Thread: Simulation Extension Paradox

My favourite simulation themed crazy theory, is that the probability that we're in a simulation right now is stupidly high. From a technology standpoint, we'll most likely be able to fully simulate reality for a person within the next 100 years, and that person would never be the wiser if they were submerged at birth, or rather, they never died and were instead just jacked into a new life, once their simulated life ended. Is crazy yo.

That's what the basis for my paradox was. I was thinking of the whole "is consciousness/awareness an illusion" an argument and wondering how a group of beings that was "sure" that they were aware and conscious would be able to test that their creations were indeed aware, and then it led me to think about how this could be applied to both create infinite realities (of which we could all observe, go back and forward in time through by watching a recording, and gain insights into our own reality, and the very real possibility of the existence of others) and then our ability to learn would skyrocket because rather than just assume things at a 99% certainty (arbitrary number), we can assume them at 99.99999999999999999% repeating, which is nearly the same as knowing something, but not quite. Nifty thought experiment, I thought.

I don't understand what is paradoxical, can you explain in more detail?

You wrote a lot of exposition which I can't see the relevance of, so possibly I am not fully understanding what you're saying. My understanding of the salient parts of your post is:

There is a set of nested simulations within simulations. The topmost simulation stops running. Do all of the other simulations stop running or do they go on forever?

My answer is that clearly the sub-simulations also stop running if their host simulation stops running. I don't see why any simulation would keep running.

If the simulation A hosts simulation B which hosts simulation C, then simulation A's simulation (simulation B) should have simulation B's time run out before its time runs out, and so on. That way all simulations shut of on their own. If we make it so each successive simulations time runs out before the host simulations time runs out, you might think that the following simulations must also terminate, but how do you do know they do? You lost your window to look through, so I guess there is no way to prove it. However, if each simulation is given enough time to allow each subsequent simulation enough time to run out before the host simulation by modifying the rate at which time moves forward, you could have simulations running infinitely (say a plank time receives an arbitrary multiplier each simulation). If this is the case, then time would still have to run out on the most-host simulation or actual reality (however you choose to look at it) eventually, but if presented with the situation I just mentioned, how then could the simulations ever quit running, they would be able to keep going on forever even though the host simulation must run out of time. I don't know if I included enough information in the original post, but given the information I just gave you now, does it make more sense? There will always be running simulations, yet the time will still run out.

So I think the question you're asking is...

Is it possible to have time run (for instance) twice as fast for each next level of simulation, indefinitely, and thus have simulations which last forever, despite some host simulations only being run for a finite time?

In that case I think the answer would be no, for a couple of reasons. Note that if 1 second passes in the topmost simulation, 2^n seconds pass in simulation n over that same time period.

So with respects to purely physical concerns, you can't nest simulations indefinitely in the first place. Your host simulation can only change a finite number of bits per second. So you can't keep making simulations inside simulations indefinitely, because eventually you'd reach a simulation where even changing bits at a rate of one per second inside that simulation would require too many bits to be changed per second in the topmost simulation than is physically possible. There's also finite memory, and as each simulation contains some information essential and unique to that level of simulation (namely whatever object is enacting the next simulation down), and probably much more information besides (the incidental world that the object is in and the simulated beings who built it), it follows that you'd eventually run out of memory. At some point there would be a simulation in which it was impossible to build the next simulator.

The other reason is purely analytical. Even if it was possible to infinitely nest simulations running at double speed... there is never a simulation which lasts forever. If the topmost simulation runs for T seconds, the nth simulation (and every simulation is an nth simulation for some n) runs for T*2^n seconds before terminating. This may be an incredibly long time, but it's still finite.

I got a question. What if time travel going backwards proved to actually be possible? Since time travel would be possible in the real world, and the simulation is a perfect copy of the real world, then the people in the simulation could time travel backwards as well. Assuming the simulation was actually perfect, would the people in the simulation actually go back in time, or would the computer just be forced to simulate such a thing?

My first thought is that it would break the program, but what if it really was like perfect simulation.

Nice, present the paradox with a possible paradox of your own, I like the way you think. Well, I would think that if the simulations can be run forward, they can be run backwards all the same. It just depends which way you run the tape. If something is recording, then you can take that recording and play it backwards, just like cassette tape or reverse something digitally. Therefore, you can go backwards if you can go forwards. Now the real question is, can you go side to side? Can you create other dimensions and explore these dimensions, or are they merely able to be observed through their interaction with the objects moving forward and backward? We already know that light isn't all we need to see. You can detect vibrations, that's sound. You can feel the heat given off by the interactions between things that produce energy, the same goes for sensing pressure (which of course is inexorably linked to heat).

My original idea about time travel was that everything happens in cycles, so presumably you could go so far forward in time that you witness at least one of the cycles of what could have happened in the past. However, what if you could take the probability of whether it was how it happened the first time out of the picture altogether? Quantum physics suggests to me that after the big bang, everything existing does so as a result of a lot of miniature bangs. Each time a moment of time occurs, things explode into existence from a state before existence. You could call it inexistence, or you could call it a super state in which all matter and therefore energy retracts into itself and then spits back outwards until it reaches a limit of potential that is achieved solely because the existence of other bits of energy and matter. The interaction between the opposing particles would thus allow for an existence of reality in a wave like pattern much like we can observe with light in the double-slit experiment. Until the next explosion happens, the next "calculation" if you will, you won't know what happens next, until you can observe particles, or single points of energy or matter. The further you try and investigate exactly where a singular piece of energy/matter/whatever is, the less likely you are to know where it will wind up, and vice versa. The uncertainty principle. The mind is like a computer with a monitor attached to it, if you will, that draws in all the information together at regular intervals and pieces the otherwise random interaction of forces, particles, energy, and matter, and strings it together into something meaningful, what we know as the flow of time. We already know that the mind waits until it gathers enough information before forming a final picture to make up for the time lag between say, feeling something touch our nose and something touch our foot, or for us to see something, decipher it, and hear something, and decipher it. The distance between neurons and the brain and the speed of electricity are factors in this that must be taken into account. It then displays this information quickly enough at regular intervals that it appears simultaneous. Much like how a CPU of a computer works, it can only ever perform a single calculation at a time, it just does so at a rate that is imperceptible to us and thus it seems to be processing multiple things at once. Therein lies the beauty of the brain: multiple processors with 3 dimensional transistors, cells/neurons. It allows for immense calculation speed. What happens in between those segments of explosive existence is as far as we are able to comprehend, incomprehensible. It may be an eternity as we would know it before things explode back into existence in what we call seconds, but it would be totally unbeknownst to us.

The question then is, if we can piece it together forward, can we piece it together backward, or does it even matter? Would we know the difference? We already see things upside down, technically, and the brain flips it right side up to make sense of it. If you give humans goggles that flip the image back upside down again, eventually the brain flips the image again to make sense of it. Could it even change the course of events if we do play it backwards? Would it cause unpredictable results like the formation of new multiverses or dimensions or black holes if we ripped spacetime apart like that? Perhaps that is what stars do. All very interesting stuff.

On a side note, I have always wondered if it was possible to escape the universe, if we were just in a simulation. Like hack the software of the universe from the inside and then hack into the internet and escape into the real world.

Originally Posted by Alric

I got a question. What if time travel going backwards proved to actually be possible? Since time travel would be possible in the real world, and the simulation is a perfect copy of the real world, then the people in the simulation could time travel backwards as well. Assuming the simulation was actually perfect, would the people in the simulation actually go back in time, or would the computer just be forced to simulate such a thing?

My first thought is that it would break the program, but what if it really was like perfect simulation.

Very interesting. The simulation would be running in it's own time as well, maybe faster, but in some way relative to ours.
So I guess, if the simulation does what the real world would do, that person would go back in time for real as well. That would also open a time paradox in the real world then, except the simulation is completely isolated, meaning not even observed.
That is if the whole simulation runs into every mind the same way. You could simulate something different into one mind, but that wouldn't equal your supposed real life time travel, wouldn't depict it correctly then.

Originally Posted by Alric

On a side note, I have always wondered if it was possible to escape the universe, if we were just in a simulation. Like hack the software of the universe from the inside and then hack into the internet and escape into the real world.

You needed to take over a matrix which is complex enough to receive your mind, though. How to get at that should be a problem, esp. if artificial life and intelligence weren't invented yet. Body snatching is out of the equation, I would guess, because it would transform "you" into another person just by the pre-existing hardware.


Sorry, Snoop - bit too lazy to read through your posts at the moment...

One thing that pops into my head as I read all this is something I have heard somewhere. I can't quite recall where. It said that, in order to create a simulation that is a copy of the universe, one would need a computer the size of the universe. If this is true, then your original paradox would require the subsequent VR's (virtual realities) to be a little smaller each time. Which would mean it cannot be infinite.

Secondly, if you speed up time in a virtual reality, what kind of consequences would that have? Do the people inside the VR realize any of it? Or does their perception of time remain the same? Can you speed up time in the first place? Or can you slow down your perception of time? I think if you can, there will probably a limit to it.

It's a fun thing to think about

Originally Posted by MrPriority

One thing that pops into my head as I read all this is something I have heard somewhere. I can't quite recall where. It said that, in order to create a simulation that is a copy of the universe, one would need a computer the size of the universe. If this is true, then your original paradox would require the subsequent VR's (virtual realities) to be a little smaller each time. Which would mean it cannot be infinite.

No, not infinitely regressable, I agree. Complexity in terms of possible matter-configurations decreases, the smaller you get, and such physical computing matrixes will be impossible to exist from a certain level down, latest when you reach subatomic particles.

But you need not start with a computer the size of our universe, if you're not actually simulating a whole universe, but only what a universe following our natural laws would be depicted as in the mind of the beings in it.
The rest need not be simulated, if it's say just humanity alone, like we feel we are anyway, you need not simulate all the aliens, who I am convinced are out there, at least not until the simulation runs towards contact.

Maybe only simulate in the direction towards which conscious attention and/or measurement devices are turned, just reaching the respective theoretical depth of registration. That would then be like some people believe reality is anyway - not existing until captured.
The borders of reality would be defined by what our collective consciousnesses have reached so far in such a scenario.

Secondly, if you speed up time in a virtual reality, what kind of consequences would that have? Do the people inside the VR realize any of it? Or does their perception of time remain the same? Can you speed up time in the first place? Or can you slow down your perception of time? I think if you can, there will probably a limit to it.

It's a fun thing to think about

I guess there's always going to be an upper limit to thought-speed depending on the physical matrix used, and maybe there's also a non-material barrier in terms of how fast you could run a mind.

Perception is illusory anyway, since we can never depict reality one on one, and hallucinations in simulation would probably have a time-tacting corresponding to thought-speed, in order to become available for conscious access. It could be theoretically possible to run "perception" faster in a simulation, since no information from the senses has to be taken into timely account, and what isn't "meant" to get into your awareness anyway, need not appear at all.

Time-dilation or speeding-up effects while dreaming haven't been demonstrable as far as I know, though.
So I guess thought-speed is bound by hardware to a tight margin and hallucinations will correspond.

The criteria to judge the creators of the simulation or it's occupants as more real than the other is arbitrary. I don't consider this a paradox but a fractal, as above so below. They are also potentially in a simulation, and so forth. In our simulation, we inherently contain the potential to make another simulation, and so does that simulation.

After thinking about it some more, I don't think it is possible to create more than a few VR's that run faster. Possibly not even one. The main problem I see is that in the universe everything holds together in a very specific way. Chaning only one aspect of that would result in a completely different universe, most likely without any life in it.

For instance, if you somehow could make a perfect simulation of reality (ignoring for a second the fact that you would have to leave out a huge portion of the universe and really only center around the earth and the solar system). And say you were to change time within that simulation, that would leave you with a lot of other things changing with them. For instance space. Since space and time really are one thing, you would have to slow down the entire space in the universe. Which would mean everything would travel slower and time would go faster. Though this would have great concequences on everything else as well, for instance gravity would also be lighter. Which in turn means you would either have completely different life forms on earth or even none at all. Or you would need a larger earth or something like that. Though that is just one of the many complications you would run into. You would have to alter how the fundamental forces work in the first place.

Even when just changing perception, you would need to speed up electrical signals in the mind eventually. Even though you have some room to work with, eventually you will break the speed of light, which is of course impossible since it then would no longer be a complete copy of the universe and it's laws.

Even if you ignore those things above, eventually computers are bound by this universal speed limit as well. Unless of course you go into quantum computing and quantum entanglement and stuff like that. Though that discussion is probably one deserving it's own thread (also I really don't know enough about it:p)

It's an interesting thought, especially since I study Software Engineering and simulations are one of the things I'm interested in very much (also from a LD point of view of course). Though I have trouble seeing how the Physics would work. Then again, I might be totally wrong, since that is not my field in the first place.

Originally Posted by MrPriority

After thinking about it some more, I don't think it is possible to create more than a few VR's that run faster. Possibly not even one. The main problem I see is that in the universe everything holds together in a very specific way. Chaning only one aspect of that would result in a completely different universe, most likely without any life in it.

For instance, if you somehow could make a perfect simulation of reality (ignoring for a second the fact that you would have to leave out a huge portion of the universe and really only center around the earth and the solar system). And say you were to change time within that simulation, that would leave you with a lot of other things changing with them. For instance space. Since space and time really are one thing, you would have to slow down the entire space in the universe. Which would mean everything would travel slower and time would go faster. Though this would have great concequences on everything else as well, for instance gravity would also be lighter. Which in turn means you would either have completely different life forms on earth or even none at all. Or you would need a larger earth or something like that. Though that is just one of the many complications you would run into. You would have to alter how the fundamental forces work in the first place.

Even when just changing perception, you would need to speed up electrical signals in the mind eventually. Even though you have some room to work with, eventually you will break the speed of light, which is of course impossible since it then would no longer be a complete copy of the universe and it's laws.

Even if you ignore those things above, eventually computers are bound by this universal speed limit as well. Unless of course you go into quantum computing and quantum entanglement and stuff like that. Though that discussion is probably one deserving it's own thread (also I really don't know enough about it:p)

It's an interesting thought, especially since I study Software Engineering and simulations are one of the things I'm interested in very much (also from a LD point of view of course). Though I have trouble seeing how the Physics would work. Then again, I might be totally wrong, since that is not my field in the first place.

I agree, I think time is merely a label. What happens is what happens, and the perception of it is what changes. Things translocate and interact with one another, and we perceive it. What our perception tells us about how quickly they are translocating and interacting is what is changing, not the actual events taking place. Due to the nature of how things interact, certain pockets of matter, energy, and particles can cause each other to accelerate and decelerate on a non-linear scale, and that is what appears to us as a "speeding up" or "slowing down" of time. It all goes back to things being relative.