Ever felt like you're living in a simulation?

Image courtesy of the Illustris simulation

No, neither have I, to be honest. But the possibility may not be as crazy as it sounds.

Not long ago, an international team of astrophysicists created a simulation of the universe that models a cube of space 350 million light-years across. That's enough to hold tens of thousands of galaxies. Not stars, but galaxies of stars.

Now this wasn't just a single point in time that was simulated. This Illustris simulation, as they called it, was actually 'started' at a time soon after the Big Bang occurred and then the simulation was run forward in time, allowing all the 'matter' to interact based on the mathematical rules of nature - as understood by the astrophysicists - all the way up until the present day.

Now that's a lot of time AND space that requires 'simulating', and so it's no surprise that they used a supercomputer to do it. Three months worth of super-computing, actually. Which is about the equivalent of 20,000 years of processing for your average desktop computer.

What resulted is a spectacularly beautiful, incredibly detailed, and surprisingly accurate slice of a 'universe' with all the same type of galaxies and super-clusters that we see in our own 'real' universe today.

It's quite an amazing thing on its own and I highly recommend watching the video in the link below.


But this is where my mind was blown a little off the reservation.

I realised that if we can do this (and we clearly can) then surely it's inevitable that we will therefore one day create a simulation so detailed and close to what we perceive as being our own reality, that it will be indistinguishable from our own.

Okay, I know that's a big call, but if you break it down to what's required to make this happen, it's not so far fetched. All we would really need is a good enough understanding of the fundamental laws of nature and an immense amount of computing processing power. That's it.

Not convinced? Well, first, let me say that the second proposition is only a matter of time. Computing capacity has been increasing exponentially for decades, and although we're reaching the limit of what silicon circuits can do, there are many new technologies currently being worked on that will get us past that sticking point. And that's not even considering the immense processing boost that quantum computing will provide, which is apparently close to being realised.

So let's assume the computing power needed to simulate even the smallest particles that would be required to make our 'realistic' simulation work will one day be available.

Which brings us to the fundamental mathematical laws of nature.

You might think that to create a realistic simulation where life could 'exist', we would need to include rules for chemistry, or biology, or even social behaviour. But the reality is (no pun intended), if you get the mathematical rules for matter and energy right at the most basic levels, the rest should just take care of itself.

That is, after all, what happens in our reality. There are no chemical or biological rules that operate independently of the fundamental laws of nature. They are simply a product of the complex interactions that are allowed within the bounds of those fundamental 'laws'.

So, if we get those fundamental laws right, matter should coalesce into atoms and then stars that eventually burn themselves out and go supernova, spreading those new atoms out into the universe where they coalesce into larger stars that create larger atoms, before going supernova and spreading the larger atoms out into the universe again, and so on and so on, until eventually, the right mix of atoms will end up in the right kind of solar system, where a planet will appear at just the right distance from its star, creating the same conditions that occurred on Earth when life spontaneously appeared.

Now, getting to this 'simulated' point would be a pretty awesome achievement in itself, but imagine what would happen if we pushed the simulation a few hundred million years into the future (and remember, this is easily done, thanks to our essentially limitless supply of computing power).

What if one of the more complex simulated species that has evolved on our simulated planet begins to learn a few tricks that make them almost appear sentient? Could we 'simulate' a new form of sentient/conscious life?

And what if (stay with me here) this new sentient species was pushed forward a few hundred thousand years again until they had their own technological revolution and possibly even surpassed our own capabilities?

Ever heard of the singularity? This is when computers will become smart enough to take over their own development and go in directions we literally cannot imagine (because we're not smart enough). Well just imagine what might happen if we pushed our 'simulated' species far beyond our own level of technological development. What would that look like? And would our simulations eventually realise they're in a simulation and perhaps even work out a way to communicate with us, their creators? Or perhaps even escape the very simulation that gave them life?

It's an intriguing possibility in so many ways, and the kind of bread-and-butter thinking that science fiction writers like myself thrive on.

It also makes me think that if you and I were living in a simulation, how would we ever know? If all of our 'interactions' with reality, on every level, are based on the fundamental rules of the universe we exist in, then how can we possibly test, let alone prove, that there might be something else beyond it?

Who knows? Not me, that's for sure.

But do you know the best thing about being human (in my humble opinion)? We may not ever be able to prove what this reality we're sharing really is, but we can imagine anything we want. Including possibilities that don't have to follow the fundamental rules that govern our universe. And that's pretty cool, if you ask me.

But that's just my two cents. What do you think?