Project Solar system generator in progress

[Drake]

Hi Artlav - awesome work!

About dual stars, IIRC, the location of everything in Orbiter is indexed off of the sun. You could add another "star" as a big planet I suppose, but one will have to be the "real" sun for location purposes. Also, dual star systems will not orbit their barycenter.

Hmm, unless you could make an invisible "star" at the barycenter, then have the two visible "stars" be planets that look like stars orbit that. But then all the planets would orbit the barycenter - not sure how that would work. I am guessing that almost certainly you would have unstable orbits, which Orbiter would not know about.

There is a gravity simulator around that one of the Orbinauts wrote - you could test generation algorithms against that.

http://www.orbitsimulator.com/gravity/articles/what.html

Most binary systems are pretty far apart, though, I think. You could make the companion star out pretty far and say that it is too far to affect the planets.

Good luck!

 

[tblaxland]

Hmm, unless you could make an invisible "star" at the barycenter, then have the two visible "stars" be planets that look like stars orbit that.

The Orbiter rending also considers all light to come from the global coordinate origin, ie the "sun", so if you fly between the your two custom suns, you will have some unexpected lighting effects.

There is a gravity simulator around that one of the Orbinauts wrote - you could test generation algorithms against that.

http://www.orbitsimulator.com/gravity/articles/what.html

Most binary systems are pretty far apart, though, I think. You could make the companion star out pretty far and say that it is too far to affect the planets.

Good idea. If Gravity Simulator had the ability to export its results, you could use that as a basis for producing ephemerides for planet modules.

 

[Artlav]

Solar system generator (Planeter) project, alpha 4:
http://orbides.1gb.ru/orbf/sway-planeter-090310.zip (222 Kb)

Improvements:
-Added Rock planet class
-Lv 6 textures export (takes a long while)
-Lv 7 textures export (takes a damn long while)

This release is aimed at the new planet class - Rock, aka "just a rock without air or anything special, like the Moon".
What about craters? Which way looks better - A or B?
What else is strikingly wrong about that planet kind?
How many craters feels better?

Also, you can test the high-res textures export, if you're ready to wait.

Case A:

Case B:

Another Case B:

More craters, darker:

Even more craters:

Much less craters:

Another colorset:

Another colorset, different craters:

 

[Hielor]

Solar system generator (Planeter) project, alpha 4:
http://orbides.1gb.ru/orbf/sway-planeter-090310.zip (222 Kb)

Improvements:
-Added Rock planet class
-Lv 6 textures export (takes a long while)
-Lv 7 textures export (takes a damn long while)

This release is aimed at the new planet class - Rock, aka "just a rock without air or anything special, like the Moon".
What about craters? Which way looks better - A or B?
What else is strikingly wrong about that planet kind?
How many craters feels better?

Also, you can test the high-res textures export, if you're ready to wait.

Case A:

Case B:

Another Case B:

More craters, darker:

Even more craters:

Much less craters:

Another colorset:

Another colorset, different craters:

I don't know, I think they all look plausible. I think it partially depends on the size of the object and its region of space. For example, I would imagine that a moon close to a large gas giant would have fewer impact events than Earth's moon. Or maybe vice versa--I don't really know.

I don't think either A or B looks "better," they just look different.

 

[Artlav]

I don't know, I think they all look plausible.
...
I don't think either A or B looks "better," they just look different.

I'd say they all go as options for the class.
The question is more of a kind of which one of them looks too bad?
If any, what to improve?

One more image, in-Orbiter shot (lv 7):

 

[Namahage]

I like them all - more choice is better.

I think the most earthmoon-like is the last color set + fewest craters.


My Opinion:

-The borders on the craters need to be thinner with less contrast.

- The width craters themselves need to be much smaller, particularly considering the relatively large size of this moon.

If the surface is dusty: any way to add radial debris to some of the craters?

----------------------------------------

Can't wait to give the new alpha a spin, alas it won't be today.

 

[destinos]

I just want to say that I think this is an amazing project. Especially as a person who was excited for years about Spore. I think what you've done here, for no pay, is turning out much better than any effort made for that game. This is what I wished that game would have been.

 

[Linguofreak]

Hitting 'E' to toggle the editor doesn't actually seem to do anything...

EDIT:

Also, some stars seem to be older than they possibly could be: I've found several stars with multi billion year ages that were massive and luminous enough that they shouldn't have ages greater than a few million years.

 

[Brycesv1]

well that should be a simple adjustment of properties or property limits apon generation

 

[cjp]

Wow.. another cool project by Artlav! :speakcool:


About the craters: I intend to make a "crater world" plugin for my own libProcTer, which will probably be the default plugin for all planets that don't have 'real-world' data available. So I already did some thinking on how to make realistic craters:

• On small bodies (like asteroids), the 3D shape of craters is so large that it should be visible from far away. On large bodies (e.g. the moon) it should only be visible from nearby (e.g. the horizon as seen from low orbit).
• Large craters on large bodies often have multiple ripples and sometimes a central mountain. Larger crater = more ripples. The very largest craters might have a lava lake inside.
• When it comes to color, craters are mostly visible because of shadow effects. For the rest, the material ejected from the crater is usually lighter than the surrounding area (and probably also lighter than the inside of the crater). The lava material is usually the darkest. I've read that the ejected material becomes darker after time (because of exposure to radiation), so it should be just like the 'highland' color for old craters, and much brighter for young craters.
• The area of ejected material is much larger than the crater itself. When going away from the crater, it slowly fades away, and shows more of the original landscape.
• Craters of older age are covered by craters of younger age. Inside young craters, the density of smaller craters is lower than in old craters.

 

[tblaxland]

I'll add that older crater walls also tend to be lower and smoother due to slumping and micro-meteorite erosion.

Another thing that would definitely improve immersion would be rilles. These could be fairly random in size/shape with some restrictions:
1. Length should be roughly proportional to width (but with some variation).
2. Narrow rilles are unlikely to cross big craters. Wide rilles could cross small craters.
3. Rilles would not normally be radial to impact features.

 

[Graham2001]

I've found this runs OK on my machine.

Athlon XP 2800, 2gb ram, Radeon 9250.

The problems I have noticed are:

1. The editing function does not seem to work.
2. It is only generating one system at present.

 

[Artlav]

1. The editing function does not seem to work.
2. It is only generating one system at present.

Press H for keyboard help.
To generate other systems use F7 menu.
I'm not sure how and where edit should or not work, this being rather old.

 

[Graham2001]

Press H for keyboard help.
To generate other systems use F7 menu.
I'm not sure how and where edit should or not work, this being rather old.

If I press 'e' to bring up the editor nothing happens. If I press 'x' to bring up the export function then it comes up.

 

[cinder1992]

I'd say they all go as options for the class.
The question is more of a kind of which one of them looks too bad?
If any, what to improve?

One more image, in-Orbiter shot (lv 7):

Waitaminute, are those moons that orbit a huge blue planet?