Evolution!! That’s what I wrote!! haha … so either we both get a pass or none of us gets one lol 
5 Likes
From a physics/science point of view, this video carries little relevance at best and is sometimes completely non-sensical (for example when talking about “trivial” and “non-trivial” eigenvalues… I have worked as a quantum physicist for many years and this is the first time I hear of something as a “binary matrix”… sounds impressive, but no).
Higher dimensions in physics are often used as mathematical tools to make sense out of our 3D+time world. For example, Einstein used time as a fourth “dimension” with a spatial extension only as a mathematical construct. Another example is string theory, using various dimensions, 11 being one of the more popular one… anyway, we should be careful when talking about these things because making it too “populistic” as in that video can be dangerous if/when people take it at face value.
Going back to what we were discussing “Can you run a simulation of a universe more complex than your own?”
Since software cannot be more complex than the hardware it is stored on, and since any hardware we built would be limited by the atoms in our universe, no, it would not be possible to run a program more complex than our own universe.
Unless we could build a processor made of matter from a higher plane of existence, it cannot be done.
1 Like
This question took me a little while to figure out how I was going to answer it. It’s fun to think that it’s possible to create a more complex universe, but I ended up with a similar answer to yours.
It kinda depends on how you define “more complex”.
We can already do simulation of basic physics. If we were to… say… add another 20 different elements to our simulated physics, it would still be a viable simulation, and be arguably more complex than our own universe.
If we take more complex to mean “have more rules”, it would be possible. If we were to say “has more simulated atoms”, again it might be possible, but it would probably run at slower speed.
Think of it like a modern computer and a 10 year old one. Both can run modern software, but the 10 year old one needs a lot more time. So without a constraint like time limiting our simulation, I do believe we could simulate things far more complex than anything “real” around us.
1 Like
There are a host of accepted definitions of “complexity” here, different definitions because of different fields of science.
This assumption is actually wrong. You can very well run software more complex than your hardware. The complexity of the hardware determines how well it is suited for certain tasks and its overall performance, but as soon as you have a turing-complete processor, the complexity of the program running on it is arbitrary.
Programming languages have been emulating a 64-bit datatype for decades now. The Hardware of the Gameboy Advance doesn’t support floating point operations at all, but you can still emulate them with software. It’s just damn slow to do something with software which the hardware has never been designed for (that’s why we have dedicated GPU’s nowadays), but it’s certainly not impossible.
Now, that “slowness” might or might not be an issue, depending on your simulation. In a matrix style “brain in the jar” simulation, it’s problematic, because the consciousness observing the simulation is not actually part of it. If we’re talking a simulated consciousness observing the simulation, however, this is not an issue, since it is part of the simulation itself. Even if there’s milenia passing between timesteps, it won’t notice, because the only reference it has for the passage of time is the simulation itself.
As far as I know, somebody built a turing-complete 8-bit processor in minecraft.
You could go on and write a program for that thing that is more complex than minecraft itself. It would be an extremely arduous task, and it would run at a completely glacial pace when observed from our reality, but there is no question on whether it would be possible in theory.
2 Likes
Morpheus to Neo “If real is what you can feel, smell, taste and see, then ‘real’ is simply electrical signals interpreted by your brain.”
2 Likes
This question has Roko’s Basilisk written all over it. Though hopefully it’s just a philosophical question, and not some malevolent Atlas AI in the future trying to suss out if we’re on its side or not.
1 Like
Exactly. They currently operate under the philosophy “who isn’t against us is with us”, but that might change in the future. Not neccessarily an outright Roko’s Basilisk, since they haven’t started punishing me yet although I answered that one with no, though. The really terrifying aspect of Roko’s Basilisk is that your thoughts could have consequences for you right now.
But if Atlas ever turns into No Man’s Skynet, I might get into trouble. I hope you guys will come looking for me…? 
1 Like
Thank you for the well thought-out reply. You’ve given me a lot to think about.
I was basing my assumption on No Man’s Sky, a simulated universe that is currently far, far less complex than our own. It does excite me, though, that the survey is asking questions like this because it could mean Hello Games is working on making NMS’s universe more complex.
1 Like
Yeah well… procedural generation is another kind of animal altogether again. Simulation and ProcGen are two intertwined, but still distinct, concepts. They can complement each other very well, as is demonstrated by constrained systems like Dwarf Fortress, but unconstrained procedural systems like NMS are inherently stateless. Simulation can pick up the thread and induce complex behavior on a locally observed level, but to have that simulation affect anything beyond the immediately observable (i.e. what’s currently in memory), is really, really difficult.
NMS currently doesn’t even employ many simulation techniques on the local level though, there could definitely be some improvement.
Huh… I think I should start getting used to this boards handling of replies vs. quotes, so my answers are properly nested in the future…
sorry if this has been answered, but i havent recieved a survey…ive been registered for a while with WT…Did I miss a step? please help?
some (on gmail) reported needing to check automated folders, like “promotions” and the like… don’t know other mail clients
found it! thanks so much for the help bcatrek
1 Like
To create a simulated universe more complex than the universe the simulation was created in doesn’t really make sense… doesn’t the very fact that the universe is capable of producing something more complex than itself put in on a higher level, in a certain manner? And, yes, I am aware of the paradox there.
2 Likes
I think you are totally right, there, from an inclusive perspective… Now I would maybe like to change my answer! Still the higher complexity could be localized into an “inside”, but the fact remains that it had to be possible within the confines of the previous frame… Brilliant.
My understanding (and that was the answer i gave on the survey) is that a simulation can be more complex than the reality because in the simulation, you can code phisical limitations out of the equation. You can create multiple instances of complex interactions, and you can program variables to interact in impossible ways that you couldnt in real life.
1 Like
That is the kind of thinking I had too… But remains the fact that the complexity created within the simulation is still a part of the “outside of the box” that it had created it… or made it possible. So a simulation can multiply the complexity of the “universe”, but it’s not detached from it.