|
 |
ok, ...
I think I've got a look for the cobblestones I can live with.
Some of you suggested real displacement, and as it was already on my list I
implemented it as a bulge across the road in addition to a bump layer (both can
be scaled separately). Added also curb stones.
consequently I still have to do:
- drains
- sidewalks
- debris and weed ;-)
- and .... THE REST OF THE TOWN
but this will have till tomorrow
... dave
Post a reply to this message
Attachments:
Download 'cobblestone_test4.jpg' (129 KB)
Preview of image 'cobblestone_test4.jpg'
|
|
|
|
David El Tom wrote:
> ok, ...
> I think I've got a look for the cobblestones I can live with.
> Some of you suggested real displacement, and as it was already on my
> list I implemented it as a bulge across the road in addition to a bump
> layer (both can be scaled separately). Added also curb stones.
>
> consequently I still have to do:
>
> - drains
> - sidewalks
> - debris and weed ;-)
> - and .... THE REST OF THE TOWN
>
> but this will have till tomorrow
>
> ... dave
>
> ------------------------------------------------------------------------
>
I like the stones in the very near foreground. (The ones that have the
spongy texture. How were they created?)
--
~Mike
Things! Billions of them!
Post a reply to this message
|
|
|
|
Mike Raiford wrote:
>
> I like the stones in the very near foreground. (The ones that have the
> spongy texture. How were they created?)
>
thanks,...
If I've got it right you mean the curbstones. The Idea was that they shouldn't
have this polished look like the cobble stones.
All stones are made via a template isosurface function which is then altered
randomly by the placing macro.
for the cobblestone:
#local COBBLESTONE_FUNC =
function
{
// BASE SHAPE
- f_superellipsoid(x,y,z,0.2,0.2) *5
// PILLOWED LOOK
- 1*cos(x*z)
#if (!(PREVIEW))
// STRUCTURE
+ 0.3*f_ridged_mf(3*(x+10*RAND1),3*(y+10*RAND2),3*(z+10*RAND3), 0.8, 4, 10,
.01, 10, 3)
#end
}
and for the curbstones:
#local CURBSTONE_FUNC =
function(x,y,z)
{
// BASE SHAPE
- f_superellipsoid(x,y/2,z/5,0.3,0.3)
#if (!(PREVIEW))
// STRUCTURE
+ .2* f_ridged_mf(x+20*RAND1,y+20*RAND2,z+20*RAND3,.2,10,12,0,6,3)
#end
}
The RAND1 ... RAND3 values are locals holding pseudo random numbers to make each
stone individual. They have to be predefined as passing rand() in function
invocation is not valid. I used the standard threshold of 0. Max Gradient was
something between 5 and 8, so it renders quit fast: ~3000 isosurfaces; standard
antialaising; 1000X500 -> ~35min on P4@3GHz/1GB.
These templates proofed to me to be quite flexible, as the superellipsoid as
base shape can be from something spherical over cylindrical to completely box
shaped. The ridged multi fractal on the other hand gives you a good control over
the surface structure.
Sometimes it is a little bit of work to find a good parameter set.
So, try it out, as I think that these functions are a good starting point for
doing any kind of stones,
.... dave
Post a reply to this message
|
|
|
|
David El Tom <dav### [at] t-online de> wrote:
> All stones are made via a template isosurface function which is then altered
> randomly by the placing macro.
The other thing you could do is to vary the colouring slightly from stone to
stone, maybe a 20-30% grey variation. Should be trivial to implement, and
should look very effective!
Very nice image though, I especially like the bulging displacement... :)
Bill
Post a reply to this message
|
|
|
|
Bill Pragnell wrote:
> David El Tom <dav### [at] t-onlinede> wrote:
>> All stones are made via a template isosurface function which is then altered
>> randomly by the placing macro.
> The other thing you could do is to vary the colouring slightly from stone to
> stone, maybe a 20-30% grey variation. Should be trivial to implement, and
> should look very effective!
>
> Very nice image though, I especially like the bulging displacement... :)
>
> Bill
>
thanks ...
the color variation is already implemented. Viewed from this particular angle it
is hard to see, 'cause the specular highlights are to dominant, but if you look
at the former post you can see it better.
... dave
Post a reply to this message
|
|
|
|
RusHHouR nous apporta ses lumieres en ce 08/03/2006 23:38:
> Yes, it is better, and I cant wait to see the outcome of the entire city
> scene.
> Will it be a view from ancient Rome? =)
>
>
>
>
Romans tended to use larger flagstones instead of cobblestones. Those where mostly
circular and oval
shaped, with smaller ones used to fill in the gaps. Cobblestones paving is much more
recent: middle
ages to mid twentith century.
--
Alain
-------------------------------------------------
WARNING: The consumption of alcohol may lead you to think people are laughing WITH
you.
Post a reply to this message
|
|
