I've finally got my planetary model working! Proper accurate topographic
information, geoid compensation, the lot. Looks good, too.

Now I want to construct a map of my moon (which was one of the reasons
behind the whole project). Ideally I want to produce something like a
spherical imagemap --- they're similar enough to a classic Mercator map
to be usable. Originally I was hoping that the spherical camera would
help here, but it turns out I'd misread it and instead of being a
spherical camera looking *inwards* it's a spherical camera looking
*outwards*.

I could place the camera at the centre of my moon, but unfortunately due
to the way I'm treating the surface (an isosurface intersected with a
sphere representing sea level) that will have the effect of reversing
the land and the sea.

Having thought about it some more, I think that what I'm looking for is
a cross between an orthographic and spherical camera --- I want Povray
to cast rays radially inwards from a spherical viewing surface towards
the centre of my model. Does such a thing exist? If not, can I fake it?

-- 
┌─── ｄｇ＠ｃｏｗｌａｒｋ．ｃｏｍ ─────
http://www.cowlark.com ─────
│
│ "Sufficiently advanced incompetence is indistinguishable from
│ malice." -- Vernon Schryver

Post a reply to this message

David Given <dg### [at] cowlarkcom> wrote:
> I could place the camera at the centre of my moon, but unfortunately due
> to the way I'm treating the surface (an isosurface intersected with a
> sphere representing sea level) that will have the effect of reversing
> the land and the sea.

  By "intersected" do you mean using the 'intersection' CSG? Or do you
simply mean that you are using an union of the isosurface and the sphere
(the two surfaces would "intersect" with each other yes, but it's not an
intersection CSG).

  If it is just a union, then make it an intersection CSG, but invert the
sphere. That way you will get a "hollow" moon surface which you can look
from the inside (with no overlapping surface pairs from the isosurface and
the sphere).

  Another possibility is to invert the height function for the isosurface
so that everything that is pointing out will point in instead.

-- 
                                                          - Warp

Post a reply to this message

On 04/01/10 19:59, Warp wrote:
[...]
>   If it is just a union, then make it an intersection CSG, but invert the
> sphere. That way you will get a "hollow" moon surface which you can look
> from the inside (with no overlapping surface pairs from the isosurface and
> the sphere).

Yeah, but it does involve a fair bit of fiddling with the model to make
work. I'd like to avoid this if possible --- too many opportunities to
break things!

Would it be feasible to use a plain orthographic camera imaging a plane,
and then have a function pigment on the plain that used the trace()
function to manually cast rays at the model? In other words, manually
building my own camera with trace()? Would this actually work?

-- 
┌─── ｄｇ＠ｃｏｗｌａｒｋ．ｃｏｍ ─────
http://www.cowlark.com ─────
│
│ "Sufficiently advanced incompetence is indistinguishable from
│ malice." -- Vernon Schryver

Post a reply to this message

David Given wrote:
> Having thought about it some more, I think that what I'm looking for is
> a cross between an orthographic and spherical camera --- I want Povray
> to cast rays radially inwards from a spherical viewing surface towards
> the centre of my model. Does such a thing exist? If not, can I fake it?

I haven't actually tried this, but perhaps you could put a spherical 
camera at the center of your moon, put everything in your scene into an 
object{...} construct with the no_image attribute, and then put a large 
perfectly reflective sphere centered at your camera around the scene?

Post a reply to this message

Kevin Wampler wrote:
> David Given wrote:
>> Having thought about it some more, I think that what I'm looking for is
>> a cross between an orthographic and spherical camera --- I want Povray
>> to cast rays radially inwards from a spherical viewing surface towards
>> the centre of my model. Does such a thing exist? If not, can I fake it?
> 
> I haven't actually tried this, but perhaps you could put a spherical 
> camera at the center of your moon, put everything in your scene into an 
> object{...} construct with the no_image attribute, and then put a large 
> perfectly reflective sphere centered at your camera around the scene?
Good idea, but make sure to set max_trace_level so you don't get 
infinite internal reflection.

Post a reply to this message

CShake wrote:
> Good idea, but make sure to set max_trace_level so you don't get 
> infinite internal reflection.

Good point. Perhaps giving the large sphere the no_reflection (and 
no_shadow just to make things easier) would work too.

Post a reply to this message

CShake <cshake+pov### [at] gmailcom> wrote:
> Good idea, but make sure to set max_trace_level so you don't get 
> infinite internal reflection.

  When have you got infinite internal reflection?

-- 
                                                          - Warp

Post a reply to this message

David Given <dg### [at] cowlarkcom> wrote:
> On 04/01/10 19:59, Warp wrote:
> [...]
> >   If it is just a union, then make it an intersection CSG, but invert the
> > sphere. That way you will get a "hollow" moon surface which you can look
> > from the inside (with no overlapping surface pairs from the isosurface and
> > the sphere).

> Yeah, but it does involve a fair bit of fiddling with the model to make
> work. I'd like to avoid this if possible --- too many opportunities to
> break things!

  I don't think adding an 'inverse' keyword will break anything.

-- 
                                                          - Warp

Post a reply to this message

Kevin Wampler <wam### [at] uwashingtonedu> wrote:
> CShake wrote:
> > Good idea, but make sure to set max_trace_level so you don't get 
> > infinite internal reflection.

> Good point. Perhaps giving the large sphere the no_reflection (and 
> no_shadow just to make things easier) would work too.

  Don't listen to him. You can't get "infinite internal reflection" with
povray.

-- 
                                                          - Warp

Post a reply to this message

Warp wrote:
>   Don't listen to him. You can't get "infinite internal reflection" with
> povray.

I was more worried that the technique I proposed would make it 
impossible for any rays to ever get to the background, since they'd just 
be bounced by the sphere until the trace_limit was hit.  I don't think 
this would be a problem in David's case since all the rays will hit the 
moon anyway, but for a different scene it might make a difference.

Post a reply to this message