.
So we stopped believing that God created the universe, and us as part of it. We believed only in science and natural processes. Following that through, we started believing that we live in a computer simulation with, presumably, some kind of programmer. That's pretty ironic.
The root distinction seems to be whether you prioritize reality or abstractions.
If reality comes first, you might not even believe that there is a set of universal mathematical laws governing nature (Feynman expresses this viewpoint, and raises the possibility that no such laws exist here[0]). Edward Nelson came up with similar ideas on the limits of abstractions in mathematics, based on Christian philosophy [1].
I'm really surprised (every time!) when someone talks about conscience or discoveries or whatever so human-centric and local and unimportant in the cosmic scale while talking about simulations. If it's simulation, it certainly isn't made because of us.
We are on the only planet we know about with life on it, so it is fairly easy to assume this is the most interesting place in the universe. Whether that's rational or not is another question, but it doesn't seem surprising.
On a semi-serious note, I've always had this nagging feeling ever since I first heard of Planck time and Planck length. Why would there be a lower limit to the resolution of the universe, unless there was something spooky going on?
I'm not a physicist, so I'm speaking as 100% layman, but I can better accept a fixed-resolution universe, where forces have a certain magnitude and beyond them (in either direction, smaller or larger) matter doesn't behave in productive ways.
For a long time, we thought atoms were these building blocks; now we think it's quarks; the Standard Model does a pretty good job at relating all observed phenomena to each other, except for, you know, gravitation, and that pesky observed-but-unaccounted-for thing we deemed 'dark matter'.
A few things about the Standard Model make my head hurt, but the idea of a fractal universe that is infinitely scalable up and down and assumptions about spacetime and forces still hold the same way is truly mindbending that I don't know how to reason around it. So perhaps conversely, can we flip the question and ask, why wouldn't the universe have a finite resolution in terms of matter actually existing and behaving like we know and love?
The Standard Model has as a subgroup the Poincaré group, which is the isometry group of flat spacetime, which is both the (global) spacetime of Special Relativity and is the local spacetime of General Relativity for a pretty narrow definition of local, although we regularly construct Local Inertial Frames which are regions of spacetime that are so approximately flat that the difference is negligible.
The Poincaré group has as a subgroup the Lorentz group, and its generators include rotations about the three spacelike axes of Minkowski (i.e., flat) spacetime.
The other symmetries of the Standard Model are invariant under these rotations.
A distant observer moving transversely relative to a particle observes a tiny rotation of the particle. The particle's fundamental properties do not change under that rotation, which can be arbitarily small.
Rather than descending into group theory to reason about a minimum translation or boost, we can look to the energy-mass-momentum equivalence E_r^2 = (m_0 c^2)^2 + (pc)^2 and focus in on photons so we can ignore the m_0 term. Here we have E = E_r = (pc) = \hbar\omega = hc/\lambda = hf. Photons come at arbitrary frequencies, and we have good blackbody radiators all across our sky. If there were a minimal length scale, we would expect that derivatives of position would incorporate that minimal length, and so we would impose observables on things like the doppler shift, for example if we accelerate in a straight line towards a radiator emitting extremely high energy photons.
So it's not so much "why wouldn't it have" but rather, "Q: does observation or experiment support a minimum scale that is large compared to the Planck scale for spacetime intervals in our universe? A: no"
The answers for minimum length et al. scales that are small compared to the Planck scale are subtler (essentially by definition of the Planck scale, at those scales quantum corrections to account for gravitation become significant, spoiling the observability of short spacetime intervals) but so far still 'no'. For every particle in the universe there can be an (ultrarelativistic) observer who sees the particle wavelength shrink below the Planck length; there is nothing special about this observer -- there is likewise a possible observer who sees the particle at some much longer wavelength -- and the point of relativity is that neither observer is more correct than the other. Furthermore, if the former type of observer manages to see a star ultra-blueshifted and extremely Lorentz-FitzGerald-contracted, that observer does not create a black hole; no event horizon forms (event horizons are a global feature of the causal structure of spacetime[0]), and so we can turn an argument about Planck length into an argument about localized Planck energy: there is nothing obviously special in spacetime about the Planck scale. However, our ultrarelativistic observer will certainly see some very strange stuff courtesy of the Unruh effect: different observers observe different particle counts, and one would expect our observer would see an explosion in the number of particles compared to a more typical observer, and those particles -- like all others -- will also interact gravitationally, and we don't know yet how to decomplicate the picture enough to do useful calculations on them.
[0] different observers may disagree on precisely where the event horizon is, what its shape is, and even the count and energy of the particles just outside the event horizon. but they will all agree that there is an event horizon.
Should we perhaps hypothesize the motive of this simulation, if any? Death seems to be a very important focal point: All living organisms are condemned to their aging bodies, eventually begin decaying, then die. Death of the universe is also inevitable, "Heat Death". Then there are other existential issues our feeble brains have to deal with. Then of course, our Earth is replete with suffering around every corner. Perhaps other terrestrial planets operate differently, for better or worse. You would think our maker(s) should have known the possible outcomes and scenarios before hitting the "start existence now" button, or perhaps it is just part of the grand plan.
You have to wonder: Are these entity/entities sadistic pranksters of some sort?
> Death of the universe is also inevitable, "Heat Death".
This is far from settled. Heat death applies only to closed systems. Although it is not unreasonable to assume our universe is a closed system, we simply do not know this to be a fact.
A cursory reading of most religious texts would certainly indicate a degree of sadism and lack of empathy with our human feelings. Human sacrifice? Plagues? Genocide? As an experiment, they would make sense. Virtual vivisection.
The old God-as-a-Programmer argument certainly has its appeal :)
Entropy could just be an inevitable factor in this simulation. As the simulation steps forward in time, with the rules set forth at the beginning, these different emergent properties of the universe begin sprouting. There may not be a motive at all for this or an other simulation, besides gathering data on the effects of different rulesets.
IMHO as a biologist infused with awe of the natural world from observation and reading, it is laughable that gods or computers could design or simulate the infinite levels of mind numbing complexity and unimaginable patterns that evolution has produced. But most people who are more focused on, and hungry for, the order created by technology and society in general, find it easy - with occasional cognitive dissonance, my favorite expression of which is the poem "Design" by Robert Frost:
A beautiful poem. But who says that the "creators" couldn't be evil, or amoral, or just have different ethics/morals than humans, or who don't think in terms of ethics/morals?
Also, while a human mind might marvel at some level of complexity, there's no reason why a superhuman mind would have to. I mean, imagine a "brain the size of a planet" (like Marvin's in The Hitchhiker's Guide), or one the size of a star, or larger. What could their limits be?
Now imagine a mind not bound by the laws of physics as we know them -- for if they're "outside" the simulation of the universe as we know it, they'd not necessarily be limited by its physical laws. Really, at that point it's anything goes.
Finally, consider that you, or any given human, or all of them, could simply be mistaken about what they think or observe. Just given that potential for error, anything could be possible.
But evolution would be part of the process. The programmer (or god) just sets the thing in motion. From an evolutionary biology perspective, evolution might be the main deal, the whole point of the simulation. It would make some sense, to create processes, systems and structures that no being could ever come up with.
As far as I know, there are some creationists that believe something along these lines already. Dawkins used to call it smuggling God in the back door.
As a BS in bio as well I learned everything in biology arises from chemistry which arises from physics which can be explained (mostly?) with math. If that isn't abundantly clear on graduation your teachers did you a disservice.
You don't have to design anything, it should just show up if its an accurate enough simulation.
You are assuming unlimited computational resources. At some point the computer required is bigger than the universe and exceeds the limits of what our current knowledge of physics and cosmology allows us to speculate about in any kind of meaningful way. I actually believe that a computer with human level intelligence and consciousness will be possible - but only because it won't need to be as messy as real human minds. To paraphrase Hitchhikers Guide, biology is messy. Really messy. You just won't believe how vastly, hugely, mind-bogglingly messy it is.
If the idea is that our universe is a simulation then its state is, by definition, represented by a finite subset of the state of the simulating universe. In other words, yes, it would be a computer bigger than our universe, in the same way we simulate a universe smaller than ours on one of our computers today.
If we live in a simulation evolution is still much more probable explanation of the life we see than creationism.
With big enough computer and interesting enough set of unerlying rules you can run a simulation to see what happens and there may be life somewhere in the simulated universe. We even do similar experiments [1] right now (but the rules are very simple for now, and the generated life isn't very impressive so far).
I find this much more likely than a Programmer designing not only the universal rules (relativity, quantum nechanics, etc), but also small details of chemistry, biology, and all the organisms (if not all - how comes they look like they evolved - that's LOTS of stuff to fake).
I can easily imagine humanity to be able to create simulation on big enough scale to create some artifical life in future.
I can't imagine all the advances in math and chaos theory and fractal design and however to even call all that fields that would be required to design artifical life working in simulated universe "by hand" in such a way, that it seems it evolved from first principles.
Additionaly - if you can design top-down and bottom-up at the same time - it means you already know what will happen. Why even do the simulation in that case?
So, if we're living in a simulation - no reason to bring Gods and ultimate whatevers into this - much more probably we're just their equivalent of the Conway's game of life: https://bitstorm.org/gameoflife/
> it is laughable that gods or computers could design or simulate the infinite levels of mind numbing complexity and unimaginable patterns that evolution has produced
This sounds like the flip side of the creationist's claim that the universe was created by a deity because it's so complex. Neither is falsifiable.
I'm not sure why people insist on assigning agency to this phenomenon. If you wanted, couldn't you invoke the 'many worlds' theory and suggest that our particular universe exhibits the same properties as what you might find in a simulation? This would do away with the need for an agent who is simulating.
Even if we live inside a simulation inside a simulation inside a simulation, ultimately there is a base reality that influences the design of the simulations contained within it, and can impinge upon the simulations at any time. Infinite regression doesn't work here - using the term "simulation" implies some special pleading when moving from the outermost simulation to the base reality. If you want to regress forever, then there is no qualitative difference between layers of reality and the word "simulation" is meaningless.
We live in the base reality. All a simulation could do is obscure complexity.
Interestingly this isn't a new idea. The gnostics of the first two centuries BCE had a theology of this, in which a flawed world, in which humans lived, was created by an imperfect creator called the demiurge. There was a perfect realm beyond this world with an intercessor named Sophia (wisdom) who would rescue us from our flawed world, the simulation created by the demiurge.
Sorry, I think I might be misunderstanding you. By "We live in the base reality", do you mean "Even if we live in a simulation, the simulation itself is hosted somewhere; following back the chain, eventually you reach the base reality. Therefore in the same sense as one should say that Microsoft Word runs/exists in our real world, one should say that (even if we're simulated) we exist in the base reality"?
You imply twice in your post that you assume there is a base reality, but why?
I could argue it makes just as much sense that every layer is simulated by the layer "above". That could continue ad infinitum or perhaps create a cycle, so eventually it's simulating itself. These may be what we refer to as simulations, like what's running on my PC, or some kind of emergent behavior. Either case seems to work.
It's natural to define the universe / all of reality / all layers as that which is self-defining, which is hard to describe without circular reasoning.
If you can recurse infinitely, then there is no need to define reality as a simulation. A simulation implies some loss of fidelity relative to the layer hosting it. If you continue this infinitely eventually you end up with a completely entropic layer. Maybe there is a discontinuity and this entropic layer crystallizes into a highly ordered "base reality". I don't know how this innermost, entropic layer can contain the outermost, ordered layer. I think "simulation" is a bad term since it implies a monotonic loss of fidelity.
A simulation only has meaning if there is a base reality. Otherwise there are only layers of reality.
A “slowing down” of a simulation would only be perceptible to those observing from without, not those who are observing from within, whose perceptive processes would be slowed in synchrony with the processes they are perceiving.
There's no reason to presume the simulation proceeds at a constant rate as measured from the outside world either: “render times” may vary with the complexity and fidelity required — for example when we innards fire up the LHC, a huge bump in resolution might be required. When we innards venture onto the moon, a whole new world has to be mapped in and professing speed might drop accordingly.
What the original poster is saying is that insofar as we do not know what the base reality is, there is no reason to argue that our reality is a lower-fidelity instantiation of it: we might as well be a flawless emulation of something totally orthogonal to the base reality.
Furthermore for obvious reasons computation cannot endlessly recurs endlessly and somehow forms into a loop: somewhere the arithmetic actually has to be performed.
A simulation could just be some states. Consider the rule, double and subtract one to get to the next world-state, over the integers, and start the world with the state 1. So the next state is 1. The one after is also 1. The one after is also 1. The universe is stuck in a static state.
As I, you, or a Python script is modeling this universe, where does the outside universe "impinge" on it?
Then let's do a different universe, starting at 2. Then the next state is 3, the state afrer is 5, the state after is 9, and so forth.
As you do this universe in your head, where does the outside universe "impinge" on it?
There is a problem, however. All future states are well-defined in the universe I've given even if nobody in our universe or any other is calculating them.
It is trivially obvious that the human mind is an emergent property of its neural firings (of the brain), and that replacing some part of it with a black box that interacted with the rest would leave it with the same emergent properties, even if this black box were modeling in a different medium.
The question is, if you can model the whole thing over a well-defined program that did not require non pseudorandom entropy, then it would be similar (just start with more complex data than 1 or 2, and more complicated rules than double and subtract one) - so in order for these states to exist, would anyone have to actually calculate it, or is it enough for these states to be mathematically defined?
What I am saying is that it is a statistical certainty that within a thousand years we can create a deterministic model of a brain in a VR, where the brain reports consciousness. (this is 100.000% absolutely necessarily true and not open to debate. The brain is a couple of pounds of stufd, with just a few hundred billion neurons or fewer - I bet we could practically have the hardware to model one today.)
Does this mean that that model already exists? (before anyone chose to calculate it)?
This is similar to asking whether the quintillionth prime exists. Yes, of course it exists.
Really? Even if nobody calculates it? Then what is to say that we might not exist without anyone calculating us?
Perhaps the states of the universe, like the quintillionth prime, exist without anyone calculating it. Perhaps there is no underlying reality at all, just as there is not an underlying reality in which the quintillionth prime is embedded.
It doesn't matter if you agree. Some of the world's 7 billion people likely believe that minds are like souls in a parallel ethereal plane and they could be severed, float around for a while, be attached to some other person and so forth. Regardless of whether anyone believes that it is just not true and conscienceness is an emergent propery of the person having it, and it doesn't really matter whether you or anyone else sees it that way.
It doesn't matter what you think conscienceness is or whether there is agreement on it. (Sorry if this is harsh.)
How can you be so confident? You have no proof of this (or at least have not demonstrated it) and are simply going on faith which is exactly what someone who says that consciousness comes from the soul is doing.
And even if consciousness is an emergent property of the structure of the brain, how could we have a simulated brain report that it is experiencing consciousness when we cant define what it is in the first place?
Suppose you couldn't define flight, and I said, of course flight is possible since birds fly so it would be possible to fly, flight is just the sum of the mechanical motions that it entails - whereas you claimed no, no, it's some kind of magic that Hermes, the god of flight deigns to imbue some creatures with. Even if you exactly copied every last muscle twitch in a bird with a mechanical version of the same, you claimed, there is no guarantee that Hermes would play along and magically make it fly, it could be an earthbound mechanical object, or even if it did appear to fly, since we don't have a good definition of flight how would we even know if it's really flying? It could just be appearing to fly but, since in our mythology Hermes doesn't make mechanical things fly, it really isn't flying...
it's not something that is even worth discussing. of course if you made an emulated version of a brain in a VR with the same structure and neurological action and it reported consciousness, then it would be conscious, of course, yes. This is like asking whether NES games can "really" be running in a browser, even if the browser is emulating an NES system and includes the code of the original cartridge. Is the game "really" running? Can you "really" play it? Meaningless questions. If the VR person with the same brain topology as a human and hundreds of billions of neurons reports consciousness then of course it has it. why would being emulated suddenly make it different from all 7 billion humans running natively in the world, especially if theoretically it could be compiled from the same source code (which we have a copy of, DNA was sequenced in full in the 90s).
Since my comment was about a thousand years out, there's no question and we don't need to discuss these things. It's open and shut. I don't need to give any citations.
So let's say that perhaps biological neurons are an antenna which picks up on a universal consciousness energy field and imbues consciousness into a brain and that thus far, or even 1000 years from now, this property has gone undetected. Simulating the known properties of the brain may make a program that functions in all respects like a brain but does not have consciousness. It might report that it has consciousness because that is some function of another property of the simulation however in this case we would know that it is not true.
I find it amusing that you think a topic as widely studied and disputed as consciousness is "open and shut" because of some assumptions that you choose to make. Nonetheless, I do appreciate hearing your perspective so thank you.
Yes, I may be butchering this (didn't read the link below) but perhaps in Descartes's philosophy the pineal gland is like a connection to the soul, sure, like an antenna to an ethereal world, in a physical organ.
I completely discount this possibility and think it's not worth discussing. So, you are right that my mind is very closed to any alternatives to what I've stated, though as in your example of an antenna, we could describe such alternatives rigorously. Not worth our time. (IMO). Thanks for the replies.
Reminds me of a short story called Luminous by Greg Egan. In that story the universe is a declarative system that does have a non-pseudo random component to it -- with actual reality being an emergent phenomenon between this declarative model and the entities within exploring that model by causing the universe to imbue the mathematical objects implied by the universe with specific reality by actually computing them, in so doing "expanding" the size of the universe and introducing new effects from this expansion into the evolution of the system ... in the story it turns out there are multiple actors active at the edge of defining the universe -- and their actions can actually influence the ability of other actors to continue in their participation in the universe creation process ...
Edit: my attempt to elaborate on the idea from the story ... I don't know if I can summarize the concept that well -- I think the framing within the story for the plot is something along the lines of:
1. Mathematical truth exists
2. There is an existential difference between a mathematical truth and a specific occurrence of the "pattern" described by that truth within the physical material of the universe.
I think the eli5 example in the story is based on the integers. You have the integers which are infinite, and you have an instantiated version of some number n if you have an occurrence of n things in nature. The reality of n is fundamentally different if you are in A universe with "n things" vs one with only m things m<n.
I think that's the motivation for the concept -- and in the story a researcher sets off to find scenarios where the universe had not yet "instantiated" one of the "laws" that could be derived from mathematical reality -- and ends up finding out that there is more than one possible way for the laws of "reality" to expand ... -- with different implications on what laws might be discovered in the future ...
This is really, really interesting. As I don't have access to the book just now, could you expand (by editing) your comment a little, your summary is very dense. I realize you might not remember it all exactly but it's really interesting to me.
EDIT: Thanks for the expansion! Fascinating and really interesting idea. The idea, "There is an existential difference between a mathematical truth and a specific occurrence of the 'pattern' described by that truth within the physical material of the universe" is a fascinating one.
The other day I saw a post on social media that how he explained spaces from 0-dimension to 10-dimention in simple drawings. At the end, he claims that 10-dimentional space is just a dot, and there is no higher dimensional space more than 10-dimention any more. And he justifies his claims based on the string theory. At that time immediately I thought about the billions of parameter space in deep neural network and fractal spaces such as Mandelbrot Set. How could one simulate the universe based on only 3D space-time lattice?
In my opinion this seems all pretty pointless. To be able to differentiate between a simulation and reality you have to know at least how one of those works in quite some detail. What if we discover floating point rounding errors in the laws of physics we observe? Who is to say that this is not a feature of reality? What if it is the other way round, what if we live in a simulation with laws of physics made out of real real or complex numbers but the reality outside of our simulation has laws of physics behaving like floating point numbers? It seems pretty impossible to ever figure out what reality or a simulation must look like so that we can determine what surrounds us. Maybe there is some incredible self-consistency law at the bottom of reality, forcing reality to be what it is because there is one and only one way reality can be. But besides something like that?
Not so pointless. Think that one of the NPC in a computer game gets intelligent. It could learn how to trigger bugs or cheats in the game or features thought only for the player. Examples: crash the game, which hopefully restarts from the last save. Get where no NPC was supposed to be, only players, and get teleported somewhere else. Trigger a stack overflow and perform actions in the host, maybe changing the programming of the game or the data in the save files.
In our case it could be the way to FTL. Did any Sci-fi author considered that yet?
But how would that NPC know that it is in a game? What if we figured out how to do teleportation or time travel? How would you determine whether that is a feature of reality or whether we discovered a bug in a simulation?
But exactly that was my point - unless you know what either reality or a simulation really looks like, you can not look around and draw conclusions whether we live in reality or a simulation. People trying to do this make the assumptions that a simulation would have properties similar to simulations we perform and that reality would have properties similar to the laws of physics as we know them, but neither assumption seems really justified.
> At fixed quark masses, the CRR of a lattice ensemble
generation (in units of petaFLOP-years) scales roughly as the dimensionless number...
The formula they give, indicates that the largest QCD calculations to date have taken 7.9e11 petaflop-years; this is consistent with the graph they give, but not with actual feasibility. What am I missing?
177 comments
[ 2197 ms ] story [ 6315 ms ] threadhttps://www.technologyreview.com/s/429561/the-measurement-th...
http://www.huffingtonpost.in/entry/physicists-may-have-evide...
[1] http://www.imdb.com/title/tt0139809/
[0] https://en.m.wikipedia.org/wiki/Bresenham%27s_line_algorithm
In General Relativity, fundamental particles are simply taking shortest paths through spacetime, and where there are multiple short paths, the one that extremizes the timelike distance is favoured.
In a perturbative theory of gravitation, you introduce corrections on the particles by way of graviton interactions.
Worse, whether a local patch of energy-density amounts to a particle is observer-dependent; in both approaches above an accelerated observer sees more particles than an unaccelerated one does. That includes more gravitons in the latter approach.
It would be a lot of work to write down a theory in which particles (whose count isn't global) decide on corrections to their own trajectories that return themselves to a path similar to one along a smooth manifold even though they are "really" travelling along a non-smooth one, so long as you are trying to match well-tested results of General Relativity. (A theory which doesn't reproduce those results is maybe an interesting toy; "cheating" by using a nearly-but-not-quite minimal length that has no direct observables is an uninteresting toy, and is not really a good argument for the existence of Simulators (or Creator Gods or whatever).)
In particular, the non-linear contributions of gravity will likely lead you into an explosion in the amount of knowledge that a particle would have to acquire in order to assess its trajectory; you'd probably also want to propose an explanation of how it goes about adjusting its trajectory when it discovers it's not on the "right" one. Don't forget to make this correct for all possible observers, including the ones that will see no particle and those that will see two or more particles.
In practice, attempts to do away with infinitesimals of length that do not clearly fail to reproduce known physics also tend to produce infinities of state in matter and/or additional gauge fields, and the infinities don't succumb to information-reduction techniques like renormalization by power-set counting.
If you're looking to encode that knowledge somewhere, like in a spacetime-permeating field, the gravitational field of General Relativity is right there and already serving that purpose in a non-toy theory that (in explorable parts of the known universe) behaves completely correctly with quantum field theories of matter, including The Standard Model. :-)
Best of all, in a block universe model, the whole thing is "pre-rendered", and there is no difference between the block universe model and an initial-values-surface formalism in the presence of determinism. So you only have to fully render one frame, and you get all the others, past and future. (Quantum mechanics as we know it is deterministic -- that's unitarity.)
http://nautil.us/blog/the-universe-expanding-symmetrically-a...
Small regions of our spacetime are locally Lorentz-invariant, meaning that a number of observable quantities (magnitude of angular momentum and mass are two) do not change for a particle held at the origin of a system of coordinates as the particle is arbitrarily rotated or boosted.
Arbitrariness is important. If we are observing a distant and predictable multifrequency radiator and define spherical coordinates with the radiator at the origin, then any movement our observation gear makes that isn't exclusively radial is equivalent to a rotation of the distant radiator. You can do this by holding the radiator at the (spacelike) origin, and the observer at its fixed (spatial) coordinates; in order to keep these coordinates constant, you have to rotate the system of coordinates to counter wholly non-radial relative movement. At large distances, the rotation at the origin becomes extremely small. Small or large, Lorentz invariance means the radiator has the same mass (it's by definition rest mass since it's always at the coordinates [0,0,0,t]) and momentum.
So arbitrarily small rotation goes hand-in-hand with arbitrarily distant observers, and also with nearer observers who can displace themselves tiny amounts.
So is there a minimum rotation?
The arbitrariness of rotation is a in conflict with "cell shape", as when your "cell" distinguishes between edges and faces, the discrete nature of rotated mass-energy-momentum exposed through the (corner-filled, discrete) cell structure leads to different observables under rotation compared to that of the (smooth, continuous) spacetime of (either theory of) relativity. This deviation is larger for objects of higher momentum; and remembering Einstein's relation for rest-massless particles, E = pc = \hbar\omega = h / \lambda, that means that for different-wavelength photons emitted from the same source, the higher-frequency photons will arrive later than the lower-frequency ones. We have good observational evidence against frequency-dependent arrival times from bright distant objects (supernovas, gamma-ray bursts, even millisecond pulsars).
We can conceive of a "cell shape" which is uniform under infinitesimal rotations, but at that point you have shifted one set of infinitesimals to another, and in the context of this topic (a simulation that among other things saves on state by abolishing real numbers in dimensions of length, rotation, boost and/or translation) is pretty much a non-winner.
(Additionaly, ignoring the simulation context, you would also almost always run into difficulties if your "cell size" -- a minimum length in space or a minimal interval in spacetime -- is large enough to produce observables.)
Ehrenfest and Weyl showed this as early as 1920 and 1922, respectively.
3+1 is also baked into the Minkowski metric by the latter's definition, which leads to the Poincaré group being the isometry group on local patches of spacetime; the Poincaré group is a subgroup of the Standard Model, and is extremely well tested in controlled, laboratory settings (and supported by an abundance of observational evidence too).
Extra dimensions of spacetime have to be small compared to the Planck length in order not to be obvious today. For all practical purposes, if such tiny dimensions exist, we can safely omit them from effective theoretical descriptions of everyday physics, just like we can ignore things like very large extra dimensions where the smallest step one can take is much bigger than the Hubble diameter.
Of course, I've also heard of the illusion of consciousness, but that's where my brain explodes.
I suppose you can point at all the artificial simulants and their sentience, but as pointed out in a sibling comment, so what.
That, or there's some interface to that platform that provides a source of intelligence. Maybe we're nothing more than Docker containers running on God's hardware.
I mean, plenty of people don't take issue with a lot of things I find reprehensible.
Are you morally prohibited from turning them off?
Can you reset to a save point, possibly undoing the creation of some of the sentient beings?
The author has commented that he regrets having overlooked the moral implications of simulated life in his previous works (see e.g. http://gregegan.customer.netspace.net.au/PERMUTATION/FAQ/FAQ..., which contains spoilers for Permutation City) so I wouldn't be surprised if that was the inspiration for this story.
https://www.amazon.com/Cyberiad-Stanislaw-Lem/dp/0156027593
It touches thoughtfully and humorously on: AI, simulations, robotics, ethical and moral questions regarding artificial life, among many other interesting topics. It's one of the funniest and greatest science fiction books I've ever read.
If someone does find out, you can simply fix the problem and revert to a point in time just before the data leading to the discovery was captured.
Speaking academically, you might be playing with fire if you probe the outer limits of the universe and we are in a simulation run by some entity concerned with keeping that a secret.
Only with classical computers, not quantum computers.
I can make 2 + 2 = 5 in a simulation. That doesn't mean 2 + 2 = 5 outside of the simulation.
I'm not sure you can. Yes you can make a chest with 2 apples contain 5 apples if you put 2 more, but it's impossible (IMHO) to make it consistent. How many apples do you have if you place 2 chests besides 2 other chests and put 1 apple in each? What if you repeat the process 2 times and then another 2 times?
Also - when some things in physics aren't intuitive (QM for example, but even adding velocities in relativity) we don't think math is wrong - we think math doesn't apply in that way to this particular phenomena.
Taking it further, what if rules and logic, and by extension, math as we know it, is just a construct that is not applicable in the "real" universe. What if, in that "real" space, 7+7=5, or mathematical operations are not possible. What if causality and time dont exist?
And if sometimes its 5 - they will do what we did with relativity when it turned out velocities doesn't add up like they "should". We haven't changed addition - we changed the equations used in physics.
I guess you could hand-made an intelligent life that is designed never to discover the true math.
The act of putting things in boxes (after first perceiving and classifying them as discrete entities) and the abstract notion of addition are only as related as you need them to be for a given purpose.
Yes. That was my point. How would you change the abstract notion invented by creatures in your simulation if not by changing reality to shape their abstractions? And how do you change enough instances that their math develops in such way that 2+2=5?
I think our basic math (logic, set theory, integer math) with some exceptions (choice theorem) is the only useful consistent basic math there is.
I do not believe that to be true. While physics is heavily dependent on math, the reverse is not true.
> I can make 2 + 2 = 5 in a simulation.
No, I am fairly confident that you absolutely can not. You can make a simulation where putting 2 oranges in a box and adding two more oranges results in 5 oranges in the box (rather like putting together certain nucleons together into a nucleus results in a mass different than the sum of their individual masses). But in that simulation it would be a peculiar property of oranges, not a property of mathematics -- just as we don't claim addition is invalidated by the behavior of nucleon masses.
But what's outside the simulation could be quite different. And humans could always be mistaken about both what they think they observe, their observations themselves, and the conclusions they make from these.
At best its a matter of estimating probabilities (ie. educated guesses), and even there they're probabilities regarding appearances. It has been argued as far back as Kant (over two hundred years ago) that one can never get to the "thing-in-itself" that is behind those appearances.
There is not even a reason to believe that the "outer universe" has such things as space and time or information as we know it, and no way to know what a "computation" might comprise in such a situation.
Maybe the situation is not that pessimal, and an outer universe is much like ours, but to prefer that believe one would need evidence, of which we have none.
Consciousness -- as in the sheer fact of experience -- is itself the base "reality," out of which everything is made. If you examine everything you're calling "the universe," all you will ever discover is sights, sounds, thoughts about sights and sounds, belief (or certainty) that those sights and sounds are caused by something external, etc. What are all of those made of? Consciousness, of course.
In this reality, "brains" are the entities that consciousness has dreamed up to demonstrate correlations with conscious properties. In our reality, we therefore say that "the brain causes consciousness."
If you investigate your experience closely, you'll discover that the epiphanies "I exist," "I am conscious," "I am alive", "a world exists" all point to the same ineffable "miracle" of existence, the one "substance" of which all realities are made.
Upon hearing this, perhaps the mind thinks "just as there are realities that exist without time and space, why can't there be some without consciousness?" There's a very interesting answer to that question, but it won't come in the form of thoughts.
Of course, there's no reason to believe any of that. On the other hand, as the Buddha said, don't believe me....
Imagine if we discover the random number generator used by quantum events!
> I think we're being virtualized in a blockchain-backed consensus reality running some variation of orchestrated objective reduction.
There is another theory which states that this has already happened."
― Douglas Adams, The Restaurant at the End of the Universe
Then one approach would be to silently and irreversibly corrupt the simulation in a way that will become noticeable. Perhaps you fill every document with secret messages that require AI as smart as the sims to detect and is only revealed years in the future.
By the time it is noticed it's too late and removing all those messages will require more resources than the simulation itself. Trying to catch them on demand will require understanding how the brains of the sims work internally, which may be unfeasible.
That idea is probably preventable, but I'm sure there are exploitable and non-reversible security holes in any imperfect simulation.
> If someone does find out, you can simply fix the problem and revert to a point in time just before the data leading to the discovery was captured.
Keep in mind that if you wanted to create an operating system, you can ensure that there are no kernel exploits.
If there is some kernel exploit, you can simply patch the kernel and reboot.
At least, that's the scenario depicted in the 1973 movie, World on a Wire. http://www.imdb.com/title/tt0070904/
I feel like no matter what we are in a 'simulation' where another dimension gave rise to ours whether it was intentional or not.
What dimension is thought in?
Not even if it tried to get that information from an external observer.
I imagine that in 1000 years from now, simulating world similar to ours will not be a problem at all. Maybe not as complex because every level you go deeper you have less resources at your disposal. Who knows which level deep we are? Maybe we are simulation in simulation in simulation...
You may be right that it will become easier to simulate worlds like ours, but it will require innovations that make a quantum computer look like an etch-a-sketch. The universe is really big and really detailed.
When I run a quantum mechanics simulation at work, I might afterwards calculate the particle positions or some other classical characteristic of the system. But I need those characteristics only because I'm trying to get a classical model of the system. The system itself, when it's running, doesn't do any of that. It interacts wavefunction-to-wavefunction. As far as we know, that's how the universe works.
As far as I know not in the De-Broglie-Bohm theory (but I'm not a physicist, so correct me if I'm wrong):
> https://en.wikipedia.org/wiki/De_Broglie%E2%80%93Bohm_theory
> http://plato.stanford.edu/entries/qm-bohm/
> http://plato.stanford.edu/entries/qm-decoherence/#PilWavThe
As a physicist I am telling you should not conclude that the De-Broglie-Bohm theory is tell you anything different than what comicjk said.
> The particle's characteristics ARE the wavefunction
This kind of statement is normally said to stress the fact that the best evidence and experiments indicate quantum mechanics accurately describe the physical world and even parts like superposition(single particle double slit experiment, tunneling etc) that defy more classical intuitions.
De-Broglie-Bohm theory includes the same set of non-intuitive behaviors that all valid interpretations of quantum mechanics have, in other words it is not testable different from other interpretations.
There exist opinions that disagree:
> https://en.wikipedia.org/w/index.php?title=De_Broglie%E2%80%...
"In 2016, Pisin Chen and Hagen Kleinert argued that the Copenhagen interpretation and the De Broglie–Bohm theory yield different results for the ratio of peak intensities in the double-slit experiment. They concluded that they are thus not mathematically equivalent."
> http://www.ejtp.com/articles/ejtpv13i35p1.pdf
Pisin Chen and Hagen Kleinert say that De-Broglie-Bohm theory does not agree with experimental results or the rest of quantum mechanics.
This is surprising since there evidence of this is from a 1979 paper, their citation [15]
[15] C. Philippidis, C. Dewdney and B. J. Hiley, Il Nuovo Cimento 52B, 15 (1979)
can be found online(republished) at: http://www.pbx-brasil.com/FisQuan/Notas/Area01/semana041/pap...
I would want to do some additional research before agreeing with Pisin Chen and Hagen Kleinert that the De-Broglie-Bohm theory has been experimentally disproven.
Other then combining the existing research double checking the numerical simulation and algorithms in [15] would be my approach.
Pisin Chen and Hagen Kleinert paper seems to lack such an examination which would have made for a more conclusive argument.
We have no absolute scale to measure against. Our 10^80 particle universe (or whatever the number is) could live in an 10^8000 particle one.
So our whole universe could literally be contained in something as "big" as a grain of sand in the host universe.
NPCs are more like walking zombies and have to fill the role they were born into.
You will either find you cannot prove it, in which case you have wasted your entire life on falsehood, or you prove it true in which you undo your dilute your existence and everyone you have ever known and loved.
Kinda like religion I guess...
Here, "Huh, the universe might have some funny symmetry" frequently seems to become "the Gods running the simulation we live in will turn you off if you ask the wrong questions", or some other similar wild leap.
I guess my issue is not faulting intellectual curiosity in general, but directed curiosity at something with so many existentially negative results.
The only time they use the word evidence is in this quote:
__ Therefore, there is a sense in which lattice QCD may be viewed as the nascent science of universe simulation, and, as will be argued in the next paragraph, very basic extrapolation of current lattice QCD resource trends into the future suggest that experimental searches for evidence that our universe is, in fact, a simulation are both interesting and logical.
If reality comes first, you might not even believe that there is a set of universal mathematical laws governing nature (Feynman expresses this viewpoint, and raises the possibility that no such laws exist here[0]). Edward Nelson came up with similar ideas on the limits of abstractions in mathematics, based on Christian philosophy [1].
[0] https://www.youtube.com/watch?v=QkhBcLk_8f0
[1] https://web.math.princeton.edu/~nelson/papers/warn.pdf
(I know they're just theoretical, but...)
For a long time, we thought atoms were these building blocks; now we think it's quarks; the Standard Model does a pretty good job at relating all observed phenomena to each other, except for, you know, gravitation, and that pesky observed-but-unaccounted-for thing we deemed 'dark matter'.
A few things about the Standard Model make my head hurt, but the idea of a fractal universe that is infinitely scalable up and down and assumptions about spacetime and forces still hold the same way is truly mindbending that I don't know how to reason around it. So perhaps conversely, can we flip the question and ask, why wouldn't the universe have a finite resolution in terms of matter actually existing and behaving like we know and love?
The Standard Model has as a subgroup the Poincaré group, which is the isometry group of flat spacetime, which is both the (global) spacetime of Special Relativity and is the local spacetime of General Relativity for a pretty narrow definition of local, although we regularly construct Local Inertial Frames which are regions of spacetime that are so approximately flat that the difference is negligible.
The Poincaré group has as a subgroup the Lorentz group, and its generators include rotations about the three spacelike axes of Minkowski (i.e., flat) spacetime.
The other symmetries of the Standard Model are invariant under these rotations.
A distant observer moving transversely relative to a particle observes a tiny rotation of the particle. The particle's fundamental properties do not change under that rotation, which can be arbitarily small.
Rather than descending into group theory to reason about a minimum translation or boost, we can look to the energy-mass-momentum equivalence E_r^2 = (m_0 c^2)^2 + (pc)^2 and focus in on photons so we can ignore the m_0 term. Here we have E = E_r = (pc) = \hbar\omega = hc/\lambda = hf. Photons come at arbitrary frequencies, and we have good blackbody radiators all across our sky. If there were a minimal length scale, we would expect that derivatives of position would incorporate that minimal length, and so we would impose observables on things like the doppler shift, for example if we accelerate in a straight line towards a radiator emitting extremely high energy photons.
So it's not so much "why wouldn't it have" but rather, "Q: does observation or experiment support a minimum scale that is large compared to the Planck scale for spacetime intervals in our universe? A: no"
The answers for minimum length et al. scales that are small compared to the Planck scale are subtler (essentially by definition of the Planck scale, at those scales quantum corrections to account for gravitation become significant, spoiling the observability of short spacetime intervals) but so far still 'no'. For every particle in the universe there can be an (ultrarelativistic) observer who sees the particle wavelength shrink below the Planck length; there is nothing special about this observer -- there is likewise a possible observer who sees the particle at some much longer wavelength -- and the point of relativity is that neither observer is more correct than the other. Furthermore, if the former type of observer manages to see a star ultra-blueshifted and extremely Lorentz-FitzGerald-contracted, that observer does not create a black hole; no event horizon forms (event horizons are a global feature of the causal structure of spacetime[0]), and so we can turn an argument about Planck length into an argument about localized Planck energy: there is nothing obviously special in spacetime about the Planck scale. However, our ultrarelativistic observer will certainly see some very strange stuff courtesy of the Unruh effect: different observers observe different particle counts, and one would expect our observer would see an explosion in the number of particles compared to a more typical observer, and those particles -- like all others -- will also interact gravitationally, and we don't know yet how to decomplicate the picture enough to do useful calculations on them.
[0] different observers may disagree on precisely where the event horizon is, what its shape is, and even the count and energy of the particles just outside the event horizon. but they will all agree that there is an event horizon.
You have to wonder: Are these entity/entities sadistic pranksters of some sort?
This is far from settled. Heat death applies only to closed systems. Although it is not unreasonable to assume our universe is a closed system, we simply do not know this to be a fact.
The old God-as-a-Programmer argument certainly has its appeal :)
I found a dimpled spider, fat and white,
On a white heal-all, holding up a moth
Like a white piece of rigid satin cloth-
Assorted characters of death and blight
Mixed ready to begin the morning right,
Like the ingredients of a witches’ broth--
A snow-drop spider, a flower like a froth,
And dead wings carried like a paper kite.
What had that flower to do with being white,
The wayside blue and innocent heal-all?
What brought the kindred spider to that height,
Then steered the white moth thither in the night?
What but design of darkness to appall?--
If design govern in a thing so small.
Also, while a human mind might marvel at some level of complexity, there's no reason why a superhuman mind would have to. I mean, imagine a "brain the size of a planet" (like Marvin's in The Hitchhiker's Guide), or one the size of a star, or larger. What could their limits be?
Now imagine a mind not bound by the laws of physics as we know them -- for if they're "outside" the simulation of the universe as we know it, they'd not necessarily be limited by its physical laws. Really, at that point it's anything goes.
Finally, consider that you, or any given human, or all of them, could simply be mistaken about what they think or observe. Just given that potential for error, anything could be possible.
As far as I know, there are some creationists that believe something along these lines already. Dawkins used to call it smuggling God in the back door.
You don't have to design anything, it should just show up if its an accurate enough simulation.
With big enough computer and interesting enough set of unerlying rules you can run a simulation to see what happens and there may be life somewhere in the simulated universe. We even do similar experiments [1] right now (but the rules are very simple for now, and the generated life isn't very impressive so far).
I find this much more likely than a Programmer designing not only the universal rules (relativity, quantum nechanics, etc), but also small details of chemistry, biology, and all the organisms (if not all - how comes they look like they evolved - that's LOTS of stuff to fake).
I can easily imagine humanity to be able to create simulation on big enough scale to create some artifical life in future.
I can't imagine all the advances in math and chaos theory and fractal design and however to even call all that fields that would be required to design artifical life working in simulated universe "by hand" in such a way, that it seems it evolved from first principles.
[1] https://en.wikipedia.org/wiki/Artificial_life
Not only to us. To any other one. To any other computer-alike.
The ultimate compiler.
Why waste to build a compiler when your mind is already the #1 thing ever?
Additionaly - if you can design top-down and bottom-up at the same time - it means you already know what will happen. Why even do the simulation in that case?
So, if we're living in a simulation - no reason to bring Gods and ultimate whatevers into this - much more probably we're just their equivalent of the Conway's game of life: https://bitstorm.org/gameoflife/
This sounds like the flip side of the creationist's claim that the universe was created by a deity because it's so complex. Neither is falsifiable.
We live in the base reality. All a simulation could do is obscure complexity.
Interestingly this isn't a new idea. The gnostics of the first two centuries BCE had a theology of this, in which a flawed world, in which humans lived, was created by an imperfect creator called the demiurge. There was a perfect realm beyond this world with an intercessor named Sophia (wisdom) who would rescue us from our flawed world, the simulation created by the demiurge.
I could argue it makes just as much sense that every layer is simulated by the layer "above". That could continue ad infinitum or perhaps create a cycle, so eventually it's simulating itself. These may be what we refer to as simulations, like what's running on my PC, or some kind of emergent behavior. Either case seems to work.
It's natural to define the universe / all of reality / all layers as that which is self-defining, which is hard to describe without circular reasoning.
A simulation only has meaning if there is a base reality. Otherwise there are only layers of reality.
A simulation is alterable in a way that a physicalists reality is not. For example by it's parent can slow it down, or alter physical constants.
There's no reason to presume the simulation proceeds at a constant rate as measured from the outside world either: “render times” may vary with the complexity and fidelity required — for example when we innards fire up the LHC, a huge bump in resolution might be required. When we innards venture onto the moon, a whole new world has to be mapped in and professing speed might drop accordingly.
What the original poster is saying is that insofar as we do not know what the base reality is, there is no reason to argue that our reality is a lower-fidelity instantiation of it: we might as well be a flawless emulation of something totally orthogonal to the base reality.
Furthermore for obvious reasons computation cannot endlessly recurs endlessly and somehow forms into a loop: somewhere the arithmetic actually has to be performed.
>and can impinge upon the simulations at any time
A simulation could just be some states. Consider the rule, double and subtract one to get to the next world-state, over the integers, and start the world with the state 1. So the next state is 1. The one after is also 1. The one after is also 1. The universe is stuck in a static state.
As I, you, or a Python script is modeling this universe, where does the outside universe "impinge" on it?
Then let's do a different universe, starting at 2. Then the next state is 3, the state afrer is 5, the state after is 9, and so forth.
As you do this universe in your head, where does the outside universe "impinge" on it?
There is a problem, however. All future states are well-defined in the universe I've given even if nobody in our universe or any other is calculating them.
It is trivially obvious that the human mind is an emergent property of its neural firings (of the brain), and that replacing some part of it with a black box that interacted with the rest would leave it with the same emergent properties, even if this black box were modeling in a different medium.
The question is, if you can model the whole thing over a well-defined program that did not require non pseudorandom entropy, then it would be similar (just start with more complex data than 1 or 2, and more complicated rules than double and subtract one) - so in order for these states to exist, would anyone have to actually calculate it, or is it enough for these states to be mathematically defined?
What I am saying is that it is a statistical certainty that within a thousand years we can create a deterministic model of a brain in a VR, where the brain reports consciousness. (this is 100.000% absolutely necessarily true and not open to debate. The brain is a couple of pounds of stufd, with just a few hundred billion neurons or fewer - I bet we could practically have the hardware to model one today.)
Does this mean that that model already exists? (before anyone chose to calculate it)?
This is similar to asking whether the quintillionth prime exists. Yes, of course it exists.
Really? Even if nobody calculates it? Then what is to say that we might not exist without anyone calculating us?
Perhaps the states of the universe, like the quintillionth prime, exist without anyone calculating it. Perhaps there is no underlying reality at all, just as there is not an underlying reality in which the quintillionth prime is embedded.
That sounds completely absurd to me given that we can't even agree on a definition for consciousness.
It doesn't matter what you think conscienceness is or whether there is agreement on it. (Sorry if this is harsh.)
And even if consciousness is an emergent property of the structure of the brain, how could we have a simulated brain report that it is experiencing consciousness when we cant define what it is in the first place?
it's not something that is even worth discussing. of course if you made an emulated version of a brain in a VR with the same structure and neurological action and it reported consciousness, then it would be conscious, of course, yes. This is like asking whether NES games can "really" be running in a browser, even if the browser is emulating an NES system and includes the code of the original cartridge. Is the game "really" running? Can you "really" play it? Meaningless questions. If the VR person with the same brain topology as a human and hundreds of billions of neurons reports consciousness then of course it has it. why would being emulated suddenly make it different from all 7 billion humans running natively in the world, especially if theoretically it could be compiled from the same source code (which we have a copy of, DNA was sequenced in full in the 90s).
Since my comment was about a thousand years out, there's no question and we don't need to discuss these things. It's open and shut. I don't need to give any citations.
I find it amusing that you think a topic as widely studied and disputed as consciousness is "open and shut" because of some assumptions that you choose to make. Nonetheless, I do appreciate hearing your perspective so thank you.
I completely discount this possibility and think it's not worth discussing. So, you are right that my mind is very closed to any alternatives to what I've stated, though as in your example of an antenna, we could describe such alternatives rigorously. Not worth our time. (IMO). Thanks for the replies.
http://plato.stanford.edu/entries/pineal-gland/
Edit: my attempt to elaborate on the idea from the story ... I don't know if I can summarize the concept that well -- I think the framing within the story for the plot is something along the lines of: 1. Mathematical truth exists 2. There is an existential difference between a mathematical truth and a specific occurrence of the "pattern" described by that truth within the physical material of the universe.
I think the eli5 example in the story is based on the integers. You have the integers which are infinite, and you have an instantiated version of some number n if you have an occurrence of n things in nature. The reality of n is fundamentally different if you are in A universe with "n things" vs one with only m things m<n.
I think that's the motivation for the concept -- and in the story a researcher sets off to find scenarios where the universe had not yet "instantiated" one of the "laws" that could be derived from mathematical reality -- and ends up finding out that there is more than one possible way for the laws of "reality" to expand ... -- with different implications on what laws might be discovered in the future ...
EDIT: Thanks for the expansion! Fascinating and really interesting idea. The idea, "There is an existential difference between a mathematical truth and a specific occurrence of the 'pattern' described by that truth within the physical material of the universe" is a fascinating one.
In our case it could be the way to FTL. Did any Sci-fi author considered that yet?
I guess there has been plenty of philosophical research on this subject.
The formula they give, indicates that the largest QCD calculations to date have taken 7.9e11 petaflop-years; this is consistent with the graph they give, but not with actual feasibility. What am I missing?