Cosmic simulation reveals how black holes grow and evolve | Hacker News Reader

Cosmic simulation reveals how black holes grow and evolve

caltech.edu

60 pointsgmays49 comments

GIZMO Users Guide - http://www.tapir.caltech.edu/~phopkins/Site/GIZMO_files/gizm...

This looks like so much fun!

Does any one know if it's feasible to run small simulations on a single desktop or is it heavily geared towards clusters?

It will run on a single computer, it uses a run of the mill MPI strategy.

The question is more like "how much can I get done with my single desktop computer in a reasonable amount of time?"

From the Users Guide:

> The code is written in standard ANSI C, and should run on all parallel platforms that support MPI. The portability of the code has been confirmed on a large number of systems ranging from a laptop to >1 million threads/cores on national super-computers.

The vast majority of what is known about black holes comes from observations of AGN/quasars (super massive black holes that are "active"), which may be as luminous as ~0.5 quadrillion suns (1). This luminosity is powered by the quasar's central engine, which converts the gravitational potential energy of infalling matter to radiative energy, through means that are still debated. In an article in Nature commemorating the 50th anniversary of the discovery of quasars, a leading AGN researcher wrote "our best hope for understanding quasars is that extraterrestrials might drop in and explain them to us" (2). AGN/quasars are an underappreciated enigma.

(1) https://www.nature.com/articles/s41550-024-02195-x (2) https://www.nature.com/articles/495165a.pdf

Yes, but he said that 19 years ago. Much has been learned since then.

when an elderly researcher says something can be done, it probably can. when an elderly researcher says something can't be done, it may or may not be possible.

I wonder if this means the standard models, and limits, on black hole growth need to be revised. A lot has been made of the weirdly early, huge black holes that JWST has seen, so if the dynamics here help black holes grow faster, that would help explain those observations. But I can't tell if that's the case from either this press release or linked article. It could be worse for all I can tell. https://arxiv.org/pdf/2309.13115v2

Anton Petrov helping wonderful people learn about electromagnetism around black holes: https://www.youtube.com/watch?v=E6qN3ovTsE0&t

will be looking forward to the Dr. Becky explainer for this one, I have a feeling she is going to be excited

but do physicists know everything about the physics of black holes?

Probably not. It's hard to be certain about a classical object with a singularity (e.g. infinities) at the center, but, so far from gravitational wave observatories, we aren't seeing anything not predicted by classical bendy space GR.

It is hoped that if we can unify GR with quantum mechanics or string theory that the singularity will go away or look different. One of my favorite crazy ideas is fuzzballs [1].

1. https://en.wikipedia.org/wiki/Fuzzball_(string_theory)

More to the point, you don't need to know what's behind the horizon to do this kind of simulation. Plain GR is probably fine for everything outside macroscopic BH's. (But yes, fuzzballs are cool)

Kerr dropped a paper last winter that is interesting about the claims of Hawking etc..

https://arxiv.org/abs/2312.00841

> In conclusion, I have tried to show that whatever the Penrose and Hawking theorems prove has nothing to do with Physics breaking down and singularities appearing. Of course, it is impossible to prove that these cannot exist, but it is extremely unlikely and goes against known physics.

No. But simulations of this nature give potentially useful hints on where physicists should look next

If it helps, consider this a more detailed prediction of our current theories, ready for attempted falsification.

If by "reveals" you mean "predicts", then yes.

Given that the predictions involved are based on long-established, result-yielding large scale astrophysics modelling programs, that's a bit too dismissive. Yes, in the strict sense "scientists" "used computer modelling" and found "a modelling parameterisation" that "seemingly correctly predicts" a formerly poorly understood part of black hole accretion disk structure.

At the same time researchers at Caltech used a combination of well-established models that revealed how electromagnetism, previously not considered, explains observed accretion disk structure.

In the science sense, this is absolutely a "reveal": the combination of models yields data that is consistent with observed reality.

I don't think its dismissive, its just the more accurate word, this entire field is about letting beliefs die whenever anything counters the model no matter how many things were built on top of the model. let the faithful die with their beliefs, this crowd lets their beliefs die.

You could say it's revealing something about the physics used as input to the simulation.

my eyebrow raises from this with all of these cosmological simulations.

Just cosmological simulations?

What about FEM and CFD? Climate models? Economics?

To be honest, I don't even know what FEM and CFD are, and I'm not familiar with economic models. As for climate models, I raise both eye brows those

FEM: Finite element method

CFD: Computation fluid dynamics

Both common in industrial design for simulating the performance of a part-to-be.

Is your skepticism directed at all climate models or just the cutting-edge planet-scale politicized ones? I find our weekly forecasts quite useful, for example.

https://ourworldindata.org/weather-forecasts

Weather models are definitely eyebrow raising. On days where there 0% chances of rain as the forecast, I receive .25". On days where it is >80% chance of rain, 0.0" of rain falls. Then there are days where it tells me with the accuracy of minutes of when the rain will start/stop. When their accurate, it's amazing. When they're not, it's very frustrating.

Feels like this is the 1000time some article states something like this. Am I the only one who feels like we're not really progressing in those fields? Just a feeling though

It's the 1000th time that Caltech unifies two of their large scale astrophysics computational models to bridge the gap between them? That seems off by a few orders of magnitude, at the very least?

Not the commenter, but I appreciate their sentiment. If the headline was something like "Caltech unifies two of their astrophysics models to bridge the gap and increase understanding of cosmology" then I'd say they're being snarky for the sake of snark. But the headline as is... Yes we've seen that plenty from University PR lately and it doesn't grab my attention anymore

Not sure I understand the "lately" part there: this is not a sudden change, big name universities have used these kind of headlines for decades, the idea that you know what the article's about without reading the article never really applied in the first place?

General Relativity is over a century old and the Standard Model was in it's current form in the mid 70's. Much of the details have shaken out from those things and while there is certainly still progress being made both in refining current theories and developing options for future ones, physics has indeed slowed down quite a bit in overall progress which was kind of an explosion between say 1875 and 1975 (let's not quibble with exact dates here).

Particle physics has slowed down, but astrophysics has accelerated.

Just to put things in perspective, the Hubble Space Telescope was only launched in 1990. Digital cameras, vastly larger telescopes, space telescopes, computers capable of analyzing much larger quantities of data and running much larger simulations have transformed astrophysics over the last 35 years or so.

I say this every time anything black hole related hits HN: I want to be yeeted into a super massive black hole. I want to know what is there. I might die at the event horizon in an instant as it might be a cosmic/causal/quatum firewall due to all the light being trapped there.

Or maybe I’ll float for a few seconds or minutes or who knows how long until I hit the singularity.

Or maybe there’s no singularity and if I — by some miracle — can avoid the singularity because it doesn’t exist and instead is a ring-ularity maybe I can be ejected into a parallel universe or something.

I wouldn’t be able to tell a soul about what I learned. But I’d at least be able to say my favorite line from all books: “… it’s full of stars!”

P.S. the universe gets bonus points if i can emerge as a monolith.

I can simulate a pink elephant but that doesn't prove its existence. These simulations probably miss important factors and thus have results between dubious and plain wrong. The wrongness of the results compared to reality might still yield interesting hints, if we're lucky.

Eventually you hope that your yet-to-be-built pink elephant detector matches the result of your pink elephant simulator.

Science doesn't happen in a vacuum. (Unless you need vacuum conditions .)

I don't really understand what you are saying - are you arguing that simulations are never useful? Or that these researchers do not understand the limits of their simulation?

> The wrongness of the results compared to reality might still yield interesting hints, if we're lucky.

Not just luck. Intelligence and luck. There is a skill involved in creating useful simulations that give clues on how to (a) reveal inconsistencies in current theories; (b) generate ideas for future theories; (c) design future experiments; (d) inspire scientists, present and future; and lots more.

If you want to educate yourself on the "why," the branch of science is called computational astrophysics.

https://en.m.wikipedia.org/wiki/Computational_astrophysics

There is a theory that says blackholes form on a border of energy differences and are act as a compensation mechanism allowing balancing the energies out between adjacent areas with different properties. When energy 'flows' out of our visible space it appears as a blackhole. When it flows in - it appears as a star.

I probably don’t understand, but how is a star energy flowing into visible space? The matter is already present and visible in space. It isn’t going anywhere or coming from anywhere mysterious. It’s visible radiation from processes occurring in stars. Black holes seem to be more than light or energy dissipating, though.

What is flowing is basic matter out of which all is made. It breaks down to more basic components as it leaves one space and flows to the other the theory says. Once these basic components enter the other space they assemble into configurations compatible with energy of the other 'space'.

Doesn't sound like any theory of black holes I've ever heard. Reference please...

Don't we pretty much understand stars though? Big ball of gas, hydrogen fusion, etc. Are there any unaccounted energy outflows that would leave room for something like this?

The main reason we look at black holes so hard is because quantum mechanics can't describe gravity (yet) and general relativity just shrugs and says "I don't do inside event horizons, buddy, but why do you care? You can't even experiment inside one". So it's one of those head-scratchers where the math isn't really helping us figure out _what_ is going on, and the darn things are really hard to observe.

A light in the sky could be a star of the kind we understand, but could it be something else? Could we say we understand all stars if we don't understand black-holes? Could the disappearance of light/matter be balancing its appearance?

General Relativity works just fine inside the event horizon. The issue with GR is that it's a classical theory, but the fundamental laws of the universe are quantum.

What theory says this? Citation?

There was a self-published book title "The Anisotropic Universe" in the early 2000. I can't find it anymore.

wow, i never thought i would be downvoted for mentioning a theory :) -4 and not once anybody wondered what theory that is.

Your theory sounds as if it was immediately preceded by a bong rip. Complaints of downvotes attract downvotes. And you've not responded to two people explicitly inquiring for details about your theory

Thanks, for pointing out. I missed those responses.

Reply on news.ycombinator.com