#version 420 #extension GL_ARB_texture_gather : 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 0bcd653c18367d59 // Used for: Restoring the native BotW Anti-Aliasing implementation const float resXScale = float($width)/float($gameWidth); const float resYScale = float($height)/float($gameHeight); #ifdef VULKAN layout(set = 0, binding = 1) uniform ufBlock { uniform ivec4 uf_remappedVS[5]; // uniform vec2 uf_windowSpaceToClipSpaceTransform; // Cemu optimized this uf_variable away in Cemu 1.15.7 }; #else uniform ivec4 uf_remappedVS[5]; // uniform vec2 uf_windowSpaceToClipSpaceTransform; // Cemu optimized this uf_variable away in Cemu 1.15.7 #endif // uf_windowSpaceToClipSpaceTransform was moved to the ufBlock TEXTURE_LAYOUT(32, 0, 0) uniform sampler2D textureUnitVS0; ATTR_LAYOUT(0, 0) in uvec4 attrDataSem0; ATTR_LAYOUT(0, 1) in uvec4 attrDataSem1; out gl_PerVertex { vec4 gl_Position; float gl_PointSize; }; layout(location = 3) out vec4 passParameterSem4; layout(location = 0) out vec4 passParameterSem0; layout(location = 1) out vec4 passParameterSem1; layout(location = 2) out vec4 passParameterSem3; layout(location = 4) out vec4 passParameterSem6; 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 R5i = ivec4(0); ivec4 R6i = ivec4(0); ivec4 R7i = ivec4(0); ivec4 R123i = ivec4(0); ivec4 R124i = ivec4(0); ivec4 R125i = ivec4(0); ivec4 R126i = ivec4(0); ivec4 R127i = ivec4(0); uvec4 attrDecoder; 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 = ivec4(gl_VertexID, 0, 0, gl_InstanceID); attrDecoder.xyzw = attrDataSem0.xyzw; attrDecoder = ((attrDecoder>>8)&0xFF)|((attrDecoder<<8)&0xFF00); attrDecoder.xyzw = floatBitsToInt(vec4(unpackHalf2x16(attrDecoder.x|(attrDecoder.y<<16)),unpackHalf2x16(attrDecoder.z|(attrDecoder.w<<16)))); R1i = ivec4(int(attrDecoder.x), int(attrDecoder.y), int(attrDecoder.z), int(attrDecoder.w)); attrDecoder.x = attrDataSem1.x; attrDecoder.x = (attrDecoder.x>>24)|((attrDecoder.x>>8)&0xFF00)|((attrDecoder.x<<8)&0xFF0000)|((attrDecoder.x<<24)); attrDecoder.y = 0; attrDecoder.z = 0; attrDecoder.w = 0; attrDecoder.xyzw = uvec4((attrDecoder.x>>0)&0x3FF,(attrDecoder.x>>10)&0x3FF,(attrDecoder.x>>20)&0x3FF,(attrDecoder.x>>30)&0x3); if( (attrDecoder.x&0x200) != 0 ) attrDecoder.x |= 0xFFFFFC00; if( (attrDecoder.y&0x200) != 0 ) attrDecoder.y |= 0xFFFFFC00; if( (attrDecoder.z&0x200) != 0 ) attrDecoder.z |= 0xFFFFFC00; attrDecoder.x = floatBitsToUint(max(float(int(attrDecoder.x))/511.0,-1.0)); attrDecoder.y = floatBitsToUint(max(float(int(attrDecoder.y))/511.0,-1.0)); attrDecoder.z = floatBitsToUint(max(float(int(attrDecoder.z))/511.0,-1.0)); attrDecoder.w = floatBitsToUint(float(attrDecoder.w)); R2i = ivec4(int(attrDecoder.x), int(attrDecoder.y), int(attrDecoder.z), int(attrDecoder.w)); // 0 backupReg0i = R0i.x; PV0i.x = (backupReg0i == int(1))?int(0xFFFFFFFF):int(0x0); R127i.y = 0x3f800000; PV0i.z = 0x3f800000; PV0i.w = 0x40400000; R127i.z = 0xbf800000; PS0i = R127i.z; // 1 R123i.x = ((PV0i.x == 0)?(PV0i.z):(0xc0400000)); PV1i.x = R123i.x; R123i.y = ((PV0i.x == 0)?(PV0i.w):(0xbf800000)); PV1i.y = R123i.y; R0i.z = 0; PV1i.w = floatBitsToInt(intBitsToFloat(uf_remappedVS[0].z) * intBitsToFloat(0x3b808081)); R2i.w = 0x3f800000; PS1i = R2i.w; // 2 R2i.x = ((R0i.x == 0)?(R127i.z):(PV1i.y)); PV0i.x = R2i.x; R2i.y = ((R0i.x == 0)?(R127i.y):(PV1i.x)); PV0i.y = R2i.y; R2i.z = floatBitsToInt(intBitsToFloat(PV1i.w) + -(0.5)); R2i.z = floatBitsToInt(intBitsToFloat(R2i.z) * 2.0); R0i.w = 0x3d2aaaab; R0i.x = 0x3e000000; PS0i = R0i.x; // 3 R5i.x = floatBitsToInt((intBitsToFloat(PV0i.x) * 0.5 + 0.5)); PV1i.x = R5i.x; R0i.y = 0; PV1i.z = floatBitsToInt(-(intBitsToFloat(PV0i.y))); R4i.w = uf_remappedVS[1].z; R4i.w = floatBitsToInt(intBitsToFloat(R4i.w) / 2.0); R3i.x = floatBitsToInt(-(intBitsToFloat(uf_remappedVS[2].z))); R3i.x = floatBitsToInt(intBitsToFloat(R3i.x) / 2.0); PS1i = R3i.x; // 4 R6i.x = floatBitsToInt(-(intBitsToFloat(uf_remappedVS[3].x))); R6i.x = floatBitsToInt(intBitsToFloat(R6i.x) * 2.0); R5i.y = floatBitsToInt((intBitsToFloat(PV1i.z) * 0.5 + 0.5)); R7i.z = PV1i.x; R3i.w = R2i.x; R3i.w = floatBitsToInt(intBitsToFloat(R3i.w) / 2.0); R6i.y = uf_remappedVS[3].y; R6i.y = floatBitsToInt(intBitsToFloat(R6i.y) * 2.0); PS0i = R6i.y; R1i.w = floatBitsToInt(texture(textureUnitVS0, intBitsToFloat(R0i.wz)).x); R1i.xyz = floatBitsToInt(texture(textureUnitVS0, intBitsToFloat(R0i.xy)).xyz); // export SET_POSITION(vec4(intBitsToFloat(R2i.x), intBitsToFloat(R2i.y), intBitsToFloat(R2i.z), intBitsToFloat(R2i.w))); // export // skipped export to semanticId 255 // 0 R126i.x = floatBitsToInt(intBitsToFloat(R3i.x) + -(0.5)); R125i.y = floatBitsToInt(intBitsToFloat(R4i.w) + -(0.5)); PV0i.z = floatBitsToInt(-(intBitsToFloat(R2i.y))); PV0i.z = floatBitsToInt(intBitsToFloat(PV0i.z) / 2.0); R7i.w = R5i.y; PV0i.w = R7i.w; R127i.y = floatBitsToInt(-(intBitsToFloat(R6i.x))); R127i.y = floatBitsToInt(intBitsToFloat(R127i.y) / 2.0); PS0i = R127i.y; // 1 R127i.x = floatBitsToInt(-(intBitsToFloat(R6i.y))); R127i.x = floatBitsToInt(intBitsToFloat(R127i.x) / 2.0); R126i.y = floatBitsToInt(intBitsToFloat(R3i.w) + intBitsToFloat(R4i.w)); PV1i.z = floatBitsToInt(intBitsToFloat(PV0i.w) + intBitsToFloat(uf_remappedVS[4].w)/resYScale); PV1i.w = floatBitsToInt(intBitsToFloat(PV0i.z) + intBitsToFloat(R3i.x)); R127i.w = floatBitsToInt(intBitsToFloat(R7i.z) + intBitsToFloat(uf_remappedVS[4].z)/resXScale); PS1i = R127i.w; // 2 backupReg0i = R127i.y; R125i.x = floatBitsToInt(intBitsToFloat(PV1i.z) + intBitsToFloat(R126i.x)); R127i.y = floatBitsToInt(intBitsToFloat(R5i.x) + intBitsToFloat(R125i.y)); R126i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R6i.y), intBitsToFloat(PV1i.w))); R126i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R6i.x),intBitsToFloat(R4i.w)) + intBitsToFloat(backupReg0i))); R125i.w = floatBitsToInt(intBitsToFloat(uf_remappedVS[4].z)/resXScale); PS0i = R125i.w; // 3 backupReg0i = R126i.y; backupReg1i = R127i.x; R127i.x = floatBitsToInt(-(intBitsToFloat(uf_remappedVS[4].w)/resYScale)); R126i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R6i.x), intBitsToFloat(backupReg0i))); R127i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R6i.y),intBitsToFloat(R3i.x)) + intBitsToFloat(backupReg1i))); R124i.w = floatBitsToInt(intBitsToFloat(uf_remappedVS[4].x)*resXScale); R124i.w = floatBitsToInt(intBitsToFloat(R124i.w) / 2.0); R125i.z = floatBitsToInt(intBitsToFloat(uf_remappedVS[4].y)*resYScale); R125i.z = floatBitsToInt(intBitsToFloat(R125i.z) / 2.0); PS1i = R125i.z; // 4 backupReg0i = R126i.x; backupReg1i = R127i.w; backupReg2i = R127i.y; R126i.x = floatBitsToInt(intBitsToFloat(uf_remappedVS[4].w)/resYScale); R126i.x = floatBitsToInt(intBitsToFloat(R126i.x) * 2.0); R127i.y = floatBitsToInt(intBitsToFloat(uf_remappedVS[4].z)/resXScale); R127i.y = floatBitsToInt(intBitsToFloat(R127i.y) * 2.0); R124i.z = floatBitsToInt(intBitsToFloat(R5i.y) + intBitsToFloat(backupReg0i)); R127i.w = floatBitsToInt(intBitsToFloat(backupReg1i) + intBitsToFloat(R125i.y)); R4i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R6i.x), intBitsToFloat(backupReg2i))); PS0i = R4i.w; // 5 backupReg0i = R127i.z; R2i.x = floatBitsToInt(intBitsToFloat(R125i.w) * 1.5); R2i.y = floatBitsToInt(intBitsToFloat(R127i.x) * 1.5); R127i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R6i.y), intBitsToFloat(R125i.x))); R6i.w = backupReg0i; R6i.z = R126i.w; PS1i = R6i.z; // 6 R3i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.x),intBitsToFloat(R124i.w)) + 0.5)); R3i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.y),intBitsToFloat(R125i.z)) + 0.5)); R5i.z = R126i.y; R5i.w = R126i.z; R3i.z = R127i.y; PS0i = R3i.z; // 7 R4i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R6i.x), intBitsToFloat(R127i.w))); R4i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R6i.y), intBitsToFloat(R124i.z))); R4i.z = R4i.w; R3i.w = R126i.x; R4i.w = R127i.z; PS1i = R4i.w; // export passParameterSem4 = vec4(intBitsToFloat(R3i.x), intBitsToFloat(R3i.y), intBitsToFloat(R3i.z), intBitsToFloat(R3i.w)); // export // skipped export to semanticId 255 // export passParameterSem0 = vec4(intBitsToFloat(R5i.x), intBitsToFloat(R5i.y), intBitsToFloat(R5i.z), intBitsToFloat(R5i.w)); // export passParameterSem1 = vec4(intBitsToFloat(R6i.x), intBitsToFloat(R6i.y), intBitsToFloat(R6i.z), intBitsToFloat(R6i.w)); // export passParameterSem3 = vec4(intBitsToFloat(R2i.x), intBitsToFloat(R2i.y), intBitsToFloat(R2i.z), intBitsToFloat(R2i.z)); // export passParameterSem6 = vec4(intBitsToFloat(R7i.x), intBitsToFloat(R7i.x), intBitsToFloat(R7i.z), intBitsToFloat(R7i.w)); }