High!

Several days have passed, and I tried to improve the appearance of my 
asteroidal moon, mostly be applying additional 3D noise to the surface 
(in the case of the image attached here, by replacing the original 
noise_3d(x, y, z) by noise_3d(pow(x, 3), pow(y, 3), pow(z, 3)), which in 
fact added subtle waves to the terrain relief).

But still the problem is, that with the crackle pattern (with form set 
to <n, 0, 0>), the craters are in fact "bubbles" of empty space 
distributed randomly inside the raw isosurface moon body and not 
necessarily cut in half by its surface, so that many craters still are 
gaping holes rather than shallow depressions.

How can I get rid of this? Is there something like UV mapping for 
isosurfaces which guarantees that only craters cut in half by the 
surface are generated?

Is it possible to write a "core function" which, on the x-y plane, 
generates a roughly symmetric (perhaps slightly turbulated) 
cross-section of a crater, with terraced walls, a rim higher than the 
surrounding terrain and even a central peak, which then is transferred 
to a circular structure by applying trigonometric functions (?), which 
in turn are distributed randomly (also with random variations in depth, 
diameter, terracing and height of central peak) across the existing 
asteroid surface?

Where can I find the mathematical resources to design such a function?

See you in Khyberspace!

Yadgar

Now playing: Pirates of Love (Zara)

Post a reply to this message

Attachments:
Download '2006-06-24 michatopia, moon .jpg' (20 KB)

Preview of image '2006-06-24 michatopia, moon .jpg'

> High!
> 
> [...]
> How can I get rid of this? Is there something like UV mapping for
> isosurfaces which guarantees that only craters cut in half by the
> surface are generated?

You could create the isosurface without holes then use trace() and
difference out the craters, but that doesn't help with the rims.

You could use trace and a blob rather than an isosurface, but again
the rims are tricky (or impossible).

You might be able to add the crackle pigment as a function to
the isosurface, but I'm not sure how that would turn out.

> Is it possible to write a "core function" which, on the x-y plane,
> generates a roughly symmetric (perhaps slightly turbulated)
> cross-section of a crater, with terraced walls, a rim higher than the
> surrounding terrain and even a central peak, which then is transferred
> to a circular structure by applying trigonometric functions (?), which
> in turn are distributed randomly (also with random variations in depth,
> diameter, terracing and height of central peak) across the existing
> asteroid surface?
> 
> Where can I find the mathematical resources to design such a function?
> 
> See you in Khyberspace!
> 
> Yadgar
> 
> Now playing: Pirates of Love (Zara)

-- 
Bill Hails
http://billhails.net/

Post a reply to this message

High!

Bill Hails wrote:

> You could create the isosurface without holes then use trace() and
> difference out the craters, but that doesn't help with the rims.

Interesting approach... perhaps I would even be able to generate halfway 
believable rims!

> You might be able to add the crackle pigment as a function to
> the isosurface, but I'm not sure how that would turn out.

This is exactly what I did in my current attempts!

See you in Khyberspace!

Yadgar

Post a reply to this message

> High!
> 

For future reference, when greeting people in English, the proper 
spelling of it is "Hi!" not "High!"  They are both pronounced the same, 
but the meanings are completely different.

Just thought you might like to know.  I understand we have a lot of 
people here that speak English as a second, or even third or fourth 
language, and I most of them want to be told when they say something 
wrong or that doesn't make sense.

-DJ

Post a reply to this message

Thought I'd have a go at this one...

// Asteroid
#version 3.6;
// ==============================================================
#default{finish{ambient 0}}
#global_settings{assumed_gamma 1.0 max_trace_level 5}

#declare f_bozo1 = // similar to noise3d
function{pigment{bozo scale 0.5 color_map{[0 color rgb 0] [1 color rgb 1]}}}

#declare f_craters =
function{
  pigment{
    crackle form <1.2, 0, 0> // never really played with this one, thaks
Yadgar!
    scale 0.3
    color_map{
      [0.00 color rgb 1.00] // centre of crater, lowest part
      [0.30 color rgb 0.80] //
      [0.35 color rgb 0.40] // rim, higher than surrounding area
      [0.40 color rgb 0.5] // level of area outside the crater
      [1.00 color rgb 0.5]
    }
  }
}

/*
the f_craters function is subtracted from the basic asteroid shape, so the
lowest points get the highest numbers, since you are subtracting more.
*/

// --------------------------------------------------------------
isosurface{
  function{
    (x*x)+(y*y*1.8)+(z*z*1.5)-1 // basic sphere function, scaled to be more
elliptical
    +f_bozo1(x,y,z).red*0.35 // deforms the ellipse shape
    +f_craters(x,y,z).red*0.05 // adjust the '*0.05' part to change the
depth of the craters
  }
  contained_by{box{<-1, -1, -1>, <1, 1, 1>}}
  threshold 0
  accuracy 0.001
  max_gradient 4
  texture{pigment{color rgb 1} finish{diffuse 0.6}}
}

light_source{<-10, 10, -5> color rgb 1}
light_source{< 10, -10, -5> color rgb<0.2, 0.4, 1.0>*0.25}

camera{
  location <0, 1, -4>
  up y*(image_height/image_width)
  right x
  sky y
  look_at <0, 0, 0>
  angle 36
}
// -------------------------------------------------- end of file

Post a reply to this message

Attachments:
Download 'asteroid.jpg' (6 KB)

Preview of image 'asteroid.jpg'

A potato-cheese with holes! Delicious! Nyam!



Doesn't looks very realistic. Might be, the crater edges are to straight. 
And the entire object (asteroid and craters) is too regular, somehow. Not in 
shape, but in surface characteristica. Too plain. Too smooth.

I believe you would be well-advised to consult photos of asteroid or moons 
surfaces to get some good ideas. Surfaces are not really plain but bear tiny 
difference in surface heights and surface roughness.

But your project is promising. Go on, and you will create a great model of 
an asteroid!

Greetings,

Sven

Post a reply to this message

High! ;-))

> For future reference, when greeting people in English, the proper spelling 
> of it is "Hi!" not "High!"

Of course I know... but I use "High!" deliberately as a kind of oxymoron to 
indicate my preferred state of mind , which is, by the way, not induced by 
any chemical additives, but by such mindblowing things like raytracing, 
electronic organs, extrasolar planets and Afghanistan!

See you in Khyberspace!

Yadgar

Post a reply to this message

High!

Sven Littkowski wrote:

> I believe you would be well-advised to consult photos of asteroid or moons 
> surfaces to get some good ideas. Surfaces are not really plain but bear tiny 
> difference in surface heights and surface roughness.

Yes, I've seen this on close-up NEAR photos of (433)Eros... and I had 
some headache to find out how to add such small-scale noise to the 
surface! Should I try noise_generator_3d rather than noise_3d?

See you in Khyberspace!

Yadgar

Post a reply to this message

> High!
> 
> DJ Wiza wrote:
> 
>> For future reference, when greeting people in English, the proper 
>> spelling of it is "Hi!" not "High!"  They are both pronounced the 
>> same, but the meanings are completely different.
> 
> Of course I know... but I deliberately use "High!" as a little pun to 
> indicate my favourite state of mind... which, by the way, is not induced 
> by hemp products, but by such mindblowing things and experiences like 
> raytracing, electronic organs, extrasolar planets, swamp mud orgies and 
> Afghanistan!
> 
> See you in Khyberspace!
> 
> Yadgar
> 
> P. S. Micro$oft Outlook Express sucks!!! I tried to post this already 
> this afternoon (CET) from a workplace computer, which of course runs 
> under Windoze... but it didn't show up at all on the server!
Still using Thunderbird 1.0! That's an antique. Newest is 1.5.0.4.
Both of your messages made it trough.

-- 
Alain
-------------------------------------------------
Never delay the ending of a meeting or the beginning of a cocktail hour.

Post a reply to this message

Dr. Egon Kasper wrote:
> High! ;-))
> 
>>For future reference, when greeting people in English, the proper spelling 
>>of it is "Hi!" not "High!"
> 
> Of course I know... but I use "High!" deliberately as a kind of oxymoron to 
> indicate my preferred state of mind , which is, by the way, not induced by 
> any chemical additives,

'Lo...     ;-)

I always understood your "High" that way.

>                          but by such mindblowing things like raytracing, 
> electronic organs, extrasolar planets and Afghanistan!
> 
> See you in Khyberspace!
> 
> Yadgar 

Besides raytracing and computers, I've always been particularly 
fascinated by pipe organs -- both classic/church style and theater style 
-- and their appropriate music.  (These pipe organ styles are quite 
different both in their sound and in their construction.)

      -=- Larry -=-

Post a reply to this message