PureBasic Forums - English

the problem is
glPushMatrix_()
and
glPopMatrix_()
before getting the ModelViewMatrix.
Therefore the matrix is always the same.

I've changed you code a little bit :wink: with a long time for changing the view
Enumeration
#Win
#Ogl
#Timer
EndEnumeration

Procedure DrawCube ...

glReadPixels_(dWinX, dWinY, 1, 1, #GL_DEPTH_COMPONENT, #GL_DOUBLE, @dWinZ) ;surely always 0.0?

only returns a float value
It should be

f you are sure that the lines are not parallel, you can shorten the code:


Procedure.d VecSpat (Array adU.d(1), Array adV.d(1), Array adW.d(1))
ProcedureReturn adU(0)*adV(1)*adW(2)+adV(0)*adW(1)*adU(2)+adW(0)*adU(1)*adV(2) - adU(0)*adW(1)*adV(2)-adV(0)*adU(1)*adW(2)-adW(0)*adV(1)*adU(2 ...

Try this:

Procedure DistancePointLine ( Array adQ.d(1), Array adP.d(1), Array adV.d(1), *dU.double, *dA.double)
Protected Dim adU.d(2)
Protected Dim adN.d(2)
Protected dL

dL = adV(0)*adV(0) + adV(1)*adV(1) + adV(2)*adV(2)
*dA\d = 1.0e20
*dU\d = 0.0

adU(0) = adQ(0) - adP(0)
adU(1) = adQ ...

The point set does not define one direction of the plane but both, the calculated direction (adV()) and the opposite direchtion (-adV(0),-adV(1),-adV(2)). To choose one of both you need additional information.

I would calculate the intersection line like this:

EnableExplicit

#DBL_EPSILON = 0.0001

Structure pt3d
x.d
y.d
z.d
EndStructure

Structure ln3d
pt1.pt3d
pt2.pt3d
EndStructure

Structure vector3d
x.d
y.d
z.d
EndStructure

Structure plane3d
VecNorm.vector3d
PtV0.pt3d
EndStructure

Procedure ...

adMinMax ist the bounding box. adSP is the centroid of all data points. This is not the same as the centroid of the bouning box. The centroid is part of the best-fit plane (see the article you linked above).
adMP is the centroid of the bounding box.
CalcMat calculates the matrix, according to which ...

If you have questions about triangles, lines, points, etc., don't hesitate to ask. I have been working on such problems for years.

I've added some comments (not for calculation of eigenvectors - this is numerical math). Maybe it makes something clearer, maybe it confuse you more . (I'm not native in this language)

I played a little bit and added zoom,center and identification functions (code above).
Hope you enjoy

Yes, a lot of code for such a "simple" problem .

I've changed the code above. It was a simple glOrtho problem. The result was correct but you couldn't see it. And I forgot a CloseFile.

It took a lot of time to convert old C-code (for calculating eigenvalues and eigenvectors) to PB.

I've put your point data into a file.

For your example the code works.


EnableExplicit

#PB_DBL_MAX = 1.7976931348623158e+308 ; max double value

#PB_DBL_EPS = 2.2204460492503131e-016 ; smallest ...

How do you choose the x-values, where the splines should start and end?

Unfortunately, I deleted all the comments because they were all in German. Therefore here are some explanations:
The code can also be used for other approximations. For this purpose, the functions LangmuirFkt and LangmuirAbl must be replaced and the array dC () must be adapted to the number of ...