///////////////////////////////////////////////////////////////////////////////
//
//      ##  ######
//       ######  ###
//  ## ###############        Shark 3D Engine (www.shark3d.com)
//   ########## # # #
//    ########                Copyright (c) 1996-2006 Spinor GmbH.
//   ######### # # #          All rights reserved.
//  ##   ##########
//      ##
//
///////////////////////////////////////////////////////////////////////////////


#define Poly2Point 1

struct PS_INPUT
{
float3 normalWorld  : TEXCOORD1;
float3 posWorld : TEXCOORD0;
};

sampler colorSamp: register(s0);
sampler distanceSamp: register(s1);

///////////////////////////////////////////////////////////////////////////////
// Pixelshader
// Profile: 3x0

#ifdef Poly2Point

float4 GetContibution(float3 L, float3 L1, float3 L2, float3 L3, float3 L4, float3 pos, float3 N, samplerCUBE cubemap)
{
float d;
d = texCUBE(cubemap, L1).r;
L1 = d * normalize(L1);
d = texCUBE(cubemap, L2).r;
L2 = d * normalize(L2);
d = texCUBE(cubemap, L3).r;
L3 = d * normalize(L3);
d = texCUBE(cubemap, L4).r;
L4 = d * normalize(L4);


    float3 r1 = normalize(L1 - pos);    
    float3 r2 = normalize(L2 - pos);
    float3 r3 = normalize(L3 - pos);
    float3 r4 = normalize(L4 - pos);
  
  float3 crossP = cross(r1, r2);
  float r = length(crossP);
  float dd = dot(r1,r2);
  float tri1 = acos(dd) * dot(crossP/r, N);
  
  crossP = cross(r2, r3);
  r = length(crossP);
  dd = dot(r1,r2);
  float tri2 = acos(dd) * dot(crossP/r, N);
  
  crossP = cross(r3, r4);
  r = length(crossP);
  dd = dot(r1,r2);
  float tri3 = acos(dd) * dot(crossP/r, N);
  
  crossP = cross(r4, r1);
  r = length(crossP);
  dd = dot(r1,r2);
  float tri4= acos(dd) * dot(crossP/r, N);
  
  
   return max(tri1 + tri2 + tri3 + tri4, 0);
}

float4 main(PS_INPUT input): COLOR0
{
    float M = 4;

    float3 N = normalize( input.normalWorld ); 
    float3 pos = input.posWorld;
    
    //return texCUBE(colorSamp, N);
      
    float4 I = 0;
    float3 L1, L2, L3, L4, L;
    float4 Le;
    float width = 1.0 / M;
    float width2 = width * 2;
    float d; 

    for (float x = 1; x < M; x++)
 for (float y = 1; y < M; y++)
 {
    float2 p, tpos; 
    tpos.x = x * width;
    tpos.y = y * width;
    
    p = tpos.xy;    
    p = 2.0 * p - 1.0; //-1..1
            
L1 = float3(p.x - width, p.y - width, 1);
L2 = float3(p.x + width, p.y - width, 1);
L3 = float3(p.x + width, p.y + width, 1);
L4 = float3(p.x - width, p.y + width, 1);
L = float3(p.x, p.y, 1);
Le = float4(texCUBE(colorSamp, L).rgb, 1);

I += 0.5 * Le * GetContibution( L, L1, L2, L3, L4, pos, N, distanceSamp);

}

for (float x = 1; x < M; x++)
 for (float y = 1; y < M; y++)
 {
float2 p, tpos; 
    tpos.x = x * width;// 0..1
    tpos.y = y * width;// 0..1
    
    p = tpos.xy;    
    p = 2.0 * p - 1.0; //-1..1
            
L4 = float3(p.x - width, p.y - width, -1);
L3 = float3(p.x + width, p.y - width, -1);
L2 = float3(p.x + width, p.y + width, -1);
L1 = float3(p.x - width, p.y + width, -1);
L = float3(p.x, p.y, -1);
Le = float4(texCUBE(colorSamp, L).rgb, 1);

I += 0.5 * Le * GetContibution( L, L1, L2, L3, L4, pos, N, distanceSamp);
 }
 
for (float x = 1; x < M; x++)
 for (float y = 1; y < M; y++)
 {
float2 p, tpos; 
    tpos.x = x * width;// 0..1
    tpos.y = y * width;// 0..1
    
    p = tpos.xy;    
    p = 2.0 * p - 1.0; //-1..1
            
L4 = float3(p.x - width, 1, p.y - width);
L3 = float3(p.x + width, 1, p.y - width);
L2 = float3(p.x + width, 1, p.y + width);
L1 = float3(p.x - width, 1, p.y + width);
L = float3(p.x, 1, p.y);
Le = float4(texCUBE(colorSamp, L).rgb, 1);

I += 0.5 * Le * GetContibution( L, L1, L2, L3, L4, pos, N, distanceSamp);
 }
 
for (float x = 1; x < M; x++)
 for (float y = 1; y < M; y++)
 {
float2 p, tpos; 
    tpos.x = x * width;// 0..1
    tpos.y = y * width;// 0..1
    
    p = tpos.xy;    
    p = 2.0 * p - 1.0; //-1..1
            
L1 = float3(p.x - width, -1, p.y - width);
L2 = float3(p.x + width, -1, p.y - width);
L3 = float3(p.x + width, -1, p.y + width);
L4 = float3(p.x - width, -1, p.y + width);
L = float3(p.x, -1, p.y);
Le = float4(texCUBE(colorSamp, L).rgb, 1);

I += 0.5 * Le * GetContibution( L, L1, L2, L3, L4, pos, N, distanceSamp);
 }
 
 for (float x = 1; x < M; x++)
 for (float y = 1; y < M; y++)
{
float2 p, tpos; 
    tpos.x = x * width;// 0..1
    tpos.y = y * width;// 0..1
    
    p = tpos.xy;    
    p = 2.0 * p - 1.0; //-1..1
            
L1 = float3(1, p.x - width, p.y - width);
L2 = float3(1, p.x + width, p.y - width);
L3 = float3(1, p.x + width, p.y + width);
L4 = float3(1, p.x - width, p.y + width);
L = float3(1, p.x, p.y);
Le = float4(texCUBE(colorSamp, L).rgb, 1);

I += 0.5 * Le * GetContibution( L, L1, L2, L3, L4, pos, N, distanceSamp);
}
 
for (float x = 1; x < M; x++)
 for (float y = 1; y < M; y++)
 {
float2 p, tpos; 
    tpos.x = x * width;// 0..1
    tpos.y = y * width;// 0..1
    
    p = tpos.xy;    
    p = 2.0 * p - 1.0; //-1..1
            
L4 = float3(-1, p.x - width, p.y - width);
L3 = float3(-1, p.x + width, p.y - width);
L2 = float3(-1, p.x + width, p.y + width);
L1 = float3(-1, p.x - width, p.y + width);
L = float3(-1, p.x, p.y);
Le = float4(texCUBE(colorSamp, L).rgb, 1);

I += 0.5 * Le * GetContibution( L, L1, L2, L3, L4, pos, N, distanceSamp);
 }
 float kd = 0.5;
return kd * I;
}



#else

float4 P2PContr(float3 N, float3 Nl, float3 pos, float3 L, samplerCUBE cubemap, samplerCUBE distmap)
{
Nl = normalize(Nl);
L = normalize(L);
float4 Le = texCUBE(cubemap, L);
float d = texCUBE(distmap, L).r;
float3 Lpos = L * d;
float3 Ldir = Lpos - pos;
float dist = 4 * dot(Ldir, Ldir);
Ldir = normalize(Ldir);

return Le * (max(dot(N, Ldir),0) * dot(Nl, -1 * Ldir)) / dist;
}

float4 main(PS_INPUT input): COLOR0
{
    float M = 4.0;

    float3 N = normalize( input.normalWorld ); 
    float3 pos = input.posWorld;

  //  return texCUBE(distanceSamp, N).r;
    
//return float4(N,1);

    float4 I = 0;
    float3 L;
    float4 Le;
    float width = 1.0 / M;
    float d;

    float kd = 3.0;

    for (float x = 0.5; x < M; x++)
 for (float y = 0.5; y < M; y++)
 {
    float2 p, tpos; 
    tpos.x = x * width;
    tpos.y = y * width;
    
    p = tpos.xy;    
    p = 2.0 * p - 1.0; //-1..1
            
I += P2PContr(N, float3(0,0,-1), pos, float3(p.x, p.y, 1), colorSamp, distanceSamp);
I += P2PContr(N, float3(0,0,1), pos, float3(-p.x, p.y, -1), colorSamp, distanceSamp);
I += P2PContr(N, float3(-1,0,0), pos, float3(1, p.y, -p.x), colorSamp, distanceSamp);
I += P2PContr(N, float3(1,0,0), pos, float3(-1, p.y, p.x), colorSamp, distanceSamp);
I += P2PContr(N, float3(0,-1,0), pos, float3(p.x, 1, -p.y), colorSamp, distanceSamp);
I += P2PContr(N, float3(0,1,0), pos, float3(p.x, -1, p.y), colorSamp, distanceSamp);
}

return kd * I;

}
#endif
///////////////////////////////////////////////////////////////////////////////