All of my aircraft models have been affected by something in MegaPov. I use
a lot of differences in order to apply images to given portions of the
planes, clip out panels, etc., and I keep seeing extraneous portions of the
difference object.

To test this I took a simple sphere mesh, and subtracted out a  box
(non-mesh). A portion of the box remains within the scene. I could post the
code, but I am sure you can reproduce this. I will post sample images, if
you request it (to binaries.images, of course).


--
GrimDude
gri### [at] isonet

Post a reply to this message

GrimDude wrote:
> All of my aircraft models have been affected by something in MegaPov. I use
> a lot of differences in order to apply images to given portions of the
> planes, clip out panels, etc., and I keep seeing extraneous portions of the
> difference object.

Are you saying these models worked in official POV? Because by 
rights they probably shouldn't have. Meshes never used to be 
reliably usable for CSG other than unions because they had no 
well defined insides and outsides, and if differencing other 
shapes from them worked at all, it proabably acted like clipping. 
MegaPOV changed this by giving meshes "insides" so differencing 
from a mesh will behave in a fundamentally different way in 
MegaPOV, but that most likely isn't a bug you're seeing. You
might have to substitute clipped_by for difference to get the
same effect now.

> To test this I took a simple sphere mesh, and subtracted out a  box
> (non-mesh). A portion of the box remains within the scene. 

Is it fundamentally different from when you do the same difference
of a box from a regular sphere primitive? If not, what you're
seeing is the difference "solid" meshes (new in MegaPOV) makes
on the operation. 

Charles

Post a reply to this message

Nope.

I have used this technique since the introduction of the mesh object.

#declare Ball = mesh { smooth_triangle.... }

difference {
   object { Ball }
   object { box ... } }

This worked in the official release, in SuperPov, but not with MegaPov. :(

Actually, it still works. There are, however, extraneous objects showing up.

--
GrimDude
gri### [at] isonet

Post a reply to this message

What happens if you use 'inverse' for either the Ball or the box or the
difference itself.

Bob

"GrimDude" <gri### [at] netfreecom> wrote in message
news:387eb79c@news.povray.org...
> Nope.
>
> I have used this technique since the introduction of the mesh object.
>
> #declare Ball = mesh { smooth_triangle.... }
>
> difference {
>    object { Ball }
>    object { box ... } }
>
> This worked in the official release, in SuperPov, but not with MegaPov. :(
>
> Actually, it still works. There are, however, extraneous objects showing up.
>
> --
> GrimDude
> gri### [at] isonet
>
>

Post a reply to this message

On Thu, 13 Jan 2000 23:43:50 -0600, GrimDude wrote:
>Nope.
>
>I have used this technique since the introduction of the mesh object.
>
>#declare Ball = mesh { smooth_triangle.... }
>
>difference {
>   object { Ball }
>   object { box ... } }
>
>This worked in the official release, in SuperPov, but not with MegaPov. :(
>
>Actually, it still works. There are, however, extraneous objects showing up.

It shouldn't work in any of them, with the possible exception of MegaPOV.  
Parts of the box that are properly inside the ball should still show up.  
If it did work, it was a lucky coincidence.

-- 
These are my opinions.  I do NOT speak for the POV-Team.
The superpatch: http://www2.fwi.com/~parkerr/superpatch/
My other stuff: http://www2.fwi.com/~parkerr/traces.html

Post a reply to this message

On 14 Jan 2000 08:28:27 -0500, Ron Parker wrote:
>On Thu, 13 Jan 2000 23:43:50 -0600, GrimDude wrote:
>>Nope.
>>
>>I have used this technique since the introduction of the mesh object.
>>
>>#declare Ball = mesh { smooth_triangle.... }
>>
>>difference {
>>   object { Ball }
>>   object { box ... } }
>>
>>This worked in the official release, in SuperPov, but not with MegaPov. :(
>>
>>Actually, it still works. There are, however, extraneous objects showing up.
>
>It shouldn't work in any of them, with the possible exception of MegaPOV.  
>Parts of the box that are properly inside the ball should still show up.  
>If it did work, it was a lucky coincidence.

er... s/inside/outside/, if you would.

-- 
These are my opinions.  I do NOT speak for the POV-Team.
The superpatch: http://www2.fwi.com/~parkerr/superpatch/
My other stuff: http://www2.fwi.com/~parkerr/traces.html

Post a reply to this message

Ron Parker wrote:
> 
> er... s/inside/outside/, if you would.

No no no.. Compile it and make a binary statemachine out of the regexp, apply to
the. . . . .

(I've been around a comp.sci. stud for far too long... he got me hooked :p )
 
-- 
//Spider    --  [ spider@bahnhof.se ]-[ http://darkmere.wanfear.com/ ]
And the devil in black dress watches over
    My guardian angel walks away
Life is short and love is always over in the morning
    Black wind come carry me far away
            --"Sisters of Mercy" -- "Temple Of Love"

Post a reply to this message

Well, so far I've come across more then 2000 lucky coincidences. :)

I worked alright. Everytime too. I know, I was warned my Ron Parker (and
Chris Colefax, too, who both recommended clipping years ago I think), that
it wasn't proper coding and shouldn't work. But, it did, so I continued the
practice. The worst part is, after I got all of the independant portions
coded I merged it with the overall code, so now I have a real mess!

Ah, well! Clipping (w/wo inverse) fixes it.

--
GrimDude
gri### [at] isonet

Post a reply to this message

The same extraneous portion shows up, but with the disected section rather
then the
host.

--
GrimDude
gri### [at] isonet
omniVERSE <inv### [at] aolcom> wrote in message
news:387f0317@news.povray.org...
> What happens if you use 'inverse' for either the Ball or the box or the
> difference itself.
>
> Bob
>
> "GrimDude" <gri### [at] netfreecom> wrote in message
> news:387eb79c@news.povray.org...
> > Nope.
> >
> > I have used this technique since the introduction of the mesh object.
> >
> > #declare Ball = mesh { smooth_triangle.... }
> >
> > difference {
> >    object { Ball }
> >    object { box ... } }
> >
> > This worked in the official release, in SuperPov, but not with MegaPov.
:(
> >
> > Actually, it still works. There are, however, extraneous objects showing
up.
> >
> > --
> > GrimDude
> > gri### [at] isonet
> >
> >
>
>

Post a reply to this message

Is there a texture (pigment at least) on the differencing object (box in this
case)?  Reason I ask is that I had a similar sounding situation a couple years
ago.

Bob

"GrimDude" <gri### [at] netfreecom> wrote in message
news:38801b38@news.povray.org...
> The same extraneous portion shows up, but with the disected section rather
> then the
> host.
>

Post a reply to this message