 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi
A few days ago I posted a thread about making a spiral, and decided to make
a to write a smooth_triangle mesh. The problem is it ain't smooth. As I
understand it, the normal vectors are just directional and their magnitude
doesn't matter (is this assumption correct?) so <0,1,0> has the same value
in terms of normals as <0,10,0>?
thanks
jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3d692039@news.povray.org>,
"James Taylor" <jim### [at] blueyonder couk> wrote:
> A few days ago I posted a thread about making a spiral, and decided to make
> a to write a smooth_triangle mesh. The problem is it ain't smooth. As I
> understand it, the normal vectors are just directional and their magnitude
> doesn't matter (is this assumption correct?) so <0,1,0> has the same value
> in terms of normals as <0,10,0>?
I'm not sure, it is possible the magnitude affects the weighting of the
normals when they are interpolated across the triangle. Try normalizing
the normal vectors you specify.
--
Christopher James Huff <cja### [at] earthlinknet>
http://home.earthlink.net/~cjameshuff/
POV-Ray TAG: chr### [at] tagpovrayorg
http://tag.povray.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> I'm not sure, it is possible the magnitude affects the weighting of the
> normals when they are interpolated across the triangle. Try normalizing
> the normal vectors you specify.
I don't know about the weighting, but the surface-vectors should
be one-unit in size. POV doesn't normalize them internally, and in
my experience, using other than unit-sized vector lead to strange
looking surfaces...
--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Tim Nikias <tim### [at] gmxde> wrote:
> POV doesn't normalize them internally, and in
> my experience, using other than unit-sized vector lead to strange
> looking surfaces...
Can you demonstrate this with a small example?
--
#macro N(D)#if(D>99)cylinder{M()#local D=div(D,104);M().5,2pigment{rgb M()}}
N(D)#end#end#macro M()<mod(D,13)-6mod(div(D,13)8)-3,10>#end blob{
N(11117333955)N(4254934330)N(3900569407)N(7382340)N(3358)N(970)}// - Warp -
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hm. I played around, and it seems that it doesn't
matter. I'll play around more when I've got more time,
I'll come back later...
--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde
> Can you demonstrate this with a small example?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Seems that it doesn't matter. I thought of some problems
about a year ago which I had when having a go on meshes,
but I guess that had to do with the way I calculated the
normals (pointing in weird directions), rather than the length
of the vectors...
Still, I think it would be best to keep uniform-vectors as surface
vectors. If you'd use subdivision-macros you'll see where
problems could arise... In the truest meaning of sense...
;-)
--
Tim Nikias
Homepage: http://www.digitaltwilight.de/no_lights/index.html
Email: Tim### [at] gmxde
> I'll come back later...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Tim Nikias" <tim### [at] gmxde> wrote in message
news:3d6cc478@news.povray.org...
> Seems that it doesn't matter. I thought of some problems
> about a year ago which I had when having a go on meshes,
> but I guess that had to do with the way I calculated the
> normals (pointing in weird directions), rather than the length
> of the vectors...
this is the funny thing, I -know- the normals are correct, I 've drawn them
in using cylinders, but I still see the triangles. for the moment I've given
up on figuring it out and gone for the brute force method - increase the
number of triangles 'til you can't see them.
jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
