6 ms·

I wonder if we’ll get to a computational space where we can completely simulate every atom in a small closed system. Are we there yet? Can you simulate at the a

I wonder if we’ll get to a computational space where we can completely simulate every atom in a small closed system. Are we there yet? Can you simulate at the atomic scale and get a perfect representation of reality or do we have to try and simulate something smaller?

simonh 3y ago
It's a bit of a stretch perfectly simulating one proton, I'm afraid. https://www.quantamagazine.org/inside-the-proton-the-most-complicated-thing-imaginable-20221019/ https://www.quantamagazine.org/inside-the-proton-the-most-co...
- huytersd 3y ago
  Oh wow. Maybe that can be a new scale measure of technological capability based on how much of reality we can perfectly simulate.
  - simonh 3y ago
    If you think about it, a computer's not going to be able to perfectly simulate it's own physical operation, so the limit of a system's ability to simulate things is it's own complexity.
    - todd8 3y ago
      This isn’t obvious to me. It doesn’t have to simulate itself in real time. It may not be able to simulate itself simulating itself. There may be proof of this using a diagonalization proof. Consider programs that are quines, programs are able to output the exact source code of the program. See [1]. And there are abstract computers that can produce any computable output. These are called Universal Turing Machines, see [2]. UTMs can be specified with a remarkably small number of internal states, see [3]. I’m not saying you are wrong, but computability is full of unintuitive results, and the answer may be more subtle than what is revealed by “just thinking about it”. [1] https://en.m.wikipedia.org/wiki/Quine_(computing) https://en.m.wikipedia.org/wiki/Quine_(computing) [2] https://en.m.wikipedia.org/wiki/Universal_Turing_machine https://en.m.wikipedia.org/wiki/Universal_Turing_machine [3] https://en.m.wikipedia.org/wiki/Wolfram%27s_2-state_3-symbol_Turing_machine https://en.m.wikipedia.org/wiki/Wolfram%27s_2-state_3-symbol...
      - simonh 3y ago
        It needs to be able to store a complete representation of different versions of it's own state, in it's own state.
adastra22 3y ago
We can get up to 100 or so lightweight atoms (the real limiter is the number of electrons, not the number of atoms. Larger atoms cost more). Maybe a thousand or so on supercomputers. Quantum computing will change this though.
- huytersd 3y ago
  Is it an absolutely perfect simulation? Like would we know the exact future of those hundreds of atoms in perpetuity?
  - adastra22 3y ago
    No, for multiple reasons. To your first question, there's still simplifying assumptions. These things usually treat the nucleus as fixed and static, for example, and often other approximating assumptions are used. Outside of the interior of a star, radioactive elements, or when relativistic energies are involved, these are good approximations though. More fundamentally, quantum systems are inherently random. You literally can't know the exact future of hundreds of atoms in perpetuity--that's not how quantum physics works. But you can get an idea of minimum energy states, the energy barriers holding those states in place, and therefore what the stable outcomes might be.
  - mjfisher 3y ago
    The other reason the answer to this is "no" is that we don't even really understand how an atom even works its entirety yet. We don't know fundamentally know what makes the potential energy of an atomic nuclear be structured in the way it is. EDIT: I thought I'd double check that since it's been a while since I studied physics formally; apparently it is possible to model light nuclei[1] (up to four nucleons) from the ground up now! [1] https://en.m.wikipedia.org/wiki/Ab_initio_methods_(nuclear_physics) https://en.m.wikipedia.org/wiki/Ab_initio_methods_(nuclear_p... [1]
    - adastra22 3y ago
      You are confusing exact solutions with the ability to solve. We know how atoms work and can solve for any property we might want to measure. Ab initio systems aren't very practical for large systems because of computational costs, but they would work if given infinite compute (or a quantum computer). We just solve them numerically instead of closed-form. It's like polynomials: there's a quadratic formula, and a cubic and even quartic formula for exact solutions (although the quartic is ridiculous and no one would ever use it). Higher order polynomials have NO exact, closed-form solutions using standard arithmetic operations. But we can still solve any polynomial no matter the degree using numerical methods. To say that we don't even know how how degree 5 polynomials work would be flat out wrong.
justinator 3y ago
There's a chance - however small - that WE live in that simulation.
- digging 3y ago
  More likely that if it's even possible, the chance is near 100%. However, it actually doesn't matter in any way whatsoever.
  - dr_dshiv 3y ago
    Except that we are living in a simulation, then it is morally wrong to be boring
    - lamontcg 3y ago
      wat
      - dr_dshiv 3y ago
        If we are very dull, then we might get switched off
        lamontcg 3y ago
        you have no idea what the purpose of the simulation is. turns out the purpose was for them to study how the borg collective spreads across galaxies, and when we wipe out the borg we've ruined their experiment and they just pull the plug, because none of it was about us. oops.
        dr_dshiv 3y ago
        The purpose of the simulation is always to learn something new.
- WalterBright 3y ago
  The simulation theory is indistinguishable from any religious explanation.
tambourine_man 3y ago
The hack is to find emergent behavior so that you don’t need to simulate every atom. That’s what we did with classic thermodynamics. Extremely simple rules that accurately describe crazy complex systems.
shagie 3y ago
From a few years ago - https://www.nytimes.com/interactive/2021/12/01/science/coronavirus-aerosol-simulation.html?unlocked_article_code=1.VU0.DZmm.wdH28KZ7CNF8&smid=url-share https://www.nytimes.com/interactive/2021/12/01/science/coron... > To better understand the coronavirus’s journey from one person to another, a team of 50 scientists has for the first time created an atomic simulation of the coronavirus nestled in a tiny airborne drop of water. > To create the model, the researchers needed one of the world’s biggest supercomputers to assemble 1.3 billion atoms and track all their movements down to less than a millionth of a second. This computational tour de force is offering an unprecedented glimpse at how the virus survives in the open air as it spreads to a new host. > ... > Once the virus was loaded into an aerosol, the scientists faced the biggest challenge of the project: bringing the drop to life. Dr. Amaro and her colleagues calculated the forces at work across the entire aerosol, taking into account the collisions between atoms as well as the electric field created by their charges. They determined where each atom would be four millionths of a billionth of a second later. > To carry out this vast set of calculations, the researchers had to take over the Summit Supercomputer at the Oak Ridge National Laboratory in Tennessee, the second most powerful supercomputer in the world. Because the machine was in high demand, they could run their simulation only a few times. “We only have so many shots to actually see if we can get this thing to actually fly,” Dr. Amaro said. > The first run was a disaster. Tiny flaws in their model caused the virtual atoms to crash into one another, and the aerosol instantly blew apart. “It basically explodes,” Dr. Amaro said. > After half a dozen rounds of adjustments, the aerosol became stable. The researchers ran the calculations all over again to see what happened inside the aerosol an instant later. All told, they created millions of frames of a movie that captured the aerosol’s activity for ten billionths of a second. ----- Aside: the "four millionths of a billionth of a second" is 4fs or 4x10^-15 seconds - https://www.wolframalpha.com/input?i=four+millionths+of+a+billionth+of+a+second https://www.wolframalpha.com/input?i=four+millionths+of+a+bi... > 0.4 x time for a typical water molecule vibration The corresponding press release: https://www.olcf.ornl.gov/2021/11/16/we-know-covidisairbornenow-we-have-the-first-ever-model-of-an-aerosolized-viral-particle/ https://www.olcf.ornl.gov/2021/11/16/we-know-covidisairborne... The abstract presented - https://sc21.supercomputing.org/proceedings/tech_paper/tech_paper_pages/gbv104.html https://sc21.supercomputing.org/proceedings/tech_paper/tech_... Preprint of paper that matches the abstract mentioned in the ornl press release - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8609898/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8609898/

simonh 3y ago

It's a bit of a stretch perfectly simulating one proton, I'm afraid. https://www.quantamagazine.org/inside-the-proton-the-most-complicated-thing-imaginable-20221019/ https://www.quantamagazine.org/inside-the-proton-the-most-co...

huytersd 3y ago
Oh wow. Maybe that can be a new scale measure of technological capability based on how much of reality we can perfectly simulate.
- simonh 3y ago
  If you think about it, a computer's not going to be able to perfectly simulate it's own physical operation, so the limit of a system's ability to simulate things is it's own complexity.
  - todd8 3y ago
    This isn’t obvious to me. It doesn’t have to simulate itself in real time. It may not be able to simulate itself simulating itself. There may be proof of this using a diagonalization proof. Consider programs that are quines, programs are able to output the exact source code of the program. See [1]. And there are abstract computers that can produce any computable output. These are called Universal Turing Machines, see [2]. UTMs can be specified with a remarkably small number of internal states, see [3]. I’m not saying you are wrong, but computability is full of unintuitive results, and the answer may be more subtle than what is revealed by “just thinking about it”. [1] https://en.m.wikipedia.org/wiki/Quine_(computing) https://en.m.wikipedia.org/wiki/Quine_(computing) [2] https://en.m.wikipedia.org/wiki/Universal_Turing_machine https://en.m.wikipedia.org/wiki/Universal_Turing_machine [3] https://en.m.wikipedia.org/wiki/Wolfram%27s_2-state_3-symbol_Turing_machine https://en.m.wikipedia.org/wiki/Wolfram%27s_2-state_3-symbol...
    - simonh 3y ago
      It needs to be able to store a complete representation of different versions of it's own state, in it's own state.

adastra22 3y ago

We can get up to 100 or so lightweight atoms (the real limiter is the number of electrons, not the number of atoms. Larger atoms cost more). Maybe a thousand or so on supercomputers. Quantum computing will change this though.

huytersd 3y ago
Is it an absolutely perfect simulation? Like would we know the exact future of those hundreds of atoms in perpetuity?
- adastra22 3y ago
  No, for multiple reasons. To your first question, there's still simplifying assumptions. These things usually treat the nucleus as fixed and static, for example, and often other approximating assumptions are used. Outside of the interior of a star, radioactive elements, or when relativistic energies are involved, these are good approximations though. More fundamentally, quantum systems are inherently random. You literally can't know the exact future of hundreds of atoms in perpetuity--that's not how quantum physics works. But you can get an idea of minimum energy states, the energy barriers holding those states in place, and therefore what the stable outcomes might be.
- mjfisher 3y ago
  The other reason the answer to this is "no" is that we don't even really understand how an atom even works its entirety yet. We don't know fundamentally know what makes the potential energy of an atomic nuclear be structured in the way it is. EDIT: I thought I'd double check that since it's been a while since I studied physics formally; apparently it is possible to model light nuclei[1] (up to four nucleons) from the ground up now! [1] https://en.m.wikipedia.org/wiki/Ab_initio_methods_(nuclear_physics) https://en.m.wikipedia.org/wiki/Ab_initio_methods_(nuclear_p... [1]
  - adastra22 3y ago
    You are confusing exact solutions with the ability to solve. We know how atoms work and can solve for any property we might want to measure. Ab initio systems aren't very practical for large systems because of computational costs, but they would work if given infinite compute (or a quantum computer). We just solve them numerically instead of closed-form. It's like polynomials: there's a quadratic formula, and a cubic and even quartic formula for exact solutions (although the quartic is ridiculous and no one would ever use it). Higher order polynomials have NO exact, closed-form solutions using standard arithmetic operations. But we can still solve any polynomial no matter the degree using numerical methods. To say that we don't even know how how degree 5 polynomials work would be flat out wrong.

justinator 3y ago

There's a chance - however small - that WE live in that simulation.

digging 3y ago
More likely that if it's even possible, the chance is near 100%. However, it actually doesn't matter in any way whatsoever.
- dr_dshiv 3y ago
  Except that we are living in a simulation, then it is morally wrong to be boring
  - lamontcg 3y ago
    wat
    - dr_dshiv 3y ago
      If we are very dull, then we might get switched off
      - lamontcg 3y ago
        you have no idea what the purpose of the simulation is. turns out the purpose was for them to study how the borg collective spreads across galaxies, and when we wipe out the borg we've ruined their experiment and they just pull the plug, because none of it was about us. oops.
        dr_dshiv 3y ago
        The purpose of the simulation is always to learn something new.
WalterBright 3y ago
The simulation theory is indistinguishable from any religious explanation.

tambourine_man 3y ago

The hack is to find emergent behavior so that you don’t need to simulate every atom. That’s what we did with classic thermodynamics. Extremely simple rules that accurately describe crazy complex systems.

shagie 3y ago

From a few years ago - https://www.nytimes.com/interactive/2021/12/01/science/coronavirus-aerosol-simulation.html?unlocked_article_code=1.VU0.DZmm.wdH28KZ7CNF8&smid=url-share https://www.nytimes.com/interactive/2021/12/01/science/coron... > To better understand the coronavirus’s journey from one person to another, a team of 50 scientists has for the first time created an atomic simulation of the coronavirus nestled in a tiny airborne drop of water. > To create the model, the researchers needed one of the world’s biggest supercomputers to assemble 1.3 billion atoms and track all their movements down to less than a millionth of a second. This computational tour de force is offering an unprecedented glimpse at how the virus survives in the open air as it spreads to a new host. > ... > Once the virus was loaded into an aerosol, the scientists faced the biggest challenge of the project: bringing the drop to life. Dr. Amaro and her colleagues calculated the forces at work across the entire aerosol, taking into account the collisions between atoms as well as the electric field created by their charges. They determined where each atom would be four millionths of a billionth of a second later. > To carry out this vast set of calculations, the researchers had to take over the Summit Supercomputer at the Oak Ridge National Laboratory in Tennessee, the second most powerful supercomputer in the world. Because the machine was in high demand, they could run their simulation only a few times. “We only have so many shots to actually see if we can get this thing to actually fly,” Dr. Amaro said. > The first run was a disaster. Tiny flaws in their model caused the virtual atoms to crash into one another, and the aerosol instantly blew apart. “It basically explodes,” Dr. Amaro said. > After half a dozen rounds of adjustments, the aerosol became stable. The researchers ran the calculations all over again to see what happened inside the aerosol an instant later. All told, they created millions of frames of a movie that captured the aerosol’s activity for ten billionths of a second. ----- Aside: the "four millionths of a billionth of a second" is 4fs or 4x10^-15 seconds - https://www.wolframalpha.com/input?i=four+millionths+of+a+billionth+of+a+second https://www.wolframalpha.com/input?i=four+millionths+of+a+bi... > 0.4 x time for a typical water molecule vibration The corresponding press release: https://www.olcf.ornl.gov/2021/11/16/we-know-covidisairbornenow-we-have-the-first-ever-model-of-an-aerosolized-viral-particle/ https://www.olcf.ornl.gov/2021/11/16/we-know-covidisairborne... The abstract presented - https://sc21.supercomputing.org/proceedings/tech_paper/tech_paper_pages/gbv104.html https://sc21.supercomputing.org/proceedings/tech_paper/tech_... Preprint of paper that matches the abstract mentioned in the ornl press release - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8609898/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8609898/