#version 420 #extension GL_ARB_texture_gather : enable #extension GL_ARB_separate_shader_objects : enable // shader 6ea8b1aa69c0b6f7 //cutscene contrasty uniform vec2 uf_fragCoordScale; 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 , $blueShadows); // [0.000 to 2.000] Adjust shadows for Red, Green and Blue. vec3 RGB_Gamma = vec3($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; } uniform ivec4 uf_remappedPS[5]; layout(binding = 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf4e13800 res 1280x720x1 dim 1 tm: 4 format 0816 compSel: 0 1 2 5 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler0 ClampX/Y/Z: 2 2 2 border: 0 layout(binding = 2) uniform sampler2D textureUnitPS2;// Tex2 addr 0x1b163800 res 512x1x1 dim 1 tm: 4 format 001a compSel: 0 1 2 3 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler2 ClampX/Y/Z: 2 2 2 border: 0 layout(binding = 3) uniform sampler2D textureUnitPS3;// Tex3 addr 0x10ea5000 res 1280x720x1 dim 1 tm: 4 format 0011 compSel: 0 0 0 0 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler3 ClampX/Y/Z: 2 2 2 border: 0 layout(binding = 4) uniform sampler2D textureUnitPS4;// Tex4 addr 0x1aff9000 res 320x180x1 dim 1 tm: 4 format 0820 compSel: 0 1 2 3 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler4 ClampX/Y/Z: 2 2 2 border: 0 layout(binding = 5) uniform sampler2D textureUnitPS5;// Tex5 addr 0x1b072000 res 320x180x1 dim 1 tm: 4 format 0820 compSel: 0 1 2 3 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler5 ClampX/Y/Z: 2 2 2 border: 0 layout(binding = 6) uniform sampler2D textureUnitPS6;// Tex6 addr 0x1b17b000 res 1x1x1 dim 1 tm: 2 format 080e compSel: 0 4 4 5 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler6 ClampX/Y/Z: 2 2 2 border: 0 layout(binding = 7) uniform sampler2D textureUnitPS7;// Tex7 addr 0x1abdd800 res 160x90x1 dim 1 tm: 4 format 041a compSel: 0 1 2 3 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler7 ClampX/Y/Z: 2 2 2 border: 0 layout(location = 0) in vec4 passParameterSem128; 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){return mix(0.0, a*b, (a != 0.0) && (b != 0.0));} void main() { ivec4 R0i = ivec4(0); ivec4 R1i = ivec4(0); ivec4 R2i = ivec4(0); ivec4 R3i = ivec4(0); ivec4 R4i = ivec4(0); ivec4 R5i = ivec4(0); ivec4 R122i = ivec4(0); ivec4 R123i = ivec4(0); ivec4 R125i = 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; vec3 cubeMapSTM; int cubeMapFaceId; R0i = floatBitsToInt(passParameterSem128); R1i.x = floatBitsToInt(texture(textureUnitPS3, intBitsToFloat(R0i.xy)).x); R4i.xyzw = floatBitsToInt(texture(textureUnitPS4, intBitsToFloat(R0i.xy)).xyzw); R3i.xyzw = floatBitsToInt(texture(textureUnitPS0, intBitsToFloat(R0i.xy)).xyzw); R2i.xyzw = floatBitsToInt(texture(textureUnitPS5, intBitsToFloat(R0i.xy)).xyzw); R5i.xyz = floatBitsToInt(texture(textureUnitPS7, intBitsToFloat(R0i.xy)).xyz); // 0 R0i.x = 0x3f000000; R1i.y = 0x3f000000; R0i.w = floatBitsToInt(intBitsToFloat(R4i.w) + -(0.5)); R1i.x = floatBitsToInt(texture(textureUnitPS2, intBitsToFloat(R1i.xy)).w); R0i.x = floatBitsToInt(texture(textureUnitPS6, intBitsToFloat(R0i.xx)).x); // 0 PV0i.x = floatBitsToInt(intBitsToFloat(R1i.x) + -(0.5)); R127i.y = ((0.5 > intBitsToFloat(R1i.x))?int(0xFFFFFFFF):int(0x0)); PV0i.y = R127i.y; PV0i.z = floatBitsToInt(max(intBitsToFloat(R0i.w), -(intBitsToFloat(R0i.w)))); PV0i.z = floatBitsToInt(intBitsToFloat(PV0i.z) * 2.0); // 1 R123i.x = ((PV0i.y == 0)?(R2i.x):(R3i.x)); PV1i.x = R123i.x; PV1i.y = floatBitsToInt(-(intBitsToFloat(PV0i.z)) + 1.0); PV1i.y = clampFI32(PV1i.y); R123i.z = ((PV0i.y == 0)?(R2i.y):(R3i.y)); PV1i.z = R123i.z; PV1i.w = floatBitsToInt(max(intBitsToFloat(PV0i.x), -(intBitsToFloat(PV0i.x)))); PV1i.w = floatBitsToInt(intBitsToFloat(PV1i.w) * 2.0); R122i.x = ((PV0i.y == 0)?(R2i.z):(R3i.z)); PS1i = R122i.x; // 2 PV0i.x = floatBitsToInt(intBitsToFloat(PV1i.y) + intBitsToFloat(0x3727c5ac)); R126i.y = floatBitsToInt(-(intBitsToFloat(R3i.y)) + intBitsToFloat(PV1i.z)); PV0i.z = floatBitsToInt(-(intBitsToFloat(PV1i.w)) + 1.0); PV0i.z = clampFI32(PV0i.z); R127i.w = floatBitsToInt(-(intBitsToFloat(R3i.x)) + intBitsToFloat(PV1i.x)); R125i.y = floatBitsToInt(-(intBitsToFloat(R3i.z)) + intBitsToFloat(PS1i)); PS0i = R125i.y; // 3 R123i.x = ((R127i.y == 0)?(R2i.w):(R3i.w)); PV1i.x = R123i.x; PV1i.y = floatBitsToInt(intBitsToFloat(PV0i.z) + intBitsToFloat(0x3727c5ac)); PS1i = floatBitsToInt(1.0 / intBitsToFloat(PV0i.x)); // 4 R127i.x = floatBitsToInt(-(intBitsToFloat(R3i.w)) + intBitsToFloat(PV1i.x)); PV0i.z = floatBitsToInt(intBitsToFloat(PS1i) + -(1.0)); PS0i = floatBitsToInt(1.0 / intBitsToFloat(PV1i.y)); // 5 PV1i.y = floatBitsToInt(intBitsToFloat(PV0i.z) + 4.0); PV1i.w = floatBitsToInt(intBitsToFloat(PS0i) + -(1.0)); // 6 PV0i.x = floatBitsToInt(intBitsToFloat(PV1i.y) + -(4.0)); PV0i.z = floatBitsToInt(intBitsToFloat(PV1i.w) + 4.0); // 7 PV1i.y = floatBitsToInt(intBitsToFloat(PV0i.z) + -(4.0)); R126i.w = floatBitsToInt(intBitsToFloat(PV0i.x) * 0.25); R126i.w = clampFI32(R126i.w); // 8 PV0i.x = floatBitsToInt(intBitsToFloat(PV1i.y) * 0.25); PV0i.x = clampFI32(PV0i.x); // 9 backupReg0i = R127i.x; backupReg1i = R126i.y; backupReg2i = R127i.w; R127i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg0i),intBitsToFloat(PV0i.x)) + intBitsToFloat(R3i.w))); PV1i.x = R127i.x; R126i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R125i.y),intBitsToFloat(PV0i.x)) + intBitsToFloat(R3i.z))); PV1i.y = R126i.y; R127i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg1i),intBitsToFloat(PV0i.x)) + intBitsToFloat(R3i.y))); PV1i.z = R127i.z; R127i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg2i),intBitsToFloat(PV0i.x)) + intBitsToFloat(R3i.x))); PV1i.w = R127i.w; // 10 PV0i.x = floatBitsToInt(intBitsToFloat(R4i.z) + -(intBitsToFloat(PV1i.y))); PV0i.y = floatBitsToInt(intBitsToFloat(R4i.y) + -(intBitsToFloat(PV1i.z))); PV0i.z = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.w))); PV0i.w = floatBitsToInt(intBitsToFloat(R4i.w) + -(intBitsToFloat(PV1i.x))); // 11 R123i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV0i.w),intBitsToFloat(R126i.w)) + intBitsToFloat(R127i.x))); PV1i.x = R123i.x; R123i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV0i.y),intBitsToFloat(R126i.w)) + intBitsToFloat(R127i.z))); PV1i.y = R123i.y; R123i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV0i.z),intBitsToFloat(R126i.w)) + intBitsToFloat(R127i.w))); PV1i.z = R123i.z; R123i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV0i.x),intBitsToFloat(R126i.w)) + intBitsToFloat(R126i.y))); PV1i.w = R123i.w; // 12 R127i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV1i.z), intBitsToFloat(R0i.x))); PV0i.x = R127i.x; PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV1i.x), intBitsToFloat(R0i.x))); R126i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV1i.w), intBitsToFloat(R0i.x))); PV0i.z = R126i.z; R127i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV1i.y), intBitsToFloat(R0i.x))); PV0i.w = R127i.w; // 13 PV1i.x = floatBitsToInt(intBitsToFloat(PV0i.x) * intBitsToFloat(0x3e6147ae)); R127i.z = floatBitsToInt(intBitsToFloat(PV0i.z) * intBitsToFloat(0x3e6147ae)); PV1i.z = R127i.z; PV1i.w = floatBitsToInt(intBitsToFloat(PV0i.w) * intBitsToFloat(0x3e6147ae)); R0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.y), intBitsToFloat(uf_remappedPS[0].x))); PS1i = R0i.w; // 14 PV0i.x = floatBitsToInt(intBitsToFloat(PV1i.w) + intBitsToFloat(0x3e99999a)); PV0i.y = floatBitsToInt(intBitsToFloat(PV1i.x) + intBitsToFloat(0x3e99999a)); PV0i.z = floatBitsToInt(intBitsToFloat(PV1i.x) + intBitsToFloat(0x3cf5c28f)); PV0i.w = floatBitsToInt(intBitsToFloat(PV1i.z) + intBitsToFloat(0x3e99999a)); R125i.z = floatBitsToInt(intBitsToFloat(PV1i.w) + intBitsToFloat(0x3cf5c28f)); PS0i = R125i.z; // 15 backupReg0i = R127i.x; R126i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R126i.z),intBitsToFloat(PV0i.w)) + intBitsToFloat(0x3d75c28f))); PV1i.y = floatBitsToInt(intBitsToFloat(R127i.z) + intBitsToFloat(0x3cf5c28f)); R123i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R127i.x),intBitsToFloat(PV0i.y)) + intBitsToFloat(0x3d75c28f))); PV1i.z = R123i.z; R126i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R127i.w),intBitsToFloat(PV0i.x)) + intBitsToFloat(0x3d75c28f))); R127i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg0i),intBitsToFloat(PV0i.z)) + intBitsToFloat(0x3b03126f))); PS1i = R127i.x; // 16 backupReg0i = R126i.z; backupReg1i = R127i.w; R126i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg0i),intBitsToFloat(PV1i.y)) + intBitsToFloat(0x3b03126f))); R127i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg1i),intBitsToFloat(R125i.z)) + intBitsToFloat(0x3b03126f))); PS0i = floatBitsToInt(1.0 / intBitsToFloat(PV1i.z)); // 17 PV1i.z = floatBitsToInt(intBitsToFloat(R127i.x) * intBitsToFloat(PS0i)); PS1i = floatBitsToInt(1.0 / intBitsToFloat(R126i.w)); // 18 PV0i.y = floatBitsToInt(intBitsToFloat(R127i.w) * intBitsToFloat(PS1i)); PV0i.w = floatBitsToInt(intBitsToFloat(PV1i.z) + -(intBitsToFloat(0x3d086595))); PS0i = floatBitsToInt(1.0 / intBitsToFloat(R126i.x)); // 19 PV1i.x = floatBitsToInt(intBitsToFloat(R126i.z) * intBitsToFloat(PS0i)); PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.w), intBitsToFloat(uf_remappedPS[0].x))); PV1i.z = floatBitsToInt(intBitsToFloat(PV0i.y) + -(intBitsToFloat(0x3d086595))); // 20 R126i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.x),intBitsToFloat(uf_remappedPS[1].x)) + intBitsToFloat(PV1i.y))); PV0i.y = floatBitsToInt(intBitsToFloat(PV1i.x) + -(intBitsToFloat(0x3d086595))); PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV1i.z), intBitsToFloat(uf_remappedPS[0].x))); // 21 R123i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.y),intBitsToFloat(uf_remappedPS[1].x)) + intBitsToFloat(PV0i.w))); PV1i.x = R123i.x; PV1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.y), intBitsToFloat(uf_remappedPS[0].x))); // 22 backupReg0i = R126i.x; R126i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg0i), intBitsToFloat(uf_remappedPS[2].x))); R126i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV1i.x), intBitsToFloat(uf_remappedPS[2].y))); R123i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.z),intBitsToFloat(uf_remappedPS[1].x)) + intBitsToFloat(PV1i.z))); PV0i.w = R123i.w; // 23 R126i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.w), intBitsToFloat(uf_remappedPS[2].z))); PV1i.z = R126i.z; // 24 tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R126i.x),intBitsToFloat(R126i.y),intBitsToFloat(PV1i.z),-0.0),vec4(intBitsToFloat(0x3e990abb),intBitsToFloat(0x3f162c13),intBitsToFloat(0x3dea747e),0.0))); PV0i.x = tempi.x; PV0i.y = tempi.x; PV0i.z = tempi.x; PV0i.w = tempi.x; R125i.z = tempi.x; // 25 PV1i.x = floatBitsToInt(intBitsToFloat(R126i.y) + -(intBitsToFloat(PV0i.x))); PV1i.y = floatBitsToInt(intBitsToFloat(R126i.x) + -(intBitsToFloat(PV0i.x))); PV1i.w = floatBitsToInt(intBitsToFloat(R126i.z) + -(intBitsToFloat(PV0i.x))); // 26 R126i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.w),intBitsToFloat(uf_remappedPS[3].z)) + intBitsToFloat(R125i.z))); R126i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.x),intBitsToFloat(uf_remappedPS[3].y)) + intBitsToFloat(R125i.z))); R123i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.y),intBitsToFloat(uf_remappedPS[3].x)) + intBitsToFloat(R125i.z))); PV0i.z = R123i.z; // 27 tempResultf = log2(intBitsToFloat(PV0i.z)); if( isinf(tempResultf) == true ) tempResultf = -3.40282347E+38F; PS1i = floatBitsToInt(tempResultf); // 28 R125i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PS1i), intBitsToFloat(uf_remappedPS[4].x))); tempResultf = log2(intBitsToFloat(R126i.y)); if( isinf(tempResultf) == true ) tempResultf = -3.40282347E+38F; PS0i = floatBitsToInt(tempResultf); // 29 R127i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PS0i), intBitsToFloat(uf_remappedPS[4].x))); tempResultf = log2(intBitsToFloat(R126i.x)); if( isinf(tempResultf) == true ) tempResultf = -3.40282347E+38F; PS1i = floatBitsToInt(tempResultf); // 30 R126i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PS1i), intBitsToFloat(uf_remappedPS[4].x))); PS0i = floatBitsToInt(exp2(intBitsToFloat(R125i.z))); // 31 R0i.x = PS0i; PS1i = floatBitsToInt(exp2(intBitsToFloat(R127i.w))); // 32 R0i.y = PS1i; R4i.w = floatBitsToInt(exp2(intBitsToFloat(R126i.x))); PS0i = R4i.w; // 33 R0i.z = R4i.w; // export passPixelColor0 = vec4(intBitsToFloat(R0i.x), intBitsToFloat(R0i.y), intBitsToFloat(R0i.z), intBitsToFloat(R0i.w)); passPixelColor0.xyz = contrasty(passPixelColor0.xyz); passPixelColor0.xyz = mix(passPixelColor0.xyz, smoothstep(0.0, 1.0, passPixelColor0.xyz), contrastCurve); float smask = lumasharping(textureUnitPS0, passParameterSem128.xy); vec3 temp3 = passPixelColor0.xyz; passPixelColor0.xyz = mix(passPixelColor0.xyz, (temp3.xyz += (smask)), sharp_mix); }