/sci/ - Science & Math

>electrons are le cloud
>can only calculate shape of cloud for monatomic hydrogen
this is the point quantum mechanics should have been dropped

>>14718409
>decades of research
>all the smartest physicists have done literally nothing else with their time
>billions spent on particle accelerators and muon detectors
>the state of the art in predicting electron configuration during chemical reactions is just train a neural network bro
knot theory yourselves physishits

>>14718409
>>14718459
this has nothing to do with quantum mechanics. the three-body problem in classical physics (e.g. how does a earth-moon system orbit if it consists of one earth and two similar size/distance moons) remains unsolved to this day. probably if we had a good classical solution to the three-body problem then the quantum analog would be a simple corollary. the problem here is not anything to do with quantum physics

honestly physicists have been working on the three-body problem since the 1700s and it’s literally foiled all the best minds in physics since the beginning. it’s arguably a much harder problem than quantum gravity or high-temperature superconductivity since at least some progress has been made in those areas. but figuring out what happens if you had uranus and neptune both had a near flyby with saturn at the same time? a well known mystery for centuries with basically no progress by any of the greatest minds in the history of physics

>>14718472
RETARD ALERT RETARD ALERT
THE THOUSAND BODY PROBLEM IS SOLVED BY RUNNING A SENSIBLE ADAPTIVE PRECISION ALGORITHM THAT DIRECTLY IMPLEMENTS THE FORCES AT PLAY
SHOW ME THE EQUIVALENT FOR ELECTRON CONFIGURATION YOU GOD DAMN NIGGER
YOU DONT EVEN SEE THAT THERE IS A PROBLEM HERE

>>14718477
>ADAPTIVE PRECISION ALGORITHM
do you know what “sensitivity to initial conditions” means?

>>14718487
there is strictly correct bounds on your position and velocity error. you can predict point masses into the future arbitrarily given more computing power
WHERE IS THE STRICTLY CORRECT BOUNDS ON ELECTRON CONFIGURATION?
WHERE IS THE ALGORITHM THAT IMPLEMENTS THE BASIC PHYSICS AND ALWAYS GIVES THE RIGHT ANSWER????
>still not seeing the problem
kill yourself and take a quantum physicist with you

>>14718496
no anon, you are wrong. with a chaotic system like the three body problem, or even a double pendulum, it turns out that in both cases even the ROUNDING ERRORS that computers necessarily make imply that their simulations diverge from reality, and continue to diverge more and more as they go on extrapolating.

the essence of a chaotic system is that any approximation has this property: the approximation starts to diverge from the real solution and gets further and further off track thereafter in general.

having some compsci idiot tell you they implemented some double-precision simulation is fucking worthless and anyone who advocates that line of research should be fired. for the three body problem the thing that is needed, and the only thing that wouldn’t diverge from reality by arbitrary amounts after finite times, is an ANALYTICAL SOLUTION. no stupid computer programmer can possibly claim any type of parity with real math here

>>14718472
Three body problem in the scientific sense is solved. Three body problem not having a closed-form solution is simply a mathematical statement. It's like if you said that quintic polynomials are unsolved because there's no formula for finding its roots,
In practice physicsts are good at making high-precision three-body models. There is no anything like that for electron configuration.

>>14718512
>no anon, you are wrong. with a chaotic system like the three body problem, or even a double pendulum, it turns out that in both cases even the ROUNDING ERRORS that computers necessarily make imply that their simulations diverge from reality, and continue to diverge more and more as they go on extrapolating.
>the essence of a chaotic system is that any approximation has this property: the approximation starts to diverge from the real solution and gets further and further off track thereafter in general.
>having some compsci idiot tell you they implemented some double-precision simulation is fucking worthless and anyone who advocates that line of research should be fired. for the three body problem the thing that is needed, and the only thing that wouldn’t diverge from reality by arbitrary amounts after finite times, is an ANALYTICAL SOLUTION. no stupid computer programmer can possibly claim any type of parity with real math here
you're totally wrong. both three body problem and double pendulum or whatever can be calculated to arbitrary precision, arbitrarily into the future given arbitrarily much computation
you're such fucking LARP just kys

>>14718472
kek what even is this bait

>>14718521
>given arbitrarily much computation
yeah, therein lies why you are wrong. “arbitrarily much computation” does not and cannot exist, and any finite amount of computation diverges from reality in finite time; any computer that exists today diverges from reality dramatically very quickly

>>14718542
this is not bait. i’m simply pointing out that computer scientists who believe that computer simulations are equivalent to actual solutions are absolute retards, and that’s a fact

>>14718512
>>14718545
kek actually don't kill yourself anon. thinking that computers can only calculate with 64 bits of floating point precision is pretty based

>>14718550
it doesn’t matter how many bits you calculate with. the calculations still diverge after finite time

>>14718554
you're so delusional it's crazy.
yeah you can predict the positons of all the planets 1000000 years into the future on a single computer and quantify exactly how wrong those positions are and refine them at will to the required precisions but it's just as much of a mystery as how the electrons behave in a molecule of fucking water.

>>14718584
that’s exactly my point. the reason physics has trouble with modeling quantum chemistry for complex molecules (i.e. anything besides hydrogen) is because it enters the territory where we don’t even understand things classically. we lose the picture even at the three-body problem. again it had nothing to do with quantum physics. it’s just that normal ass newtonian physics goes chaotic if you have more than 3 particles/bodies

>>14718512
Compsci actually solves the whole problem of finding mathfag solution equations that won't be ever needed.

Quanty Sointist have really little brain in the skull, they don't realize once forces are known you can solve it by simulating the particles all together.
Limit the precision to planck length or more don't care, but if your shit never ends up building up hydrogen atom by itself it should be thrown out of the window period.

We have computational era now, you little paper number drawing turds should just die already, science is now in programmers hands, accept the reality and pass out from copium, you'll sleep better.

>>14718550
Don't tell him that 256-bit floating point already covers up whole universe up to planck length unit.
It's only 4x more from current fp64 hardware and either way we can already calculate that via software, he's the ultimate copium idiot thinking you can't be discrete to replicate real world physics, just sad.
At certain time frame your quantum whatever effects disappear, we don't immedietly need matrix.
By example we simulated the COVID protein by molecular dynamics, do you think any mathfag writing equation for COVID was involved? That's literaly 200 idea of science.

>>14718605
No that basically means that not a single scientist actually understand proton, neutron and electrons.
We're literally stuck in 1900 single long time and they don't realize it

>>14718605
the three body problem simulator always gives you a mathematically correct answer as to how big the error is it lies within and can keep that error vanishingly small on a huge ass system simulated to huge ass timescales.
it does this by doing an approximation and then rigorously quantifying how that approximation (including the floating point error) is affecting the result.
As I said earlier the best approximation for electron configuration is
>train a neural network bro
or
>run some "best practice" algorithm that doesn't know shit about if the results are wrong or right
maybe it's a communications problem between chemists and physicists, maybe they have all the answers.
right now the comparison between 3 body problem and electron configuration is unbelievably shitty

We literally know nothing about le particles, do we?

>>14718644
We know nothing more than looking at cake with heavy -10 eye myopia and claiming how it tastes

>>14718542
He's been at it for several months. See that double pendulum thread in the catalog? It's by him

>>14718472
>Three-body problem is not solvable pseud and billion body problem is just a conspiracy theory, not real okay!!!

>>14718699
this is cgi. nasa lies. earth is flat. christ is king

>>14718514
the area of an ellipsis also doesn't have a closed form solution, the circle also doesn't, that's why we invented pi.