mirror of
https://github.com/cemu-project/cemu_graphic_packs.git
synced 2024-11-22 09:39:17 +01:00
ac5d1a770f
This should play nice with the script update. Nasty merge conflict :( Local had folder changes and settings I never got around to commit.
246 lines
8.4 KiB
Plaintext
246 lines
8.4 KiB
Plaintext
#version 430
|
|
#extension GL_ARB_texture_gather : enable
|
|
#extension GL_ARB_separate_shader_objects : enable
|
|
// shader c14019840473ff86
|
|
// start of shader inputs/outputs, predetermined by Cemu. Do not touch
|
|
#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 GET_FRAGCOORD() vec4(gl_FragCoord.xy*uf_fragCoordScale.xy,gl_FragCoord.zw)
|
|
#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 GET_FRAGCOORD() vec4(gl_FragCoord.xy*uf_fragCoordScale,gl_FragCoord.zw)
|
|
#endif
|
|
#ifdef VULKAN
|
|
layout(set = 1, binding = 2) uniform ufBlock
|
|
{
|
|
uniform vec4 uf_fragCoordScale;
|
|
};
|
|
#else
|
|
uniform vec2 uf_fragCoordScale;
|
|
#endif
|
|
TEXTURE_LAYOUT(0, 1, 0) uniform sampler2D textureUnitPS0;
|
|
TEXTURE_LAYOUT(1, 1, 1) uniform sampler2D textureUnitPS1;
|
|
layout(location = 0) in vec4 passParameterSem0;
|
|
layout(location = 0) out vec4 passPixelColor0;
|
|
// end of shader inputs/outputs
|
|
|
|
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;
|
|
}
|
|
|
|
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 R123f = 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;
|
|
vec3 cubeMapSTM;
|
|
int cubeMapFaceId;
|
|
float scaleRes = uf_fragCoordScale.x;
|
|
R0f = passParameterSem0;
|
|
R0f.w = (textureGather(textureUnitPS1, R0f.xy).y);
|
|
// 0
|
|
R1f.x = R0f.x + intBitsToFloat(0xba088889);
|
|
R1f.y = R0f.y + intBitsToFloat(0xba72b9d6);
|
|
R2f.z = R0f.x;
|
|
R2f.y = R0f.y + intBitsToFloat(0xba72b9d6);
|
|
PS0f = R2f.y;
|
|
// 1
|
|
R3f.x = R0f.x + intBitsToFloat(0xba088889);
|
|
R3f.y = R0f.y;
|
|
R1f.xyzw = (textureGather(textureUnitPS1, R1f.xy).xyzw);
|
|
R2f.yz = (textureGather(textureUnitPS1, R2f.zy).yz);
|
|
R3f.xy = (textureGather(textureUnitPS1, R3f.xy).xy);
|
|
// 0
|
|
PV0f.x = R2f.z + R3f.x;
|
|
PV0f.y = R1f.w + R0f.w;
|
|
PV0f.z = R1f.z + R3f.y;
|
|
PV0f.w = R1f.x + R2f.y;
|
|
// 1
|
|
R123f.x = (R1f.y * 2.0 + -(PV0f.x));
|
|
PV1f.x = R123f.x;
|
|
R123f.y = (R1f.y * 2.0 + -(PV0f.y));
|
|
PV1f.y = R123f.y;
|
|
R123f.z = (R1f.y * 2.0 + -(PV0f.z));
|
|
PV1f.z = R123f.z;
|
|
R123f.w = (R1f.y * 2.0 + -(PV0f.w));
|
|
PV1f.w = R123f.w;
|
|
// 2
|
|
PV0f.x = max(PV1f.x, -(PV1f.x));
|
|
PV0f.y = max(PV1f.y, -(PV1f.y));
|
|
PV0f.z = max(PV1f.z, -(PV1f.z));
|
|
PV0f.w = max(PV1f.w, -(PV1f.w));
|
|
// 3
|
|
PV1f.x = PV0f.x + intBitsToFloat(0xb400d959);
|
|
PV1f.y = PV0f.y + intBitsToFloat(0xb400d959);
|
|
PV1f.z = PV0f.z + intBitsToFloat(0xb400d959);
|
|
PV1f.w = PV0f.w + intBitsToFloat(0xb400d959);
|
|
// 4
|
|
R123f.x = intBitsToFloat(((PV1f.w >= 0.0)?(floatBitsToInt(1.0)):(0)));
|
|
PV0f.x = R123f.x;
|
|
R123f.y = intBitsToFloat(((PV1f.z >= 0.0)?(floatBitsToInt(1.0)):(0)));
|
|
PV0f.y = R123f.y;
|
|
R123f.z = intBitsToFloat(((PV1f.y >= 0.0)?(floatBitsToInt(1.0)):(0)));
|
|
PV0f.z = R123f.z;
|
|
R123f.w = intBitsToFloat(((PV1f.x >= 0.0)?(floatBitsToInt(1.0)):(0)));
|
|
PV0f.w = R123f.w;
|
|
// 5
|
|
tempf.x = dot(vec4(PV0f.x,PV0f.y,PV0f.z,PV0f.w),vec4(0.25,0.25,0.25,0.25));
|
|
PV1f.x = tempf.x;
|
|
PV1f.y = tempf.x;
|
|
PV1f.z = tempf.x;
|
|
PV1f.w = tempf.x;
|
|
// 6
|
|
R127f.z = PV1f.x * intBitsToFloat(0x3f400000)* scaleRes;
|
|
PV0f.z = R127f.z;
|
|
// 7
|
|
R1f.x = (PV0f.z * intBitsToFloat(0xba088889) * scaleRes + R0f.x);
|
|
R1f.y = R0f.y;
|
|
R2f.z = R0f.x;
|
|
R2f.y = (PV0f.z * intBitsToFloat(0xba72b9d6) * scaleRes + R0f.y);
|
|
PS1f = R2f.y;
|
|
// 8
|
|
R3f.x = (R127f.z * intBitsToFloat(0x3a088889) * scaleRes + R0f.x);
|
|
R3f.y = R0f.y;
|
|
R0f.w = (R127f.z * intBitsToFloat(0x3a72b9d6) * scaleRes + R0f.y);
|
|
R1f.xyzw = (texture(textureUnitPS0, R1f.xy).xyzw);
|
|
R2f.xyzw = (texture(textureUnitPS0, R2f.zy).xyzw);
|
|
R3f.xyzw = (texture(textureUnitPS0, R3f.xy).xyzw);
|
|
R0f.xyzw = (texture(textureUnitPS0, R0f.xw).xyzw);
|
|
// 0
|
|
PV0f.x = R1f.w + R2f.w;
|
|
PV0f.y = R1f.z + R2f.z;
|
|
PV0f.z = R1f.y + R2f.y;
|
|
PV0f.w = R1f.x + R2f.x;
|
|
// 1
|
|
PV1f.x = R3f.w + PV0f.x;
|
|
PV1f.y = R3f.z + PV0f.y;
|
|
PV1f.z = R3f.y + PV0f.z;
|
|
PV1f.w = R3f.x + PV0f.w;
|
|
// 2
|
|
backupReg0f = R0f.y;
|
|
backupReg1f = R0f.x;
|
|
PV0f.x = R0f.w + PV1f.x;
|
|
PV0f.y = R0f.z + PV1f.y;
|
|
PV0f.z = backupReg0f + PV1f.z;
|
|
PV0f.w = backupReg1f + PV1f.w;
|
|
// 3
|
|
R0f.x = PV0f.w * 0.25;
|
|
R0f.y = PV0f.z * 0.25;
|
|
R0f.z = PV0f.y * 0.25;
|
|
R0f.w = PV0f.x * 0.25;
|
|
// export
|
|
|
|
R0f.xyz = contrasty(R0f.xyz);
|
|
R0f.xyz = mix(R0f.xyz, smoothstep(0.0, 1.0, R0f.xyz), contrastCurve);
|
|
float smask = lumasharping(textureUnitPS0, passParameterSem0.xy);
|
|
vec3 temp3 = R0f.xyz;
|
|
R0f.xyz = mix(R0f.xyz, (temp3.xyz += (smask)), sharp_mix);
|
|
|
|
passPixelColor0 = vec4(R0f.x, R0f.y, R0f.z, R0f.w);
|
|
}
|