Latest wip here with more heavily transformed rocks (1 mesh2 object), an 
eagle, a media in the middleground. Total render time about 14.5 hours 
altogether.

Thomas

Post a reply to this message

Attachments:
Download 'Journey to an Unknown Region_06.jpg' (206 KB)

Preview of image 'Journey to an Unknown Region_06.jpg'

The rocks look great!  And the media addition is a nice touch, too.  Something I
didn't mention before: I really like the ice-like glints on the snow, as if it's
slightly frozen over.

The texture or pigment on the closer, steeper mountain sides still doesn't look
quite right to my eye--something about those little gray(?) speckles. Or perhaps
the surface detail there is still too smooth.

Ken

Post a reply to this message

Kenneth wrote:
> The rocks look great!  And the media addition is a nice touch, too.  Something I
> didn't mention before: I really like the ice-like glints on the snow, as if it's
> slightly frozen over.
> 
> The texture or pigment on the closer, steeper mountain sides still doesn't look
> quite right to my eye--something about those little gray(?) speckles. Or perhaps
> the surface detail there is still too smooth.
> 
> Ken
> 
> 
> 
Seconded.  Mysterious and surprising as nature can be, that glass-like 
sheer face in the foreground just doesn't carry.  Especially with all 
the evidence of freeze-melt cycle around it.  It is the only element in 
the picture I don't find believable.

Post a reply to this message

"Kenneth" <kdw### [at] earthlinknet> schreef in bericht 
news:web.4b2e3b80242e0a9065f302820@news.povray.org...
> The rocks look great!  And the media addition is a nice touch, too. 
> Something I
> didn't mention before: I really like the ice-like glints on the snow, as 
> if it's
> slightly frozen over.
>
> The texture or pigment on the closer, steeper mountain sides still doesn't 
> look
> quite right to my eye--something about those little gray(?) speckles. Or 
> perhaps
> the surface detail there is still too smooth.
>

To Ken and Jim, Thanks a lot indeed.

I agree about the rockface in the foreground. It is frustratingly not coming 
out as I want it to be. The reason is of course that the height_field is too 
smooth at that distance/angle, which cannot be entirely compensated by a 
texture. What I need to do is follow Robert McGregor's tip about adding 
irregularities to the mountain in the foreground. I can also scale them down 
with distance and making them merge with the rocks.

The snow/ice is all to be credited to a nice snow texture by Gail Shaw, she 
presented here a couple of years ago:

//snow texture by Gail Shaw:
#declare T_Snow_close=
texture {
  pigment {rgb <0.95,0.95,1>}
  finish {
    ambient 0.0
    diffuse 1.0
    reflection {0,0.2}
    specular 0.6 roughness 0.008
  }
  normal {granite 0.1 scale 0.01}
}

For the snow in the distance, I simply commented out the normal.

Thomas

Post a reply to this message

"Thomas de Groot" <tDOTdegroot@interDOTnlANOTHERDOTnet> wrote:

> I agree about the rockface in the foreground. It is frustratingly not coming
> out as I want it to be. The reason is of course that the height_field is too
> smooth at that distance/angle, which cannot be entirely compensated by a
> texture.

Another, more radical idea might be to make that foreground mountain a separate
height_field, with its own (finer) detail. Of course, visually blending that one
into your main HF could be a tricky (if not impossible) chore.

Another idea: Perhaps apply an image_map overlay of some kind--one with finer
'detail' so to speak-- to the nearer mountain slope, to give it some visual
variation.

As you can probably tell, I have no problem 'cheating' the structural make-up of
a scene, if I can make it look the way I want.

Ken

Post a reply to this message

"Kenneth" <kdw### [at] earthlinknet> schreef in bericht 
news:web.4b2f8289242e0a9065f302820@news.povray.org...
> Another, more radical idea might be to make that foreground mountain a 
> separate
> height_field, with its own (finer) detail. Of course, visually blending 
> that one
> into your main HF could be a tricky (if not impossible) chore.
>
> Another idea: Perhaps apply an image_map overlay of some kind--one with 
> finer
> 'detail' so to speak-- to the nearer mountain slope, to give it some 
> visual
> variation.
>
> As you can probably tell, I have no problem 'cheating' the structural 
> make-up of
> a scene, if I can make it look the way I want.

Your two ideas are interesting to pursue further. The first would be 
difficult to get really right, but I shall nevertheless give it a thought. 
The second one could be simpler to do. I have to dig into my images (or make 
one; or look for one) that would be appropriate. I could also use Jaime's 
proceduralize macro to blend it better with the scene...

Thanks for the suggestions, Ken. You have put in motion a couple of 
interesting challenges  :)

Thomas

Post a reply to this message

Thomas de Groot wrote:

> Your two ideas are interesting to pursue further. The first would be 
> difficult to get really right, but I shall nevertheless give it a thought. 

I just played a bit with adding sub-height_fields
on top of a main height_field and came up with this:

// *** SUB_HF *** Add detail to part of another height_field
//
// F_BASE:    Function describing the main height field
// F_SUB:     Structure of the overlaid subfield.
// F_MERGE:   Function for merging the seams. This should be smooth
//            and evaluate to 0 on the border of the unit square.
// SUB_SCALE: Strength of overlaid substructure for convenience
// SUB_?_???: Rectangle specifying the portion of the main field.
#macro SUB_HF(F_BASE,F_SUB,F_MERGE,SUB_SCALE,
               SUB_X_MIN,SUB_X_MAX,SUB_Z_MIN,SUB_Z_MAX)

   #local sub_x_min = SUB_X_MIN;
   #local sub_x_max = SUB_X_MAX;
   #local sub_z_min = 1-SUB_Z_MIN;
   #local sub_z_max = 1-SUB_Z_MAX;
   #local sub_x_ext = sub_x_max - sub_x_min;
   #local sub_z_ext = sub_z_max - sub_z_min;

   height_field
   {
     function 512, 512
     {
       0.5*f1(sub_x_min + x*sub_x_ext,sub_z_min+y*sub_z_ext) +
       0.5*SUB_SCALE*f_merge(x,y) * f2(x,y)
     }
     scale     <sub_x_ext,2,sub_z_ext>
     translate <sub_x_min,0.001,1-sub_z_max>
   }

#end



Sample usage:

#declare f1 = function(x,z) {f_bozo(x*10,0,z*10)};
#declare f2 = function(x,z) {f_bozo(x*50,0,z*50)};
#declare f_merge = function(x,z){f_spherical(2*(x-0.5),0,2*(z-0.5))};

union
{
   height_field {function 512, 512 {f1(x,y)}}
   object {SUB_HF(f1,f2,f_merge,0.3,0.6,0.7,0.1,0.3) pigment {color Red}}
   object {SUB_HF(f1,f2,f_merge,0.2,0.2,0.3,0.6,0.8) pigment {color 
Red}}
   scale 0.5 * y
   pigment {color White}
}

Post a reply to this message

Thanks Christian, I have to play with this a while to see if it works...

Thomas

Post a reply to this message

High!

Thomas de Groot wrote:
> Latest wip here with more heavily transformed rocks (1 mesh2 object), an 
> eagle,

Could it be Gwaihir carrying Gandalf above Mt. Caradhras? Sorry, but I'm 
somewhat "Tolkienized" at the moment...

See you in Khyberspace!

Yadgar

Now playing: Sad Songs (Elton John)

Post a reply to this message

Wow! You heard me, even before i finished to write... Even before i started 
to think...   ;-)

Sven

Post a reply to this message