Author Topic: Celestial Objects Thread  (Read 324080 times)

0 Members and 1 Guest are viewing this topic.

Offline Rodo

  • Custom tittle
  • 212
  • stargazer
    • Minecraft
    • Steam
Re: Celestial Objects Thread
and 2.



It's too blurry IMO, if you use gimp you can take a look at this tut: http://gtuts.com/design/the-ultimate-gimp-planet-tutorial

Hope it will be as helpfull to you as it was for me.
el hombre vicio...

 

Offline TopAce

  • Stalwart contributor
  • 212
  • FREDder, FSWiki editor, and tester
Re: Celestial Objects Thread
It's as if a portion of the lit part were under a giant overarching glass.
My community contributions - Get my campaigns from here.

I already announced my retirement twice, yet here I am. If I bring up that topic again, don't believe a word.

 

Offline watsisname

Re: Celestial Objects Thread
Lighting weirdness aside (could it be an alpha-blending problem between separate layers?), I'm curious to see more of that surface texture.  It looks funky, in a good way. :)  Also is this a planet with an atmosphere, or a moon?
In my world of sleepers, everything will be erased.
I'll be your religion, your only endless ideal.
Slowly we crawl in the dark.
Swallowed by the seductive night.

 

Offline ShadowWolf_IH

  • A Real POF Guy
  • 211
    • CoW
Re: Celestial Objects Thread
Can't get atmosphere to work right yet, but when I do it will be a planet.  Latest incarnation, 2 views.

You can't take the sky from me.  Can't take that from me.

Casualties of War

 

Offline Droid803

  • Trusted poster of legit stuff
  • 213
  • /人 ◕ ‿‿ ◕ 人\ Do you want to be a Magical Girl?
    • Skype
    • Steam
Re: Celestial Objects Thread
Big one pixellated. Small one good.
(´・ω・`)
=============================================================

  

Offline watsisname

Re: Celestial Objects Thread
I concur.  The texture looks pretty stretched on the larger render, but the small one looks great. :yes:
In my world of sleepers, everything will be erased.
I'll be your religion, your only endless ideal.
Slowly we crawl in the dark.
Swallowed by the seductive night.

 

Offline TopAce

  • Stalwart contributor
  • 212
  • FREDder, FSWiki editor, and tester
Re: Celestial Objects Thread
Both seem good to me.
My community contributions - Get my campaigns from here.

I already announced my retirement twice, yet here I am. If I bring up that topic again, don't believe a word.

 

Offline Sushi

  • Art Critic
  • 211
Re: Celestial Objects Thread
I concur.  The texture looks pretty stretched on the larger render, but the small one looks great. :yes:

There's a set of funny-looking dark patches on the small one, but other than that, it does indeed look awesome.

 

Offline Trivial Psychic

  • 212
  • Snoop Junkie
Re: Celestial Objects Thread

Not to nit-pick (but I guess I will), but shouldn't both celestial bodies in that image have the same amount of illumination, in that the larger has a half-phase and the smaller has a quarter phase.  I believe that they should both have the same phase.
The Trivial Psychic Strikes Again!

 

Offline Droid803

  • Trusted poster of legit stuff
  • 213
  • /人 ◕ ‿‿ ◕ 人\ Do you want to be a Magical Girl?
    • Skype
    • Steam
Re: Celestial Objects Thread
Well, I don't think they're supposed to exist together like that.
I think he was just showing off the bitmap (in which case phase doesn't really matter cause you just have to reposition the sun accordingly :) )
(´・ω・`)
=============================================================

 

Offline Aardwolf

  • 211
  • Posts: 16,384
    • Minecraft
Re: Celestial Objects Thread


Behold, the first presentable planet created using a planet maker program I've been programming.

Note: this is a view from an altitude of .5x the radius of the planet, i.e. the distance from the center of the planet is 1.5x its radius.
« Last Edit: May 10, 2010, 09:04:13 pm by Aardwolf »

 

Offline watsisname

Re: Celestial Objects Thread
Very nice work, Aard.  The surface color and texture looks great, as does the lighting.  One suggestion I'd make might be to play with the atmospheric effect.  It seems like the intensity drops off too gradually as you go farther from the surface, (also it's veeery faint overall) but maybe it's just me.

Do wanna see more of these, keep it up! :):yes:
In my world of sleepers, everything will be erased.
I'll be your religion, your only endless ideal.
Slowly we crawl in the dark.
Swallowed by the seductive night.

 

Offline Rodo

  • Custom tittle
  • 212
  • stargazer
    • Minecraft
    • Steam
Re: Celestial Objects Thread
It looks pretty good, so this program makes the planet all by itself or it uses textures as reference?

For the next step maybe adding normal maps or some elevation effects to the surface will make the overall planet look better.
el hombre vicio...

 

Offline Aardwolf

  • 211
  • Posts: 16,384
    • Minecraft
Re: Celestial Objects Thread
I supplied the textures.

Take two:



...Featuring more pronounced atmo glow, also experimental normal maps. I think i got the tangent-space math wrong :doubt: ... but whatever. Also, I doubled the resolution of the base map, and added a tiny bit of noise to it.

 

Offline Thaeris

  • Can take his lumps
  • 211
  • Away in Limbo
Re: Celestial Objects Thread
That's pretty neat, Aard. Does your program draw on a 3D model or is it just a 2D application?
"trolls are clearly social rejects and therefore should be isolated from society, or perhaps impaled."

-Nuke



"Look on the bright side, how many release dates have been given for Doomsday, and it still isn't out yet.

It's the Duke Nukem Forever of prophecies..."


"Jesus saves.

Everyone else takes normal damage.
"

-Flipside

"pirating software is a lesser evil than stealing but its still evil. but since i pride myself for being evil, almost anything is fair game."


"i never understood why women get the creeps so ****ing easily. i mean most serial killers act perfectly normal, until they kill you."


-Nuke

 

Offline Aardwolf

  • 211
  • Posts: 16,384
    • Minecraft
Re: Celestial Objects Thread
I provide a model with texture coordinates and it finds what point on that model each texel corresponds to; then it fires a ray from the origin through that point, figures out what color it needs to be, and puts that in the texel.

But in these images the model is just a plane.

 

Offline Rodo

  • Custom tittle
  • 212
  • stargazer
    • Minecraft
    • Steam
Re: Celestial Objects Thread
Good Good, atmosphere is a little glowy on the dark part, but overall it looks amazing, good app you got there mate  :yes:
el hombre vicio...

 

Offline Aardwolf

  • 211
  • Posts: 16,384
    • Minecraft
Re: Celestial Objects Thread
Good Good, atmosphere is a little glowy on the dark part, but overall it looks amazing, good app you got there mate  :yes:

Well, it's important to note that the viewpoint is only 1/2 the radius of the planet above the surface, so the visible portion of the surface isn't that big. So although the transition from light to dark may look like it's stretched wider than it ought to be, that might be in part because it's showing up that much bigger.

Blowfish has noticed there's some 'banding' going on in the dark area, which I've determined has to do with the way I'm discretizing the atmosphere. Increasing the minimum number of iterations might provide improvement, but it's the brute-force way to do it, and it seems to slow it down somewhat. That said, so do the normal maps, and the tangent-space math is probably wrong there anyway. Also I've got no texture-filtering, since it's all software and I haven't implemented anything like that. It would probably also make it slower.

If there's going to be much more discussion of this, I might have to start another thread in the programming board. Or be (voluntarily) thread-split. I reckon the images ought to remain here though.

 
Re: Celestial Objects Thread
Yeah, looking pretty good! Can you do something about the aliasing of the edge, too?

 

Offline Aardwolf

  • 211
  • Posts: 16,384
    • Minecraft
Re: Celestial Objects Thread
I've added linear texture filtering. Also, I'm thinking about switching to a height map instead of a tangent-space normal map, and then using the derivative of the height (numerically) to determine the surface normal vectors.

Edit without bump (*gasp*):

« Last Edit: May 11, 2010, 06:10:04 pm by Aardwolf »