BuRnInG_IcE nous apporta ses lumieres en ce 01-02-2007 00:31:
> Alain <ele### [at] netscapenet> wrote:
>> BuRnInG_IcE nous apporta ses lumieres en ce 31-01-2007 20:52:
>>> I have this screen name on many different web site, and since I use POV-Ray,
>>> I thought "Hey! I'll make an avater in POV-Ray!"
>>> My ire is this: I can not find a good ice . . .
>>> texture/interior/normal/what-have-you. I know the ice IOR, but it still
>>> looks like glass, because it is so smooth. A texture doesn't help, nor do
>>> normals. I can't understand exactly what to do. I was hoping someone might
>>> know where source for good ice might be. Specifically, I'm looking for a
>>> natural, sorta frosty, ripply looking ice, not a ice cube look. But right
>>> now, I'm not picky. Any ice would work, and I could just fiddle with it.
>>> P.S. I got the flame worked out. Just need ice.
>> Natural ice often have air bubbles or other inclusions. Make them by using
>> several small spheres with the ior of air (1) placed inside your ice. Add some
>> opaque and dark bits "floating" inside. Differencing them can be very slow,
>> beter use an union. Another way (not tested), wrap them in a merge and diference
>> that from your block: may be faster for you then have only one bounding box.
>> Layer some whitish, partly transparent, pattern over your ice texture. Give the
>> white part a strong normal scaled small. granite may be a good start.
>> Add some dispersion.
>> If it's not the case now, try using an iso-surface. Carve some grooves in it by
>> adding some bump function, unevenly scaled.

>> --
>> Alain
>> -------------------------------------------------
>> Don't try so hard, the best things come when you least expect them to.

> Thank You. Using difference helps. Right now I have 2500 clear spheres of
> size 0.015 randomly placed inside a size 1 sphere. O_o it's working, but a
> little slow. . .

> i'll try the normal/pattern idea once i get the interior set up.



The slowness is due to the fact that all differenced bits have overlaping 
bounding box, and that every ray have to be tested against all of them. By 
merging all those small bubbles, you force them to only have 1 bounding box. 
Usualy, a merge is slower than an union, but in this case, it may be faster. 
Those been perty small, you may be able to get by using small white spheres in 
an union, this could be faster.

-- 
Alain
-------------------------------------------------
   My wife's such a bad cook, the dog begs for Alka-Seltzer.
   	Rodney Dangerfield

Post a reply to this message