Can a computer simulation simulate another computer running another simulation

[edy]

There's a fundamental lack of understanding on how our brain works that will need to have a breakthrough before we get anywhere close to modelling AI. I'm not talking about biochemistry and neural networks. I'm talking about deeper hardwired firmware or even basic building block architecture from which to work.

I present to you the "simple" brain of a fruit fly or ant. They are built with a ton of complex networks by DNA (innate) to accomplish 90% or more of all the tasks they will need. With a bit of plasticity to learn. Human brains, while much more complex, still come pre-built with an enormous amount of structure yet can learn and adapt when young enough.

We are trying to model also a massively parallel system that has chaotic oscillating character which may be easily perturbed from initial conditions and feedback, and "fuzzy" logic... and we still are in the infancy of this science. Exciting though!

[Carl_Smith]

Years ago I started with an idea that began as a joke to a friend, but I have filled in so many details over the years that I'm starting to actually believe it.

And that joke is that we are part of a computer simulation, and we are already seeing the limits of the simulation.  It's called quantum mechanics.

I joked to a friend that we were part of a simulation, and the quantized nature of electromagnetic energy is the evidence.  A quantum of energy is simply the smallest amount that can be represented by the least significant bit of whatever number representation is used to represent energy in our simulation.

The Theory of Relativity is actually a bug patch on the simulation by the programmers.    Whatever number representation they use for speed has a limit of the speed of light.  They set it that way because, of course, light should be faster than anything else.  When they realized that light emitting from a moving object would travel faster than the maximum speed they could represent, and overflow the variable for speed, they programmed the system to actually alter the shape of space and the passage of time to make sure that wouldn't happen.   

The double slit experiment, where light behaves as a wave and causes an interference pattern, continues to work even when you fire individual photons far apart enough in time that you know one has hit the target before the next one is fired at the target, shows another "bug patch" on the simulation.

Then there is polarized light.  Strange that a photon of unpolarized light has a 50/50 chance of passing through a polarized filter.  You would think that pretty much no light would go through, because the chance that any randomly polarized photons were perfectly lined up with the filter would be close to zero.   Another patch on the system to fix something.  The LCD screen I'm looking at right how wouldn't work without this "fix."

Of course, consider how electrons exist only in certain orbits in an atom, and move from one orbit to another without ever existing at any point in between.  This is simply because the variable that represents orbit numbers in our atomic simulation is an integer variable.   

The creators of our simulation never planned for the fact that we would someday become smart enough to see the limits of our simulation environment.

Someday I am going to write a small book on these ideas.  Maybe I can start a cult and get people to shovel money in my direction.

[daqq]

Someday I am going to write a small book on these ideas.  Maybe I can start a cult and get people to shovel money in my direction.

It would be amusing to one day rename Planck length to 1 LSB of length.

[Kilrah]

I joked to a friend that we were part of a simulation, and the quantized nature of electromagnetic energy is the evidence.  A quantum of energy is simply the smallest amount that can be represented by the least significant bit of whatever number representation is used to represent energy in our simulation.

The Theory of Relativity is actually a bug patch on the simulation by the programmers.   

I feel that way too. IMO the simulation actually does not simulate every quark in the universe and work from the bottom up to atoms, molecules etc to in the end result in the high level concepts we see every day for the reasons explained by Artlav a few posts above, it would be computationally ridiculously heavy and also useless as no one is actually looking at those things most of the time.
The simulation started and usually runs at a higher functional level of objects, concepts and actions, and the details are only created the first time we reach a level of "ability" where we should be finding something new to keep us happy - the new elements fitting more or less good with the established concepts/rules (and sometimes requiring patches if it turns out it wasn't that great...). That's worth for both ends, in the infinitely small just like at the level of the universe when we "find" new galaxies and other things "up there", and also includes new concepts. Then those things are only simulated when required because someone's looking at them.

My previous post about the brain being an example of that, the reason we can’t really understand how the brain works is that it probably hasn’t even been defined yet by the simulation. We « know » there’s a thing in there that supposedly has a role, we can recognise correlations between activity in there and things we do as well as externally act on it and see a result on a test subject, but that’s where the current detail level of the simulation ends. Or there is more detail but it’s buggy and needs a patch before it makes sense and its behavior can be replicated.

[elgonzo]

The creators of our simulation never planned for the fact that we would someday become smart enough to see the limits of our simulation environment.

"They" might not have planned for it specifically, but they might have installed sufficient capacity / resources to allow the simulation to go that way. On the other hand, perhaps "they" didn't install sufficient resources and the simulation is already operating at maximum capacity (which could explain the incomprehensible levels of stupidity individual members of the general population exhibit at an ever-increasing rate...)

Someday I am going to write a small book on these ideas.  Maybe I can start a cult and get people to shovel money in my direction.

Do that. You'll get lots of simulated money for all your your simulated efforts...   
(By the way, read Dicks "The Electric Ant" and "Time Out of Joint", if you haven't done so yet...)

Ah, all these simulated forum posts here. It's all a giant bot net... 

[xrunner]

But if we're a simulation - why? Whatever created it - why do it? What's the point? Are we like them, or are we some sort of crazy fantasy life form?

(Even if we aren't a simulation and a "god" created us, there's still the same question - why do it at all?  )

[Artlav]

Someday I am going to write a small book on these ideas.

I've read that book already.
The story was around the protagonist, a player in a game who was assaulted by NPCs that figured out the world is a simulation by observing differences in the level of quantum imperfections, and noticed that the errors were not the same everywhere.
Specifically, the errors were the least around the two persons - the protagonist and the antagonist, both PCs in the game, and increased with distance from them (some sort of LOD).

A short, but rather interesting story.
I can see potential for a full-length story on the similar idea set.

But if we're a simulation - why? Whatever created it - why do it? What's the point?

Have you ever played computer games?

[xrunner]

Have you ever played computer games?

LOL - well yea, but I'm not a member of a Super-Race of beings so far advanced we can't imagine it. So why would they do a simulation? Is somebody "playing" my character - me? My Gawd if they are playing me they have got to be bored, so bored I can't even comprehend the magnitude of their boredom.   

[botcrusher]

We could be a scientific experiment in evolution and AI, there could be beings figthing over if the simulation (us) has the right to exist and are debating pulling the plug

[Carl_Smith]

IMO the simulation actually does not simulate every quark in the universe and work from the bottom up to atoms, molecules etc to in the end result in the high level concepts we see every day for the reasons explained by Artlav a few posts above, it would be computationally ridiculously heavy and also useless as no one is actually looking at those things most of the time.

Sort of like how 3D game engines only send to the graphics card the polygons for objects that should be in your field of view.  And the graphics card only renders polygons that aren't hidden behind other polygons.

The simulation started and usually runs at a higher functional level of objects, concepts and actions, and the details are only created the first time we reach a level of "ability" where we should be finding something new to keep us happy

Hmmm.  And those details aren't actually "rendered" by the simulation until we do something to observe them.  This is an angle for working  Heisenberg's uncertainty principle and Schrödinger's quantum superposition into the simulation theory.

"They" might not have planned for it specifically, but they might have installed sufficient capacity / resources to allow the simulation to go that way. On the other hand, perhaps "they" didn't install sufficient resources and the simulation is already operating at maximum capacity (which could explain the incomprehensible levels of stupidity individual members of the general population exhibit at an ever-increasing rate...)

This reminds me of a story I once heard about how Facebook was buying truckloads of servers and installing them in data centers as fast as they could every day just to stay ahead of increasing demand on their system.

I like this idea though.  There is a maximum level of intelligence that the system can simulate, and we have sometime recently hit that limit, so now as more people are created, the average intelligence level decreases.   But somehow there are people that end up with a higher than average allocation of CPU time to their intelligence process.

Someday I am going to write a small book on these ideas.  Maybe I can start a cult and get people to shovel money in my direction.

Do that. You'll get lots of simulated money for all your your simulated efforts...   

Fine by me as long as the average person thinks the simulated money is real.       So if our "real" money is just part of our simulation like everything else, what's that make BitCoin?   Virtual simulated money?   Sort of like Virtual Virtual Skeeball?

[botcrusher]

Well, there must be expansions being made to said hypothetical server, if i was making a simulation, I'd control the birthrate to match what hardware was being installed.

Lowering intelligence as a measure to free up cpu time seems like a bad idea in terms of corrupting the value of the simulation's scientific data.

[Kilrah]

But if we're a simulation - why? Whatever created it - why do it? What's the point?

Well, same question at another level - why do we do things at all in our "lives"? If you think about it there is absolutely no point to it either

Sort of like how 3D game engines only send to the graphics card the polygons for objects that should be in your field of view.  And the graphics card only renders polygons that aren't hidden behind other polygons.

Exactly

I like this idea though.  There is a maximum level of intelligence that the system can simulate, and we have sometime recently hit that limit, so now as more people are created, the average intelligence level decreases.   But somehow there are people that end up with a higher than average allocation of CPU time to their intelligence process.

I don't think it's a matter of capacity but rather a scenario "experiment". Low intelligence people who are pretty happy in their lives and don't ask questions are a strong stabilizing factor, and enable the others to live the lives they expect by doing the dirty jobs without complaining.
In history there have been several significantly different kinds of balance in society that worked more or less well, and that's just how the current experiment goes.

I've travelled a lot (if that's a thing) and seeing the different ways of living is one of the things that make me believe in the simulation concept with a "supervisor" who tweaks parameters to see their influence, because looking at it there's no way things that are so diverse and sometimes very F-ed up could just work together that well without external input to balance things.

[xrunner]

Well, same question at another level - why do we do things at all in our "lives"? If you think about it there is absolutely no point to it either

Well of course there is a point to me doing "things". Even if I'm a simulated person, I have to do things or the simulation is going to make my simulated life miserable. If I do not go to the simulated bathroom, I'll have simulated pee and poop in my pants. Sure, I could just keep saying to myself "I'm not real and so what's in my pants isn't real either" How long do you think that's going to last?

So if we are simulations, we do things in our lives because it seems real, and we are subject to the rules of the simulator, whether we like it or not. Simulated or "real" -"Cogito ergo sum".

[Carl_Smith]

So if we are simulations, we do things in our lives because it seems real, and we are subject to the rules of the simulator, whether we like it or not. Simulated or "real" -"Cogito ergo sum".

That reminded me of this Bloom County comic from a couple decades ago:

[Kilrah]

Well of course there is a point to me doing "things". Even if I'm a simulated person, I have to do things or the simulation is going to make my simulated life miserable. If I do not go to the simulated bathroom, I'll have simulated pee and poop in my pants.

I mean "activities" you choose to do... a job, a hobby, etc... Whatever hobby you have you do because you somehow "like it" even if it's "useless" - and other people like doing things you have zero interest in. That can very well include "running a simulation and look at stupid guys virtually running around".

[hendorog]

In theory ...

Could an Intel 386 PC simulate the human brain, given enough memory and time, or is the brain not subject to being modeled in lines of code? Do you think human level AI can be created by programming at all?

What am I trying to ask, I guess, is can human consciousness be created and reside in a computer? Maybe we really don't know but what do you all think?

No, I don't think so - and I hope not.
Not due to the complexity, instead for the simple reason that would mean that the brain cannot make its own independent decisions.
A 386 is deterministic - it doesn't have true randomness, and so therefore given any starting state, any future state can be predicted.

So if the 386 truely emulated a brain it would mean that our path is also deterministic. Life would be pre-ordained and could not be changed. This is 'destiny'.

I remember figuring this out back at uni - what if there was a computer large enough to model the universe. Then program in all of the 'laws of nature'. Then any future state of the universe could be computed given enough time. Therefore the computer could predict the future.

Fortunately for the people running Lotto, the Heisenberg Uncertainty Principle comes to the rescue.

[Kilrah]

So if the 386 truely emulated a brain it would mean that our path is also deterministic. Life would be pre-ordained and could not be changed. This is 'destiny'.

I don't believe in "destiny" in the usual way (i.e. everything is hard written before birth and cannot be changed), yet am sure that we still don't have any real choice in what our life is made of. It's just done in a flexible way - the simulation is deterministic, but some of its parameters can be tweaked at run time by the "supervisor".

[hendorog]

So if the 386 truely emulated a brain it would mean that our path is also deterministic. Life would be pre-ordained and could not be changed. This is 'destiny'.

I don't believe in "destiny" in the usual way (i.e. everything is hard written before birth and cannot be changed), yet am sure that we still don't have any real choice in what our life is made of. It's just done in a flexible way - the simulation is deterministic, but some of its parameters can be tweaked at run time by the "supervisor".

I hope the BOFH isn't the 'supervisor'

[Artlav]

that would mean that the brain cannot make its own independent decisions.

The problem is that we think of a choice, a decision, in terms of
You: -i want to do that
Your jailer: -nope, do this instead.

While in practice a world with destiny is indistinguishable from the world without one, in our perception.

To truly be able to predict the future in a deterministic world you would need to know everything. Every state of every particle everywhere.
And the problem is not the uncertainty principle, but the fact that many causes are approaching at the speed of light, and a tiny prick you'd feel when a cosmic ray hits your hand can change the entire course of your day.

Back to the decision, it would be your decision, determined by the set of memories and habits, with no external will guiding it.
Only with every part of the above being a static mathematical pattern that an external observer can take a moving slice across and call that slice "now".

[HighVoltage]

I was talking with some guys on the local repeater, and we got into "what is reality" sort of discussions. I mentioned that some people hypothesize that we and our reality is simply some unknown intelligence's version of the "Sims", i.e. we are just a simulation "running" on some system we can't investigate. But yet we also have computers (simulated?) that run simulations ...

I was drawn in to this discussion through a movie, called "welt an draht" when I was 13 in 1973
https://de.wikipedia.org/wiki/Welt_am_Draht
from 1973
Probably the first one of this kind of movie

English title: "world on a wire"
https://en.wikipedia.org/wiki/World_on_a_Wire

For a wile I was 100% convinced that I was living inside a simulation.
But no matter what I did, I could not proof or disproof it.

[hendorog]

that would mean that the brain cannot make its own independent decisions.

The problem is that we think of a choice, a decision, in terms of
You: -i want to do that
Your jailer: -nope, do this instead.

While in practice a world with destiny is indistinguishable from the world without one, in our perception.

Yes I agree - we would not know that our world was pre-determined. I said 'I hope not' partly for that reason

To truly be able to predict the future in a deterministic world you would need to know everything. Every state of every particle everywhere.

Yep - and if you ignore some practical limitations, the uncertainty principle prevents you from knowing everything.

[Artlav]

Yep - and if you ignore some practical limitations, the uncertainty principle prevents you from knowing everything.

But only from inside the universe, since any observation would be an interaction.
From the outside of the universe you can observe without interacting, so it won't apply.

Then again, you can't get data from outside of the universe (by definition), so this doesn't really change anything.

[xrunner]

No, I don't think so - and I hope not.
Not due to the complexity, instead for the simple reason that would mean that the brain cannot make its own independent decisions.
A 386 is deterministic - it doesn't have true randomness, and so therefore given any starting state, any future state can be predicted.

Well, why isn't the human brain deterministic like the 386? Aren't both made out of matter?

[Carl_Smith]

Well, why isn't the human brain deterministic like the 386? Aren't both made out of matter?

I think the brain is deterministic because it is made out of matter.  I see no reason to think that our consciousness exists in any way outside of our brain, which is made up of an assemblage of molecules, atoms, subatomic particles, and electromagnetism.   If you could map all of this and simulate it, and present the simulation with the exact same inputs as what we think is the real world, it would always create the same outputs, just like a computer.   Unless there is some sort of randomness involved, perhaps from the probabilistic nature of quantum mechanics.  But a bit of randomness introduced into the equations still does not imply free will or independent decision making.  It just makes the future outputs predictable only with probabilities.

And my opinion is that if we are really part of a simulation, the simulation is at the subatomic level, and therefore our consciousness is still running on our "brain computer" even if it isn't "real".

[hendorog]

Yep - and if you ignore some practical limitations, the uncertainty principle prevents you from knowing everything.

But only from inside the universe, since any observation would be an interaction.
From the outside of the universe you can observe without interacting, so it won't apply.

Then again, you can't get data from outside of the universe (by definition), so this doesn't really change anything.

I think you are confusing the uncertainty principle with the observer effect. The UP states that two complementary properties cannot be known simultaneously. I had to look it up, its been nearly 30 years