complexity of computing the location of particles that make up the universe
For this you’d need a quantum computer which is just around the corner. Its more a question of how much energy does the thing need to run on. Does it need a universe sized power supply?
We would need at least 1 bit or power per 1 bit of information.
Physics says we aren’t in a simulation, says a bunch of people who learned simulated physics in a simulation.
This whole article is premised on simulating a quantum universe on a classical computer. But quantum computers are a thing, and one of the reasons physicists are interested in them is trying to simulate larger quantum systems than classically feasible.
There is no way to predict whether you are standing on the upslope of a power-law curve or an S-curve that’s about to level off, until you get farther along the curve.
Yes there is. The latter. Nothing grows forever. This is the first error of market capitalism.
Clearly Trump believes that Donald Trump is the point of the current simulation.
Great article, Cory! There are far too many popular non-falsifiable propositions out there already. They’re called “religions.”
I really don’t understand the argument at all. Why does living in a simulation require an entire universe to be calculated?
Wouldn’t it be much more likely that we live in a procedural generated “sky box” of limited scope and vastly limited granularity of detail? The simulated location of subatomic particles wouldn’t even be needed unless called upon by some sub-routine. (Function (ObserverParticleAccellerator))?
Humans are ridiculously limited when it comes to physical awareness of their environment at even a macro level. Why would the simulation need to provide more than necessary to maintain that coarse, low res level of local detail with the ability to procedurally generate a zoom in once in a while?
So they did the equivalent of the IEEE 754 test of computing round off error in the real world. Try this in your computing device: 0.1 + 0.2 - 0.3 and see that it doesn’t equal zero. Do it in reality and it’s zero.
Not this fucking bullshit again! Wait, let me get my shoe to stamp this cockroach out…again!
The probability of any single binary event is fifty-fifty, meaning it either does or does not occur. Only over multiple iterations do probabilities hold sway. If someone lives in a given space and time (their hour upon the stage), then someone has to be that someone. They’re no more or less likely to exist than anyone else. They simply do or they do not. Statistics is not about what does or does not occur, it’s about the mathematical relationship between events. Moreover, statistics and probability are related, but they are not the same thing.
That is one reason. It’s far from the only reason we work on the technology.
If the universe is a simulation, it has to be able to calculate probability amplitudes of quantum states. The only type of computer we know of that can even do this in theory is a general quantum computer, therefore the universe cannot be a classical computational simulation. The caveat is that we do not know if the same calculations can be carried out by a large but finite classical bits, that is we do not know if quantum amplitudes, and thus the physical universe as we observe it, are truly analog or at some level digital. They may simply be so large as to be practically indistinguishable from infinity, or they may be truly infinite. Either way, we know that any classical computer that could simulate even a modest number of quantum bits – even just the ones we measure (that is, even if the only parts of the universe that are simulated are those we happen to observe), “procedurally generated” in the current buzzword-happy rhetoric – is at least many many orders of magnitude larger than our universe. Therefore we can conclude that a simulation equivalent to our observed reality would have to either be a quantum computer, a classical computer in a universe quite different from the one we observe, or something else entirely.
There is simply no way to rule out a simulation for exactly the same reason there is no way to rule out philosophical zombies and all other manner of solipsism. A mind computed to believe something would believe it, regardless, because we can never know reality but through the synthesis of our minds. Descartes tried to get around this and, in spite of his need to believe otherwise in his quest to “prove” his God, he failed. Many atheists have tried to prove reality in the same intellectually dishonest tradition. They have all failed. It’s, as the article points out, a non-falsifiable hypothesis, and therefore scientifically a difference that makes no difference.
But it’s important to note that efforts to prove we are not in a simulation are as pointless as efforts to prove that we are in a simulation. We do not know and, barring discovering a bug in the hypothetical simulation that we can exploit to escape it, or a change in it to reveal it to us, we never will know. Spilling ink and pixels on it might be an entertaining distraction, but practically it’s a waste of time, philosophical junk food with zero scientific value. By giving Musk’s interest in this ancient and by this point largely played out line of inquiry any attention, you’re only fixing other people’s otherwise potentially valuable focus on it. If you want people to stop wasting time on it, you’re doing it wrong. But if you want to feed off the hype with click-bait engagement, which i rather suspect is the main motivation of most of these paid writers, then carry on.
The article is by author Andrew Masterson, Cory merely summarized it. Credit, or blame, where it’s due.
Final note, I suspect the universe may be a type of natural quantum computer, and I’m not alone among physicists in that hypothesis which, unlike the simulation hypothesis, might be falsifiable. I very much doubt we live in what we call a simulation. I suspect it is yet another example of humankind’s long, torrid and foolish love affair with anthropomorphize nature. As worthy of ridicule as any supernatural mythology.
To simulate a universe we would need simulate every sub atomic particle in the universe. There seems to be a logistics problem here.
And what about the uncertainty principle and Schrodinger’s cat?
And what about quantization itself? What if “real” reality is continuous?
(Of course this kind of thinking actually points to an obvious problem with the simulation argument, that we necessary don’t know anything about the simulating universe, and hence the question of whether a universe like ours can be simulated is impossible to answer)
The likelihood that you are alive at any given moment is thus a tiny number divided by a number so large that the result is largely indistinguishable from zero, and remains zero-ish unless you are actually immortal.
Therefore you are immortal.
No no no. Totally wrong.
The likelihood of me existing at any given moment is infinitesimally low unless I’m immortal… therefore I don’t exist.
From now on, I will imagine the ghosts in Pacman having this conversation before heading out into the[ir] world.
“storing information about even a couple of Pac-Dots would require a computer memory that would physically require more atoms than exist in the universe.”
Yes, I looked up what ‘Pac-Dots’ are called.
I mean you don’t need to render or track any subatomic particles at all, or anything else invisible to the naked eye, for that matter. All you really need to keep the generated world manageable are enough NPCs to convince PCs that the world is made of “atoms” and that there is an expansive universe beyond you immediate, myopic field of vision. In later versions, even throw in a website where people you kinda “know” post pictures of their babies, breakfasts and vacations to stretch the impression of an inhabited space. To avoid any of this being tested, you then make the weight of simply existing so heavy that active participants don’t venture beyond the rendered zone…
I love this extension of a Solipsism… The universe only needs to be as big as the ‘hero’ characters who study it.
Edit: Of course, this is presupposing that the point of the simulation is to trick the inhabitants, rather than the inhabitants being an unintended consequence of a general universe sim.
Absolutely false. In fact, situations where something increases or decreases forever (asypmtotically) are quite common, both in abstract math and physics, including many common, real-life examples. See also Xeno’s Paradox of the Tortoise and Achilles for a perfect example:
http://platonicrealms.com/encyclopedia/zenos-paradox-of-the-tortoise-and-achilles
your estimate of the size of the containing universe is too small.
Or maybe the quantum effects they are trying to simulate are an artifact of the hypothetical higher simulation and not the desired behavior of the higher simulation. Maybe the simulation doesn’t keep track of protons, neutrons, and electrons unless something is actually paying attention to them. Just keep the location of the atom, and the counts of it’s elements and if someone (or some other part of the simulation) checks just spring those elements into existence until they look away again.
That is the type of optimization a lot of computer models make for the simulations we run of our own world. Need to model the movement of air molecules? Just use random movement instead of modeling actual collision physics, for many and possibly the vast majority of simulations it produces results close enough to be indistinguishable form the real calculations. There are cases where you really do need to simulate down to the smallest detail, and then you do, but if you can avoid the work most people writing models will.
Regardless of all that, the article grossly misrepresents the actual scientific paper it references. The real paper discusses how some problems solvable on a quantum computer would be impossible to solve on a traditional computer. This has almost nothing to do with if our reality could be simulated, it only means that if it can be simulated then that simulation must be running on a quantum computer.
As the great philosopher Douglas Adams once said, the average population of the universe is 0.
“The simulation argument” recently popularized by Bostrom suggests that one of three statements is very likely to be true. The first two statements get overlooked because they’re mundane, they’re obvious possibilities, they’re boring:
The fraction of human-level civilizations that reach a posthuman stage (that is, one capable of running high-fidelity ancestor simulations) is very close to zero
The fraction of posthuman civilizations that are interested in running ancestor-simulations is very close to zero
The shock, the surprise, and thus the popularity of the thought experiment comes from the third possibility (3. We’re probably simulated). But even so, it’s really only “mind-blowing” if you decide #3 alone is the conclusion of some chain of reasoning, which it isn’t.
