#version 420
#extension GL_ARB_texture_gather : enable
#extension GL_ARB_separate_shader_objects : enable
#ifdef VULKAN
#define ATTR_LAYOUT(__vkSet, __location) layout(set = __vkSet, location = __location)
#define UNIFORM_BUFFER_LAYOUT(__glLocation, __vkSet, __vkLocation) layout(set = __vkSet, binding = __vkLocation, std140)
#define TEXTURE_LAYOUT(__glLocation, __vkSet, __vkLocation) layout(set = __vkSet, binding = __vkLocation)
#define SET_POSITION(_v) gl_Position = _v; gl_Position.z = (gl_Position.z + gl_Position.w) / 2.0
#define GET_FRAGCOORD() vec4(gl_FragCoord.xy*uf_fragCoordScale.xy,gl_FragCoord.zw)
#define gl_VertexID gl_VertexIndex
#define gl_InstanceID gl_InstanceIndex
#else
#define ATTR_LAYOUT(__vkSet, __location) layout(location = __location)
#define UNIFORM_BUFFER_LAYOUT(__glLocation, __vkSet, __vkLocation) layout(binding = __glLocation, std140) 
#define TEXTURE_LAYOUT(__glLocation, __vkSet, __vkLocation) layout(binding = __glLocation)
#define SET_POSITION(_v) gl_Position = _v
#define GET_FRAGCOORD() vec4(gl_FragCoord.xy*uf_fragCoordScale,gl_FragCoord.zw)
#endif
// This shaders was auto-converted from OpenGL to Cemu so expect weird code and possible errors.

// shader a7f4801a8d29e333 

#ifdef VULKAN
layout(set = 1, binding = 1) uniform ufBlock
{
uniform vec4 uf_fragCoordScale;
uniform ivec4 uf_remappedPS[3];
};
#else
uniform vec2 uf_fragCoordScale;
uniform ivec4 uf_remappedPS[3];
#endif

const float hazeFactor = 0.1;

const float gamma = $gamma; // 1.0 is neutral  Botw is already colour graded at this stage
const float exposure = $exposure; // 1.0 is neutral
const float vibrance = $vibrance;  // 0.0 is neutral  
const float crushContrast = $crushContrast; // 0.0 is neutral. Use small increments, loss of shadow detail 
const float contrastCurve = $contrastCurve;


vec3 RGB_Lift = vec3($redShadows, $greenShadows , $blueSadows);		// [0.000 to 2.000] Adjust shadows for Red, Green and Blue.
vec3 RGB_Gamma = vec3($redMid ,$greenMid, $blueMid);				// [0.000 to 2.000] Adjust midtones for Red, Green and Blue
vec3 RGB_Gain = vec3($redHilight, $greenHilight, $blueHilight);		// [0.000 to 2.000] Adjust highlights for Red, Green and Blue
//lumasharpen
const float sharp_mix = $sharp_mix;
const float sharp_strength = 2.0; 
const float sharp_clamp = 0.75;
const float offset_bias = 1.0;
float Sigmoid (float x) {

	return 1.0 / (1.0 + (exp(-(x - 0.5) * 5.5))); 
}


#define px (1.0/1920.0*uf_fragCoordScale.x)
#define py (1.0/1080.0*uf_fragCoordScale.y)
#define CoefLuma vec3(0.2126, 0.7152, 0.0722)  

float lumasharping(sampler2D tex, vec2 pos) {
	vec4 colorInput = texture(tex, pos);

	vec3 ori = colorInput.rgb;

	// -- Combining the strength and luma multipliers --
	vec3 sharp_strength_luma = (CoefLuma * sharp_strength);

	// -- Gaussian filter --
	//   [ .25, .50, .25]     [ 1 , 2 , 1 ]
	//   [ .50,   1, .50]  =  [ 2 , 4 , 2 ]
	//   [ .25, .50, .25]     [ 1 , 2 , 1 ]

	vec3 blur_ori = texture(tex, pos + vec2(px, -py) * 0.5 * offset_bias).rgb; // South East
	blur_ori += texture(tex, pos + vec2(-px, -py) * 0.5 * offset_bias).rgb;  // South West
	blur_ori += texture(tex, pos + vec2(px, py) * 0.5 * offset_bias).rgb; // North East
	blur_ori += texture(tex, pos + vec2(-px, py) * 0.5 * offset_bias).rgb; // North West

	blur_ori *= 0.25;  // ( /= 4) Divide by the number of texture fetches

	// -- Calculate the sharpening --
	vec3 sharp = ori - blur_ori;  //Subtracting the blurred image from the original image

	// -- Adjust strength of the sharpening and clamp it--
	vec4 sharp_strength_luma_clamp = vec4(sharp_strength_luma * (0.5 / sharp_clamp), 0.5); //Roll part of the clamp into the dot

	float sharp_luma = clamp((dot(vec4(sharp, 1.0), sharp_strength_luma_clamp)), 0.0, 1.0); //Calculate the luma, adjust the strength, scale up and clamp
	sharp_luma = (sharp_clamp * 2.0) * sharp_luma - sharp_clamp; //scale down

	return sharp_luma;
}

vec3 LiftGammaGainPass(vec3 colorInput)
{ 	//reshade BSD https://reshade.me ,  Alexkiri port
	vec3 color = colorInput;
	color = color * (1.5 - 0.5 * RGB_Lift) + 0.5 * RGB_Lift - 0.5;
	color = clamp(color, 0.0, 1.0); 
	color *= RGB_Gain;	
	color = pow(color, 1.0 / RGB_Gamma);
	return clamp(color, 0.0, 1.0);
}

vec3 contrasty(vec3 colour){
	vec3 fColour = (colour.xyz);
	//fColour = LiftGammaGainPass(fColour);
	
	fColour = clamp(exposure * fColour, 0.0, 1.0);
	fColour = pow(fColour, vec3(1.0 / gamma));
	float luminance = fColour.r*0.299 + fColour.g*0.587 + fColour.b*0.114;
	float mn = min(min(fColour.r, fColour.g), fColour.b);
	float mx = max(max(fColour.r, fColour.g), fColour.b);
	float sat = (1.0 - (mx - mn)) * (1.0 - mx) * luminance * 5.0;
	vec3 lightness = vec3((mn + mx) / 2.0);
	fColour = LiftGammaGainPass(fColour);
	// vibrance
	fColour = mix(fColour, mix(fColour, lightness, -vibrance), sat);
	fColour = max(vec3(0.0), fColour - vec3(crushContrast));
	return fColour;
}

const float resScale = 3.0;  //AA in PS 
TEXTURE_LAYOUT(0, 1, 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf470a000 res 1280x720x1 dim 1 tm: 4 format 0816 compSel: 0 1 2 5 mipView: 0x0 (num 0x1
layout(location = 0) in vec4 passParameterSem136;
layout(location = 1) in vec4 passParameterSem137;
layout(location = 0) out vec4 passPixelColor0;
//uniform vec2 uf_fragCoordScale;
int clampFI32(int v)
{
if( v == 0x7FFFFFFF )
	return floatBitsToInt(1.0);
else if( v == 0xFFFFFFFF )
	return floatBitsToInt(0.0);
return floatBitsToInt(clamp(intBitsToFloat(v), 0.0, 1.0));
}
float mul_nonIEEE(float a, float b){ if( a == 0.0 || b == 0.0 ) return 0.0; return a*b; }
void main()
{
ivec4 R0i = ivec4(0);
ivec4 R1i = ivec4(0);
ivec4 R2i = ivec4(0);
ivec4 R3i = ivec4(0);
ivec4 R4i = ivec4(0);
ivec4 R123i = ivec4(0);
ivec4 R126i = ivec4(0);
ivec4 R127i = ivec4(0);
int backupReg0i, backupReg1i, backupReg2i, backupReg3i, backupReg4i;
ivec4 PV0i = ivec4(0), PV1i = ivec4(0);
int PS0i = 0, PS1i = 0;
ivec4 tempi = ivec4(0);
float tempResultf;
int tempResulti;
ivec4 ARi = ivec4(0);
bool predResult = true;
bool activeMaskStack[2];
bool activeMaskStackC[3];
activeMaskStack[0] = false;
activeMaskStackC[0] = false;
activeMaskStackC[1] = false;
activeMaskStack[0] = true;
activeMaskStackC[0] = true;
activeMaskStackC[1] = true;
vec3 cubeMapSTM;
int cubeMapFaceId;
R0i = floatBitsToInt(passParameterSem136);
R1i = floatBitsToInt(passParameterSem137);
if( activeMaskStackC[1] == true ) {
R2i.y = floatBitsToInt(textureLod(textureUnitPS0, intBitsToFloat(R1i.zy),0.0).y);
R0i.w = floatBitsToInt(textureLod(textureUnitPS0, intBitsToFloat(R1i.xy),0.0).y);
R0i.z = floatBitsToInt(textureLod(textureUnitPS0, intBitsToFloat(R1i.xw),0.0).y);
R2i.x = floatBitsToInt(textureLod(textureUnitPS0, intBitsToFloat(R1i.zw),0.0).y);
R1i.xyzw = floatBitsToInt(textureLod(textureUnitPS0, intBitsToFloat(R0i.xy),0.0).xyzw);
}
if( activeMaskStackC[1] == true ) {
activeMaskStack[1] = activeMaskStack[0];
activeMaskStackC[2] = activeMaskStackC[1];
// 0
backupReg0i = R2i.y;
R127i.x = floatBitsToInt(max(intBitsToFloat(R0i.w), intBitsToFloat(R0i.z)));
R2i.y = floatBitsToInt(intBitsToFloat(backupReg0i) + intBitsToFloat(0x3b2aaaa8));
PV0i.y = R2i.y;
R127i.z = floatBitsToInt(min(intBitsToFloat(R0i.w), intBitsToFloat(R0i.z)));
// 1
PV1i.x = floatBitsToInt(max(intBitsToFloat(R2i.x), intBitsToFloat(PV0i.y)));
PV1i.w = floatBitsToInt(min(intBitsToFloat(R2i.x), intBitsToFloat(PV0i.y)));
// 2
R4i.z = floatBitsToInt(min(intBitsToFloat(R127i.z), intBitsToFloat(PV1i.w)));
PV0i.z = R4i.z;
R4i.w = floatBitsToInt(max(intBitsToFloat(R127i.x), intBitsToFloat(PV1i.x)));
PV0i.w = R4i.w;
// 3
PV1i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.w), intBitsToFloat(uf_remappedPS[0].x)));
PV1i.y = floatBitsToInt(max(intBitsToFloat(R1i.y), intBitsToFloat(PV0i.w)));
PV1i.z = floatBitsToInt(min(intBitsToFloat(R1i.y), intBitsToFloat(PV0i.z)));
// 4
PV0i.z = floatBitsToInt(-(intBitsToFloat(PV1i.z)) + intBitsToFloat(PV1i.y));
PV0i.w = floatBitsToInt(max(intBitsToFloat(uf_remappedPS[0].y), intBitsToFloat(PV1i.x)));
// 5
PV1i.y = ((intBitsToFloat(PV0i.w) > intBitsToFloat(PV0i.z))?int(0xFFFFFFFF):int(0x0));
// 6
backupReg0i = R1i.y;
backupReg1i = R1i.z;
backupReg2i = R1i.w;
backupReg3i = R1i.x;
R3i.x = ((PV1i.y == 0)?(0):(0x3f800000));
R1i.y = ((PV1i.y == 0)?(0):(backupReg0i));
R1i.z = ((PV1i.y == 0)?(0):(backupReg1i));
R1i.w = ((PV1i.y == 0)?(0):(backupReg2i));
R1i.x = ((PV1i.y == 0)?(0):(backupReg3i));
PS0i = R1i.x;
// 7
predResult = (R3i.x == 0);
activeMaskStack[1] = predResult;
activeMaskStackC[2] = predResult == true && activeMaskStackC[1] == true;
}
else {
activeMaskStack[1] = false;
activeMaskStackC[2] = false;
}
if( activeMaskStackC[2] == true ) {
// 0
backupReg0i = R0i.w;
PV0i.y = floatBitsToInt(intBitsToFloat(R0i.z) + -(intBitsToFloat(R2i.y)));
R127i.z = 0;
PV0i.w = floatBitsToInt(-(intBitsToFloat(backupReg0i)) + intBitsToFloat(R2i.x));
// 1
R127i.x = floatBitsToInt(intBitsToFloat(PV0i.y) + intBitsToFloat(PV0i.w));
PV1i.x = R127i.x;
R127i.y = floatBitsToInt(intBitsToFloat(PV0i.y) + -(intBitsToFloat(PV0i.w)));
PV1i.y = R127i.y;
// 2
tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(R127i.z),-0.0),vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(R127i.z),0.0)));
PV0i.x = tempi.x;
PV0i.y = tempi.x;
PV0i.z = tempi.x;
PV0i.w = tempi.x;
// 3
tempResultf = 1.0 / sqrt(intBitsToFloat(PV0i.x));
PS1i = floatBitsToInt(tempResultf);
// 4
backupReg0i = R127i.y;
R127i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg0i), intBitsToFloat(PS1i)));
PV0i.y = R127i.y;
R127i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PS1i)));
PV0i.z = R127i.z;
// 5
PV1i.x = floatBitsToInt(max(intBitsToFloat(PV0i.z), -(intBitsToFloat(PV0i.z))));
PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.z) / resScale, intBitsToFloat(uf_remappedPS[1].x))); // sharpen pass?
PV1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.y)/ resScale, intBitsToFloat(uf_remappedPS[1].y))); //sharpen pass?
PV1i.w = floatBitsToInt(max(intBitsToFloat(PV0i.y), -(intBitsToFloat(PV0i.y))));
// 6
R1i.x = floatBitsToInt(intBitsToFloat(R0i.x) + -(intBitsToFloat(PV1i.y)));
R1i.y = floatBitsToInt(intBitsToFloat(R0i.y) + -(intBitsToFloat(PV1i.z)));
PV0i.z = floatBitsToInt(min(intBitsToFloat(PV1i.x), intBitsToFloat(PV1i.w)));
R2i.w = floatBitsToInt(intBitsToFloat(R0i.x) + intBitsToFloat(PV1i.y) / resScale);
R2i.y = floatBitsToInt(intBitsToFloat(R0i.y) + intBitsToFloat(PV1i.z) / resScale);
PS0i = R2i.y;
// 7
PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.z), intBitsToFloat(uf_remappedPS[2].w))); //?
// 8
PS0i = floatBitsToInt(1.0 / intBitsToFloat(PV1i.y));
// 9
PV1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.y), intBitsToFloat(PS0i)));
PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.z), intBitsToFloat(PS0i)));
// 10
PV0i.x = floatBitsToInt(max(intBitsToFloat(PV1i.z), intBitsToFloat(0xc0000000)));
PV0i.y = floatBitsToInt(max(intBitsToFloat(PV1i.w), intBitsToFloat(0xc0000000)));
// 11
PV1i.z = floatBitsToInt(min(intBitsToFloat(PV0i.x), 2.0));
PV1i.w = floatBitsToInt(min(intBitsToFloat(PV0i.y), 2.0));
// 12
PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[1].w) , intBitsToFloat(PV1i.z)));
PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[1].z) , intBitsToFloat(PV1i.w)));
// 13
backupReg0i = R0i.x ;
backupReg1i = R0i.y;
backupReg0i = R0i.x;
backupReg1i = R0i.y;
R0i.xyz = floatBitsToInt(vec3(intBitsToFloat(backupReg0i),intBitsToFloat(backupReg1i),intBitsToFloat(backupReg0i)) + vec3(-(intBitsToFloat(PV0i.w) / resScale),-(intBitsToFloat(PV0i.z) / resScale),intBitsToFloat(PV0i.w) / resScale));
R0i.w = floatBitsToInt(intBitsToFloat(backupReg1i) + intBitsToFloat(PV0i.z) / resScale);
}
if( activeMaskStackC[2] == true ) {
R1i.xyzw = floatBitsToInt(textureLod(textureUnitPS0, intBitsToFloat(R1i.xy),0.0).xyzw);
R2i.xyzw = floatBitsToInt(textureLod(textureUnitPS0, intBitsToFloat(R2i.wy),0.0).xyzw);
R3i.xyzw = floatBitsToInt(textureLod(textureUnitPS0, intBitsToFloat(R0i.xy),0.0).xyzw);
R0i.xyzw = floatBitsToInt(textureLod(textureUnitPS0, intBitsToFloat(R0i.zw),0.0).xyzw);
}
if( activeMaskStackC[2] == true ) {
// 0
R127i.xyz = floatBitsToInt(vec3(intBitsToFloat(R1i.y),intBitsToFloat(R1i.x),intBitsToFloat(R1i.w)) + vec3(intBitsToFloat(R2i.y),intBitsToFloat(R2i.x),intBitsToFloat(R2i.w)));
PV0i.y = R127i.y;
R127i.w = floatBitsToInt(intBitsToFloat(R1i.z) + intBitsToFloat(R2i.z));
// 1
backupReg0i = R0i.x;
backupReg1i = R0i.z;
PV1i.x = floatBitsToInt(intBitsToFloat(R3i.y) + intBitsToFloat(R0i.y));
PV1i.y = floatBitsToInt(intBitsToFloat(R3i.x) + intBitsToFloat(backupReg0i));
PV1i.z = floatBitsToInt(intBitsToFloat(R3i.w) + intBitsToFloat(R0i.w));
PV1i.w = floatBitsToInt(intBitsToFloat(R3i.z) + intBitsToFloat(backupReg1i));
R126i.z = PV0i.y;
R126i.z = floatBitsToInt(intBitsToFloat(R126i.z) / 2.0);
PS1i = R126i.z;
// 2
PV0i.x = floatBitsToInt(intBitsToFloat(R127i.x) + intBitsToFloat(PV1i.x));
PV0i.y = floatBitsToInt(intBitsToFloat(R127i.y) + intBitsToFloat(PV1i.y));
PV0i.z = floatBitsToInt(intBitsToFloat(R127i.z) + intBitsToFloat(PV1i.z));
PV0i.w = floatBitsToInt(intBitsToFloat(R127i.w) + intBitsToFloat(PV1i.w));
R126i.y = R127i.x;
R126i.y = floatBitsToInt(intBitsToFloat(R126i.y) / 2.0);
PS0i = R126i.y;
// 3
backupReg0i = R127i.w;
R127i.x = floatBitsToInt(intBitsToFloat(PV0i.x) * 0.25 );
PV1i.x = R127i.x;
R127i.y = floatBitsToInt(intBitsToFloat(PV0i.y) * 0.25);
R127i.z = floatBitsToInt(intBitsToFloat(PV0i.z) * 0.25);
R127i.w = floatBitsToInt(intBitsToFloat(PV0i.w) * 0.25);
R126i.x = backupReg0i;
R126i.x = floatBitsToInt(intBitsToFloat(R126i.x) / 2.0);
PS1i = R126i.x;
// 4
PV0i.x = ((intBitsToFloat(PV1i.x) > intBitsToFloat(R4i.w))?int(0xFFFFFFFF):int(0x0));
PV0i.y = ((intBitsToFloat(R4i.z) > intBitsToFloat(PV1i.x))?int(0xFFFFFFFF):int(0x0));
// 5
R123i.w = ((PV0i.y == 0)?(PV0i.x):(int(-1)));
PV1i.w = R123i.w;
// 6
R1i.x = ((PV1i.w == 0)?(R127i.y):(R126i.z));
R1i.y = ((PV1i.w == 0)?(R127i.x):(R126i.y));
R1i.z = ((PV1i.w == 0)?(R127i.w):(R126i.x));
R1i.w = ((PV1i.w == 0)?(R127i.z):(R127i.z));
}
activeMaskStackC[1] = activeMaskStack[0] == true && activeMaskStackC[0] == true;
// export

vec3 R0f =  vec3(intBitsToFloat(R1i.x), intBitsToFloat(R1i.y), intBitsToFloat(R1i.z));

R0f.xyz = contrasty(R0f.xyz);
R0f.xyz = mix(R0f.xyz, smoothstep(0.0, 1.0, R0f.xyz), contrastCurve);
float smask = lumasharping(textureUnitPS0, passParameterSem136.xy);
vec3 temp3 = R0f.xyz;
R0f.xyz = mix(R0f.xyz, (temp3.xyz += (smask)), sharp_mix);

passPixelColor0 = vec4(R0f.xyz, intBitsToFloat(R1i.w));

//passPixelColor0 = vec4(intBitsToFloat(R1i.x), intBitsToFloat(R1i.y), intBitsToFloat(R1i.z), intBitsToFloat(R1i.w));
}