The way to determine if we are in a simulation is to build one ourselves; If the universe that the simulation hardware exists in has approximately the same complexity as our own, then the construction of a vast computer to run simulations on-
perhaps a Matrioshka Brain -
will cause the simulating entities to rapidly increase the necessary size of the simulation, and probably slow it down as well. This might eventually cause their simulation to become too big to run without tell-tale errors.
but if the universe in which the simulating hardware is significantly more complex than the one we are in, then nothing we can do would ever overload the hardware.
Come on- all that empty space between worlds; and all that empty space between nuclei?
Bit of a give away isn’t it?
I think the simulation of the laws of physics is a real necessity. If the simulatiors were only interested in stone-age societies (or anything up to when humans first began seeing experimental data that implied quantum level granularity) then they could get away with a very cheap simulation - it would pretty much just need Newton’s laws and thermodynamics. This would indeed allow them to simulate our less technological ancestors over and over and over.
This kind of simulation would necessarily take the form of the simplified environment and a number of BrainsInnaJar[sup]TM[/sup] (either literally, or as simulations of human thought processes in a computer) with their senses hooked up to the simulation output. The environmental simulation’s detail would be insufficient to model the function of a mind, so the minds must be external to it. This could suggest that if we were a pre-technological race then the chance of us being simulated goes up.
However, as technological expertise increases in the inhabitants of the simulation the simulation detail would have to be increased to the point at which it simulates the quantum level. At this point it would be possible for the simulation inhabitants to function entirely as part of the environmental simulation; it would not be necessary for their minds to run on separate hardware. The computational overhead of running the minds inside the simulation would not be that great since the total mass of all human brains is insignificant relative to the sum of all other mass in the simulation.
You could draw the conclusion that the more technologically advanced we are the less likely we are to be in an ancestor simulation.
The problem with the simulation builders running lots and lots of low detail simulations means they would only be simulating a portion of their ancestors. How big a portion we don’t know; how often we can’t guess. It makes a nonsense of Bostrom’s reasoning - his equations are far, far too simplified to be anything other than a curiosity. It’s like trying to work out the size of the universe while assuming that Pi is four, up is down and light travels at 250mph except in built up areas. Sure, you get an answer, but its relation to reality is coincidental at best.
Just a layman here, but a very intelligent layman. I was lost after the second posting. So let’s go back to the basics, for my sake atleast.
In my opinion, option 1 is the most probable. I chalk this up to Human Nature, if such a thing really exists and wasn’t a programmed line of code created by some being.
Option 2 cannot be determined, because how would we know what a future civilization would do? But I would put this in the improbable category since one of the most popular computer games on the market today is the Civilization series. In that sense, some being is running a simulation of us running a simulation of a previous civilization.
Given what I previously said about options 1 and 2, 3 is just too far removed to make any reasonable guess on the validity of this. A lot of steps are missing in between options 2 and 3, so much that the leap seems ridiculous.
We take so much for granted in what we know, that in discussing this topic it is too easy to get caught up in our perceptions. You have to assume that everything we know, gravity, the moon, light, electricity, the infinite universe, is not true. There may be no such thing as gravity or light. We may not even trully have the shape and form of a human being.
Having said that, here is my thought on the mass simulation idea:
If everything around us is a simulation, it is a pretty good one. The only clue that we are in a simulation was addressed by Mangetout: it is only when we explore the extremes of the universe and its laws, that they start to break down and exceptions occur to fill the gaps that are made. Mangetout gave some examples, but a couple others come to mind: the size of the universe, what exactly is time?, what is the actual smallest particle?
The overhead for such a simulation would be the details. The beads of sweat on my forehead, the crease in the leaf of a tree.
But before we get to the resolution of what we are seeing, the underlying base code would be pretty simple. I have always thought of this as a fractal. You create a simple line of code, an equation, call it a law of nature, and put it in motion. The replications ripple out through the universe, creating the universe as it expands.
You would only need a handful of laws. Gravity, light, time, relativity to get started. These laws beget physics and chemistry, which begets life, and so on. Pretty soon they cross over each other, and layers of equations start to form, making the simulation more complicated and appear more real to the inhabitant. Eventually, there is no reason to doubt it.
One thing people forget is that the simulation doesnt have to be fantastically detailed;
for instance- consider the Fovea.
This is the only part of your eye which has sufficient resolution to read small print; to check this out try to read print at the bottom of the computer screen while you are looking at the top.
We see, feel, taste etc the world through a very limited
sensorium; all the simulating demons have to do is provide enough bandwidth to be convincing.
Chicago
It’s easy to be pessimistic about humanity’s future. Perhaps a little too easy…
I think the point is that either options 1, 2 or 3 will occur with a very high degree of probability. (Or at least that is the author’s argument.)
Furthermore, given that option 1 is by no means certain or even extremely likely (Prob<.9, say), we must consider options 2 and 3. For those who assign a low probability to option 2, it follows that option 3 cannot be ruled out.
I have argued for putting a heavier weight on option 2. Armilla has argued for the low likelihood of option 3, at least for technologically advanced civilizations.
eburacum45:
We are indeed creatures of limited bandwidth, but that understates our Overlord’s problem: the bandwidth they send us has to make sense: it has to be internally consistent. This is a smaller task for a Stone Age society and a larger task for ours.
Nonsense. We understand the world with models. All the Overlords need to do is arrange it so their models are somewhat more complex than ours are forecasted to be over the next 100 years or so. Anomalies are ok: the Overlords merely have to make sure that no extra-natural patterns appear, so that scientists stay focused on the curtain, rather than the being or mechanism behind it.
Next step: Provide an example of a 21st century phenomenon that would be extraordinarily difficult to simulate or provide a patch for, relative to hunter-gatherer life-experiences.
First of all, I don’t think that the moral argument–that our creators would have to be some bastards to let us suffer like this–holds too much value. The truth, as any good experience of drug-induced hyperconsciosness reveals, is sadly that life is completely meaningless and there’s no point at all worrying about such things as pain and death. If our creators are worth a grain of salt in intelligence, they too would plainly realize this.
As for the mother universe: i very much like the idea that it is somehow fundamentally different from our own. The simplest solution to the problem of calculation is to simulate a universe that is one or more dimensions less than yours. You’d even still fully have the same laws of physics (and i think this is important because i have a hunch, for reasons that are too much to describe here, that the laws are the only logically possible ones–except i guess, though very much doubt, that logic itself might be arbitrary).
Another thing to note: a given volume of space has a finite informational content (ie there is a finite number of ways in which it can be configured). I don’t know exactly how physicists came up with this, but it has something to do with being able to calculate the possible information content inside a black hole simply from its surface area. Or something like that. Anyway, if space has a limited information content (and that statement implies and includes the granulities of the various plank measurements and the quantum nature of the world and means that the universe isn’t infinitely complex), then the whole simulation becomes far more feasable and practical. More specifically, it becomes CALCULABLE (at least i speculate that it does). And probably quite easily calculable if the number of dimensions in it is less than of the simulator.
There is one point of Armilla’s that I think I should address;
** If they can build a computer sufficient to simulate the Earth for us, why bother at all? The sim-builders probably live in a perfectly good universe already - why not just genetically engineer a few thousand humans, dump them on a terraformed planet and watch the action unfold from your majestic starships? The ability to simulate obviates the need to do so - you’ll already be powerful enough to create engineered environments using the raw materials in your own universe with all the advantages that brings (it’ll run in real-time, won’t comsume as much energy and you won’t have to keep getting the tech-support guys out of bed at 3am when the whole thing crashes becasue the guys at CERN just did something unexpected).**
You see, there aren’t that many terraformable planets up there, and the ones that are will take a long time to get into a habitable state; there might be a planet as difficult to terraform as Mars in every system, but (IMO) you won’t find an Earth like planet if you survey a thousand worlds…
not to mention the incredible difficulty of delivering living people to solar systems light years distant.
On the other hand every solar system has solar energy and metals to manufacture computer processors and support systems;
each solar system could support thousands of virtual worlds and quadrillions of virtual intelligences, even if none of the planets can support human life.
A robot computer-builder ship with no life support requirements could be a lot smaller and lighter, saving a great deal in fuel.
Hey, even though I would like the universe to be colonised by human meat-bags it seems more likely that the people of the future will be solid state rather that protein and lipid.
