I have a bit map image , http://www.cosy.com/cosycube.gif , where I've made
the background transparent ,

  #declare CCube = object { box { < 0, 0 , -.2 > , < 1 , 1 , .2 >
                             pigment { image_map { png "CoSyCubeFULL.png"
                                                   transmit 2 , 1.0 } }  }

but it appears the figure is projected only on the front and back of planes
of the box . How do I make it solid thru the thickness of the box which is
what I expected by default ?

Thanks .

Post a reply to this message

"Bob Armstrong" <bob### [at] cosycom> wrote in message 
news:web.44174d8d9ca30044945ca8690@news.povray.org...
>I have a bit map image , http://www.cosy.com/cosycube.gif , where I've made
> the background transparent ,
>
>  #declare CCube = object { box { < 0, 0 , -.2 > , < 1 , 1 , .2 >
>                             pigment { image_map { png "CoSyCubeFULL.png"
>                                                   transmit 2 , 1.0 } }  }
>
> but it appears the figure is projected only on the front and back of 
> planes
> of the box . How do I make it solid thru the thickness of the box which is
> what I expected by default ?

You've encountered a common misconception. Objects in POV are not solid, 
merely surfaces, until acted upon by other objects not just textures.

A possible solution could be to use height_field for the box-like object 
after creating a second image especially for that, and apply the image_map 
pigment onto it.
Should be fairly easy in your case since that cosycube.gif is already of the 
right form. You'll just need to do an inverse of the colors if you need the 
black part to exist instead of the white part (white is high point, black 
low). The black portion can then be set to not show up at all by using (in 
the height_field) keyword water_level followed by a value greater than 0 but 
less than 1 (as long as the white is fully white).
Oh, and scale things accordingly, of course, to match the image file's 
aspect ratio or resolution. Again, yours appears to be square already to not 
a problem. Would only need a certain depth scale. Just be sure to apply the 
image_map (if you even need one for this) with a rotate 90*x so it lays 
horizontally onto the HF.

Hope that helps.

-- 
Bob H  www.3digitaleyes.com
http://3digitaleyes.com/imagery/

Post a reply to this message

"Bob Hughes" <omniverse@charter%net> wrote:
> "Bob Armstrong" <bob### [at] cosycom> wrote in message
> news:web.44174d8d9ca30044945ca8690@news.povray.org...
> > the figure is projected only on the front and back of planes
> > of the box . How do I make it solid thru the thickness of the box which is
> > what I expected by default ?
>
> You've encountered a common misconception. Objects in POV are not solid,
> merely surfaces, until acted upon by other objects not just textures.

> A possible solution could be to use height_field for the box-like object
> after creating a second image especially for that, and apply the image_map
> pigment onto it.
>...
> Hope that helps.

Thanks. After messing around with some problem with the .png file , perhaps
because it had white set as transparent ( inverting the colors in MS paint
caused the image to disappear ) I got it to work . I'm impressed it handles
the few bit wide ridges so well .

Boy , some of the images on your web site are intimidating .
Please tell me some of the images like
http://3digitaleyes.com/imagery/povrenders1/lcsrule_std were not
constructed from scratch .

Bob A http://CoSy.com

Post a reply to this message

"Bob Armstrong" <bob### [at] cosycom> wrote in message 
news:web.4417ca9fe453bbb1945ca8690@news.povray.org...
>
> Boy , some of the images on your web site are intimidating .
> Please tell me some of the images like
> http://3digitaleyes.com/imagery/povrenders1/lcsrule_std were not
> constructed from scratch .

*blush*  Ha! That one uses image_map on what is otherwise fairly basic POV 
primitives in CSG stuff. I didn't dare try to add all the little lines and 
numerals-- maybe I could have, guess I'll never know!

Glad you liked those, getting a bit outdated now. A lot of it has to do with 
perserverence, I believe. I just never seem to get much rendered anymore, 
even less done at a resolution and quality for showing. I don't (or can't 
very well) use modeller programs, although the Seattle Space Needle does 
have a part in the concrete tripod where it curves that I used one for. Oh 
yeah, and the mountain terrain in a souple of those is height_field's from 
elevation images.

Anyway, I'm much more intimidated by many other POV-Ray users. ;)

Bob

Post a reply to this message

Hey, Bob.

Heck. The moment I pressed that send button I remembered that car at the 
4-way stop is a free model from someplace. I should move it out of the first 
place in the album but I can't help liking how it looks there. Fact is, I 
prefer to code things myself in POV almost exclusively unless there's 
something interesting somebody else has made or an item I can't readily make 
would be good for a scene. Realistic trees is another of those things I'd 
rather add than try to create.

And since I'm replying again, good to know you figured out how to get that 
logo done in POV. I wondered about those fine lines and the image size being 
okay. The bigger the image file the better.

Bob H

Post a reply to this message

"Bob Hughes" <omniverse@charter%net> wrote:
> "Bob Armstrong" <bob### [at] cosycom> wrote in message
> news:web.4417ca9fe453bbb1945ca8690@news.povray.org...
> >
> > Boy , some of the images on your web site are intimidating .
> > Please tell me some of the images like
> > http://3digitaleyes.com/imagery/povrenders1/lcsrule_std were not
> > constructed from scratch .
>
> *blush*  Ha! That one uses image_map on what is otherwise fairly basic POV
> primitives in CSG stuff. I didn't dare try to add all the little lines and
> numerals-- maybe I could have, guess I'll never know!
>
That reminds me of a linear slide rule I did not finish about two years ago.
I used a heightfield for the scales. It took some doing first I tried to
subtract the heightfield from the stock but that gave poor results. I
finally got something looking ok by using a union.

Stephen

Post a reply to this message

Bob Hughes <omniverse@charter%net> wrote:
> *blush*  Ha! That one uses image_map on what is otherwise fairly basic POV 
> primitives in CSG stuff. I didn't dare try to add all the little lines and 
> numerals-- maybe I could have, guess I'll never know!

  It should be rather trivial with some simple #while loops.

  Of course getting the slightly blurred look may prove a bit of a challenge,
but should be doable.

-- 
                                                          - Warp

Post a reply to this message