top | item 17410570

The Universe Is Not a Simulation, but We Can Now Simulate It

54 points| he0001 | 7 years ago |quantamagazine.org | reply

105 comments

order
[+] comboy|7 years ago|reply
I'm so disappointed by quanta lately. Clickbaits and worsening quality.

> The Universe Is Not a Simulation

That's based on what? Author of the article opinion? That's cool but then it should be "In my opinion.." and "here's my reasoning".

[+] contravariant|7 years ago|reply
Arguably it would've been better to say "The universe may not be a simulation" but calling this clickbait and evidence of a decline in quality because of such a minor detail seems an overreaction. Unless there's more you object to than the first 6 words of the article.
[+] nothis|7 years ago|reply
Is the "Universe is a Simulation" more than a dreamy internet theory, though? Like, are astronomers and physicists taking it seriously? I don't know either way, I just assumed the headline was basically saying "no, this isn't some sensational new post about how we're all in a simulation, this is hard data leading to realistic results". Kinda defensive more so than click-bait-y.
[+] saas_sam|7 years ago|reply
Thank you. I observe the same sentiment of unjustifiable certainty with so many deep issues among popular journalists and scientists. Simulation theory, the potential disaster(s) of AI, not to mention political topics. When even smart people are blind to weaknesses in their position, I despair, and go back to funny gifs.
[+] sbuttgereit|7 years ago|reply
OK, I'll bite.

If the Universe is a simulation (or might be) what is it simulating and who's simulating it? I'd be more interested to know about -that- than the arbitrary details of some measly little simulation itself. You might say something hand-wavy like, "it's simulating itself"... but OK, when I look at our "simulations" they are always highly referential to the external real... er... possibly real... world. Dwarf Fortress could be called a simulation (perhaps one of the better ones!) ... and while I'd be hard pressed to point to the outside world and say it's simulating any one thing with high fidelity, I can point to trees, mountains, water animals, etc. where the creative element is really just tweaking the attributes of these otherwise externally real things. It's hard to simulate what you're you're completely unaware of, unconscious of ... in fact to be conscious at all is to be conscious "of something" and insofar as simulations are conscious efforts at creation, it's difficult to not have some concrete reference that exists outside of a "simulation".

Of course, perhaps, it's the great mystery of God which I'm always told is beyond mere human understanding. Or maybe its simulations all the way down. Or maybe any explanation talking about simulations requires greater leaps of faith and greater acceptance of more and more unlikely scenarios of reality than just saying the Universe is perhaps a metaphysically real place that doesn't require an appeal to "simulation" for explanation.

I think I'll just stick to the simple minded idea of the Universe is a real place and not a simulation. That part of the title is so likely to be true that the author need not feel bad about the assumption.

[+] rbanffy|7 years ago|reply
Actually, when you found out it was a simulation, we scrapped the current data, we had to restore it to the last valid snapshot and adjust local simulation parameters so that you wouldn't find it out. It's all reported in the post-mortem of the incident. Service levels are now fully restored.
[+] earenndil|7 years ago|reply
In the beginning, the universe was simulated. This has made a lot of people very unhappy and is widely regarded as a bad move.
[+] stabbles|7 years ago|reply
There is another theory which states that this has already happened
[+] decebalus1|7 years ago|reply
Skimmed through the article and couldn't find any mention of the first part of the title which was the initial reason for reading it as it's a subject of great interest to me. I wasn't expecting this from quanta. Very disappointed.
[+] psetq|7 years ago|reply
It's still an extremely interesting article imo.

I sometimes read hn as the discussions here can add to articles on topics I find interesting but have limited knowledge on, but the 3+ top comment threads (so far) discussing the title and not content of this article is somewhat disappointing.

[+] gattr|7 years ago|reply
In Charles Stross's "Accelerando" (strongly recommended!) astronomers suspect that the inhabitants of Andromeda Galaxy converted all its baryonic matter to computronium (here: configuration of matter capable of the most efficient computation as per the laws of physics) and are using it to run a massive side-channel attack, trying to verify if the Universe runs on a virtual machine (and I guess maybe break out of it).

On a different note, here's a sobering thought for the Universe-as-simulation optimists: while it's possible the simulation's Creators wish us well, e.g. will store your soul/mind and let you live carefree in a Paradise... it might rather work like our current massive simulations (e.g. the Millennium Run [0]). It generates enormous amounts of data, but only some small amount (like a state snapshot every 100M years) is actually captured and analyzed. I.e. the "dumb" computing power at the Creators' disposal is far greater than their mental capacity. The whole existence of humanity might go unnoticed, just a side-effects of the simulation's fidelity.

[0] https://en.wikipedia.org/wiki/Millennium_Run

[+] ur-whale|7 years ago|reply
Is it just me or all that talk of "codes" combined with all-uppercase names is exuding a potent whiff of fortran :D ?

And if so, I'm wondering if any of that stuff runs on GPUs or if we're still talking about "traditional" supercomputers?

Anyone knowledgeable in the field care to share?

[+] freshhawk|7 years ago|reply
Not knowledgeable in the field really, but I have a bunch of old school friends who do this for a living.

Your ability to detect Fortran is apparently excellent.

[+] gojomo|7 years ago|reply
There is nothing in the article to refute the conjecture that our universe is a simulation – and in fact the rapid progress it reports in simulating simplified universes could be seen as further support for the idea that universes like ours may be simulated.
[+] NVRM|7 years ago|reply
Because you can chroot doesn't mean that you are not in a virtualized container. Actually this tend to say the opposite...
[+] rococode|7 years ago|reply
Could someone elaborate on how accurate the results of these simulations are generally considered to be by the broader scientific community? It seems to me that the universe is so insanely complex that any simulations we can currently make would be fairly inaccurate because of lack of knowledge. Couldn't that cause a problem of building a whole base of theory on unproven guesses? I guess the researchers all know that but I'm just curious what the actual attitude towards this kind of work is.
[+] bpicolo|7 years ago|reply
> The Universe Is Not a Simulation

Proof left as exercise for the reader?

[+] harshalizee|7 years ago|reply
Is there any of the code for the simulation or similar available in the public domain? Would love to be able to take a look.
[+] stillbourne|7 years ago|reply
Not down to the planck level. Simulating the universe at the planck level would require at least 10^157 bytes of storage.
[+] gattr|7 years ago|reply
Right, but adaptive refinement is bread and butter of today's physics simulations. I imagine that in order to simulate our Universe, you'd go down to elementary particle level only when some of the simulated folks run their accelerators, etc. But for an in-simulation billiard game you'd use a plain impulse-based rigid body dynamics.
[+] ctb9|7 years ago|reply
How does one come to that figure?
[+] Simon_says|7 years ago|reply
You're probably making some unfounded assumptions about the size of our universe to come up with that number.
[+] krrishd|7 years ago|reply
this is highly naive of me to ask, but "in theory" if we could simulate the universe, would familiar stuff like living organisms/'societies'/'humans' potentially just emerge within such simulations?

just super cool to think about even if not the case/unrealistic.

[+] AnIdiotOnTheNet|7 years ago|reply
Depends on how you define living, really. It's a philosophical problem, what makes one collection of matter "alive" and another "not alive"?

Since at least Thomas S. Ray's work on Tierra in the 90s we've been creating universes in which things that could be called "life" have developed.

[+] ada1981|7 years ago|reply
If we could simulate it past a certain fidelity threshold, it would seem so.