Camera Space Light Animator

[Klunk]

a simple light animator that keeps the light fixed in camera space and always looking at the origin. Like having an object tracking torch just above your right shoulder (if you want it on the left add a minus sign to the csoffset.X initializer ). For those who know it, it will mimic the 3ds max single default light.

code, (i know its just a snippet)

Code: Select all

#include <irrlicht.h>
#include <math.h>
 
using namespace irr;
using namespace core;
using namespace scene;
 
class CMaxDefLightAnim : public ISceneNodeAnimator
{
public:
 
        CMaxDefLightAnim(ICameraSceneNode* cam) : camera(cam), csoffset(0.5 * sin(PI/6.0), 0.5 * cos(PI/6.0),-cos(PI/6.0)) {}
        void animateNode(ISceneNode* node, u32 timeMs)
        {
                matrix4 camTM, offsetTM;
                vector3df wsoffset;
                camera->getViewMatrix().getInverse(camTM);
                camTM.setTranslation(vector3df(0.0,0.0,0.0));
                camTM.transformVect(wsoffset,csoffset);
                TransformFromDirVector(-wsoffset,  offsetTM);
                node->setRotation(offsetTM.getRotationDegrees());
        }
        virtual void serializeAttributes(io::IAttributes* out, io::SAttributeReadWriteOptions* options=0) const {}
        virtual void deserializeAttributes(io::IAttributes* in, io::SAttributeReadWriteOptions* options=0) {}
        virtual ESCENE_NODE_ANIMATOR_TYPE getType() const { return ESNAT_UNKNOWN; }
        virtual ISceneNodeAnimator* createClone(ISceneNode* node, ISceneManager* newManager=0) { return new CMaxDefLightAnim(camera); }
 
private:
 
        void TransformFromDirVector(const vector3df& dir, matrix4& mat)
        {
                vector3df up(0.0f,1.0f,0.0f), zaxis, xaxis, yaxis;
                zaxis = dir;
                zaxis.normalize();
                xaxis = up.crossProduct(zaxis);
                xaxis.normalize();
                yaxis = zaxis.crossProduct(xaxis);
                mat[0] = xaxis.X; mat[1] = xaxis.Y; mat[2] = xaxis.Z; mat[3] = 0;
                mat[4] = yaxis.X; mat[5] = yaxis.Y; mat[6] = yaxis.Z; mat[7] = 0;
                mat[8] = zaxis.X; mat[9] = zaxis.Y; mat[10] = zaxis.Z; mat[11] = 0;
                mat[12] = 0; mat[13] = 0; mat[14] = 0; mat[15] = 1;
        }
        vector3df                       csoffset;
        ICameraSceneNode*       camera;
};

usage.

Code: Select all

ICameraSceneNode* cam = smgr->addCameraSceneNodeMaya();
ILightSceneNode* lnode = smgr->addLightSceneNode(smgr->getRootSceneNode(),vector3df(0,0,0));
lnode->setLightType(ELT_DIRECTIONAL);
 
ISceneNodeAnimator* anim = new CMaxDefLightAnim(cam);
if(anim)
{
        lnode->addAnimator(anim);
        anim->drop();
}

[randomMesh]

Please don't use the using namespace directive in headers, as it will pollute your namespace and can lead to name clashes.
You include the C header (math.h) rather than the C++ one (cmath).

[Klunk]

oh no, what shall I do.

[randomMesh]

Easy. Fix your code.

[Klunk]

yeah, no.

[Klunk]

I have a question about my own code, is there any advantage binding this sort of thing to a fixed frame rate/update rate ? with something like

Code: Select all

void CMaxDefLightAnim::animateNode(ISceneNode* node, u32 timeMs)
{
    if((timeMs - oldtime) * updaterate >= 1000)
    {
           //... animator code
 
           oldtime = timeMs;
    }
}

where oldtime is a class member

[Mel]

Is there any advantage over linking the light to the camera as a child node, and move the camera on its own?

[Klunk]

the light would work differently, it's more of a sort of "orientation constraint" to the camera.