[Code-Beta]

Hello. I am persionaly Globe Earther, but I found one odd thing. Modern Astronomy claims that Earth and other planeds were created form asteroids colliding. More specificly, asteroids came form gases form Nebula, and then they collided and made planets. But, if you hit something in motion, it will lose some of its speed.A Planet needs a constant/near constant speed to otbit somenting. But according to modern astronomy, earh was made by millions of asteroids hitting. So why didn't original Earth slow up just lose its orbit?

And did we mananged to get rebbutal to N-body problem?

[JSS]

Hello. I am persionaly Globe Earther, but I found one odd thing. Modern Astronomy claims that Earth and other planeds were created form asteroids colliding. More specificly, asteroids came form gases form Nebula, and then they collided and made planets. But, if you hit something in motion, it will lose some of its speed.A Planet needs a constant/near constant speed to otbit somenting. But according to modern astronomy, earh was made by millions of asteroids hitting. So why didn't original Earth slow up just lose its orbit?

The solar system began as a large gas cloud, where all the particles were moving randomly.  Once the cloud started to condense, everything began to rotate as the overall gravity of the cloud pulled the random motions into curves.  You can see this process when draining a sink or bathtub, the water starts to rotate as everything is pulled inward.

That's where the rotation comes from, and small bits collect into larger ones and larger ones are pulled into smaller.

By the time a planetoid starts to form, everything is all rotating in the same general direction, so you don't have a lot of head on collisions. Especially early on, things are not smashing at all, just nearby clumps slowly drifting closer and forming larger clumps.

Once you have the Earth mostly formed, the impacts are more violent due to the gravity of the planet involved. Even if some asteroids slam into it head on, it won't slow it down much.

But you are right, there is some rotational velocity lost if two asteroids hit head on, but the amount of rotational energy in the entire system is so massive that it only loses a tiny fraction to collisions.

And did we mananged to get rebbutal to N-body problem?

I've written n-body simulations before as a hobby so I've got direct experience with them and have answered a few questions on that here, but in a nutshell it's not a problem.

Every measurement in reality is going to limited in it's accuracy. You can measure a length of wood with a ruler, then use a micrometer and more and more specialized tools but you will never, ever know it's EXACT length. No matter how accurate you get, there is always going to be some margin of error.

But if you measured a piece of wood and found it was 100.00001 cm long, nobody is going to convince you it's actually 2,000 cm long just because you can't measure it down to a trillionth of a cm.

Same with Newtons and Einsteins laws of gravity.  The math is easy with just two bodies or point fields, and we can calculate those perfectly. But when you add in more, we simply don't have the math to solve it perfectly. It's still an open question of there is a solution to n-body like math problems, or if they just have to be iterated.

But it in no way can be said that those laws are wrong. Every demonstration proves them correct. The Voyager probes were launched and traveled via gravitational slingshots to all the outer planets. We landed probes on moons of Jupiter, landed robots on Mars, can calculate in advance the orbit of planets at very high precision.

All of that is strong evidence gravity exists, and we understand how it works to a very high degree.

Do we know EVERYTHING about gravity? No, of course not. There are still many mysteries, and one day someone will come up with an even better theory than Einstein. But to claim that all those accomplishments and all our knowledge is 100% wrong just because there are still unknowns is just silly.

If you throw out every bit of knowledge that still has unanswered questions, you are left with NOTHING AT ALL because we will never know EVERYTHING.

It's what makes searching for answers so fun and rewarding.

[Tom Bishop]

If the solar system can be simulated with Newton's laws alone, then why does New Scientist say the opposite?

See the New Scientist article at the bottom of this link: https://wiki.tfes.org/Symplectic_Integrators

[JSS]

If the solar system can be simulated with Newton's laws alone, then why does New Scientist say the opposite?

See the New Scientist article at the bottom of this link: https://wiki.tfes.org/Symplectic_Integrators

Thanks for proving my point.  From that article: "Consequently, the calculations a computer makes are not absolutely precise, so it can only provide us with an approximate picture of what will happen in the real world. Normally the errors are so small that they go unnoticed, but when computers are set to work on the enormously long string of calculations needed to simulate the movement of the planets round the Sun, tiny errors in each step can build up to make the final result wildly inaccurate."

Yes, trying to simulate the solar system over a large enough time scale will introduce errors due to the limits of the computing power we have. More computer power will reduce those problems but no matter how accurate your simulation, if you run it long enough errors will creep in. Lots of clever math has been developed to try and improve accuracy. 

Still, it's accurate enough to land probes on asteroids, comets, other planets and decades long trips through the solar system. It's accurate enough that you can't find one planet that has ever moved in a way not predicted or explained by the current laws of gravity.

Again, just because there are limits to how accurate we can measure or calculate or simulate something doesn't mean it's wrong. This is the real world, we can't ever measure it perfectly.

Thinking you found 'proof' that a theory is wrong just because it has limits only proves you don't understand how science, or the real world works.

[Tom Bishop]

The article says in the first sentence in the byline "Conventional calculations of the future of the Solar System quickly degenerate into disarray as computer errors build up."

So, the people thinking that it could be done were wrong. They can't do it. The simulations quickly degenerate into disarray. Science cannot simulate the Solar System or even keep the Sun, Earth, and Moon together with Newton's laws. It cannot be simulated. How embarrassing for science.

[JSS]

The article says in the first sentence in the byline "Conventional calculations of the future of the Solar System quickly degenerate into disarray as computer errors build up."

So, the people thinking that it could be done were wrong. They can't do it. The simulations quickly degenerate into disarray. Science cannot simulate the Solar System or even keep the Sun, Earth, and Moon together with Newton's laws. It cannot be simulated. How embarrassing for science.

Please quote where they defined the term ''quickly', thank you. Quickly in astronomical terms can be a million years. What are you assuming they mean by quickly? Five minutes? A day? A hundred years? You are making some big assumptions and leaps of faith here.

Again, we land robots on other planets and send probes on twenty year tours through the solar systems with these simulations you claim don't work at all. There are multiple spacecraft moving through the solar system RIGHT NOW using Newtons laws and simulations that are not perfect, but are perfect enough to visit dozens of moons of Jupiter 500 million miles away. Good enough to land people on the moon and return them safely to Earth. Clearly, Newton is good enough for decades long predictions. I don't consider any of those achievements embarrassing.

You can claim the math doesn't work, but reality proves you wrong.

[GreatATuin]

The article says in the first sentence in the byline "Conventional calculations of the future of the Solar System quickly degenerate into disarray as computer errors build up."

So, the people thinking that it could be done were wrong. They can't do it. The simulations quickly degenerate into disarray. Science cannot simulate the Solar System or even keep the Sun, Earth, and Moon together with Newton's laws. It cannot be simulated. How embarrassing for science.

Have you actually read and understood the article? It degenerates "quickly" when you simulate a million years every second. The article textually says that for a "short" duration of a few hundreds of thousands of years, simulations using conventional computer methods are accurate.

Doesn't look very embarrassing to me.

[Tom Bishop]

"Quickly doesn't mean quickly"

"Newton is good enough for decades long predictions." - Source: Myself

Not very convincing, guys.

[GreatATuin]

"Quickly doesn't mean quickly"

"Newton is good enough for decades long predictions." - Source: Myself

Not very convincing, guys.

Source : https://wiki.tfes.org/Symplectic_Integrators

"The Earth’s ellipse, for example, takes 112 000 years to rotate once—the sort of period that conventional computer methods can easily simulate".

Sure, it's not decades - it's much more than that.

[Tom Bishop]

That's talking about the Earth's ellipse.

Here is the full quote: "The first thing you notice is that the planets’ perihelia—the points at which their elliptical orbits are closest to the Sun—are not fixed, but rotate slowly. This phenomenon, which is known as perihelion precession, has been known for around a century. The Earth’s ellipse, for example, takes 112 000 years to rotate once—the sort of period that conventional computer methods can easily simulate."

If the Moon can't rotate around the Earth more than a few times before shooting off into space, why should we assume that Newton's laws can keep the Solar System together for years?

https://wiki.tfes.org/Symplectic_Integrators

Computing the long term evolution of the solar system with geometric numerical integrators

Abstract

  “ Simulating the dynamics of the Sun–Earth–Moon system with a standard algorithm yields a dramatically wrong solution, predicting that the Moon is ejected from its orbit. In contrast, a well chosen algorithm with the same initial data yields the correct behavior. We explain the main ideas of how the evolution of the solar system can be computed over long times by taking advantage of so-called geometric numerical methods. Short sample codes are provided for the Sun–Earth–Moon system. ”

The standard algorithms produce 'wrong solution' because the Moon is ejected from its orbit. A different algorithm is necessary to keep it together, and produces the 'correct' behavior. The paper describes that the algorithm which keeps it together is the symplectic integrator.

Figure 7 from the paper shows a comparison between a non-symplectic and symplectic integrator:

The Sun-Earth-Moon System



  “ Figure 7: Comparison of the explicit Euler method (left) and the symplectic Euler method (right) for the Sun-Earth-Moon system simulated over one year. The distance between the Moon (blue trajectory) and the Earth (black trajectory) is scaled by a factor of 100 in the plots, to better distinguish the Earth and the Moon. ”

[JSS]

"Quickly doesn't mean quickly"

"Newton is good enough for decades long predictions." - Source: Myself

Not very convincing, guys.

You avoided answering my questions I see.  Luckily someone else answered it for you. You also avoided commenting on the fact we use Newtons math to explore the solar system. It's easy to be convinced when you simply ignore entire responses. It shows you have no actual rebuttal to make.

Source : https://wiki.tfes.org/Symplectic_Integrators

"The Earth’s ellipse, for example, takes 112 000 years to rotate once—the sort of period that conventional computer methods can easily simulate".

Sure, it's not decades - it's much more than that.

So conventional computer methods can easily simulate 112,000 years of orbital calculations.

So what's quickly mean in the context used in that paper? You still haven't answered the question.

You uncovered math meant to help push calculations past a hundred centuries worth of predictions. And claim that somehow, it doesn't work at all?

[Tom Bishop]

Your quote says that they can simulate perihelion precession out to that amount of time. It says nothing about a full gravitational model of the Solar System.

To simulate the precession you only need two bodies, the Earth and Sun. Two body orbit simulations are possible, whereas three or more body simulations are not.

You have provided literally no sources, and are shouting that Newton's laws can simulate the solar system with no other sources except for yourself and whatever can be misinterpreted.

https://wiki.tfes.org/Three_Body_Problem

A dozen physicists telling you that the Three Body Problem falls apart, and you can find no sources in favor of your position except for your own self.

Lets see ONE physicist or astrophyscist who says that the Three Body Problem can simulate the Sun-Earth-Moon system.

[JSS]

Your quote says that they can simulate perihelion precession out to that amount of time. It says nothing about a full gravitational model of the Solar System.

I assume you are replying to me?

Would you like to move the goalposts out any further right away, or wait before demanding we simulate all the bodies in the galaxy, and then the entire universe before you will accept that these laws work?

At least it's progress you can accept that some parts of the solar system can be simulated for long periods of time. Just a little more open-mindedness might get you to accept that we can simulate the major bodies for long periods as well, in enough precision to do real work.

You also still have yet to comment on the fact we use these formula to explore the solar system, and how that squares with your insistence that they are completely wrong and don't work?

[Code-Beta]

I know that Moon gets away form earth few centimenters each year. But shoudn't Gravity of Sun pull it much, much more drasticly? Any calculations of Moon's orbit and Sun's effects on it?

[Groit]

You also still have yet to comment on the fact we use these formula to explore the solar system, and how that squares with your insistence that they are completely wrong and don't work?

FE'rs don't believe in any space exploration. 

[GreatATuin]

Your quote says that they can simulate perihelion precession out to that amount of time. It says nothing about a full gravitational model of the Solar System.

To simulate the precession you only need two bodies, the Earth and Sun.

Wrong. With only two bodies, there is no precession : http://farside.ph.utexas.edu/teaching/336k/Newtonhtml/node115.html ("if we neglect the relatively weak interplanetary gravitational interactions then the perihelia of the various planets (i.e., the points on their orbits at which they are closest to the Sun) remain fixed in space. However, once these interactions are taken into account, it turns out that the planetary perihelia all slowly precess around the Sun")

Also, the paragraph explicitly mentions a "model of the Solar System", not just the Earth and Sun.

Not my fault if you try to argue the model doesn't work, quoting an article that says it works extremely well for thousands of centuries.

[Tom Bishop]

Your quote says that they can simulate perihelion precession out to that amount of time. It says nothing about a full gravitational model of the Solar System.

To simulate the precession you only need two bodies, the Earth and Sun.

Wrong. With only two bodies, there is no precession : http://farside.ph.utexas.edu/teaching/336k/Newtonhtml/node115.html ("if we neglect the relatively weak interplanetary gravitational interactions then the perihelia of the various planets (i.e., the points on their orbits at which they are closest to the Sun) remain fixed in space. However, once these interactions are taken into account, it turns out that the planetary perihelia all slowly precess around the Sun")

Also, the paragraph explicitly mentions a "model of the Solar System", not just the Earth and Sun.

Not my fault if you try to argue the model doesn't work, quoting an article that says it works extremely well for thousands of centuries.

From your link: "Thus, by treating the other planets as rings, we can calculate the mean gravitational perturbation due to these planets, and, thereby, determine the desired precession rate."

If the other planets are rings and only the mean gravitational perturbation is calculated then there are only two orbiting bodies. That's not a full simulation of gravity. It's a limited simulation. If this is your idea of a full gravity simulation, it fails.

From a vague sentence you found about simulating a precession you conclude that there is, somewhere out there, a fully working dynamical gravitational simulation of the solar system in which the four hundred year old three body and n-body problems are working fine.

You guys act as if there are simulations, but no physcists ever talk about them and only confine their discussions to how the three body problem does not work.

Once again, please provide ONE quote from a Physcist who says that the three body problem can simulate the Sun-Earth-Moon system.

[GreatATuin]

Your quote says that they can simulate perihelion precession out to that amount of time. It says nothing about a full gravitational model of the Solar System.

To simulate the precession you only need two bodies, the Earth and Sun.

Wrong. With only two bodies, there is no precession : http://farside.ph.utexas.edu/teaching/336k/Newtonhtml/node115.html ("if we neglect the relatively weak interplanetary gravitational interactions then the perihelia of the various planets (i.e., the points on their orbits at which they are closest to the Sun) remain fixed in space. However, once these interactions are taken into account, it turns out that the planetary perihelia all slowly precess around the Sun")

Also, the paragraph explicitly mentions a "model of the Solar System", not just the Earth and Sun.

Not my fault if you try to argue the model doesn't work, quoting an article that says it works extremely well for thousands of centuries.

From your link: "Thus, by treating the other planets as rings, we can calculate the mean gravitational perturbation due to these planets, and, thereby, determine the desired precession rate."

If the other planets are rings and only the mean gravitational perturbation is calculated then there are only two orbiting bodies. That's not a full simulation of gravity. It's a limited simulation. If this is your idea of a full gravity simulation, it fails.

From a vague sentence you found about simulating a precession you conclude that there is, somewhere out there, a fully working dynamical gravitational simulation of the solar system in which the four hundred year old three body and n-body problems are working fine.

The sentence is not vague at all, it clearly says that your statement "To simulate the precession you only need two bodies, the Earth and Sun" is wrong.

You guys act as if there are simulations, but no physcists ever talk about them and only confine their discussions to how the three body problem does not work.

Once again, please provide ONE quote from a Physcist who says that the three body problem can simulate the Sun-Earth-Moon system.

Once again, you fail to understand that the fact the n-body doesn't have a formal, analytical closed-form solution doesn't mean it cannot be solved with numerical methods. Just repeating it does not work will not make that true.

In this article astrophysicists Michele Tenti and Piet Hut tell us "gravitational N-body simulations, that is numerical solutions of the equations of motions for N particles interacting gravitationally, are widely used tools in astrophysics, with applications from few body or solar system like systems all the way up to galactic and cosmological scales". Not only Sun-Earth-Moon, but much more complex systems.

[AATW]

I genuinely don’t know if Tom really doesn’t understand this stuff or whether he just pretends to, but it’s a common tactic of his.

It amounts to “your model isn’t perfect, ergo you don’t have one that works”.
It’s a common FE argument and a silly one. A model doesn’t have to be perfect in order to be useful.

It’s obvious to anyone who understands how things work in real life that masses and velocities and distances are not going to be integers. And they’re not going to be known exactly. Small errors, or limits in accuracy of representation, will build up over time, especially if you’re going to try and model something over huge timespans. And there are more subtle things like how some fractions cannot be represented perfectly in binary which is how computers represent numbers under the bonnet. A for example is 0.1 which in binary can’t be represented perfectly because it’s a recurring decimal in that base.

The point is our models are good and accurate enough to send craft to the outer planets or to land them on Mars or land men on the moon. And they were good enough to lead to the discovery of Neptune. I’d say that’s a pretty impressive validation of our models. The only FE rebuttal to any of that is “didn’t”, just denial. Not a strong counter argument.

And this line of argument is particularly silly when you think that while FE are pointing out the speck in the eye of RE models they are ignoring the plank in the eye of FE ones which is that there IS no FE model that has any predictive power.

[Tim Alphabeaver]

[...]It's a limited simulation.[...]

I've realised that this discussion is actually a non-discussion. So reality isn't analytically solvable: so what? This doesn't provide an argument for or against either the globe or the flat earth as both are equally not analytically solvable.

It's a chance for all of us (myself included) to show how much we know about numeric solvers but that's about it in terms of actual content.