I keep telling them that there is a limit to speed and acceleration. Like, for light, the maximum acceleration sends the photon in the other direction.
Gravitational acceleration calculation
Author
@Nagle Care to elaborate? How does it not work like that in a two body simulation? If an object falls straight inward and passes through the center of gravity of the object it will accelerate outwards until its acceleration is canceled out and it falls back in.
Author
taby said:
Why are you doing this?
There are too many conversations so now I don't know what “this” is exactly.
jonny-b said:
There are too many conversations so now I don't know what “this” is exactly.
Sorry, I mean: why are you studying the gravitation of galaxies?
Author
@taby , I'm just doing this for fun. I wanted to see if I could create a simulation that would produce a spiral galaxy and it grew from there.
jonny-b said:
If an object falls straight inward and passes through the center of gravity of the object it will decelerate outwards until its velocity is reversed and it falls back in.
If i had to nitpick, i would rephrase the bold parts like so. But that's probably not what he meant, and i guess his concerns from page one are now outdated on page 12.
jonny-b said:
I wanted to see if I could create a simulation that would produce a spiral galaxy and it grew from there.
That's still an interesting goal. Any progress? Did spirals form while you tried many settings? Or eventually just some stable orbits?
I guess @taby already tried to help on this. If we place random suns at some average density, is something simple like that already enough to get something that looks like the real thing?
The question reminds me on my own crusade of terrain simulation in 3D.
I never was happy with my erosion simulation. It worked, but it did not generate patterns as nice as in 2D simulators or nature. I was not happy and have tried a dozen of different erosion algorithms.
Now, finally it seems to work. I still need to test at larger scales to be sure.
But the point is: I do not know why the latest algorithm is better than the others.
Author
JoeJ said:
If i had to nitpick,
Thanks for pointing that out. I think your suggestions is better.
Any progress?
Not yet. But it's really hard to tell. there are only 700 points and they are moving so fast it's hard to tell. It's also only a 2d simulation and it's missing 1000's of variables that a real galaxy has. Plus my math might be wrong somewhere. I think I may reach out to a professor from The Ohio State university and see if they would be willing to offer their input.
Yes there are waves that form. I really really think that you really really need to buy yourself a copy of Galactic Dynamics.
A fair simulation would have a black hole at the centre of the galaxy, and a disc made up of stars and interstellar gas. Both can be approximated as fields of point particles.
Author
@taby , at least there is a black hole in the galaxy sim. One real problem in making a more accurate simulation is computing power. I have to allow for some pretty insane velocities in order for the simulation to do anything in real time since a more accurate (as far as strength of gravity and velocity constraints) simulation takes way to long to do anything. Since this is running in a browser I'm only getting so many processes per second. I thought about writing it to run outside of the browser and then create a file that could be imported. That way you could run the simulation for a couple days and then let it play in a min or two.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
