Hello,

I have released two new objects today: Beveled_Box(); Beveled_Rounded_Box().
They could be useful!

The objects contain no internal geometry, so they can be made transparent, and
will even hold media. They render at a fair clip, and would probably be even
faster if clipping was disabled. Unfortunately I didn't develop CSG-friendly
versions, so they can really only be used in unions for now.

The link:
http://news.povray.org/povray.text.scene-files/thread/%3Cweb.5224f564360cd137510f5afc0%40news.povray.org%3E/

Happy POVving!

Sam

Post a reply to this message

Attachments:
Download 'beveled_rounded_box11m_50s.png' (481 KB)

Preview of image 'beveled_rounded_box11m_50s.png'

On 09/02/2013 04:54 PM, Samuel Benge wrote:
> Hello,
>
> I have released two new objects today: Beveled_Box(); Beveled_Rounded_Box().
> They could be useful!
>
> The objects contain no internal geometry, so they can be made transparent, and
> will even hold media. They render at a fair clip, and would probably be even
> faster if clipping was disabled. Unfortunately I didn't develop CSG-friendly
> versions, so they can really only be used in unions for now.
>
> The link:
>
http://news.povray.org/povray.text.scene-files/thread/%3Cweb.5224f564360cd137510f5afc0%40news.povray.org%3E/
>
> Happy POVving!
>
> Sam
>
Thanks - and a very cool example image.
Bill P.

Post a reply to this message

William F Pokorny <ano### [at] anonymousorg> wrote:
> Thanks - and a very cool example image.

Thanks, Bill.

I don't know why it won't work in difference, intersection and merge statements,
though. I very carefully defined all objects' insides and outsides. Clipping
works. Proximity patterns don't :(

Sam

Post a reply to this message

"Samuel Benge" <stb### [at] hotmailcom> wrote:
> I don't know why it won't work in difference, intersection and merge statements,
> though. I very carefully defined all objects' insides and outsides. Clipping
> works. Proximity patterns don't :(
>
> Sam

I'm still guessing a bit too. But as all I can see, you have a union of a lot of
"two-dimensional" meshes which are all not closed. Only the union of this meshes
is closed. Usually csg operations work with meshes only if they are closed, have
no holes and an inside vector. You are using a collection of meshes
which all have a "hole", since they are not closed. I know the maths will be
very much harder, but try to create only one mesh putting the transforms of your
"quads" into the calculation of the vertices of the quads themselves. This will
result in a lot of duplicate vertices but may work.

Another idea is to have one (and a second for the rounded version) reference
mesh and transform the vertices within proper groups due to your desired
proportions. You can create even an uv-mapping which musn't be changed due to
the transformations.

Anyway, a very good work.

Best regards,
Michael

Post a reply to this message

"MichaelJF" <mi-### [at] t-onlinede> wrote:
>
> Another idea is to have one (and a second for the rounded version) reference
> mesh and transform the vertices within proper groups due to your desired
> proportions. You can create even an uv-mapping which musn't be changed due to
> the transformations.
>
I forgot to mention that in this case the mesh2-syntax is more helpful. Store
the vertices in an array, transform them and put them out as a mesh2. Once you
have crafted the other entries (faces indices, normals, uv-vertices, uv-indices
and so on) you must not change them.

Best regards,
Michael

Post a reply to this message

"MichaelJF" <mi-### [at] t-onlinede> wrote:
> "MichaelJF" <mi-### [at] t-onlinede> wrote:
> >
> > Another idea is to have one (and a second for the rounded version) reference
> > mesh and transform the vertices within proper groups due to your desired
> > proportions. You can create even an uv-mapping which musn't be changed due to
> > the transformations.
> >
> I forgot to mention that in this case the mesh2-syntax is more helpful. Store
> the vertices in an array, transform them and put them out as a mesh2. Once you
> have crafted the other entries (faces indices, normals, uv-vertices, uv-indices
> and so on) you must not change them.
>
> Best regards,
> Michael

Sounds like it will work. I'll give that a try... sooner or later :) Or I could
just go the same route as whoever wrote the Rounded_Box macros, and just make
union and merge versions.

Post a reply to this message