Hi!
Im trying to use this radial shader in spheres, but unsuccessfully. the problem is related with screen coordinates. The shader only works if the sphere is in the center of the screen, just like show this pictures:

if its not, this is what happen:

i discovered that if i convert the position of the sphere to screen coordinates the shader works fine, just like i show in this code, using the function cam.worldToScreen():

fbo.begin();
ofClear(0);
cam.begin();

shader.begin();

sphere.setRadius(10);
sphere.setPosition(v.x, v.y, v.z);
sphere.draw();
shader.end();

cam.end();
fbo.end();



depthFbo.begin();
ofClear(0);
fbo.getDepthTexture().draw(0,0);
ofSetColor(255);

depthFbo.end();


radialBuffer.begin();
ofClear(0);
radial.begin();

f = cam.worldToScreen(v);
radial.setUniform3f("ligthPos", f.x, f.y, f.z);

radial.setUniformTexture("depth", depthFbo.getTexture(), 1);
fbo.draw(0,0);
radial.end();
radialBuffer.end();

radialBuffer.draw(0,0);

and this is the shader:

#version 150

in vec2 varyingtexcoord;
uniform sampler2DRect tex0;
uniform sampler2DRect depth;

uniform vec3 ligthPos;
 float exposure = 0.19;
 float decay = 0.9;
 float density = 2.0;
 float weight = 1.0;
 int samples = 25;

out vec4 fragColor;
const int MAX_SAMPLES = 100;
void main()
{
//        float sampleDepth = texture(depth, varyingtexcoord).r;
//    
//        vec4 H = vec4 (varyingtexcoord.x *2 -1, (1- varyingtexcoord.y)*2 -1, sampleDepth, 1);
//    
//    vec4 = mult(H, gl_ViewProjectionMatrix);

    vec2 texCoord = varyingtexcoord;
    vec2 deltaTextCoord = texCoord - ligthPos.xy;
    deltaTextCoord *= 1.0 / float(samples) * density;
    vec4 color = texture(tex0, texCoord);
    float illuminationDecay = .6;
    for(int i=0; i < MAX_SAMPLES; i++)
    {
        if(i == samples){
            break;
        }
        texCoord -= deltaTextCoord;
        vec4 sample = texture(tex0, texCoord);
        sample *= illuminationDecay * weight;
        color += sample;
        illuminationDecay *= decay;
    }
    fragColor = color * exposure;
}

the problem is that i want to apply this effect to multiples spheres, and i dont know how to do it, because i pass the position of the sphere like a uniform variable to the radial shader post processing.

but what i want in fact, is discover the way that apply this or others shader to the meshes and not to deal with this problem everytime (what about is the shader have not a uniform to determine the position of the object? which is the way to make a shader works in screen coordinates? how can do? i need some light here, becouse i feel that im missing something.

this is an illustrative picture for what i want to do :

but i want post processing shaders that can have the information of the objects and not always compute the effect in the center of the screen (in the center of the object is what i want). please, any suggestion will be apprecciate

#version 150



// these are for the programmable pipeline system and are passed in

// by default from OpenFrameworks

uniform mat4 modelViewMatrix;

uniform mat4 projectionMatrix;

uniform mat4 textureMatrix;





uniform mat4 modelViewProjectionMatrix;







in vec3 position;

in vec4 color;

in vec4 normal;

in vec2 texcoord;



vec4 temp;

// this is the end of the default functionality



// this is something we're creating for this shader

out vec2 varyingtexcoord;



// this is coming from our C++ code

uniform float mouseX;



void main()

{

    // here we move the texture coordinates

    varyingtexcoord = vec2(texcoord.x, texcoord.y);



    // send the vertices to the fragment shader

    gl_Position = projectionMatrix * modelViewMatrix * vec4( position, 1.0 );



}

thanks, i actually fix the problem of the coordinates, but now i have another problem. because i dont know how to avoid calculate the samples once that has been calculate. this is my problem, for each loop the function calculates the shader for each coordiantes, plus the others coordinates... this is what happens: 

notice howthe halo is compute well for each point, plus others iterations... any idea? this is the shader now:

 

#version 150



in vec2 varyingtexcoord;

uniform sampler2DRect tex0;



uniform int size;





 float exposure = 0.79;

 float decay = 0.9;

 float density = .9;

 float weight = .1;

 int samples = 25;





out vec4 fragColor;

const int MAX_SAMPLES = 25;

const int N = 3;

uniform vec2 ligthPos [N];





int a = 1;





vec4 halo(vec2 pos){

    

    

    float illuminationDecay = 1.2;

    vec2 texCoord = varyingtexcoord;

    vec2 current = pos.xy;

    vec2  deltaTextCoord = texCoord - current;

    

    deltaTextCoord *= 1.0 / float(samples) * density;

    vec4 color = texture(tex0, texCoord);



    

    for(int i=0; i < MAX_SAMPLES; i++){

        

        texCoord -= deltaTextCoord;



        vec4 sample = texture(tex0, texCoord);

        sample *= illuminationDecay * weight;

        color += sample;

        illuminationDecay *= decay;

        

    }

    return color;



}





void main(){

    

    vec2 uv = varyingtexcoord;

    

    vec4 color = texture(tex0, uv);

   



        

        vec4 accum = vec4(0.0);

        

        for(int e = 0; e < N;e++){

            

            vec2 current =ligthPos[e];



            accum += halo(current);



            

            

        }

        

        fragColor = (accum) * exposure;

    







}

i need a way to avoid this extra calculations. i understand the problem, cos the fragment computes all the pixels on the screen, i want a way to avoid that. any clues?

the function "halo" makes the calculation to generate that "glow" effect/ radial blur. it takes an origin point. mi attemp is to have a shader with multiples origin points. for that,  pass an array of origin positions, as you can see in the pictures, the shader calculates the different origin points well, but for each loop, they keep generate the samples for the others points.