//Lens flares VS 2.
// ---- Created with geo-11 v0.6.193 on Sun May 19 23:52:41 2024

cbuffer PER_BATCH : register(b0)
{
  float4 wposAndSize : packoffset(c0);
  row_major float3x4 xform : packoffset(c1);
  float4 externTint : packoffset(c4);
  float4 dynamics : packoffset(c5);
  float4 meshCenterAndBrt : packoffset(c6);
}



// 3Dmigoto declarations
#define cmp -
Texture1D<float4> IniParams : register(t120);
Texture2D<float4> StereoParams : register(t125);
// Depth buffer copied to this input with 3Dmigoto:
// Texture2D<float> DepthBuffer : register(t110);
Texture2DArray<float> DepthBuffer : register(t110);

static const float near = 0.00001;
static const float far = 1;

float world_z_from_depth_buffer(float x, float y, int eye)
{
    uint width, height, elements = 0;
    float z;

    DepthBuffer.GetDimensions(width, height, elements);

    x = min(max((x / 2 + 0.5) * width, 0), width - 1);
    y = min(max((-y / 2 + 0.5) * height, 0), height - 1);
    z = DepthBuffer.Load(int4(x, y, eye, 0)).x;
    if (z == 0)
        return 0;

  return (far*near/((z)*near) + (far*z));
}

float adjust_from_depth_buffer(float x, float y, float numsamples, float convmult)
{
    float4 stereo = StereoParams.Load(0);
    float separation = stereo.x; float convergence = stereo.y;
    float old_offset, offset, w, sampled_w, distance;
    uint i;

    // Stereo cursor: To improve the accuracy of the stereo cursor, we
    // sample a number of points on the depth buffer, starting at the near
    // clipping plane and working towards original x + separation.
    //
    // You can think of this as a line in three dimensional space that
    // starts at each eye and stretches out towards infinity. We sample 255
    // points along this line (evenly spaced in the X axis) and compare
    // with the depth buffer to find where the line is first intersected.
    //
    // Note: The reason for sampling 255 points came from a restriction in
    // DX9/SM3 where loops had to run a constant number of iterations and
    // there was no way to set that number from within the shader itself.
    // I'm not sure if the same restriction applies in DX11 with SM4/5 - if
    // it doesn't, we could change this to check each pixel instead for
    // better accuracy.
    //
    // Based on DarkStarSword's stereo crosshair code originally developed
    // for Miasmata, adapted to Unity, then translated to HLSL.

    offset = (near - convergence) * separation*1.0; // Z = X offset from center
    distance = separation - offset;         // Total distance to cover (separation - starting X offset)

    old_offset = offset;
    for (i = 0; i < numsamples; i++) {
        offset += distance / numsamples;

        // Calculate depth for this point on the line:
        w = (separation * convergence*convmult) / (separation - offset);
		
		//sampled_w = world_z_from_depth_buffer(x + offset, y);
        //sampled_w = world_z_from_depth_buffer(x + offset * stereo.z, y);
		
		float left = min((x + offset * stereo.z), 0);
        float right = max((x - offset * stereo.z), 0);
        float left_sampled_w = world_z_from_depth_buffer(left, y, 0);
        float right_sampled_w = world_z_from_depth_buffer(right, y, 1);
        sampled_w = max(left_sampled_w, right_sampled_w);
		
        if (sampled_w == 0)
            return separation;

        // If the sampled depth is closer than the calculated depth,
        // we have found something that intersects the line, so exit
        // the loop and return the last point that was not intersected:
        if (w > sampled_w)
            break;

        old_offset = offset;
    }

    return old_offset;
}

void main(
  float3 v0 : POSITION0,
  float2 v1 : TEXCOORD0,
  float4 v2 : COLOR0,
  out float4 o0 : SV_Position0,
  out float4 o1 : TEXCOORD0,
  out float3 o2 : TEXCOORD1,
  out float4 o3 : COLOR0)
{
  float4 r0;
  uint4 bitmask, uiDest;
  float4 fDest;
  
  float4 stereo = StereoParams.Load(0);
  
  r0.xy = wposAndSize.ww * v0.xy;
  r0.yz = xform._m10_m11 * r0.yy;
  r0.xy = r0.xx * xform._m00_m01 + r0.yz;
  r0.xy = xform._m20_m21 + r0.xy;
  r0.xy = r0.xy * float2(0.5,0.5) + meshCenterAndBrt.xy;
  r0.z = 1 + -r0.y;
  r0.xy = r0.xz * float2(2,2) + float2(-1,-1);
  o0.xy = r0.xy;
  o1.zw = r0.xy;
  o0.zw = float2(0,1);
  
  if (o0.w==1) {
    o0.x+=adjust_from_depth_buffer(o0.x,o0.y,300,10.0);
  }
  
  o1.xy = v1.xy;
  r0.xy = meshCenterAndBrt.xy * float2(1,-1) + float2(0,1);
  o2.xy = r0.xy * float2(2,2) + float2(-1,-1);
  o2.z = meshCenterAndBrt.z;
  r0.xyzw = meshCenterAndBrt.wwww * v2.zyxw;
  r0.xyzw = externTint.xyzw * r0.xyzw;
  o3.xyzw = dynamics.xxxx * r0.xyzw;
  return;
}

/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
//
// Generated by Microsoft (R) HLSL Shader Compiler 10.0.10011.16384
//
//   using 3Dmigoto v0.6.193 on Sun May 19 23:52:41 2024
//
//
// Buffer Definitions:
//
// cbuffer PER_BATCH
// {
//
//   float4 wposAndSize;                // Offset:    0 Size:    16
//   row_major float3x4 xform;          // Offset:   16 Size:    48
//   float4 externTint;                 // Offset:   64 Size:    16
//   float4 dynamics;                   // Offset:   80 Size:    16
//   float4 meshCenterAndBrt;           // Offset:   96 Size:    16
//
// }
//
//
// Resource Bindings:
//
// Name                                 Type  Format         Dim Slot Elements
// ------------------------------ ---------- ------- ----------- ---- --------
// PER_BATCH                         cbuffer      NA          NA    0        1
//
//
//
// Input signature:
//
// Name                 Index   Mask Register SysValue  Format   Used
// -------------------- ----- ------ -------- -------- ------- ------
// POSITION                 0   xyz         0     NONE   float   xy
// TEXCOORD                 0   xy          1     NONE   float   xy
// COLOR                    0   xyzw        2     NONE   float   xyzw
//
//
// Output signature:
//
// Name                 Index   Mask Register SysValue  Format   Used
// -------------------- ----- ------ -------- -------- ------- ------
// SV_Position              0   xyzw        0      POS   float   xyzw
// TEXCOORD                 0   xyzw        1     NONE   float   xyzw
// TEXCOORD                 1   xyz         2     NONE   float   xyz
// COLOR                    0   xyzw        3     NONE   float   xyzw
//
vs_5_0
dcl_globalFlags refactoringAllowed
dcl_constantbuffer CB0[7], immediateIndexed
dcl_input v0.xy
dcl_input v1.xy
dcl_input v2.xyzw
dcl_output_siv o0.xyzw, position
dcl_output o1.xyzw
dcl_output o2.xyz
dcl_output o3.xyzw
dcl_temps 1
mul r0.xy, v0.xyxx, cb0[0].wwww
mul r0.yz, r0.yyyy, cb0[2].xxyx
mad r0.xy, r0.xxxx, cb0[1].xyxx, r0.yzyy
add r0.xy, r0.xyxx, cb0[3].xyxx
mad r0.xy, r0.xyxx, l(0.500000, 0.500000, 0.000000, 0.000000), cb0[6].xyxx
add r0.z, -r0.y, l(1.000000)
mad r0.xy, r0.xzxx, l(2.000000, 2.000000, 0.000000, 0.000000), l(-1.000000, -1.000000, 0.000000, 0.000000)
mov o0.xy, r0.xyxx
mov o1.zw, r0.xxxy
mov o0.zw, l(0,0,0,1.000000)
mov o1.xy, v1.xyxx
mad r0.xy, cb0[6].xyxx, l(1.000000, -1.000000, 0.000000, 0.000000), l(0.000000, 1.000000, 0.000000, 0.000000)
mad o2.xy, r0.xyxx, l(2.000000, 2.000000, 0.000000, 0.000000), l(-1.000000, -1.000000, 0.000000, 0.000000)
mov o2.z, cb0[6].z
mul r0.xyzw, v2.zyxw, cb0[6].wwww
mul r0.xyzw, r0.xyzw, cb0[4].xyzw
mul o3.xyzw, r0.xyzw, cb0[5].xxxx
ret
// Approximately 18 instruction slots used

//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
