cemu_graphic_packs/Enhancements/WindWakerHD_Contrasty/ff71dcd2ad4defdc_00000000000003c9_ps.txt

#version 420
#extension GL_ARB_texture_gather : enable
// shader ff71dcd2ad4defdc
//AA ps

uniform vec2 uf_fragCoordScale;

const float hazeFactor = $hazeFactor;

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 = 2.0;
uniform ivec4 uf_remappedPS[4];
layout(binding = 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf5807800 res 1920x1080x1 dim 1 tm: 4 format 0019 compSel: 0 1 2 3 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler0 ClampX/Y/Z: 2 2 2 border: 1
layout(binding = 1) uniform sampler2D textureUnitPS1;// Tex1 addr 0xf4000800 res 1920x1080x1 dim 1 tm: 4 format 0001 compSel: 0 4 4 5 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler1 ClampX/Y/Z: 2 2 2 border: 1
layout(location = 0) in vec4 passParameterSem2;
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()
{
vec4 R0f = vec4(0.0);
vec4 R1f = vec4(0.0);
vec4 R2f = vec4(0.0);
vec4 R3f = vec4(0.0);
vec4 R4f = vec4(0.0);
vec4 R123f = vec4(0.0);
vec4 R126f = vec4(0.0);
vec4 R127f = vec4(0.0);
float backupReg0f, backupReg1f, backupReg2f, backupReg3f, backupReg4f;
vec4 PV0f = vec4(0.0), PV1f = vec4(0.0);
float PS0f = 0.0, PS1f = 0.0;
vec4 tempf = vec4(0.0);
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;
R0f = passParameterSem2;
if( activeMaskStackC[1] == true ) {
R1f.xyzw = (textureGather(textureUnitPS1, R0f.xy).wzxy);
R2f.xyzw = (texture(textureUnitPS0, R0f.xy).xyzw);
}
if( activeMaskStackC[1] == true ) {
activeMaskStack[1] = activeMaskStack[0];
activeMaskStackC[2] = activeMaskStackC[1];
// 0
PV0f.x = R1f.y + R1f.x;
R127f.y = intBitsToFloat(uf_remappedPS[0].z) * 0.25;
PV0f.z = R1f.w + -(R1f.x);
PV0f.w = R1f.z + -(R1f.y);
R127f.z = mul_nonIEEE(R2f.x, intBitsToFloat(uf_remappedPS[1].x));
PS0f = R127f.z;
// 1
R127f.x = PV0f.w + -(PV0f.z);
PV1f.x = R127f.x;
R126f.y = PV0f.w + PV0f.z;
PV1f.y = R126f.y;
PV1f.z = R1f.z + PV0f.x;
R127f.w = min(R1f.z, R1f.x);
R126f.w = min(R1f.w, R1f.y);
PS1f = R126f.w;
// 2
PV0f.x = R1f.w + PV1f.z;
PV0f.y = max(PV1f.x, -(PV1f.x));
PV0f.z = max(PV1f.y, -(PV1f.y));
PV0f.w = max(R1f.z, R1f.x);
PS0f = max(R1f.w, R1f.y);
// 3
PV1f.x = min(PV0f.z, PV0f.y);
R123f.y = (mul_nonIEEE(R2f.y,intBitsToFloat(uf_remappedPS[1].y)) + R127f.z);
PV1f.y = R123f.y;
R127f.z = min(R127f.w, R126f.w);
PV1f.z = R127f.z;
R123f.w = (mul_nonIEEE(R127f.y,PV0f.x) + intBitsToFloat(uf_remappedPS[0].w));
PV1f.w = R123f.w;
R127f.w = max(PV0f.w, PS0f);
PS1f = R127f.w;
// 4
PV0f.x = mul_nonIEEE(PS1f, intBitsToFloat(uf_remappedPS[2].x));
PV0f.y = max(PV1f.y, PS1f);
PV0f.z = min(PV1f.y, PV1f.z);
PV0f.w = max(PV1f.w, PV1f.x);
R4f.z = -(PV1f.z) + PS1f;
PS0f = R4f.z;
// 5
R3f.x = max(PV0f.x, intBitsToFloat(uf_remappedPS[2].y));
R3f.y = -(PV0f.z) + PV0f.y;
R4f.w = R127f.z + R127f.w;
PS1f = 1.0 / PV0f.w;
// 6
PV0f.x = mul_nonIEEE(R127f.x, PS1f); //p
PV0f.y = mul_nonIEEE(R126f.y, PS1f);
// 7
PV1f.z = max(PV0f.x, -(intBitsToFloat(uf_remappedPS[0].y))); //p
PV1f.w = max(PV0f.y, -(intBitsToFloat(uf_remappedPS[0].y)));
// 8
R1f.x = min(PV1f.w, intBitsToFloat(uf_remappedPS[0].y));
R1f.y = min(PV1f.z, intBitsToFloat(uf_remappedPS[0].y));
// 9
predResult = (R3f.y > R3f.x);
activeMaskStack[1] = predResult;
activeMaskStackC[2] = predResult == true && activeMaskStackC[1] == true;
}
else {
activeMaskStack[1] = false;
activeMaskStackC[2] = false;
}
if( activeMaskStackC[2] == true ) {
// 0
R3f.x = (mul_nonIEEE(R1f.x,-(intBitsToFloat(uf_remappedPS[3].z) / resScale)) + R0f.x);
R3f.y = (mul_nonIEEE(R1f.y,-(intBitsToFloat(uf_remappedPS[3].w) / resScale)) + R0f.y);
R0f.z = (mul_nonIEEE(R1f.x,intBitsToFloat(uf_remappedPS[3].z)) + R0f.x); //leave out, looks better on avg
R0f.w = (mul_nonIEEE(R1f.y,intBitsToFloat(uf_remappedPS[3].w)) + R0f.y);//leave out, looks better on avg
R4f.x = (mul_nonIEEE(R1f.x,-(intBitsToFloat(uf_remappedPS[3].x) / resScale)) + R0f.x);
PS0f = R4f.x;
// 1
R4f.y = (mul_nonIEEE(R1f.y,-(intBitsToFloat(uf_remappedPS[3].y)/resScale)) + R0f.y);
R3f.z = (mul_nonIEEE(R1f.x,intBitsToFloat(uf_remappedPS[3].x) / resScale) + R0f.x);
R3f.w = (mul_nonIEEE(R1f.y,intBitsToFloat(uf_remappedPS[3].y) / resScale) + R0f.y);
}
if( activeMaskStackC[2] == true ) {
R1f.xyzw = (texture(textureUnitPS0, R0f.zw).xyzw);
R0f.xyzw = (texture(textureUnitPS0, R3f.xy).xyzw);
R2f.xyzw = (texture(textureUnitPS0, R4f.xy).xyzw);
R3f.xyzw = (texture(textureUnitPS0, R3f.zw).xyzw);
}
if( activeMaskStackC[2] == true ) {
// 0
R127f.xyz = vec3(R0f.x,R0f.w,R0f.z) + vec3(R1f.x,R1f.w,R1f.z);
PV0f.x = R127f.x;
PV0f.y = R127f.y;
PV0f.z = R127f.z;
R127f.w = R0f.y + R1f.y;
PV0f.w = R127f.w;
// 1
PV1f.x = R2f.x + PV0f.x;
PV1f.y = R2f.w + PV0f.y;
PV1f.z = R2f.z + PV0f.z;
PV1f.w = R2f.y + PV0f.w;
// 2
R126f.x = R3f.x + PV1f.x;
R126f.x /= 2.0;
PV0f.x = R126f.x;
R126f.y = R3f.w + PV1f.y;
R126f.y /= 2.0;
PV0f.y = R126f.y;
R126f.z = R3f.z + PV1f.z;
R126f.z /= 2.0;
PV0f.z = R126f.z;
R126f.w = R3f.y + PV1f.w;
R126f.w /= 2.0;
PV0f.w = R126f.w;
// 3
backupReg0f = R127f.y;
R123f.x = (mul_nonIEEE(PV0f.x,intBitsToFloat(uf_remappedPS[1].x)) + -(R4f.w));
PV1f.x = R123f.x;
R127f.yzw = vec3(R127f.z,R127f.w,R127f.x) + vec3(-(PV0f.z),-(PV0f.w),-(PV0f.x));
R127f.x = backupReg0f + -(PV0f.y);
PS1f = R127f.x;
// 4
R123f.w = (mul_nonIEEE(R126f.w,intBitsToFloat(uf_remappedPS[1].y)) + PV1f.x);
PV0f.w = R123f.w;
// 5
PV1f.z = max(PV0f.w, -(PV0f.w));
// 6
PV0f.y = -(R4f.z) + PV1f.z;
// 7
R123f.x = intBitsToFloat(((PV0f.y >= 0.0)?(floatBitsToInt(1.0)):(0)));
PV1f.x = R123f.x;
// 8
R2f.x = (mul_nonIEEE(R127f.w,PV1f.x) + R126f.x)/2.0;
R2f.y = (mul_nonIEEE(R127f.z,PV1f.x) + R126f.w)/2.0;
R2f.z = (mul_nonIEEE(R127f.y,PV1f.x) + R126f.z)/2.0;
R2f.w = (mul_nonIEEE(R127f.x,PV1f.x) + R126f.y)/2.0;
}
activeMaskStackC[1] = activeMaskStack[0] == true && activeMaskStackC[0] == true;
// export

R2f.xyz = contrasty(R2f.xyz);
R2f.xyz = mix(R2f.xyz, smoothstep(0.0, 1.0, R2f.xyz), contrastCurve);
float smask = lumasharping(textureUnitPS1, passParameterSem2.xy);
vec3 temp3 = R2f.xyz;
R2f.xyz = mix(R2f.xyz, (temp3.xyz += (smask)), sharp_mix);

passPixelColor0 = vec4(R2f.x, R2f.y, R2f.z, R2f.w);
}