> However, a quadric should do the job, if an infinite cylinder really is
> necessary. Maybe a "infinite" keyword could be added to cones and
> cylinders...all it would have to do is remove bounding and not clip the
> ends and render disks there.

Thanks, the quadrics work pretty fine!

-- 
+-------------------------+----------------------------------+
| Simon Lemieux           | Website : http://www.666Mhz.net  |
| Email : Sin### [at] 666Mhznet | POV-Ray, OpenGL, C++ and more... |
+-------------------------+----------------------------------+

Post a reply to this message

Chris Huff wrote:
> 
> In article <3A0EB1DE.C830462B@peak.edu.ee>, Margus Ramst
> <mar### [at] peakeduee> wrote:
> 
> > this too will have to be contained by an infinite object.
> 
> As I recall, the list of possible contained_by objects is very
> short...box and sphere last I heard.
> 

*ack* I meant to say "finite object"

-- 
Margus Ramst

Personal e-mail: mar### [at] peakeduee
TAG (Team Assistance Group) e-mail: mar### [at] tagpovrayorg

Post a reply to this message

Simon Lemieux wrote:
> 
> I was wondering if it was possible to simply have an infinite
> cylinder?
> Without any tricks... this should normally be trivial for the
> renderer... but was this coded?

I think this ought to work.  It uses the Reorient() macro from my
Thoroughly Useful Macros file:

#macro InfiniteCylinder(PointA,PointB,Radius)
quadric { <1,0,1>,<0,0,0>,<0,0,0>,-Radius*Radius
  Reorient(y,vnormalize(PointB-PointA))
  translate PointA
}
#end

Hope this helps,
John
-- 
ICQ: 46085459

Post a reply to this message

On Sun, 12 Nov 2000 02:59:58 -0500, Simon Lemieux
<lem### [at] yahoocom> wrote:

>Hi,
>	I was wondering if it was possible to simply have an infinite cylinder? 
>Without any tricks... this should normally be trivial for the renderer... but
>was this coded?

A quadric (anyone remember these?) should do it. Look at shapes.inc
and shapesq.inc in the standard POV-Ray package.


Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] usanet
TAG      e-mail : pet### [at] tagpovrayorg

Post a reply to this message

Peter Popov <pet### [at] usanet> wrote:
: A quadric (anyone remember these?) should do it.

  Isosurfaces killed quartics and other poly objects... RIP.

  ... Or perhaps not. You can create infinite surfaces with polys but not
with isosurfaces. Polys also are more accurate (but usually slower).

  Btw, see http://iki.fi/warp/polytutorial/

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message

On 13 Nov 2000 09:27:03 -0500, Warp <war### [at] tagpovrayorg> wrote:

>Peter Popov <pet### [at] usanet> wrote:
>: A quadric (anyone remember these?) should do it.
>
>  Isosurfaces killed quartics and other poly objects... RIP.
>
>  ... Or perhaps not. You can create infinite surfaces with polys but not
>with isosurfaces. Polys also are more accurate (but usually slower).

That's exactly what I had in mind. With all the iso frenzy, people
seem to have forgotten about the good ole infinite primitives. They
still have their uses.


Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] usanet
TAG      e-mail : pet### [at] tagpovrayorg

Post a reply to this message

Peter Popov wrote:
> 
> That's exactly what I had in mind. With all the iso frenzy, people
> seem to have forgotten about the good ole infinite primitives. They
> still have their uses.
> 

Out of curiosity, what uses?
I.e. how is a "truly infinite" quadric cylinder better than a regular cylinder
scaled so long that POV considers it infinite?

-- 
Margus Ramst

Personal e-mail: mar### [at] peakeduee
TAG (Team Assistance Group) e-mail: mar### [at] tagpovrayorg

Post a reply to this message

Margus Ramst <mar### [at] peakeduee> wrote:
: I.e. how is a "truly infinite" quadric cylinder better than a regular cylinder
: scaled so long that POV considers it infinite?

  There might be some accuracy problems if you scale an object too much
in one direction (remember that it's actually the ray that is
inverse-transformed; scaling the cylinder extremely long in one direction
means actually that you are scaling the rays inside an extremely narrow
area; might be so narrow that you get accuracy problems).
  Also specifying too big endpoints for the cylinder might cause accuracy
problems.

  These accuracy problems can be specially true if you are using an
isosurface. The bigger the contained_by object, the less accurate the
isosurface.

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message

>   These accuracy problems can be specially true if you are using an
> isosurface. The bigger the contained_by object, the less accurate the
> isosurface.

Really, this should probably be relative to the camera... ie if the camera is at
2 nanounits of the isosurface that has size of (1000, 1000, 1000)...  that
should also give the problems you mentionned, I think!...  This is probably
because povray uses a maximum of 32 bits or 64 to determine a number and if you
zoom too much you will overload that 32-64 bits number and then will have a few
precision bugs...

Is this true, I never worked with isosurfaces...

-- 
+-------------------------+----------------------------------+
| Simon Lemieux           | Website : http://www.666Mhz.net  |
| Email : Sin### [at] 666Mhznet | POV-Ray, OpenGL, C++ and more... |
+-------------------------+----------------------------------+

Post a reply to this message

On Thu, 16 Nov 2000 03:44:40 +0200, Margus Ramst <mar### [at] peakeduee>
wrote:

>Peter Popov wrote:
>> 
>> That's exactly what I had in mind. With all the iso frenzy, people
>> seem to have forgotten about the good ole infinite primitives. They
>> still have their uses.
>> 
>Out of curiosity, what uses?
>I.e. how is a "truly infinite" quadric cylinder better than a regular cylinder
>scaled so long that POV considers it infinite?

Attitude :) *I* will know that I've cheated with the latter.


Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] usanet
TAG      e-mail : pet### [at] tagpovrayorg

Post a reply to this message