Why you can't simulate an universe - interesting paradox

[Master Disaster]

2 minutes ago, straight_stewie said:

And just to add some real science to this, if quantum mechanics is correct, then every possible universe that might exist is nothing but a closed form probability function. In fact, every universe is just different points along the same probability function. And that, is surely something that can be calculated within the universe for which you are doing the calculation.

My favourite quantum mechanics theory is that reality is just an illusion. Since we kind of know that down at the sub atomic scale its possible for particles to "borrow" energy from the future and seemingly come into existence from nothing its theorised that everything we experience in our day to day is just an illusion created by the sub atomic world.

 

Quantum mechanics are freaking awesome, totally confusing but awesome.

[Wictorian]

4 minutes ago, straight_stewie said:

You can do a complete simulation of a modern cell phone, in near real time (but not quite), of the physical circuits, including those in the integrated circuits, on even a relatively modest computer.

You can even get the tools to do so, for free, from device makers (like Microchip Inc. for example).

You can even simulate/emulate a portion of the hardware, completely accurately including electrical response timing, on a desktop computer while the rest of the circuit is physically built and operating, from a cheap laptop. This is called In Circuit Emulation.

I'm not trying to be to anti-Phil Wheaton here, but your estimations of the amount of data involved in doing things leave a lot to be desired.

And just to add some real science to this, if quantum mechanics is correct, then every possible universe that might exist is nothing but a closed form probability function. In fact, every universe is just different points along the same probability function. And that, is surely something that can be calculated within the universe for which you are doing the calculation.

You can do a simulation of a cell phone, I never said you can't. I said you can't simulate the universe itself. And the kind of simulation you are talking about isn't fully realistic. 

 

I don't really know about quantum mechanics, it may be possible that way.

[XWAUForceflow]

17 hours ago, Wictorian said:

So in order to simulate an atom, you would need more data points than that can be stored in an atom. Thus, you would need bigger servers than the universe itself to simulate it.

There is one obvious flaw in that sentiment. The vast majority of the universe is empty, so in terms of pure size you could store all data points of a universe in a system that has less size (though not mass) that the entire universe.

Though all in all this is more of a philosophical discussion than a technical one. You do not necessarily need all data points of a system to still simulate it depending on the error rate you accept.

[minibois]

Would a simulation of the universe, contain itself?

If so, could a full simulation ever exist?

[Master Disaster]

57 minutes ago, minibois said:

Would a simulation of the universe, contain itself?

If so, could a full simulation ever exist?

What if we're already in a simulation, is it like inception where every layer you go deeper time slows down?

[Sauron]

Define "universe". You can't simulate the entirety of our universe within itself simply because as far as we know it is infinite; however you can hypothetically simulate a limited one.

 

As for how much data you would need to store per atom... you wouldn't actually need to copy the same data for each individual one, I hope you realize. All you'd need to know is its position, speed and type which you could compare with a list of atom types in order to get specifics. A bigger problem would be the uncertainty principle but because it is a simulation you would not need everything to be an exact representation of the real world, you would just need it to behave in the same way under observation. For the purpose of a simulation, if you can't observe it it might as well not exist.

[tikker]

1 hour ago, minibois said:

Would a simulation of the universe, contain itself?

If so, could a full simulation ever exist?

Nice brain-breaker and if a simulated universe contains itself then that simulation should contain the simulation which contains the simulation... universe-ception.

 

1 hour ago, Wictorian said:

I said you can't simulate the universe itself.

We can and we do, just not in real time. There's an entire branch in astronomy that focuses on simulating universes on supercomputers with the end-goal of obtaining something that represents our Universe. These are very complex and demanding simulations, but we can do them. To first order you are right. There isn't enough computational power and storage to track every single proton, electron or object in the Universe. That would be nigh impossible (look at the n-body problem), 

2 hours ago, Wictorian said:

I am just talking about simulating a universe as detailed as our universe, which involves simulating the quarks in the atoms and even more.

but here's the thing: we don't have to. We have scientific theories that tell us how (we think) forces and other concepts behave. Where necessary we'll use n-body and hydrodynamical simulations to model evolution, but plenty of times we can get away with evaluating a bunch of equations instead. One still has to calculate these equations at many many timesteps, but it's much faster and simpler than a full blown simulation. This is called semi-analytic modeling. For example, simplified, you do a presumably fully-accurate n-body simulation for the gravtitational interaction of dark matter. Then on top of this you evaluate your computationally-cheap equations for stars, galaxies etc. and you have a simulated universe.

 

One reason this is possible is simply scale related. It doesn't matter what an iron molecule on Earth is doing when you are simulating the solar system, not even all life on Earth matters for that. All that matters is you have something the mass of the Earth. So indeed we can't really simulate the largest and the smallest scales all at the same time, but we don't have to either because those things don't really affect each other and can be described by other means.

 

1 hour ago, Wictorian said:

And the kind of simulation you are talking about isn't fully realistic.

What do you mean with "fully realistic". If you construct a simulation that correctly predicts how a something behaves and evolves, than by definition you have a realistic simulation. If the simulations @straight_stewie mentions would not be realistic and could not accurately predict how circuits would behave, all that software would be completely worthless and unusable! Those simulations also (I would assume) don't simulate every single particle interacting with every single other particle. They use equations for circuits or electrodynamics and solve/evaluate those over time, because they either perfectly describe (in the simplest cases probably), or approximate to high enough accuracy, reality for that purpose.

[minibois]

37 minutes ago, Master Disaster said:

What if we're already in a simulation, is it like inception where every layer you go deeper time slows down?

slow down, Elon!

 

[Wictorian]

1 hour ago, XWAUForceflow said:

There is one obvious flaw in that sentiment. The vast majority of the universe is empty, so in terms of pure size you could store all data points of a universe in a system that has less size (though not mass) that the entire universe.

 

Wow, I totally didn't take it into account! But isn't there dark matter ?

[Master Disaster]

36 minutes ago, minibois said:

slow down, Elon!

Pun intended or accidental?

[XWAUForceflow]

42 minutes ago, Wictorian said:

Wow, I totally didn't take it into account! But isn't there dark matter ?

Theoretically yes, though how much there is and how dense it is is unclear. It's not everywhere though, so the simulation would still be smaller.

[minibois]

2 hours ago, Master Disaster said:

Pun intended or accidental?

 

actually it's just accidental, as I don't really associate with "speed". I suppose he's at the top of two companies, which make cars and spaceships, but "speed" doesn't seem like the most important thing in either of them.

[_princess_]

11 hours ago, Wictorian said:

It wouldn't be realistic enough.there are many things in space we don't see, it would fuck up the orbits.

hmm this could be used to disprove/prove we are in a simulation. If we get everyone on earth to look in the same direction, after many years we might see some error in the orbits. The error would be very small because I'd assume the sim would still calculate the orbits even if noone is looking at the planets.

[Wictorian]

3 hours ago, _princess_ said:

hmm this could be used to disprove/prove we are in a simulation. If we get everyone on earth to look in the same direction, after many years we might see some error in the orbits. The error would be very small because I'd assume the sim would still calculate the orbits even if noone is looking at the planets.

Also for example the light of stars take millions of years to travel to earth so they can't be only calculated only when they are being observed. Of course you can surprass the speed of the light in a simulation but I guess we would notice it or it could cause issues.

[Trinopoty]

I'll chime in my two cents. And I'll add that I haven't read all the previous comments.

Most physics simulators that run on desktop grade hardware are just trimmed down (very trimmed down) emulations that get the job done good enough.

If you need more accuracy, you need to consider much more data points and need that much more processing power to compute those data points. Processing power that you can't find in a desktop level computer. That's why people spend billions of dollars building supercomputers.

You can essentially take the same models that supercomputers run, remove about 99 percent of the data points and have it run "good enough" on your home PC.

Professional engineering simulations take into account data points that they deem necessary, which does not cover all possible data points, and even those take a beefy systems to run.

So yes, at the current stage, with our current understanding and technology, it's difficult to accurately simulate any universes on an atomic level. But we can kind of make good enough simulations without going down to the atomic level.

[tikker]

3 hours ago, _princess_ said:

hmm this could be used to disprove/prove we are in a simulation. If we get everyone on earth to look in the same direction, after many years we might see some error in the orbits. The error would be very small because I'd assume the sim would still calculate the orbits even if noone is looking at the planets.

It probably wouldn't matter if people were looking left or right (we likely don't really affect Earth's orbit as we weigh next to nothing), but you've got the right idea. We'd need to somehow know or assume something about the nature of the simulation we might live in. You would need to predict "if we live in this simulation, the Earth will be at position X in 20 years" and then measure where the Earth is in 20 years. This is a bit similar to how GR was tested. Planets precess, but Newtonian physics could not predict the correct rate. When GR was used, the predicted value matched the observed value of e.g. Mercury's precession.

 

22 minutes ago, Wictorian said:

Also for example the light of stars take millions of years to travel to earth so they can't be only calculated only when they are being observed. Of course you can surprass the speed of the light in a simulation but I guess we would notice it or it could cause issues.

If we lived in a simulation with a different speed of light our speed of light would simply be different. We would probably be none the wiser. We can say that a different speed of light will produce a universe different from the one we live in, but it won't help us determine if we live in a simulated universe that happens to have our speed of light.

 

In the context of simulating a universe, I think the Illustris project is interesting to take a look at:

https://www.illustris-project.org/

It's a state of the art simulation trying to reproduce what we see in reality and they're producing impressive stuff.

 

 

 

[piratemonkey]

As others have mentioned the logic is flawed. Not only from a simulation standpoint, but also because how we understand the universe. It is ever expanding. If we wanted to perfectly simulate the universe (wouldn't be possible with our current level of knowledge), we would probably not have enough energy to sustain a simulation. I don't believe quantum computing would be able to, or at least not this century. It is not possible to simulate the universe even with what we know. We can't even simulate the weather properly

[_princess_]

On 1/22/2021 at 3:50 PM, Wictorian said:

Also for example the light of stars take millions of years to travel to earth so they can't be only calculated only when they are being observed. Of course you can surprass the speed of the light in a simulation but I guess we would notice it or it could cause issues.

 

On 1/22/2021 at 4:14 PM, Trinopoty said:

I'll chime in my two cents. And I'll add that I haven't read all the previous comments.

Most physics simulators that run on desktop grade hardware are just trimmed down (very trimmed down) emulations that get the job done good enough.

If you need more accuracy, you need to consider much more data points and need that much more processing power to compute those data points. Processing power that you can't find in a desktop level computer. That's why people spend billions of dollars building supercomputers.

You can essentially take the same models that supercomputers run, remove about 99 percent of the data points and have it run "good enough" on your home PC.

Professional engineering simulations take into account data points that they deem necessary, which does not cover all possible data points, and even those take a beefy systems to run.

So yes, at the current stage, with our current understanding and technology, it's difficult to accurately simulate any universes on an atomic level. But we can kind of make good enough simulations without going down to the atomic level.

 

On 1/22/2021 at 4:38 PM, piratemonkey said:

As others have mentioned the logic is flawed. Not only from a simulation standpoint, but also because how we understand the universe. It is ever expanding. If we wanted to perfectly simulate the universe (wouldn't be possible with our current level of knowledge), we would probably not have enough energy to sustain a simulation. I don't believe quantum computing would be able to, or at least not this century. It is not possible to simulate the universe even with what we know. We can't even simulate the weather properly

If we are in a simulation, it is not that it is simulating the universe, but that it is simulating a universe.

 

We need not create a simulation that simulates our universe in order for our universe to actually be a simulation. The original universe that created this universe might not be anything like this universe at all. Rather this universe is more like a game of sorts (not a good one imo.)

 

A game doesn't need to simulate all that much, all it has to do is produce consistency and immersion. Ie. when we go to do something IRL whatever happens should be within the expected framework of the game. If we drop a ball it falls, and it always will fall, and so forth. Not that much computing power is actually needed. For instance, when we look at bacteria in a microscope, all the game has to render is a few bacteria cells, at no point does the game ever actually compute millions of bacteria cells. The game uses more generalized equations to predict behaviors of small entities: hence quantum physics. That is why when you try to measure something precisely, you cannot be sure of the outcome: the game's equations predict behaviors of large masses and lacks the computing power to make precise predictions of movement on the individual molecular level. This is why, theoretically you could have air particles fill only 1 side of a room due to random probability, but actually air particles will never fill only 1 side of a room.

 

In terms of quality of the game, already stated imo its not a good one. It may be more of an emulation than a simulation. For instance in some fps games there is a hitbox on the head and body to simulate shooting but actually no shooting is occurring and its just an illusion that lacks realism. That could describe the relationship between this universe and the universe that created this universe. You would not know if this universe is simulating anything since you don't know what the universe that created this is. All that this universe would need is consistent behaviors and so the illusion could feel real. However it doesn't even provide that, a lot of people can do telekinesis and other paranormal stuff that isn't explained by physics laws, this universe doesn't provide behavior that is really that consistent, each year old physics discoveries are being challenged contested or broken, and though that could be attributed to errors in human reasoning of possibly both parties, the actual physics of the universe might be slowly changing and accumulating errors over time.

Elon Musk is not yet disproven, QED.

[piratemonkey]

2 hours ago, gabrielcarvfer said:

I think the statement could be more specific. What do you mean by that?

That we don't have the computational power to simulate complex models or that we don't have a precise model due to the lack of historic/diverse/precise data?

I would say the later option is way more accurate than the first.

A bit of both. We can't simulate weather cuz we don't have enough data, and just can't simulate everything needed to properly do it. Weather is complex and has a ton of variables ig. Compare that with doing space stuff, its much easier to simulate where a planet will be in x number of years

[tikker]

10 hours ago, gabrielcarvfer said:

I think the statement could be more specific. What do you mean by that?

That we don't have the computational power to simulate complex models or that we don't have a precise model due to the lack of historic/diverse/precise data?

I would say the later option is way more accurate than the first.

 

8 hours ago, piratemonkey said:

A bit of both. We can't simulate weather cuz we don't have enough data, and just can't simulate everything needed to properly do it. Weather is complex and has a ton of variables ig. Compare that with doing space stuff, its much easier to simulate where a planet will be in x number of years

A big issue with weather, if I'm not mistaken, is that it's a chaotic system. Small changes in initial conditions can produce very different outcomes, so uncertainties in measurements have a big impact on predictions.

[WihGlah]

On 1/21/2021 at 4:29 PM, Wictorian said:

So in order to simulate an atom, you would need more data points than that can be stored in an atom. Thus, you would need bigger servers than the universe itself to simulate it.

 

Computing a Universe Simulation - YouTube

[Wictorian]

18 hours ago, _princess_ said:

 

 

If we are in a simulation, it is not that it is simulating the universe, but that it is simulating a universe.

 

We need not create a simulation that simulates our universe in order for our universe to actually be a simulation. The original universe that created this universe might not be anything like this universe at all. Rather this universe is more like a game of sorts (not a good one imo.)

 

A game doesn't need to simulate all that much, all it has to do is produce consistency and immersion. Ie. when we go to do something IRL whatever happens should be within the expected framework of the game. If we drop a ball it falls, and it always will fall, and so forth. Not that much computing power is actually needed. For instance, when we look at bacteria in a microscope, all the game has to render is a few bacteria cells, at no point does the game ever actually compute millions of bacteria cells. The game uses more generalized equations to predict behaviors of small entities: hence quantum physics. That is why when you try to measure something precisely, you cannot be sure of the outcome: the game's equations predict behaviors of large masses and lacks the computing power to make precise predictions of movement on the individual molecular level. This is why, theoretically you could have air particles fill only 1 side of a room due to random probability, but actually air particles will never fill only 1 side of a room.

 

In terms of quality of the game, already stated imo its not a good one. It may be more of an emulation than a simulation. For instance in some fps games there is a hitbox on the head and body to simulate shooting but actually no shooting is occurring and its just an illusion that lacks realism. That could describe the relationship between this universe and the universe that created this universe. You would not know if this universe is simulating anything since you don't know what the universe that created this is. All that this universe would need is consistent behaviors and so the illusion could feel real. However it doesn't even provide that, a lot of people can do telekinesis and other paranormal stuff that isn't explained by physics laws, this universe doesn't provide behavior that is really that consistent, each year old physics discoveries are being challenged contested or broken, and though that could be attributed to errors in human reasoning of possibly both parties, the actual physics of the universe might be slowly changing and accumulating errors over time.

Elon Musk is not yet disproven, QED.

> That's why I said simulating the universe is impossible.

 

> The weather example is exactly why a simulation wouldn't work. For example to determine a baby's gender, you could choose one with 50% chance, but that's actually not how it works and with enough data, you could know it beforehand. 
 

> Wow, the idea that the laws of physics are changing is crazy! It would be a pain in the ass for us. At this point I would like to quote Murray Gell- Mann, "Imagine how hard physics would be if particles could think". But I would rather think our understanding of the universe is flawed mostly.

 

> That we are living in a simulation could be proven, but I don't think it could be disproven.

[_princess_]

18 minutes ago, Wictorian said:

> That's why I said simulating the universe is impossible.

 

> The weather example is exactly why a simulation wouldn't work. For example to determine a baby's gender, you could choose one with 50% chance, but that's actually not how it works and with enough data, you could know it beforehand. 
 

> Wow, the idea that the laws of physics are changing is crazy! It would be a pain in the ass for us. At this point I would like to quote Murray Gell- Mann, "Imagine how hard physics would be if particles could think". But I would rather think our understanding of the universe is flawed mostly.

 

> That we are living in a simulation could be proven, but I don't think it could be disproven.

The weather example is explained if the human brains themselves are computers running the simulation. For instance, the simulation could be a network of computers. The human brains could actually be creating the cloud data and transferring cloud data from brain to brain. As soon as one human turns on the TV, the data of the weather is uploaded into the computer of their brain. The weatherman going to school and learning how to analyze the clouds, sends the data to the tv which is then sent to other brains/computers of the network. A hallucination could be a computer error. Either it is corrected by the other brains in the network (like blockchain) or else it might become a group hallucination.

 

Quote

That we are living in a simulation could be proven, but I don't think it could be disproven.

Please list all possible ways we can definitely prove we are in a simulation.

[Wictorian]

On 1/22/2021 at 8:27 PM, _princess_ said:

hmm this could be used to disprove/prove we are in a simulation. If we get everyone on earth to look in the same direction, after many years we might see some error in the orbits. The error would be very small because I'd assume the sim would still calculate the orbits even if noone is looking at the planets.

You said yourself. 
 

We can't really know it but proving it would be a lot easier than disproving it.

[_princess_]

1 hour ago, Wictorian said:

You said yourself. 
 

We can't really know it but proving it would be a lot easier than disproving it.

yes but that was just one vague example. I think there might be some difference but I don't know what difference. And what other tests could we run also?