 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Now the result is the same like with my original version (see
> attachment)...
I can't help but think that the torus and cylinder ends might be bogus.
Do you have some code for me to toy around with?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
clipka <ano### [at] anonymous org> wrote:
>
> > Now the result is the same like with my original version (see
> > attachment)...
>
> I can't help but think that the torus and cylinder ends might be bogus.
>
> Do you have some code for me to toy around with?
Also, an orthographic view from above could be illuminating. Even if your
angles are correct it looks like the issue could be due to a miss calculation of
the torus center or radius.
Cheers,
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 02.10.2011 00:44, schrieb clipka:
>
>> Now the result is the same like with my original version (see
>> attachment)...
>
> I can't help but think that the torus and cylinder ends might be bogus.
>
> Do you have some code for me to toy around with?
O.k., here it is:
#declare rfh = 1; // Roof railing total height
#declare ur = 0.025; // Upper rail radius
#declare or = 0.018; // Other rails' radius
#declare pr = 0.018; // Railing post radius
#declare cp = 0.08; // Rail cornerpieces' radius
#declare fh = 0.02; // Railing post foot height
#declare fpr = 0.005; // Railing post foot outer rounding radius
#declare jlt = 0.012; // Railing middle join loop thickness
#declare jlr = 0.005; // Railing middle join loop outer rounding radius
#declare jlw = 2*(jlr+pr); // Railing middle join loop width
#declare mr = 2; // Number of middle rails
#declare rvsp = rfh/(mr+1); // Vertical spacing of rails
#declare F_Shiny =
finish
{
diffuse 1
brilliance 0.6
specular 0.8
roughness 0.05
}
#declare T_Red_Metal =
texture
{
pigment { color rgb <0.8, 0, 0> }
finish { F_Shiny }
}
cylinder
{
<2.1+cp, sh+2*rfh/(mr+1), 4.1>, <4.9, sh+2*rfh/(mr+1), 4.1>, or
texture { T_Red_Metal }
}
sphere
{
// sqrt(sqr(3.6)-sqr(0.1))
// <4.9+(3.6-cos(atan(0.1/3.6))*3.6), sh+rfh/(mr+1), 4.1>, or
<4.9+(3.6-sqrt(pow(3.6, 2)-pow(0.1, 2))), sh+rfh/(mr+1), 4.1>, or
texture { T_Red_Metal }
}
difference // ground floor round cutout railing, lower rail
{
torus
{
3.6, or
}
box
{
<0, -(or+df), -(3.6+or+df)>, <3.6+or+df, or+df, 3.6+or+df>
}
box
{
<-(3.6+or+df), -(or+2*df), -(3.6+or+2*df)>, <3.6+or+2*df, or+2*df, 0>
rotate <0, degrees(atan(0.1/3.5)), 0>
}
translate <8.5, sh+rfh/3, 4>
texture { T_Red_Metal }
}
See you in Khyberspace!
Yadgar
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi(gh)!
Am 02.10.2011 02:41, schrieb Solar Mike:
> Also, an orthographic view from above could be illuminating. Even if your
> angles are correct it looks like the issue could be due to a miss calculation of
> the torus center or radius.
O.k.... with and without the spherical cornerpiece!
See you in Khyberspace!
Yadgar
Post a reply to this message
Attachments:
Download 'orthoview1.png' (66 KB)
Download 'orthoview2.png' (65 KB)
Preview of image 'orthoview1.png'
Preview of image 'orthoview2.png'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> cylinder
> {
> <2.1+cp, sh+2*rfh/(mr+1), 4.1>, <4.9, sh+2*rfh/(mr+1), 4.1>, or
~~~
Note that this does /not/ coincide with the sphere's center. You need to
apply the small correction here as well to fix this...
> sphere
> {
> // sqrt(sqr(3.6)-sqr(0.1))
> // <4.9+(3.6-cos(atan(0.1/3.6))*3.6), sh+rfh/(mr+1), 4.1>, or
> <4.9+(3.6-sqrt(pow(3.6, 2)-pow(0.1, 2))), sh+rfh/(mr+1), 4.1>, or
> texture { T_Red_Metal }
> }
> difference // ground floor round cutout railing, lower rail
> {
> torus
> {
> 3.6, or
> }
> box
> {
> <0, -(or+df), -(3.6+or+df)>, <3.6+or+df, or+df, 3.6+or+df>
> }
> box
> {
> <-(3.6+or+df), -(or+2*df), -(3.6+or+2*df)>, <3.6+or+2*df, or+2*df, 0>
> rotate <0, degrees(atan(0.1/3.5)), 0>
~~~~ ~~~
... and here you're miscalculating the angle, using both the wrong
values (should be 0.1/3.6 rather than 0.1/3.5) and the wrong formula
(should be asin rather than atan).
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
High!
Am 02.10.2011 13:13, schrieb clipka:
>
>> cylinder
>> {
>> <2.1+cp, sh+2*rfh/(mr+1), 4.1>, <4.9, sh+2*rfh/(mr+1), 4.1>, or
> ~~~
> Note that this does /not/ coincide with the sphere's center. You need to
> apply the small correction here as well to fix this...
Yes, that's obvious...
>> <-(3.6+or+df), -(or+2*df), -(3.6+or+2*df)>, <3.6+or+2*df, or+2*df, 0>
>> rotate <0, degrees(atan(0.1/3.5)), 0>
> ~~~~ ~~~
> ... and here you're miscalculating the angle, using both the wrong
> values (should be 0.1/3.6 rather than 0.1/3.5) and the wrong formula
> (should be asin rather than atan).
Oh...
After your suggested corrections, it looks almost perfect... but there
are still those parallel rows of dots, just like there were
coincindental surfaces!
See you in Khyberspace!
Yadgar
Post a reply to this message
Attachments:
Download 'railing_corner2.png' (25 KB)
Preview of image 'railing_corner2.png'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I think the problem here lies in the assumption the use of 0.1, 3.5 and 3.6 as
the basis for triangle with the center of the torus, the bend in the rail.
Keeping the radius and 0.1, your delta X is more like 3.5986. I think you will
be fine if you follow clipka's suggestion but the change you are calculating
should be subtracted from 4.9. Also, I tend to have better success with these
things when I declare local points or values and share that between the objects
like:
#local bendX = 8.5 - sqrt(pow(3.6, 2)-pow(0.1, 2));
or
#local pBend = <8.5 - sqrt(pow(3.6, 2)-pow(0.1, 2)), sh+2*rfh/(mr+1), 4.1>;
Cheers,
Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> After your suggested corrections, it looks almost perfect... but there
> are still those parallel rows of dots, just like there were
> coincindental surfaces!
>
> See you in Khyberspace!
>
> Yadgar
Is it possible that those parallel rows of dots are from the railing on the
other side?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Please ignore my comment about subtracting below.... I really should not post
with a head cold.
"Solar Mike" <wem### [at] yahooca> wrote:
> I think the problem here lies in the assumption the use of 0.1, 3.5 and 3.6 as
> the basis for triangle with the center of the torus, the bend in the rail.
>
> Keeping the radius and 0.1, your delta X is more like 3.5986. I think you will
> be fine if you follow clipka's suggestion but the change you are calculating
> should be subtracted from 4.9. Also, I tend to have better success with these
> things when I declare local points or values and share that between the objects
> like:
>
> #local bendX = 8.5 - sqrt(pow(3.6, 2)-pow(0.1, 2));
> or
> #local pBend = <8.5 - sqrt(pow(3.6, 2)-pow(0.1, 2)), sh+2*rfh/(mr+1), 4.1>;
>
> Cheers,
> Mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi(gh)!
On 02.10.2011 23:01, Solar Mike wrote:
> Is it possible that those parallel rows of dots are from the railing on the
> other side?
As the view is from the roof's edge inwards and the texture does not
contain reflection, this is not possible!
See you in Khyberspace!
Yadgar
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
