(Stupid me, got the header wrong the first time...)

There's something I don't understand about this function.

Here it is, stripped down to only reflect a non-transformed plane:

static int Intersect_Plane (RAY *Ray, PLANE *Plane, DBL *Depth)
{
  DBL NormalDotOrigin, NormalDotDirection;
  VECTOR P, D;
  Increase_Counter(stats[Ray_Plane_Tests]);

  VDot(NormalDotDirection, Plane->Normal_Vector, Ray->Direction);

  if (fabs(NormalDotDirection) < EPSILON)
    return(FALSE);

  VDot(NormalDotOrigin, Plane->Normal_Vector, Ray->Initial);

  *Depth = -(NormalDotOrigin + Plane->Distance) / NormalDotDirection;

  if ((*Depth >= DEPTH_TOLERANCE) && (*Depth <= Max_Distance)) {
    Increase_Counter(stats[Ray_Plane_Tests_Succeeded]);
    return (TRUE);
  }
  else
    return (FALSE);
}

First of all, I want to note that Plane->Distance has the opposite
sign of what is parsed from the scene file. This is seen in
Parse_Plane in parse.c.

The only test made is if the ray is parallel to the plane. The value
stored in *Depth is such that, when passed to Evaluate_Ray, always
results in a point on the plane.

Take a very simple case. Let's test a ray whose origin is at 0 and
points at +x against a plane { -x, -5 }. Then:

NormalDotOrigin = 0
NormalDotDirection = -1
Plane->Distance = 5

Then *Depth evaluates to -(0+5)/(-1) = 5
When passed to Evaluate_Ray, this will result in
<0,0,0> + 5*<1,0,0> = <5,0,0>

Now, let's suppose the ray points at -x. Then:

NormalDotOrigin = 0
NormalDotDirection = 1
Plane->Distance = 5

*Depth = -(0+5)/1 = -5

which, when evaluated, gives:

<0,0,0> + (-5)*<-1,0,0> = <5,0,0>

But this ray does not intersect the plane, and no tests are being done
to test if it does!

Am I not seeing something?
Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] vipbg
TAG      e-mail : pet### [at] tagpovrayorg

Post a reply to this message

On Wed, 01 Aug 2001 19:31:02 +0300, Peter Popov wrote:
>  if ((*Depth >= DEPTH_TOLERANCE) && (*Depth <= Max_Distance)) {
         ^^^^^^^^^^^^^^^^^^^^^^^^
...
>*Depth = -(0+5)/1 = -5
...
>But this ray does not intersect the plane, and no tests are being done
>to test if it does!

Are you sure?

-- 
#local R=<7084844682857967,0787982,826975826580>;#macro L(P)concat(#while(P)chr(
mod(P,100)),#local P=P/100;#end"")#end background{rgb 1}text{ttf L(R.x)L(R.y)0,0
translate<-.8,0,-1>}text{ttf L(R.x)L(R.z)0,0translate<-1.6,-.75,-1>}sphere{z/9e3
4/26/2001finish{reflection 1}}//ron.parker@povray.org My opinions, nobody else's

Post a reply to this message

In article <jjbgmts2tkrhkqh0lb8dn5i5i5n97hfnak@4ax.com> , Peter Popov 
<pet### [at] vipbg>  wrote:

> *Depth = -(0+5)/1 = -5

I haven't really looked at any of your code, but I am confident that the
plane intersection algorithm works.  So the only obvious explanation is that
the code does actually do line/plane rather than a ray/plane intersection.
I assume later on only positive values will be used.

Of course, without looking at the code (which I have no time to do right
now) this is just a guess - nothing more, nothing less.

    Thorsten

____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde

Visit POV-Ray on the web: http://mac.povray.org

Post a reply to this message

On 1 Aug 2001 12:48:05 -0400, ron### [at] povrayorg (Ron Parker)
wrote:

>On Wed, 01 Aug 2001 19:31:02 +0300, Peter Popov wrote:
>>  if ((*Depth >= DEPTH_TOLERANCE) && (*Depth <= Max_Distance)) {
>         ^^^^^^^^^^^^^^^^^^^^^^^^

>>no tests are being done to test if [the ray intersects the plane]!

>Are you sure?

Gee...

I've been skipping the DEPTH_TOLERANCE stuff for so long I just missed
that one. And I swear I spent more than an hour dissecting this
function byte by byte.

Thanks for the heads-up.


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

Post a reply to this message