#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. // shader 840947e29015aa9a //BB cliff const float dither = $dither ; const float scaleShader = $scaleShader; const float scaleBlur = $scaleBlur; const int sampleScale = 4; float lineRand(vec2 co) { float a = 12.9898; float b = 78.233; float c = 43758.5453; float dt = dot(co.xy, vec2(a, b)); float sn = mod(dt, 3.14); return fract(sin(sn) * c); } #ifdef VULKAN layout(set = 1, binding = 2) uniform ufBlock { uniform ivec4 uf_remappedPS[3]; uniform vec4 uf_fragCoordScale; }; #else uniform ivec4 uf_remappedPS[3]; 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 = 1) in vec4 passParameterSem1; layout(location = 0) out vec4 passPixelColor0; // uf_fragCoordScale was moved to the ufBlock // FabriceNeyret2 CC, single shader gaussian by intermediate MIPmap level. www.shadertoy.com/view/ltScRG const int samples = 8 * sampleScale, //8 or 4 balances xy position LOD = 2, // gaussian done on MIPmap at scale LOD sLOD = 1 << LOD; // tile size = 2^LOD const float sigma = float(samples) * .25; float gaussian(vec2 i) { return exp(-.5* dot(i /= sigma, i)) / (6.28 * sigma*sigma); } vec4 blur(sampler2D sp, vec2 U, vec2 scale) { vec4 O = vec4(0); int s = samples / sLOD; for (int i = 0; i < s*s; i++) { vec2 d = vec2(i%s, i / s)*float(sLOD) - float(samples) / 2.; O += gaussian(d) * textureLod(sp, U + scale * d, float(LOD)); } return O / O.a; } 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 min(a*b,min(abs(a)*3.40282347E+38F,abs(b)*3.40282347E+38F)); } 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 R5f = vec4(0.0); vec4 R6f = vec4(0.0); vec4 R7f = vec4(0.0); vec4 R8f = 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; R0f = passParameterSem0; R1f = passParameterSem1; R0f.xy = R0f.xy - (lineRand(gl_FragCoord.xy)*0.0012 *dither); R0f.xy = R0f.xy + (lineRand(gl_FragCoord.xy*vec2(0.1, 0.09))*0.0011 *dither); // 0 R2f.x = R1f.x + intBitsToFloat(uf_remappedPS[0].x); R2f.y = R1f.y + intBitsToFloat(uf_remappedPS[0].y); R0f.z = R1f.x + intBitsToFloat(uf_remappedPS[0].z); R0f.w = R1f.y + intBitsToFloat(uf_remappedPS[0].w); // 1 backupReg0f = R1f.x; backupReg1f = R1f.y; backupReg0f = R1f.x; backupReg1f = R1f.y; R1f.xyz = vec3(backupReg0f,backupReg1f,backupReg0f) + vec3(intBitsToFloat(uf_remappedPS[1].x),intBitsToFloat(uf_remappedPS[1].y),intBitsToFloat(uf_remappedPS[1].z)); R1f.w = backupReg1f + intBitsToFloat(uf_remappedPS[1].w); vec2 coord = passParameterSem0.xy*textureSize(textureUnitPS0, 0); vec2 ps = vec2(1.0) / textureSize(textureUnitPS0, 0); vec2 uv = coord * ps; R3f.xyz = (texture(textureUnitPS1, R2f.xy).xyz); R4f.xyz = (texture(textureUnitPS1, R0f.zw).xyz); R5f.xyz = (texture(textureUnitPS1, R1f.xy).xyz); R6f.xyz = (texture(textureUnitPS1, R1f.zw).xyz); R2f.xyz = blur(textureUnitPS0, R2f.xy, ps*scaleBlur).xyz; R7f.xyz = R2f.xyz; R8f.xyz = R2f.xyz; R1f.xyz = R2f.xyz; //R2f.xyz = (texture(textureUnitPS0, R2f.xy).xyz); //R7f.xyz = (texture(textureUnitPS0, R0f.zw).xyz); //R8f.xyz = (texture(textureUnitPS0, R1f.xy).xyz); //R1f.xyz = (texture(textureUnitPS0, R1f.zw).xyz); // 0 tempf.x = dot(vec4(R3f.x,R3f.y,R3f.z,-0.0),vec4(intBitsToFloat(0x3e000000),intBitsToFloat(0x41ff0000),intBitsToFloat(0x45fe0100),0.0)); PV0f.x = tempf.x; PV0f.y = tempf.x; PV0f.z = tempf.x; PV0f.w = tempf.x; R127f.w = tempf.x; R127f.z = R2f.x + R7f.x; PS0f = R127f.z; // 1 tempf.x = dot(vec4(R4f.x,R4f.y,R4f.z,-0.0),vec4(intBitsToFloat(0x3e000000),intBitsToFloat(0x41ff0000),intBitsToFloat(0x45fe0100),0.0)); PV1f.x = tempf.x; PV1f.y = tempf.x; PV1f.z = tempf.x; PV1f.w = tempf.x; R127f.y = R2f.y + R7f.y; PS1f = R127f.y; // 2 tempf.x = dot(vec4(R5f.x,R5f.y,R5f.z,-0.0),vec4(intBitsToFloat(0x3e000000),intBitsToFloat(0x41ff0000),intBitsToFloat(0x45fe0100),0.0)); PV0f.x = tempf.x; PV0f.y = tempf.x; PV0f.z = tempf.x; PV0f.w = tempf.x; PS0f = R127f.w + PV1f.x; // 3 tempf.x = dot(vec4(R6f.x,R6f.y,R6f.z,-0.0),vec4(intBitsToFloat(0x3e000000),intBitsToFloat(0x41ff0000),intBitsToFloat(0x45fe0100),0.0)); PV1f.x = tempf.x; PV1f.y = tempf.x; PV1f.z = tempf.x; PV1f.w = tempf.x; PS1f = PS0f + PV0f.x; // 4 PV0f.y = PS1f + PV1f.x; PV0f.z = R127f.y + R8f.y; PV0f.w = R2f.z + R7f.z; // 5 R123f.x = (PV0f.y * 0.25 + -(R0f.x)); PV1f.x = R123f.x; PV1f.y = R127f.z + R8f.x; PV1f.z = PV0f.z + R1f.y; PV1f.w = PV0f.w + R8f.z; // 6 PV0f.x = PV1f.y + R1f.x; PV0f.y = PV1f.w + R1f.z; PV0f.z = PV1f.z * 0.25; PV0f.w = max(PV1f.x, -(PV1f.x)); // 7 backupReg0f = R0f.y; PV1f.x = PV0f.x * 0.25; PV1f.y = PV0f.y * 0.25; PV1f.z = backupReg0f * PV0f.w; PV1f.z = clamp(PV1f.z, 0.0, 1.0); PV1f.w = max(PV0f.z, 0.0); // 8 PV0f.x = max(PV1f.y, 0.0); R0f.y = min(PV1f.w, 4.0); PV0f.z = max(PV1f.x, 0.0); tempResultf = log2(PV1f.z); if( isinf(tempResultf) == true ) tempResultf = -3.40282347E+38F; PS0f = tempResultf; // 9 PV1f.x = PS0f * intBitsToFloat(uf_remappedPS[2].w); R0f.z = min(PV0f.x, 4.0); R0f.x = min(PV0f.z, 4.0); PS1f = R0f.x; // 10 PS0f = exp2(PV1f.x); // 11 R0f.w = PS0f * intBitsToFloat(uf_remappedPS[2].z); // export passPixelColor0 = vec4(R0f.x, R0f.y, R0f.z, R0f.w); }