I'm fairly sure that this has been reported before, but I can't see it mentioned
in the most recent known bugs list. The red box specified last in the following
code will disappear unless either Top_BC or Bottom_BC (or both) is/are commented
out of the final union.

The problem disappears with -UV

#version 3.5;

#local Cam_Dist = 5;
camera {
  orthographic
  location 0
  right x*Cam_Dist * 4/3
  up y*Cam_Dist
  look_at<0,0,1>
}

#declare TopBack_BC =
difference{
  sphere{0,1 scale <2,0.5,1.5>}
  plane{x,0}
}

#declare Top_BC =
difference{
  object{TopBack_BC}
  plane{y,0}
}

#declare Bottom_BC =
difference{
  object{TopBack_BC scale<1,0.5,1>}
  plane{-y,0}
}

union{
  object{Top_BC} // comment either this or next out to get red box (below)
  object{Bottom_BC}
  pigment{rgb 1}
  finish{ambient 1}
  translate z*4
}

box{0,<1.5,0.25,1.5> translate<0,-0.5,-0.75> pigment{rgb<1,0,0>} finish{ambient
1}}

Post a reply to this message

On Mon, 25 Feb 2002 13:47:11 -0000, "Tom Melly" <tom### [at] tomandlucouk> wrote:

> I'm fairly sure that this has been reported before, but I can't see it mentioned
> in the most recent known bugs list. The red box specified last in the following
> code will disappear unless either Top_BC or Bottom_BC (or both) is/are commented
> out of the final union.
>
> The problem disappears with -UV

I found similiar but very strange problem today with sphere_sweeps. 

Here is scene to test this, look at comments:

camera{orthographic}

light_source{-100*z 1}

#local Show_Example=0; // render with 0..7, for 7 set -UV

#macro First(T)
  sphere_sweep{
    catmull_rom_spline 7
    <-0.50,0.25,0.50>.002
    <-0.25,0.00,0.50>.002
    < 0.00,0.25,0.50>.002
    <-0.25,0.50,0.50>.002
    <-0.50,0.25,0.50>.002
    <-0.25,0.00,0.50>.002
    < 0.00,0.25,0.50>.002
    pigment{color rgb 1}
    translate T*.5
  }
#end

#switch(Show_Example)
  #case(0)
    First(0)  // appear
  #break
  #case(1)
    First(x)  // appear
  #break
  #case(2)
    First(-y)  // appear
  #break
  #case(3)
    First(0)  // appear
    First(x)  // appear
  #break
  #case(4)
    First(0)  // appear
    First(-y)  // appear
  #break
  #case(5)
    First(x)  // appear
    First(-y)  // appear
  #break
  #case(6)
    First(0)  // disappear
    First(x)  // disappear
    First(-y)  // disappear
  #break
  #case(7)
    // with -UV
    First(0)  // appear
    First(x)  // appear
    First(-y)  // appear
  #break
#end

Any confirmation ?

POV 3.5 b 11 icl on WinNT Sp 6 PII 233 with 128 MB

ABX

Post a reply to this message

news:r6hk7usts1qgik8fiohe5r4gvb937hnj9e@4ax.com...

Hey! Get your own bug-report! ;)

Post a reply to this message

On Mon, 25 Feb 2002 15:27:32 -0000, "Tom Melly" <tom### [at] tomandlucouk> wrote:
> Hey! Get your own bug-report! ;)

I have a lot and I don't need another one ;)

ABX

Post a reply to this message

"Tom Melly" <tom### [at] tomandlucouk> wrote in message
news:3c7a4060@news.povray.org...

<snip>

Anyone want to confirm this/point out I'm an idiot for writing bad code/point
out I'm an idiot for reporting a known bug/other ?

Post a reply to this message

Tom Melly <tom### [at] tomandlucouk> wrote:
> Anyone want to confirm this/point out I'm an idiot for writing bad code/point
> out I'm an idiot for reporting a known bug/other ?

  Wasn't the effect of disappearing objects with orthographic camera a known
issue (which was related to the fact that with orthographic camera the rays
don't start from the camera location but from the projection plane)?

-- 
#macro N(D)#if(D>99)cylinder{M()#local D=div(D,104);M().5,2pigment{rgb M()}}
N(D)#end#end#macro M()<mod(D,13)-6mod(div(D,13)8)-3,10>#end blob{
N(11117333955)N(4254934330)N(3900569407)N(7382340)N(3358)N(970)}//  - Warp -

Post a reply to this message

On 26 Feb 2002 06:49:15 -0500, Warp <war### [at] tagpovrayorg> wrote:
>  Wasn't the effect of disappearing objects with orthographic camera a known
> issue (which was related to the fact that with orthographic camera the rays
> don't start from the camera location but from the projection plane)?

At least in my case it can be. What is current state of this problem ? Is this
a bug ? Design hole ? Feature ?
I wonder why -UV influences this.

ABX

Post a reply to this message

"Warp" <war### [at] tagpovrayorg> wrote in message news:3c7b763a@news.povray.org...
>
>   Wasn't the effect of disappearing objects with orthographic camera a known
> issue (which was related to the fact that with orthographic camera the rays
> don't start from the camera location but from the projection plane)?
>

That was my (very vague) recollection. Is this now considered a known problem,
rather than a bug? I couldn't see it on the bug list.

Post a reply to this message

news:9vtm7usbihgpqca9of5cshv7ieh94iacro@4ax.com...
>
> I wonder why -UV influences this.
>

IMHO POV thinks the objects are not visible, so doesn't render them. With the UV
flag, POV renders all objects whether visible or not, correcting the bug.

Hmm, if this bug cannot be resolved, is it possible to have POV automatically
render with "-UV" if the camera is orthographic (unless the user specifically
sets "UV")? If not, should something be added to the docs to warn users to use
"-UV" with ortho. to avoid problems (or if they experience problems)?

Post a reply to this message

On Tue, 26 Feb 2002 13:04:20 -0000, "Tom Melly" <tom### [at] tomandlucouk> wrote:
> IMHO POV thinks the objects are not visible, so doesn't render them. With the UV
> flag, POV renders all objects whether visible or not, correcting the bug.

Note UV flag turns vista buffer, right ? Documentation says it is "(...)
projecting the bounding box hierarchy onto the screen and determining the
rectangular areas that are covered by each of the elements in the hierarchy."
And it is correct with my case. Projection (screen) is parallel to xy and
cross point <0,0,1>. my object has z coordinates lower than 1 so vista buffer
exclude them. IIRC vista buffer is off by default when there is less than
three objects. So when I use two or one sphere sweep then only raw
intersection is checked. But then what Warp said? "(...) with orthographic
camera the rays don't start from the camera location but from the projection
plane". Then my sphere_sweeps shouldn't appear in any case.

ABX

Post a reply to this message