SoylentNews is people
SoylentNews
from the that's-what-we-want-you-to-think dept.
Physicists have "confirmed" that we "aren't" "living" in a computer "simulation":
Scientists have discovered that it's impossible to model the physics of our universe on even the biggest computer.
What that means is that we're probably not living in a computer simulation.
Theoretical physicists Zohar Ringel and Dmitry Kovrizhin from the University of Oxford and the Hebrew University in Israel applied Monte Carlo simulations (computations used to generate probabilities) to quantum objects moving through various dimensions and found that classical systems cannot create the mathematics necessary to describe quantum systems. They showed this by proving that classical physics can't erase the sign problem, a particular quirk of quantum Monte Carlo simulations of gravitational anomalies (like warped spacetime, except in this case the researchers used an analogue from condensed matter physics).
Therefore, according to Ringel and Kovrizhin, classical computers most certainly aren't controlling our universe.
Which type of computers are we being simulated on?
Also at Newsweek.
Quantized gravitational responses, the sign problem, and quantum complexity (open, DOI: 10.1126/sciadv.1701758) (DX)
(Score: 2) by TheRaven on Thursday October 05 2017, @09:50AM (2 children)
And the notion of 'classical computing' is related to the physical laws of our universe, so (without reading the paper) this result just tells us that either we're not simulated, we're simulated on a quantum computer, or we're simulated on a computer that has different laws of physics to ours. Within our universe, simulations always have at least some simplifications of physics to make the simulation more efficient. Quantum mechanics may just be the simulation artefacts from quantising various continuous functions.
This isn't the first time physics has been down this rabbit hole. When black holes were first discovered, it became clear that close to the singularity you'd have no information coming in from outside and the laws of physics could be radically different. Various people hypothesised that you might have potentially unbounded layers of nesting of universes, with each one appearing as a small bubble akin to a black hole in the others. Much the same logic applies here as to simulations: if simulations or universe nesting are possible, then there is one top-level universe and an infinite number of those (recursively) contained within it, so the probability of ours being the top-level universe is vanishingly small.
sudo mod me up
(Score: 0) by Anonymous Coward on Thursday October 05 2017, @02:00PM (1 child)
Yes, a classical computer in an extra-dimensional universe would have a CPU laid out in "flat" 3 dimensions, or heck say 8 dimensions. Who would think that exponentially complex things cannot be solved in linear time in that universe?
(Score: 3, Interesting) by hendrikboom on Thursday October 05 2017, @03:11PM
You'd get a polynimial speedup from parallelism at the most. For exponential, you'd need a universe with hyperbolic geometry and at least two dimensions.