new update

[Airforce]

when will the next update be ready

[Dan Dixon]

I'll be releasing a new update later today with some minor bug fixes, but I doubt that's what you're asking.

What would you like to see in the next update?

[Airforce]

i would like to be able to make new planets out of dust

[Dan Dixon]

You can do a version of this with the current version of Universe Sandbox:

1 - Open the simulation: Moons, 75 orbiting Earth.
2 - Right click and change the collision mode to "Combine" (in the group at the top center)
3 - Watch as the moons pull themselves together and create larger bodies.

Also try the activity Perpetual Motion (click on "Fun Things to Do" from the main menu) to watch new planets form from smaller bodies.

Feel free to share feedback on this:
- How would you like to be able to make new planets out of dust?
- What kind of controls are you looking for?
- Describe how the perfect version of this feature would work.

[Airforce]

my feedback is i wish that dust have weight and would collect together and form new bodies
that rings could make moons

[Dan Dixon]

Forming a moon from a ring of dust is an interesting idea. I've got some new features planned for Universe Sandbox that will touch on this idea.

Thanks again for the feedback.

[Airforce]

how would you make it so that dust forms a new planet

[Dan Dixon]

Good question.

There are two types of objects in Universe Sandbox: Bodies/spheres and dust/particles. Bodies can have a size and mass, but dust does not. Dust has no size or mass; it is a point. Dust is what makes up the ring of planets and is kicked up whenever planets collide.

If computers were infinitely fast then there would be no need for dust; everything could happen with bodies (you could just make them really small and weight very little) but because bodies have more features and options they're not as efficient as using dust for things like the rings around Saturn.

But I'd imagine you're not too concered about the difference; you just want it to work.

I think the solution is going to be to mix dust with other small bodies so that there is mass in the dust that can be used for planet building. And give dust a mass to add to other bodes when they collide.

[APODman]

I made a simulation of the formation of some moons ( some with the massa of our moon) around the Earth from the accretion of a disk of bodies orbiting around them.

I don´t run the simulation for much time to see how many resulting moons remain in stable orbits.

If the simulation can help you Airforce, open the attached file in Universe Sandbox.


[ ]´s

[APODman]

Something interesting to do in this simulation is to observe the moment when the particles become gravitational attracted by other particles that coming more massive.

To observe this I suggest that take off the functions  "Trails" and "Labels", enable "Color Mode" and "Vivid" and enable "Glow" and "Relation".

So watch where the lines of Earth's relationship changing to some neighbor massive object when a particle is close enough of it.

Note that not all particles that establishes a gravitational connection with the neighbor more massive particle are aggregated by it, sometimes the high speed of the smaller mass particles prevent this aggregation and it returns to show a relationship line with the Earth ( despite the fact that its orbit was obviously affected by this short gravitational tug of more massive particles )

Thanks to these gravitational interactions the orbital characteristics of the particles on the disk can change constantly increasing numbers of collisions and ejection the system.


Example:

Capture, accretion and escape:





[ ]´s

[Dan Dixon]

Very cool...

I will confess though that the relation lines show the greatest value found by comparing every body using this formula:

Force for Relation Lines between A & B (where A is larger) = Gravity * (Mass of Body A) / (Distance between A & B)^3

as opposed to Newton's actually equation which is:

Force of A on B= Gravity * (Mass of Body A) / (Distance between A & B)^2

I found that cubing the distance just worked and looked better for showing the connecting lines.

[VurX]

Forming a moon from a ring of dust is an interesting idea. I've got some new features planned for Universe Sandbox that will touch on this idea.

Thanks again for the feedback.

In my mind adding mass to the dust motes would just be a matter of a variable... Colliding a particle with an object, fairly simple. Colliding it with another particle, well, they will rarely (if at all) occupy exactly the same space, and yet you still want them to count as collided. Not to mention setting the threshold of where do the clumps count as actual (model) bodies. (I'm assuming that your system supports dust particles attracting one another, based on mass only)

But I agree with the OP that it would be an awesome feature. I myself would love to see an accretion disk around earth form into the moon (or more than one )

One additional feature in "Combine" collision mode. If the colliding body is more than, say, 1/2 the size of the base body. The base body and the colliding body should explode (or at least fracture). Just my dream

[FGFG]

Another idea:

Particles don't interact with other particles
Particles have no volume.
Particles have no mass.


To make them creating an object they need mass.

You can add a value to the particles (say 1000 kg), but without setting the size.
When a particle goes near enough to another particle you can collide them, but still without size (so still a particle; this is to not overload the cpu). You can create size (so make them an object) just when the mass is more than a set value (e.g. 100000kg). So the particles remain particles (with no interaction with other particles) till they become more "heavy".

[Dan Dixon]

I appreciate the ideas... Thanks to both of you.

I do expect to improve the collisions so that they will shatter into multiple pieces. It's my dream too. And the hybrid idea is good too.

Dust was created to be a way of adding lots of particles that would be fairly computationally efficient... Since they are single points (their diameter is 0) there is no need to do collision checks between each one another (only with normal bodies) and since they don't have mass they don't effect anything else.

It would be fairly trivial to convert dust into normal bodies or have an option to create rings and collision particles out of normal bodies (that have a mass and size). The simulation would run slower (than it would with dust particles), but this might satisfy the need for dust to collect together.

[VurX]

Well that's my reasoning. Only I'd keep them as dust particles for as long as possible. (making the sim run faster) and convert them into normal bodies when their mass reaches a certain threshold.

Now I've no idea how you've implemented collisions, however, one off the cuff idea for particle / particle collision checks might be to do proximity checks. (exhaustive, and slow), there's probably a better way.
Keeping particles (not substituting them with bodies), would allow for higher numbers of said particles and thus increase the accuracy of the sim.

[Dan Dixon]

It's fair to say that there are many opportunities for optimization. In fact the next release will load large systems/simulations twice as fast as previous versions.

My next focus will be improving the interface but I look forward to improving the collisions, explosions, and particles in the near future.