Hi,

I went searching for a macro to round out the corners of linear 
sphere_sweeps, but was not successful. So I made my own :) It's easy to 
use: you just provide the spline points (and optionally the arc radii) 
for each point. It automatically joins the first and last points if they 
are the same. Since the object is made out of cylinders, tori and discs, 
the render time is pretty good.

The include file and an example scene are available over at p.t.scene-files:
http://news.povray.org/povray.text.scene-files/thread/%3C4c3d20db%40news.povray.org%3E/

Happy Ray Tracing!

Sam

Post a reply to this message

Attachments:
Download 'rounded_lsweep_render.jpg' (114 KB)

Preview of image 'rounded_lsweep_render.jpg'

You did it again, Sam! Congratulations and thank you indeed!

Something for the Object Collection, no doubt.

Thomas

Post a reply to this message

Thomas de Groot wrote:
> You did it again, Sam! Congratulations and thank you indeed!
> 
> Something for the Object Collection, no doubt.


Cool! Thanks for looking beyond the horribly sterile image :) The macro 
does not perform any bounds checking to properly fit the arcs to the 
segments. Doing so would have been too complex. The current behavior 
actually gives you more control, I think.

Can you think of a better name than "rounded_lsweep"? I tried to choose 
a name that would conform to the Object Collection's naming convention, 
just in case I decided to post it there.

I can already think of a couple more useful macros based on this 
technique. Hopefully I can make progress on at least one of them, to 
finish a modeling project.

Post a reply to this message

stbenge wrote:
> Thomas de Groot wrote:
>> You did it again, Sam! Congratulations and thank you indeed!
>>
>> Something for the Object Collection, no doubt.
> 
> The macro does not perform any bounds checking to properly fit the arcs to the 
> segments.

To clarify, the arc segments *do* conform to the angles as they should. 
But if they are too large they will extend past the arcs from another 
corner, so the arc radius will not automatically adjust itself to keep 
this from happening... which amounts to you having to give the offending 
corner a smaller arc radius. I hope that makes sense. As usual I didn't 
express myself properly :(

Post a reply to this message

"stbenge" <myu### [at] hotmailcom> schreef in bericht 
news:4c3dd9af$1@news.povray.org...
> Cool! Thanks for looking beyond the horribly sterile image :) The macro 
> does not perform any bounds checking to properly fit the arcs to the 
> segments. Doing so would have been too complex. The current behavior 
> actually gives you more control, I think.

It looks very reasonable to me indeed and useful for most cases I am sure. 
Purists might want something more sophisticated, but generalist like me are 
very happy with it!

>
> Can you think of a better name than "rounded_lsweep"? I tried to choose a 
> name that would conform to the Object Collection's naming convention, just 
> in case I decided to post it there.

This seems acceptable to me, simple and descriptive. I cannot come up with 
something better...

>
> I can already think of a couple more useful macros based on this 
> technique. Hopefully I can make progress on at least one of them, to 
> finish a modeling project.

Waiting impatiently for further contributions ;-)

Thomas

Post a reply to this message

"stbenge" <myu### [at] hotmailcom> schreef in bericht 
news:4c3ddd79@news.povray.org...
> To clarify, the arc segments *do* conform to the angles as they should. 
> But if they are too large they will extend past the arcs from another 
> corner, so the arc radius will not automatically adjust itself to keep 
> this from happening... which amounts to you having to give the offending 
> corner a smaller arc radius. I hope that makes sense. As usual I didn't 
> express myself properly :(

Clear enough for me, Sam. Good to keep in mind. I see that you have put this 
in your caveats.

Thomas

Post a reply to this message

Thomas de Groot wrote:
> "stbenge" <myu### [at] hotmailcom> schreef in bericht 
> news:4c3dd9af$1@news.povray.org...
>> Cool! Thanks for looking beyond the horribly sterile image :) The macro 
>> does not perform any bounds checking to properly fit the arcs to the 
>> segments. Doing so would have been too complex. The current behavior 
>> actually gives you more control, I think.
> 
> It looks very reasonable to me indeed and useful for most cases I am sure. 
> Purists might want something more sophisticated, but generalist like me are 
> very happy with it!

Glad to hear it. The purists can be such snobs sometimes ;)

>> I can already think of a couple more useful macros based on this 
>> technique. Hopefully I can make progress on at least one of them, to 
>> finish a modeling project.
> 
> Waiting impatiently for further contributions ;-)

It all depends on how stuck I get. I found myself stumped for hours 
yesterday on a single problem. I finally found the solution by analyzing 
the problem carefully, but now I need the implementation... I ended up 
learning the basics of Coordinate Geometry to figure out the problem, 
since I was never taught higher math in school... I can make a line and 
find the y-intercept, but now I need to find it for three dimensions. If 
I take it a step at a time, the pieces should fall into place like they 
did for rounded_lsweep.

Post a reply to this message

stbenge wrote:

> I can make a line and find the y-intercept, but now I need to find it
>  for three dimensions.

If you have a problem just ask in the newsgroups and I'm sure
people will be queueing up for the honor of answering ;)

Post a reply to this message

Christian Froeschlin wrote:
> stbenge wrote:
> 
>> I can make a line and find the y-intercept, but now I need to find it
>>  for three dimensions.
> 
> If you have a problem just ask in the newsgroups and I'm sure
> people will be queueing up for the honor of answering ;)

I know, I know. And it's appreciated! Usually I try to figure these 
things out for myself, asking questions only when I'm stuck. Which I'm 
not this time, because right when I decided to start hammering out a 
method for 3D line/plane intersection, a thought occurred to me: Luke, 
use the trace(). So now I'm unstuck :D

Post a reply to this message

stbenge wrote:
> method for 3D line/plane intersection, 

I thought a cheat-sheet kind of thing for this sort of basic geometric 
equations would be handy.  A book that goes into the basics of the geometric 
math in a way that's easy to look things up, for (say) inside/outside calcs, 
intersections of various simple shapes, calculating angles in various 
situations, and so on.

-- 
Darren New, San Diego CA, USA (PST)
    C# - a language whose greatest drawback
    is that its best implementation comes
    from a company that doesn't hate Microsoft.

Post a reply to this message