From 91e6222239856fc872a2bf7f770facec06331500 Mon Sep 17 00:00:00 2001 From: Valery Guskov Date: Wed, 15 Nov 2017 00:52:43 +0200 Subject: [PATCH 01/10] added AMD light circles experimental fix shamelessly based on @alexkiri work --- .../cf2e9b7aab93e342_000000000001fe49_ps.txt | 523 ++++++++++++++++++ .../e95d54571be1c0a9_000000000001fe49_ps.txt | 470 ++++++++++++++++ .../BreathOfTheWild_!AMDLights/rules.txt | 8 + 3 files changed, 1001 insertions(+) create mode 100644 Workaround/BreathOfTheWild_!AMDLights/cf2e9b7aab93e342_000000000001fe49_ps.txt create mode 100644 Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt create mode 100644 Workaround/BreathOfTheWild_!AMDLights/rules.txt diff --git a/Workaround/BreathOfTheWild_!AMDLights/cf2e9b7aab93e342_000000000001fe49_ps.txt b/Workaround/BreathOfTheWild_!AMDLights/cf2e9b7aab93e342_000000000001fe49_ps.txt new file mode 100644 index 00000000..699569ae --- /dev/null +++ b/Workaround/BreathOfTheWild_!AMDLights/cf2e9b7aab93e342_000000000001fe49_ps.txt @@ -0,0 +1,523 @@ +#version 420 +#extension GL_ARB_texture_gather : enable +// shader cf2e9b7aab93e342 +// Sheikah Tower - inactive, reflection pass +// Glow Lights Color Mod +// Credit for hsv functions http://lolengine.net/blog/2013/07/27/rgb-to-hsv-in-glsl + +#define HUE_ROTATION 0 //[0, 360] where 0 and 360 is unchanged Hue and 180 is completely opposite Hue. Check http://dba.med.sc.edu/price/irf/Adobe_tg/models/images/hsl_top.JPG +#define SATURATION_FACTOR 0.9 //[0.0, 1.0] 1.0 means unchanged Saturation, 0.0 means completely desaturated. Values above 1.0 are accepted, but they may cause clipping +#define VALUE_FACTOR 0.65 //same as above; applies to Value +#define ALPHA_FACTOR 1.0 //same as above; applies to Transparency + +const float hueRotation = HUE_ROTATION / 360.0; +uniform ivec4 uf_remappedPS[10]; +layout(binding = 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf5054000 res 640x360x1 dim 1 tm: 4 format 080e compSel: 0 4 4 5 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 0xf494a800 res 1280x720x1 dim 1 tm: 4 format 001a compSel: 0 1 2 3 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler1 ClampX/Y/Z: 2 2 2 border: 1 +layout(binding = 2) uniform sampler2D textureUnitPS2;// Tex2 addr 0x3df6f800 res 64x64x1 dim 1 tm: 4 format 0806 compSel: 0 4 4 5 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler2 ClampX/Y/Z: 2 2 2 border: 1 +layout(location = 1) noperspective in vec4 passParameterSem0; +layout(location = 0) out vec4 passPixelColor0; +layout(location = 1) out vec4 passPixelColor1; +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; } +vec3 rgb2hsv(vec3 c) { + vec4 K = vec4(0.0, -1.0 / 3.0, 2.0 / 3.0, -1.0); + vec4 p = mix(vec4(c.bg, K.wz), vec4(c.gb, K.xy), step(c.b, c.g)); + vec4 q = mix(vec4(p.xyw, c.r), vec4(c.r, p.yzx), step(p.x, c.r)); + + float d = q.x - min(q.w, q.y); + float e = 1.0e-10; + return vec3(abs(q.z + (q.w - q.y) / (6.0 * d + e)), d / (q.x + e), q.x); +} +vec3 hsv2rgb(vec3 c) { + vec4 K = vec4(1.0, 2.0 / 3.0, 1.0 / 3.0, 3.0); + vec3 p = abs(fract(c.xxx + K.xyz) * 6.0 - K.www); + return c.z * mix(K.xxx, clamp(p - K.xxx, 0.0, 1.0), c.y); +} +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 R8i = ivec4(0); +ivec4 R9i = ivec4(0); +ivec4 R10i = ivec4(0); +ivec4 R11i = ivec4(0); +ivec4 R12i = ivec4(0); +ivec4 R13i = ivec4(0); +ivec4 R122i = ivec4(0); +ivec4 R123i = ivec4(0); +ivec4 R124i = 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(vec4(gl_FragCoord.xy*uf_fragCoordScale,gl_FragCoord.zw)); +R1i = floatBitsToInt(passParameterSem0); +// 0 +R123i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.x),intBitsToFloat(R1i.x)) + 1.0)); +PV0i.x = R123i.x; +R6i.y = floatBitsToInt(-(intBitsToFloat(R1i.y)) + intBitsToFloat(uf_remappedPS[0].y)); +R0i.z = 0; +R2i.w = floatBitsToInt(-(intBitsToFloat(R1i.x)) + intBitsToFloat(uf_remappedPS[0].x)); +PS0i = floatBitsToInt(1.0 / intBitsToFloat(uf_remappedPS[1].x)); +// 1 +R2i.x = floatBitsToInt(intBitsToFloat(R0i.x) * intBitsToFloat(PS0i)); +R0i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.y),intBitsToFloat(R1i.y)) + intBitsToFloat(PV0i.x))); +PS1i = floatBitsToInt(1.0 / intBitsToFloat(uf_remappedPS[1].y)); +// 2 +R2i.y = floatBitsToInt(intBitsToFloat(R0i.y) * intBitsToFloat(PS1i)); +R13i.w = 0; +R1i.w = floatBitsToInt(1.0 / intBitsToFloat(uf_remappedPS[2].w)); +PS0i = R1i.w; +R3i.xyzw = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R2i.xy)).xyzw); +R2i.x = floatBitsToInt(texture(textureUnitPS0, intBitsToFloat(R2i.xy)).x); +// 0 +R127i.x = floatBitsToInt((intBitsToFloat(R3i.x) * 2.0 + -(1.0))); +PV0i.x = R127i.x; +R127i.y = floatBitsToInt((intBitsToFloat(R3i.y) * 2.0 + -(1.0))); +PV0i.y = R127i.y; +R127i.z = floatBitsToInt((intBitsToFloat(R3i.z) * 2.0 + -(1.0))); +PV0i.z = R127i.z; +R127i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(uf_remappedPS[3].y),intBitsToFloat(R2i.x)) + intBitsToFloat(uf_remappedPS[3].x))); +PV0i.w = R127i.w; +tempResultf = 1.0 / sqrt(intBitsToFloat(R0i.w)); +R126i.w = floatBitsToInt(tempResultf); +PS0i = R126i.w; +// 1 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),-0.0),vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),0.0))); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +R126i.y = floatBitsToInt(-(intBitsToFloat(PV0i.w))); +PS1i = R126i.y; +// 2 +backupReg0i = R127i.w; +R125i.x = floatBitsToInt(-(intBitsToFloat(PS1i)) + intBitsToFloat(uf_remappedPS[4].z)); +R9i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R1i.x), intBitsToFloat(R126i.w))); +R126i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R1i.x), -(intBitsToFloat(R127i.w)))); +PV0i.z = R126i.z; +R127i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R1i.y), -(intBitsToFloat(backupReg0i)))); +PV0i.w = R127i.w; +tempResultf = 1.0 / sqrt(intBitsToFloat(PV1i.x)); +PS0i = floatBitsToInt(tempResultf); +// 3 +R4i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PS0i))); +R11i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.y), intBitsToFloat(PS0i))); +R6i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.z), intBitsToFloat(PS0i))); +R124i.w = floatBitsToInt(-(intBitsToFloat(PV0i.z)) + intBitsToFloat(uf_remappedPS[4].x)); +PV1i.w = R124i.w; +R125i.y = floatBitsToInt(-(intBitsToFloat(PV0i.w)) + intBitsToFloat(uf_remappedPS[4].y)); +PS1i = R125i.y; +// 4 +tempi.x = floatBitsToInt(dot(vec4(-(intBitsToFloat(PV1i.w)),-(intBitsToFloat(PS1i)),-(intBitsToFloat(R125i.x)),-0.0),vec4(intBitsToFloat(uf_remappedPS[2].x),intBitsToFloat(uf_remappedPS[2].y),intBitsToFloat(uf_remappedPS[2].z),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R10i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R1i.y), intBitsToFloat(R126i.w))); +PS0i = R10i.y; +// 5 +R0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R1i.z), intBitsToFloat(R126i.w))); +PV1i.x = R0i.x; +R3i.y = floatBitsToInt(-(intBitsToFloat(R3i.w)) + 1.0); +PV1i.y = R3i.y; +R127i.z = floatBitsToInt(intBitsToFloat(PV0i.x) * intBitsToFloat(R1i.w)); +R127i.z = clampFI32(R127i.z); +PV1i.z = R127i.z; +R0i.w = floatBitsToInt(-(intBitsToFloat(uf_remappedPS[0].x)) + intBitsToFloat(uf_remappedPS[5].x)); +R5i.y = floatBitsToInt(-(intBitsToFloat(uf_remappedPS[0].y)) + intBitsToFloat(uf_remappedPS[5].y)); +PS1i = R5i.y; +// 6 +R123i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(uf_remappedPS[2].x),intBitsToFloat(PV1i.z)) + intBitsToFloat(uf_remappedPS[4].x))); +PV0i.x = R123i.x; +R123i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(uf_remappedPS[2].y),intBitsToFloat(PV1i.z)) + intBitsToFloat(uf_remappedPS[4].y))); +PV0i.y = R123i.y; +R125i.z = floatBitsToInt((intBitsToFloat(PV1i.y) * 0.5 + 0.5)); +R123i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.y)) + intBitsToFloat(0x3d4ccccd))); +R123i.w = clampFI32(R123i.w); +PV0i.w = R123i.w; +PS0i = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R6i.z), intBitsToFloat(PV1i.x))); +// 7 +R126i.x = floatBitsToInt(intBitsToFloat(R126i.z) + -(intBitsToFloat(PV0i.x))); +R127i.y = floatBitsToInt(intBitsToFloat(R127i.w) + -(intBitsToFloat(PV0i.y))); +PV1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.w), intBitsToFloat(PV0i.w))); +R123i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(uf_remappedPS[2].z),intBitsToFloat(R127i.z)) + intBitsToFloat(uf_remappedPS[4].z))); +PV1i.w = R123i.w; +R122i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R11i.y),intBitsToFloat(R10i.y)) + intBitsToFloat(PS0i))); +PS1i = R122i.x; +// 8 +backupReg0i = R125i.z; +backupReg0i = R125i.z; +R127i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R4i.x),intBitsToFloat(R9i.y)) + intBitsToFloat(PS1i))); +PV0i.x = R127i.x; +R12i.y = floatBitsToInt(intBitsToFloat(PV1i.z) + -(1.0)); +R125i.z = floatBitsToInt(intBitsToFloat(R126i.y) + -(intBitsToFloat(PV1i.w))); +PV0i.z = R125i.z; +R4i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg0i), intBitsToFloat(backupReg0i))); +R4i.w = floatBitsToInt(intBitsToFloat(R4i.w) / 2.0); +R8i.z = floatBitsToInt(intBitsToFloat(PV1i.z) * intBitsToFloat(0x3ea2f983)); +PS0i = R8i.z; +// 9 +tempi.x = floatBitsToInt(dot(vec4(-(intBitsToFloat(R126i.x)),-(intBitsToFloat(R127i.y)),-(intBitsToFloat(PV0i.z)),-0.0),vec4(-(intBitsToFloat(R126i.x)),-(intBitsToFloat(R127i.y)),-(intBitsToFloat(PV0i.z)),0.0))); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +R3i.z = tempi.x; +PS1i = PV0i.x; +PS1i = floatBitsToInt(intBitsToFloat(PS1i) * 2.0); +// 10 +R1i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R4i.x),intBitsToFloat(PS1i)) + -(intBitsToFloat(R9i.y)))); +R1i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R11i.y),intBitsToFloat(PS1i)) + -(intBitsToFloat(R10i.y)))); +R1i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R6i.z),intBitsToFloat(PS1i)) + -(intBitsToFloat(R0i.x)))); +R127i.w = R127i.x; +R127i.w = clampFI32(R127i.w); +tempResultf = 1.0 / sqrt(intBitsToFloat(PV1i.x)); +PS0i = floatBitsToInt(tempResultf); +// 11 +PV1i.x = floatBitsToInt(mul_nonIEEE(-(intBitsToFloat(R126i.x)), intBitsToFloat(PS0i))); +PV1i.y = floatBitsToInt(mul_nonIEEE(-(intBitsToFloat(R127i.y)), intBitsToFloat(PS0i))); +R5i.z = floatBitsToInt(mul_nonIEEE(-(intBitsToFloat(R125i.z)), intBitsToFloat(PS0i))); +PV1i.z = R5i.z; +R125i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[6].w), intBitsToFloat(PS0i))); +PV1i.w = R125i.w; +R2i.x = floatBitsToInt(-(intBitsToFloat(R4i.w)) + 1.0); +PS1i = R2i.x; +// 12 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R4i.x),intBitsToFloat(R11i.y),intBitsToFloat(R6i.z),-0.0),vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.z),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R126i.w = tempi.x; +R126i.y = floatBitsToInt(1.0 / intBitsToFloat(PV1i.w)); +PS0i = R126i.y; +// 13 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(uf_remappedPS[2].x),intBitsToFloat(uf_remappedPS[2].y),intBitsToFloat(uf_remappedPS[2].z),-0.0),vec4(intBitsToFloat(R1i.x),intBitsToFloat(R1i.y),intBitsToFloat(R1i.z),0.0))); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +R127i.y = tempi.x; +R122i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R127i.w),intBitsToFloat(R2i.x)) + intBitsToFloat(R4i.w))); +PS1i = R122i.x; +// 14 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R124i.w),intBitsToFloat(R125i.y),intBitsToFloat(R125i.x),-0.0),vec4(intBitsToFloat(uf_remappedPS[2].x),intBitsToFloat(uf_remappedPS[2].y),intBitsToFloat(uf_remappedPS[2].z),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R127i.w = tempi.x; +R7i.z = floatBitsToInt(1.0 / intBitsToFloat(PS1i)); +PS0i = R7i.z; +// 15 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R124i.w),intBitsToFloat(R125i.y),intBitsToFloat(R125i.x),-0.0),vec4(intBitsToFloat(R1i.x),intBitsToFloat(R1i.y),intBitsToFloat(R1i.z),0.0))); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +R125i.z = tempi.x; +R126i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.x), intBitsToFloat(R127i.y))); +PS1i = R126i.x; +// 16 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R124i.w),intBitsToFloat(R125i.y),intBitsToFloat(R125i.x),-0.0),vec4(intBitsToFloat(R124i.w),intBitsToFloat(R125i.y),intBitsToFloat(R125i.x),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +PS0i = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.w), intBitsToFloat(PV1i.x))); +// 17 +backupReg0i = R126i.x; +R126i.x = floatBitsToInt(max(intBitsToFloat(R126i.w), -(1.0))); +R123i.y = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(uf_remappedPS[2].w)),intBitsToFloat(R125i.z)) + intBitsToFloat(backupReg0i))); +PV1i.y = R123i.y; +R125i.z = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(PV0i.x)),intBitsToFloat(R127i.y)) + intBitsToFloat(PS0i))); +R126i.w = floatBitsToInt(intBitsToFloat(R126i.y) * intBitsToFloat(0x40400000)); +R5i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R125i.w), intBitsToFloat(R125i.w))); +PS1i = R5i.w; +// 18 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R0i.w),intBitsToFloat(R5i.y),intBitsToFloat(R0i.z),-0.0),vec4(intBitsToFloat(R0i.w),intBitsToFloat(R5i.y),intBitsToFloat(R5i.y),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R127i.z = tempi.x; +PS0i = floatBitsToInt(1.0 / intBitsToFloat(PV1i.y)); +// 19 +backupReg0i = R126i.x; +R126i.x = floatBitsToInt(min(intBitsToFloat(backupReg0i), 1.0)); +R0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(uf_remappedPS[7].y))); +R4i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R126i.y),intBitsToFloat(uf_remappedPS[8].x)) + intBitsToFloat(uf_remappedPS[8].y))); +PV1i.w = floatBitsToInt(intBitsToFloat(R125i.z) * intBitsToFloat(PS0i)); +PV1i.w = clampFI32(PV1i.w); +PS1i = floatBitsToInt(1.0 / intBitsToFloat(R126i.w)); +// 20 +R3i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(uf_remappedPS[2].x),intBitsToFloat(PV1i.w)) + intBitsToFloat(R124i.w))); +PV0i.x = R3i.x; +R8i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(uf_remappedPS[2].y),intBitsToFloat(PV1i.w)) + intBitsToFloat(R125i.y))); +PV0i.y = R8i.y; +R2i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(uf_remappedPS[2].z),intBitsToFloat(PV1i.w)) + intBitsToFloat(R125i.x))); +PV0i.z = R2i.z; +R3i.w = floatBitsToInt(intBitsToFloat(R3i.y) + intBitsToFloat(PS1i)); +R3i.w = clampFI32(R3i.w); +R2i.y = floatBitsToInt(1.0 / intBitsToFloat(R127i.z)); +PS0i = R2i.y; +// 21 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R1i.x),intBitsToFloat(R1i.y),intBitsToFloat(R1i.z),-0.0),vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),0.0))); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +R1i.w = tempi.x; +R4i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R126i.x),intBitsToFloat(uf_remappedPS[8].z)) + intBitsToFloat(uf_remappedPS[8].w))); +PS1i = R4i.y; +R7i.y = floatBitsToInt(texture(textureUnitPS2, intBitsToFloat(R4i.yy)).x); +// 0 +R126i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.x),intBitsToFloat(R1i.w)) + -(intBitsToFloat(R3i.x)))); +PV0i.x = R126i.x; +R125i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.y),intBitsToFloat(R1i.w)) + -(intBitsToFloat(R8i.y)))); +PV0i.y = R125i.y; +R127i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.z),intBitsToFloat(R1i.w)) + -(intBitsToFloat(R2i.z)))); +PV0i.z = R127i.z; +R124i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.z), intBitsToFloat(uf_remappedPS[9].y))); +R124i.w = clampFI32(R124i.w); +PS0i = floatBitsToInt(1.0 / intBitsToFloat(R3i.w)); +// 1 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),-0.0),vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),0.0))); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +R126i.w = floatBitsToInt(intBitsToFloat(R3i.y) * intBitsToFloat(PS0i)); +PS1i = R126i.w; +// 2 +backupReg0i = R0i.z; +tempi.x = floatBitsToInt(dot(vec4(-(intBitsToFloat(R2i.w)),-(intBitsToFloat(R6i.y)),-(intBitsToFloat(R6i.y)),-0.0),vec4(intBitsToFloat(R0i.w),intBitsToFloat(R5i.y),intBitsToFloat(backupReg0i),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +PS0i = floatBitsToInt(sqrt(intBitsToFloat(PV1i.x))); +// 3 +backupReg0i = R0i.y; +R125i.x = floatBitsToInt(intBitsToFloat(PV0i.x) * intBitsToFloat(R2i.y)); +R125i.x = clampFI32(R125i.x); +PV1i.x = R125i.x; +PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R7i.y), intBitsToFloat(backupReg0i))); +R123i.z = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R124i.w)),intBitsToFloat(R124i.w)) + 1.0)); +PV1i.z = R123i.z; +PV1i.w = ((intBitsToFloat(R4i.z) > intBitsToFloat(uf_remappedPS[7].x))?int(0xFFFFFFFF):int(0x0)); +PS1i = floatBitsToInt(1.0 / intBitsToFloat(PS0i)); +// 4 +R123i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R0i.w),intBitsToFloat(PV1i.x)) + intBitsToFloat(R2i.w))); +PV0i.x = R123i.x; +R126i.y = ((PV1i.w == 0)?(R7i.y):(PV1i.y)); +R125i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.y),intBitsToFloat(PV1i.x)) + intBitsToFloat(R6i.y))); +PV0i.w = floatBitsToInt(intBitsToFloat(uf_remappedPS[6].w) * intBitsToFloat(PS1i)); +PV0i.w = clampFI32(PV0i.w); +R5i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV1i.z), intBitsToFloat(PV1i.z))); +PS0i = R5i.y; +// 5 +backupReg0i = R126i.x; +backupReg1i = R125i.y; +backupReg2i = R127i.z; +R126i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg0i),intBitsToFloat(PV0i.w)) + intBitsToFloat(R3i.x))); +PV1i.x = R126i.x; +R125i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg1i),intBitsToFloat(PV0i.w)) + intBitsToFloat(R8i.y))); +PV1i.y = R125i.y; +R127i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg2i),intBitsToFloat(PV0i.w)) + intBitsToFloat(R2i.z))); +PV1i.z = R127i.z; +R124i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.x), intBitsToFloat(PV0i.x))); +PS1i = floatBitsToInt(-(intBitsToFloat(uf_remappedPS[0].z)) + intBitsToFloat(uf_remappedPS[5].z)); +// 6 +backupReg0i = R125i.x; +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.z),-0.0),vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.z),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R125i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PS1i),intBitsToFloat(backupReg0i)) + intBitsToFloat(uf_remappedPS[0].z))); +PS0i = R125i.x; +// 7 +R127i.x = floatBitsToInt(max(-(intBitsToFloat(R5i.z)), 0.0)); +R127i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[6].y), intBitsToFloat(R126i.y))); +R123i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R125i.z),intBitsToFloat(R125i.z)) + intBitsToFloat(R124i.w))); +PV1i.z = R123i.z; +R124i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[6].x), intBitsToFloat(R126i.y))); +tempResultf = 1.0 / sqrt(intBitsToFloat(PV0i.x)); +PS1i = floatBitsToInt(tempResultf); +// 8 +backupReg0i = R126i.x; +backupReg1i = R127i.z; +backupReg2i = R126i.y; +R126i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg0i), intBitsToFloat(PS1i))); +PV0i.x = R126i.x; +R126i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R125i.y), intBitsToFloat(PS1i))); +PV0i.y = R126i.y; +R127i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg1i), intBitsToFloat(PS1i))); +PV0i.z = R127i.z; +R125i.w = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R125i.x)),intBitsToFloat(PV1i.z)) + 1.0)); +R125i.w = clampFI32(R125i.w); +R124i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[6].z), intBitsToFloat(backupReg2i))); +PS0i = R124i.y; +// 9 +backupReg0i = R127i.x; +R127i.x = floatBitsToInt(intBitsToFloat(R9i.y) + intBitsToFloat(PV0i.x)); +PV1i.x = R127i.x; +R125i.y = floatBitsToInt(intBitsToFloat(R10i.y) + intBitsToFloat(PV0i.y)); +PV1i.y = R125i.y; +R125i.z = floatBitsToInt(intBitsToFloat(R0i.x) + intBitsToFloat(PV0i.z)); +PV1i.z = R125i.z; +PV1i.w = floatBitsToInt(min(intBitsToFloat(backupReg0i), 1.0)); +PS1i = floatBitsToInt(-(intBitsToFloat(R5i.y)) + 1.0); +// 10 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.z),-0.0),vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.z),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R8i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.w),intBitsToFloat(PS1i)) + intBitsToFloat(R5i.y))); +PS0i = R8i.y; +// 11 +backupReg0i = R126i.y; +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R4i.x),intBitsToFloat(R11i.y),intBitsToFloat(R6i.z),-0.0),vec4(intBitsToFloat(R126i.x),intBitsToFloat(backupReg0i),intBitsToFloat(R127i.z),0.0))); +tempi.x = clampFI32(tempi.x); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +R126i.y = tempi.x; +tempResultf = 1.0 / sqrt(intBitsToFloat(PV0i.x)); +PS1i = floatBitsToInt(tempResultf); +// 12 +backupReg0i = R127i.x; +backupReg1i = R125i.y; +backupReg2i = R125i.z; +R127i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg0i), intBitsToFloat(PS1i))); +PV0i.x = R127i.x; +R125i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg1i), intBitsToFloat(PS1i))); +PV0i.y = R125i.y; +R125i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg2i), intBitsToFloat(PS1i))); +PV0i.z = R125i.z; +R123i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.x),intBitsToFloat(R2i.x)) + intBitsToFloat(R4i.w))); +PV0i.w = R123i.w; +R125i.x = floatBitsToInt(intBitsToFloat(R124i.w) * intBitsToFloat(0x3ea2f983)); +PS0i = R125i.x; +// 13 +backupReg0i = R0i.x; +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R9i.y),intBitsToFloat(R10i.y),intBitsToFloat(backupReg0i),-0.0),vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),0.0))); +tempi.x = clampFI32(tempi.x); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +R126i.x = floatBitsToInt(1.0 / intBitsToFloat(PV0i.w)); +PS1i = R126i.x; +// 14 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R4i.x),intBitsToFloat(R11i.y),intBitsToFloat(R6i.z),-0.0),vec4(intBitsToFloat(R127i.x),intBitsToFloat(R125i.y),intBitsToFloat(R125i.z),0.0))); +tempi.x = clampFI32(tempi.x); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R125i.z = floatBitsToInt(-(intBitsToFloat(PV1i.x)) + 1.0); +PS0i = R125i.z; +// 15 +backupReg0i = R127i.y; +PV1i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.x), intBitsToFloat(PV0i.x))); +R127i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PS0i), intBitsToFloat(PS0i))); +PV1i.y = R127i.y; +PV1i.z = floatBitsToInt(intBitsToFloat(backupReg0i) * intBitsToFloat(0x3ea2f983)); +R124i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R7i.z), intBitsToFloat(R126i.x))); +PS1i = floatBitsToInt(intBitsToFloat(R124i.y) * intBitsToFloat(0x3ea2f983)); +// 16 +PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R125i.z), intBitsToFloat(PV1i.y))); +PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.y), intBitsToFloat(PS1i))); +PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.y), intBitsToFloat(R125i.x))); +R123i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R12i.y),intBitsToFloat(PV1i.x)) + 1.0)); +PV0i.w = R123i.w; +PS0i = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.y), intBitsToFloat(PV1i.z))); +// 17 +R13i.x = floatBitsToInt(min(intBitsToFloat(PV0i.z), intBitsToFloat(0x42000000))); +PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.w), intBitsToFloat(PV0i.w))); +R13i.z = floatBitsToInt(min(intBitsToFloat(PV0i.y), intBitsToFloat(0x42000000))); +PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.y), intBitsToFloat(PV0i.x))); +R13i.y = floatBitsToInt(min(intBitsToFloat(PS0i), intBitsToFloat(0x42000000))); +PS1i = R13i.y; +// 18 +R125i.z = floatBitsToInt((intBitsToFloat(PV1i.w) * intBitsToFloat(0x3f75c28f) + intBitsToFloat(0x3d23d70a))); +PS0i = floatBitsToInt(1.0 / intBitsToFloat(PV1i.y)); +// 19 +PV1i.w = floatBitsToInt(intBitsToFloat(R8i.z) * intBitsToFloat(PS0i)); +// 20 +PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV1i.w), intBitsToFloat(R125i.z))); +// 21 +PV1i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R124i.w), intBitsToFloat(PV0i.y))); +// 22 +PV0i.w = floatBitsToInt(intBitsToFloat(PV1i.x) * 0.25); +// 23 +PV1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R126i.y), intBitsToFloat(PV0i.w))); +// 24 +PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(PV1i.z))); +// 25 +PV1i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R126i.w), intBitsToFloat(PV0i.y))); +// 26 +PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[6].z), intBitsToFloat(PV1i.x))); +PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[6].y), intBitsToFloat(PV1i.x))); +PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[6].x), intBitsToFloat(PV1i.x))); +// 27 +PV1i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R125i.w), intBitsToFloat(PV0i.w))); +PV1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R125i.w), intBitsToFloat(PV0i.y))); +PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R125i.w), intBitsToFloat(PV0i.z))); +// 28 +backupReg0i = R8i.y; +backupReg0i = R8i.y; +R4i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R8i.y), intBitsToFloat(PV1i.w))); +R8i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg0i), intBitsToFloat(PV1i.x))); +R5i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg0i), intBitsToFloat(PV1i.z))); +// 0 +backupReg0i = R4i.x; +R4i.x = floatBitsToInt(min(intBitsToFloat(R8i.y), intBitsToFloat(0x42000000))); +R4i.y = floatBitsToInt(min(intBitsToFloat(backupReg0i), intBitsToFloat(0x42000000))); +R4i.z = floatBitsToInt(min(intBitsToFloat(R5i.w), intBitsToFloat(0x42000000))); +// 1 +R0i.xyz = ivec3(R13i.x,R13i.y,R13i.z); +R0i.w = R13i.w; +// 2 +R1i.xyz = ivec3(R4i.x,R4i.y,R4i.z); +R1i.w = R4i.w; +// export +passPixelColor0 = vec4(intBitsToFloat(R0i.x), intBitsToFloat(R0i.y), intBitsToFloat(R0i.z), intBitsToFloat(R0i.w)); +passPixelColor1 = vec4(intBitsToFloat(R1i.x), intBitsToFloat(R1i.y), intBitsToFloat(R1i.z), intBitsToFloat(R1i.w)); +vec3 colhsv = rgb2hsv(passPixelColor0.rgb); +passPixelColor0.rgb = hsv2rgb(vec3(mod(colhsv.x + hueRotation, 1.0), colhsv.y*SATURATION_FACTOR, colhsv.z*VALUE_FACTOR)); +passPixelColor0.a *= ALPHA_FACTOR; +colhsv = rgb2hsv(passPixelColor1.rgb); +colhsv = rgb2hsv(passPixelColor1.rgb); +passPixelColor1.rgb = hsv2rgb(vec3(mod(colhsv.x + hueRotation, 1.0), colhsv.y*SATURATION_FACTOR, colhsv.z*VALUE_FACTOR)); +passPixelColor1.a *= ALPHA_FACTOR; +} diff --git a/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt b/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt new file mode 100644 index 00000000..50e24577 --- /dev/null +++ b/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt @@ -0,0 +1,470 @@ +#version 420 +#extension GL_ARB_texture_gather : enable +// shader e95d54571be1c0a9 +// Shrine - balls, reflection pass +// Glow Lights Color Mod +// Credit for hsv functions http://lolengine.net/blog/2013/07/27/rgb-to-hsv-in-glsl + +#define HUE_ROTATION 0 //[0, 360] where 0 and 360 is unchanged Hue and 180 is completely opposite Hue. Check http://dba.med.sc.edu/price/irf/Adobe_tg/models/images/hsl_top.JPG +#define SATURATION_FACTOR 0.9 //[0.0, 1.0] 1.0 means unchanged Saturation, 0.0 means completely desaturated. Values above 1.0 are accepted, but they may cause clipping +#define VALUE_FACTOR 0.65 //same as above; applies to Value +#define ALPHA_FACTOR 1.0 //same as above; applies to Transparency + +const float hueRotation = HUE_ROTATION / 360.0; +uniform ivec4 uf_remappedPS[8]; +layout(binding = 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf5054000 res 640x360x1 dim 1 tm: 4 format 080e compSel: 0 4 4 5 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 0xf494a800 res 1280x720x1 dim 1 tm: 4 format 001a compSel: 0 1 2 3 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler1 ClampX/Y/Z: 2 2 2 border: 1 +layout(binding = 2) uniform sampler2D textureUnitPS2;// Tex2 addr 0x3df6f800 res 64x64x1 dim 1 tm: 4 format 0806 compSel: 0 4 4 5 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler2 ClampX/Y/Z: 2 2 2 border: 1 +layout(location = 1) noperspective in vec4 passParameterSem0; +layout(location = 0) out vec4 passPixelColor0; +layout(location = 1) out vec4 passPixelColor1; +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; } +vec3 rgb2hsv(vec3 c) { + vec4 K = vec4(0.0, -1.0 / 3.0, 2.0 / 3.0, -1.0); + vec4 p = mix(vec4(c.bg, K.wz), vec4(c.gb, K.xy), step(c.b, c.g)); + vec4 q = mix(vec4(p.xyw, c.r), vec4(c.r, p.yzx), step(p.x, c.r)); + + float d = q.x - min(q.w, q.y); + float e = 1.0e-10; + return vec3(abs(q.z + (q.w - q.y) / (6.0 * d + e)), d / (q.x + e), q.x); +} +vec3 hsv2rgb(vec3 c) { + vec4 K = vec4(1.0, 2.0 / 3.0, 1.0 / 3.0, 3.0); + vec3 p = abs(fract(c.xxx + K.xyz) * 6.0 - K.www); + return c.z * mix(K.xxx, clamp(p - K.xxx, 0.0, 1.0), c.y); +} +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 R8i = ivec4(0); +ivec4 R123i = ivec4(0); +ivec4 R124i = 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(vec4(gl_FragCoord.xy*uf_fragCoordScale,gl_FragCoord.zw)); +R1i = floatBitsToInt(passParameterSem0); +// 0 +PV0i.x = floatBitsToInt(intBitsToFloat(R1i.x) + -(intBitsToFloat(uf_remappedPS[0].x))); +R127i.y = floatBitsToInt(intBitsToFloat(R1i.y) + -(intBitsToFloat(uf_remappedPS[0].y))); +R123i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.x),intBitsToFloat(R1i.x)) + 1.0)); +PV0i.w = R123i.w; +PS0i = floatBitsToInt(1.0 / intBitsToFloat(uf_remappedPS[1].x)); +// 1 +R2i.x = floatBitsToInt(intBitsToFloat(R0i.x) * intBitsToFloat(PS0i)); +PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.x), intBitsToFloat(PV0i.x))); +R123i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.y),intBitsToFloat(R1i.y)) + intBitsToFloat(PV0i.w))); +PV1i.z = R123i.z; +R5i.w = floatBitsToInt((0.0 * intBitsToFloat(uf_remappedPS[2].w) + 1.0)); +PS1i = floatBitsToInt(1.0 / intBitsToFloat(uf_remappedPS[1].y)); +// 2 +R0i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R127i.y),intBitsToFloat(R127i.y)) + intBitsToFloat(PV1i.y))); +R2i.y = floatBitsToInt(intBitsToFloat(R0i.y) * intBitsToFloat(PS1i)); +R8i.w = 0; +tempResultf = 1.0 / sqrt(intBitsToFloat(PV1i.z)); +R0i.y = floatBitsToInt(tempResultf); +PS0i = R0i.y; +R3i.xyzw = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R2i.xy)).xyzw); +R2i.x = floatBitsToInt(texture(textureUnitPS0, intBitsToFloat(R2i.xy)).x); +// 0 +R127i.x = floatBitsToInt((intBitsToFloat(R3i.x) * 2.0 + -(1.0))); +PV0i.x = R127i.x; +R127i.y = floatBitsToInt((intBitsToFloat(R3i.y) * 2.0 + -(1.0))); +PV0i.y = R127i.y; +R127i.z = floatBitsToInt((intBitsToFloat(R3i.z) * 2.0 + -(1.0))); +PV0i.z = R127i.z; +R127i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(uf_remappedPS[3].y),intBitsToFloat(R2i.x)) + intBitsToFloat(uf_remappedPS[3].x))); +PV0i.w = R127i.w; +R126i.w = floatBitsToInt(intBitsToFloat(R3i.w) * intBitsToFloat(0x437f0000)); +PS0i = R126i.w; +// 1 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),-0.0),vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),0.0))); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +R124i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.x),-(intBitsToFloat(PV0i.w))) + -(intBitsToFloat(uf_remappedPS[4].x)))); +PS1i = R124i.x; +// 2 +R4i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R1i.x), intBitsToFloat(R0i.y))); +R2i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.y),-(intBitsToFloat(R127i.w))) + -(intBitsToFloat(uf_remappedPS[4].y)))); +R125i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.z),-(intBitsToFloat(R127i.w))) + -(intBitsToFloat(uf_remappedPS[4].z)))); +R0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R1i.y), intBitsToFloat(R0i.y))); +tempResultf = 1.0 / sqrt(intBitsToFloat(PV1i.x)); +PS0i = floatBitsToInt(tempResultf); +// 3 +R5i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PS0i))); +R1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.y), intBitsToFloat(PS0i))); +R4i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.z), intBitsToFloat(PS0i))); +R1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R1i.z), intBitsToFloat(R0i.y))); +PS1i = int(intBitsToFloat(R126i.w)); +// 4 +tempi.x = floatBitsToInt(dot(vec4(-(intBitsToFloat(R124i.x)),-(intBitsToFloat(R2i.y)),-(intBitsToFloat(R125i.z)),-0.0),vec4(-(intBitsToFloat(R124i.x)),-(intBitsToFloat(R2i.y)),-(intBitsToFloat(R125i.z)),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R1i.x = PS1i & 0x00000002; +PS0i = R1i.x; +// 5 +R125i.x = floatBitsToInt(max(intBitsToFloat(PV0i.x), intBitsToFloat(0x3a83126f))); +PV1i.x = R125i.x; +R0i.y = floatBitsToInt(-(intBitsToFloat(R3i.w)) + 1.0); +R7i.z = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(uf_remappedPS[0].z)),intBitsToFloat(R0i.x)) + 1.0)); +R7i.z = clampFI32(R7i.z); +PS1i = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R4i.z), intBitsToFloat(R1i.w))); +// 6 +R126i.x = floatBitsToInt(dot(vec4(intBitsToFloat(R5i.x),intBitsToFloat(R1i.y),intBitsToFloat(PS1i),-0.0),vec4(intBitsToFloat(R4i.x),intBitsToFloat(R0i.w),1.0,0.0))); +PV0i.x = R126i.x; +PV0i.y = R126i.x; +PV0i.z = R126i.x; +PV0i.w = R126i.x; +tempResultf = 1.0 / sqrt(intBitsToFloat(PV1i.x)); +PS0i = floatBitsToInt(tempResultf); +// 7 +PV1i.x = floatBitsToInt(mul_nonIEEE(-(intBitsToFloat(R124i.x)), intBitsToFloat(PS0i))); +PV1i.y = floatBitsToInt(mul_nonIEEE(-(intBitsToFloat(R2i.y)), intBitsToFloat(PS0i))); +R5i.z = floatBitsToInt(mul_nonIEEE(-(intBitsToFloat(R125i.z)), intBitsToFloat(PS0i))); +PV1i.z = R5i.z; +PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[5].w), intBitsToFloat(PS0i))); +R126i.w = PV0i.x; +R126i.w = floatBitsToInt(intBitsToFloat(R126i.w) * 2.0); +PS1i = R126i.w; +// 8 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R5i.x),intBitsToFloat(R1i.y),intBitsToFloat(R4i.z),-0.0),vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.z),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R126i.z = tempi.x; +R124i.y = floatBitsToInt(min(intBitsToFloat(PV1i.w), 1.0)); +PS0i = R124i.y; +// 9 +R127i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.x),intBitsToFloat(R126i.w)) + -(intBitsToFloat(R4i.x)))); +PV1i.x = R127i.x; +R127i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.y),intBitsToFloat(R126i.w)) + -(intBitsToFloat(R0i.w)))); +PV1i.y = R127i.y; +R127i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R4i.z),intBitsToFloat(R126i.w)) + -(intBitsToFloat(R1i.w)))); +PV1i.z = R127i.z; +R123i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R0i.y),intBitsToFloat(R0i.y)) + intBitsToFloat(0x3d4ccccd))); +R123i.w = clampFI32(R123i.w); +PV1i.w = R123i.w; +R126i.y = floatBitsToInt((intBitsToFloat(R0i.y) * 0.5 + 0.5)); +PS1i = R126i.y; +// 10 +tempi.x = floatBitsToInt(dot(vec4(-(intBitsToFloat(R124i.x)),-(intBitsToFloat(R2i.y)),-(intBitsToFloat(R125i.z)),-0.0),vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.z),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R125i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV1i.w), intBitsToFloat(PV1i.w))); +PS0i = R125i.y; +// 11 +backupReg0i = R127i.x; +backupReg1i = R127i.z; +R127i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg0i),intBitsToFloat(PV0i.x)) + intBitsToFloat(R124i.x))); +PV1i.x = R127i.x; +R3i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R127i.y),intBitsToFloat(PV0i.x)) + intBitsToFloat(R2i.y))); +PV1i.y = R3i.y; +R127i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg1i),intBitsToFloat(PV0i.x)) + intBitsToFloat(R125i.z))); +PV1i.z = R127i.z; +R2i.w = floatBitsToInt(intBitsToFloat(PS0i) + -(1.0)); +R127i.y = floatBitsToInt(max(intBitsToFloat(R126i.z), -(1.0))); +PS1i = R127i.y; +// 12 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.z),-0.0),vec4(intBitsToFloat(PV1i.x),intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.z),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R2i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R126i.y), intBitsToFloat(R126i.y))); +R2i.x = floatBitsToInt(intBitsToFloat(R2i.x) / 2.0); +PS0i = R2i.x; +// 13 +R3i.x = floatBitsToInt(-(intBitsToFloat(PS0i)) + 1.0); +PV1i.x = R3i.x; +PV1i.y = R126i.x; +PV1i.y = clampFI32(PV1i.y); +R6i.z = floatBitsToInt(intBitsToFloat(R125i.y) * intBitsToFloat(0x3ea2f983)); +PV1i.w = floatBitsToInt(min(intBitsToFloat(R127i.y), 1.0)); +PS1i = floatBitsToInt(sqrt(intBitsToFloat(PV0i.x))); +// 14 +backupReg0i = R125i.x; +R125i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.x)) + intBitsToFloat(R2i.x))); +R127i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg0i), intBitsToFloat(uf_remappedPS[2].y))); +R127i.y = clampFI32(R127i.y); + + + + + + + +R126i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.w),intBitsToFloat(uf_remappedPS[6].z)) + intBitsToFloat(uf_remappedPS[6].w))); + + + + + + + +PV0i.z = R126i.z; +R4i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R124i.y), intBitsToFloat(R124i.y))); +PV0i.w = R4i.w; +PS0i = floatBitsToInt(1.0 / intBitsToFloat(PS1i)); +// 15 +backupReg0i = R124i.y; +R0i.x = ((intBitsToFloat(PV0i.z) > 0.0)?int(0xFFFFFFFF):int(0x0)); +R124i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.z), intBitsToFloat(PV0i.z))); +R2i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.w), intBitsToFloat(uf_remappedPS[7].y))); +PV1i.w = floatBitsToInt(intBitsToFloat(uf_remappedPS[5].w) * intBitsToFloat(PS0i)); +PV1i.w = clampFI32(PV1i.w); +R126i.x = floatBitsToInt(1.0 / intBitsToFloat(backupReg0i)); +PS1i = R126i.x; +// 16 +backupReg0i = R127i.x; +backupReg1i = R127i.z; +R127i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg0i),intBitsToFloat(PV1i.w)) + -(intBitsToFloat(R124i.x)))); +PV0i.x = R127i.x; +R125i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R3i.y),intBitsToFloat(PV1i.w)) + -(intBitsToFloat(R2i.y)))); +PV0i.y = R125i.y; +R127i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(backupReg1i),intBitsToFloat(PV1i.w)) + -(intBitsToFloat(R125i.z)))); +PV0i.z = R127i.z; +PV0i.w = floatBitsToInt(intBitsToFloat(PS1i) * intBitsToFloat(0x40400000)); +R3i.y = floatBitsToInt(1.0 / intBitsToFloat(R125i.x)); +PS0i = R3i.y; +// 17 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),-0.0),vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),0.0))); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +PS1i = floatBitsToInt(1.0 / intBitsToFloat(PV0i.w)); +// 18 +PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R126i.z), intBitsToFloat(R124i.y))); +R6i.y = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R127i.y)),intBitsToFloat(R127i.y)) + 1.0)); + + + + +//!!!!!!!!!!!!!!! +R3i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R126i.x),intBitsToFloat(uf_remappedPS[6].x)) + intBitsToFloat(uf_remappedPS[6].y))); + + + + + + +PV0i.w = floatBitsToInt(intBitsToFloat(R0i.y) + intBitsToFloat(PS1i)); +PV0i.w = clampFI32(PV0i.w); +tempResultf = 1.0 / sqrt(intBitsToFloat(PV1i.x)); +PS0i = floatBitsToInt(tempResultf); +// 19 +backupReg0i = R0i.x; +R0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PS0i))); +PV1i.x = R0i.x; +R2i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R125i.y), intBitsToFloat(PS0i))); +PV1i.y = R2i.y; +R1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.z), intBitsToFloat(PS0i))); +PV1i.z = R1i.z; +R123i.w = ((backupReg0i == 0)?(0):(PV0i.x)); +PV1i.w = R123i.w; +PS1i = floatBitsToInt(1.0 / intBitsToFloat(PV0i.w)); +// 20 +backupReg0i = R1i.x; +backupReg1i = R0i.y; +R1i.x = floatBitsToInt(intBitsToFloat(R4i.x) + intBitsToFloat(PV1i.x)); +R0i.y = floatBitsToInt(intBitsToFloat(R0i.w) + intBitsToFloat(PV1i.y)); +R0i.z = floatBitsToInt(intBitsToFloat(R1i.w) + intBitsToFloat(PV1i.z)); + + + + + + + + +R3i.w = R126i.z; + + + + + + + + + + + +R4i.y = floatBitsToInt(intBitsToFloat(backupReg1i) * intBitsToFloat(PS1i)); +PS0i = R4i.y; +R5i.y = floatBitsToInt(texture(textureUnitPS2, intBitsToFloat(R3i.ww)).x); +// 0 +backupReg0i = R0i.y; +backupReg0i = R0i.y; +backupReg1i = R0i.z; +backupReg1i = R0i.z; +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R1i.x),intBitsToFloat(backupReg0i),intBitsToFloat(backupReg1i),-0.0),vec4(intBitsToFloat(R1i.x),intBitsToFloat(backupReg0i),intBitsToFloat(backupReg1i),0.0))); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R127i.y = ((intBitsToFloat(R3i.z) > intBitsToFloat(uf_remappedPS[7].x))?int(0xFFFFFFFF):int(0x0)); +PS0i = R127i.y; +// 1 +R125i.x = floatBitsToInt(dot(vec4(intBitsToFloat(R5i.x),intBitsToFloat(R1i.y),intBitsToFloat(R4i.z),-0.0),vec4(intBitsToFloat(R0i.x),intBitsToFloat(R2i.y),intBitsToFloat(R1i.z),0.0))); +R125i.x = clampFI32(R125i.x); +PV1i.x = R125i.x; +PV1i.y = R125i.x; +PV1i.z = R125i.x; +PV1i.w = R125i.x; +tempResultf = 1.0 / sqrt(intBitsToFloat(PV0i.x)); +PS1i = floatBitsToInt(tempResultf); +// 2 +R127i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R1i.x), intBitsToFloat(PS1i))); +PV0i.x = R127i.x; +R125i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R0i.y), intBitsToFloat(PS1i))); +PV0i.y = R125i.y; +R126i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R0i.z), intBitsToFloat(PS1i))); +PV0i.z = R126i.z; +R123i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.x),intBitsToFloat(R3i.x)) + intBitsToFloat(R2i.x))); +PV0i.w = R123i.w; +R127i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.y), intBitsToFloat(R2i.z))); +PS0i = R127i.w; +// 3 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R4i.x),intBitsToFloat(R0i.w),intBitsToFloat(R1i.w),-0.0),vec4(intBitsToFloat(PV0i.x),intBitsToFloat(PV0i.y),intBitsToFloat(PV0i.z),0.0))); +tempi.x = clampFI32(tempi.x); +PV1i.x = tempi.x; +PV1i.y = tempi.x; +PV1i.z = tempi.x; +PV1i.w = tempi.x; +R126i.w = floatBitsToInt(1.0 / intBitsToFloat(PV0i.w)); +PS1i = R126i.w; +// 4 +tempi.x = floatBitsToInt(dot(vec4(intBitsToFloat(R5i.x),intBitsToFloat(R1i.y),intBitsToFloat(R4i.z),-0.0),vec4(intBitsToFloat(R127i.x),intBitsToFloat(R125i.y),intBitsToFloat(R126i.z),0.0))); +tempi.x = clampFI32(tempi.x); +PV0i.x = tempi.x; +PV0i.y = tempi.x; +PV0i.z = tempi.x; +PV0i.w = tempi.x; +R125i.y = floatBitsToInt(-(intBitsToFloat(PV1i.x)) + 1.0); +PS0i = R125i.y; +// 5 +R127i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PS0i), intBitsToFloat(PS0i))); +PV1i.x = R127i.x; +R126i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R4i.y), intBitsToFloat(R4i.y))); +R126i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.y), intBitsToFloat(R126i.w))); +PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.x), intBitsToFloat(PV0i.x))); +PS1i = floatBitsToInt(max(-(intBitsToFloat(R5i.z)), 0.0)); +// 6 +R126i.x = ((R127i.y == 0)?(R5i.y):(R127i.w)); +PV0i.x = R126i.x; +R127i.y = floatBitsToInt(min(intBitsToFloat(PS1i), 1.0)); +R123i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R2i.w),intBitsToFloat(PV1i.w)) + 1.0)); +PV0i.z = R123i.z; +PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R125i.y), intBitsToFloat(PV1i.x))); +// 7 +PV1i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.z), intBitsToFloat(PV0i.z))); +R124i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R6i.y), intBitsToFloat(R6i.y))); +PV1i.y = R124i.y; +PV1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PV0i.w))); +PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[5].x), intBitsToFloat(PV0i.x))); +R127i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[5].y), intBitsToFloat(PV0i.x))); +PS1i = R127i.x; +// 8 +PV0i.x = floatBitsToInt(-(intBitsToFloat(PV1i.y)) + 1.0); +R125i.y = floatBitsToInt((intBitsToFloat(PV1i.z) * intBitsToFloat(0x3f75c28f) + intBitsToFloat(0x3d23d70a))); +PV0i.z = floatBitsToInt(intBitsToFloat(PV1i.w) * intBitsToFloat(0x3ea2f983)); +PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[5].z), intBitsToFloat(R126i.x))); +PS0i = floatBitsToInt(1.0 / intBitsToFloat(PV1i.x)); +// 9 +backupReg0i = R127i.x; +R127i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R127i.y),intBitsToFloat(PV0i.x)) + intBitsToFloat(R124i.y))); +PV1i.y = floatBitsToInt(intBitsToFloat(backupReg0i) * intBitsToFloat(0x3ea2f983)); +PV1i.z = floatBitsToInt(intBitsToFloat(R6i.z) * intBitsToFloat(PS0i)); +PV1i.w = floatBitsToInt(intBitsToFloat(PV0i.w) * intBitsToFloat(0x3ea2f983)); +PS1i = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R124i.y), intBitsToFloat(PV0i.z))); +// 10 +PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV1i.z), intBitsToFloat(R125i.y))); +PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R124i.y), intBitsToFloat(PV1i.w))); +PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R124i.y), intBitsToFloat(PV1i.y))); +PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(PS1i))); +// 11 +PV1i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(PV0i.z))); +PV1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(PV0i.y))); +PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R126i.z), intBitsToFloat(PV0i.x))); +R8i.x = floatBitsToInt(min(intBitsToFloat(PV0i.w), intBitsToFloat(0x42000000))); +PS1i = R8i.x; +// 12 +R8i.y = floatBitsToInt(min(intBitsToFloat(PV1i.x), intBitsToFloat(0x42000000))); +PV0i.z = floatBitsToInt(intBitsToFloat(PV1i.w) * 0.25); +R8i.z = floatBitsToInt(min(intBitsToFloat(PV1i.z), intBitsToFloat(0x42000000))); +PS0i = R8i.z; +// 13 +PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R125i.x), intBitsToFloat(PV0i.z))); +// 14 +PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R4i.w), intBitsToFloat(PV1i.y))); +// 15 +PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R126i.y), intBitsToFloat(PV0i.x))); +// 16 +PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[5].z), intBitsToFloat(PV1i.w))); +PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[5].y), intBitsToFloat(PV1i.w))); +PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(uf_remappedPS[5].x), intBitsToFloat(PV1i.w))); +// 17 +PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R7i.z), intBitsToFloat(PV0i.x))); +PV1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R7i.z), intBitsToFloat(PV0i.y))); +PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R7i.z), intBitsToFloat(PV0i.z))); +// 18 +PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PV1i.w))); +PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PV1i.y))); +PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PV1i.z))); +// 19 +PV1i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(PV0i.w))); +PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(PV0i.x))); +PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(PV0i.z))); +// 20 +R7i.x = floatBitsToInt(min(intBitsToFloat(PV1i.y), intBitsToFloat(0x42000000))); +R7i.y = floatBitsToInt(min(intBitsToFloat(PV1i.x), intBitsToFloat(0x42000000))); +R7i.z = floatBitsToInt(min(intBitsToFloat(PV1i.w), intBitsToFloat(0x42000000))); +// 21 +R0i.xyz = ivec3(R8i.x,R8i.y,R8i.z); +R0i.w = R8i.w; +// 22 +R1i.xyz = ivec3(R7i.x,R7i.y,R7i.z); +R1i.w = R7i.w; +// export +passPixelColor0 = vec4(intBitsToFloat(R0i.x), intBitsToFloat(R0i.y), intBitsToFloat(R0i.z), intBitsToFloat(R0i.w)); +vec3 colhsv = rgb2hsv(passPixelColor0.rgb); +passPixelColor0.rgb = hsv2rgb(vec3(mod(colhsv.x + hueRotation, 1.0), colhsv.y*SATURATION_FACTOR, colhsv.z*VALUE_FACTOR)); +passPixelColor0.a *= ALPHA_FACTOR; +passPixelColor1 = vec4(intBitsToFloat(R1i.x), intBitsToFloat(R1i.y), intBitsToFloat(R1i.z), intBitsToFloat(R1i.w)); +colhsv = rgb2hsv(passPixelColor1.rgb); +passPixelColor1.rgb = hsv2rgb(vec3(mod(colhsv.x + hueRotation, 1.0), colhsv.y*SATURATION_FACTOR, colhsv.z*VALUE_FACTOR)); +passPixelColor1.a *= ALPHA_FACTOR; +} diff --git a/Workaround/BreathOfTheWild_!AMDLights/rules.txt b/Workaround/BreathOfTheWild_!AMDLights/rules.txt new file mode 100644 index 00000000..223ed328 --- /dev/null +++ b/Workaround/BreathOfTheWild_!AMDLights/rules.txt @@ -0,0 +1,8 @@ +[Definition] +titleIds = 00050000101C9300,00050000101C9400,00050000101C9500 +name = "The Legend of Zelda: Breath of the Wild - AMD Light Circle fix" +version = 2 + +# credit of identifying shaders go to Kiri +# these two shaders are based on Glow Lights Color Mod +# works Cemu 1.11.0c and BotW 1.3.1 \ No newline at end of file From 3cfd4de13248d8215cfc5f2ef07a41021a4cb0ae Mon Sep 17 00:00:00 2001 From: Valery Guskov Date: Wed, 15 Nov 2017 00:54:24 +0200 Subject: [PATCH 02/10] whitespace cleanup --- .../e95d54571be1c0a9_000000000001fe49_ps.txt | 44 ------------------- 1 file changed, 44 deletions(-) diff --git a/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt b/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt index 50e24577..7002600e 100644 --- a/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt +++ b/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt @@ -219,21 +219,7 @@ backupReg0i = R125i.x; R125i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.y),intBitsToFloat(PV1i.x)) + intBitsToFloat(R2i.x))); R127i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(backupReg0i), intBitsToFloat(uf_remappedPS[2].y))); R127i.y = clampFI32(R127i.y); - - - - - - - R126i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(PV1i.w),intBitsToFloat(uf_remappedPS[6].z)) + intBitsToFloat(uf_remappedPS[6].w))); - - - - - - - PV0i.z = R126i.z; R4i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R124i.y), intBitsToFloat(R124i.y))); PV0i.w = R4i.w; @@ -269,18 +255,7 @@ PS1i = floatBitsToInt(1.0 / intBitsToFloat(PV0i.w)); // 18 PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R126i.z), intBitsToFloat(R124i.y))); R6i.y = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R127i.y)),intBitsToFloat(R127i.y)) + 1.0)); - - - - -//!!!!!!!!!!!!!!! R3i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R126i.x),intBitsToFloat(uf_remappedPS[6].x)) + intBitsToFloat(uf_remappedPS[6].y))); - - - - - - PV0i.w = floatBitsToInt(intBitsToFloat(R0i.y) + intBitsToFloat(PS1i)); PV0i.w = clampFI32(PV0i.w); tempResultf = 1.0 / sqrt(intBitsToFloat(PV1i.x)); @@ -302,26 +277,7 @@ backupReg1i = R0i.y; R1i.x = floatBitsToInt(intBitsToFloat(R4i.x) + intBitsToFloat(PV1i.x)); R0i.y = floatBitsToInt(intBitsToFloat(R0i.w) + intBitsToFloat(PV1i.y)); R0i.z = floatBitsToInt(intBitsToFloat(R1i.w) + intBitsToFloat(PV1i.z)); - - - - - - - - R3i.w = R126i.z; - - - - - - - - - - - R4i.y = floatBitsToInt(intBitsToFloat(backupReg1i) * intBitsToFloat(PS1i)); PS0i = R4i.y; R5i.y = floatBitsToInt(texture(textureUnitPS2, intBitsToFloat(R3i.ww)).x); From 61bf0415c705bf460246861a0369853c53462938 Mon Sep 17 00:00:00 2001 From: getdls Date: Mon, 20 Nov 2017 01:53:39 +0100 Subject: [PATCH 03/10] [XCX} Probe view, Blur fix Fix - Scale map view so probe doesn't break Fix - Inital commit of blur fix --- .../840947e29015aa9a_00000000000003c9_ps.txt | 174 +++++++++----- .../8c1e55fd967b0496_0000000000000079_ps.txt | 54 ++--- .../d8e69e8df8c227f5_00000000000003c9_ps.txt | 219 ++++++++++++++---- Source/XenobladeX/rules.txt | 6 + 4 files changed, 310 insertions(+), 143 deletions(-) diff --git a/Source/XenobladeX/840947e29015aa9a_00000000000003c9_ps.txt b/Source/XenobladeX/840947e29015aa9a_00000000000003c9_ps.txt index 375a6d9d..68d2f41a 100644 --- a/Source/XenobladeX/840947e29015aa9a_00000000000003c9_ps.txt +++ b/Source/XenobladeX/840947e29015aa9a_00000000000003c9_ps.txt @@ -8,9 +8,10 @@ $scaleFactorY = always_decimal_format($fullHeight / 720.0); #version 420 #extension GL_ARB_texture_gather : enable -// shader 840947e29015aa9a +// shader 840947e29015aa9a // fix circle // cutscene DOF, reveal talk BG -// To-do Review multi pass performance and bokeh depth +// To-do. Tweak blur scaling to resolution. Weights are funky, too bright + const float resScale = ; //const float resScale = 3; uniform ivec4 uf_remappedPS[3]; @@ -21,78 +22,127 @@ layout(location = 1) in vec4 passParameterSem1; layout(location = 0) out vec4 passPixelColor0; uniform vec2 uf_fragCoordScale; -vec4 blur9(sampler2D image, vec2 uv, vec2 resolution, vec2 direction) { - vec4 color = vec4(0.0); - vec2 off1 = vec2(1.3846153846) * direction; - vec2 off2 = vec2(3.2307692308) * direction; - color += texture2D(image, uv) * 0.2270270270; - color += texture2D(image, uv + (off1 / resolution)) * 0.3162162162; - color += texture2D(image, uv - (off1 / resolution)) * 0.3162162162; - color += texture2D(image, uv + (off2 / resolution)) * 0.0702702703; - color += texture2D(image, uv - (off2 / resolution)) * 0.0702702703; - return color; +vec2 Circle(float Start, float Points, float Point) +{ + float Rad = (3.141592 * 2.0 * (1.0 / Points)) * (Point + Start); + return vec2(sin(Rad), cos(Rad)); +} + +highp float lineRand(vec2 co) +{ + highp float a = 12.9898; + highp float b = 78.233; + highp float c = 43758.5453; + highp float dt = dot(co.xy, vec2(a, b)); + highp float sn = mod(dt, 3.14); + return fract(sin(sn) * c); +} + +vec4 circleBlur(sampler2D image, vec2 vuv, vec2 res) { + + + vec2 uv = vuv; + vec2 PixelOffset = 1.25 / res.xy; //1.0 + + float Start = 2.0 / 14.0; + vec2 Scale = 0.66 * 4.0 * 2.0 * PixelOffset.xy; + + vec3 N0 = texture(image, uv + Circle(Start, 14.0, 0.0) * Scale).rgb; + vec3 N1 = texture(image, uv + Circle(Start, 14.0, 1.0) * Scale).rgb; + vec3 N2 = texture(image, uv + Circle(Start, 14.0, 2.0) * Scale).rgb; + vec3 N3 = texture(image, uv + Circle(Start, 14.0, 3.0) * Scale).rgb; + vec3 N4 = texture(image, uv + Circle(Start, 14.0, 4.0) * Scale).rgb; + vec3 N5 = texture(image, uv + Circle(Start, 14.0, 5.0) * Scale).rgb; + vec3 N6 = texture(image, uv + Circle(Start, 14.0, 6.0) * Scale).rgb; + vec3 N7 = texture(image, uv + Circle(Start, 14.0, 7.0) * Scale).rgb; + vec3 N8 = texture(image, uv + Circle(Start, 14.0, 8.0) * Scale).rgb; + vec3 N9 = texture(image, uv + Circle(Start, 14.0, 9.0) * Scale).rgb; + vec3 N10 = texture(image, uv + Circle(Start, 14.0, 10.0) * Scale).rgb; + vec3 N11 = texture(image, uv + Circle(Start, 14.0, 11.0) * Scale).rgb; + vec3 N12 = texture(image, uv + Circle(Start, 14.0, 12.0) * Scale).rgb; + vec3 N13 = texture(image, uv + Circle(Start, 14.0, 13.0) * Scale).rgb; + vec3 N14 = texture(image, uv).rgb; + + + vec3 color = vec3(0, 0, 0); + + color.rgb = + (N0 * 0.01478636) + + (N1 * 0.03696590) + + (N2 * 0.07393179) + + (N3 * 0.07393179) + + (N4 * 0.12013917) + + (N5 * 0.16018555) + + (N6 * 0.17620411) + + (N7 * 0.17620411) + + (N8 * 0.16018555) + + (N9 * 0.12013917) + + (N10 * 0.07393179) + + (N11 * 0.07393179) + + (N12 * 0.03696590) + + (N13 * 0.03696590) + + (N14 * 0.01478636); + + return vec4(color.rgb, 1.0); + } 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)); + 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)); } +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; -// 0 -R2f.x = R1f.x + (intBitsToFloat(uf_remappedPS[0].x)/ resScale); -R2f.y = R1f.y + (intBitsToFloat(uf_remappedPS[0].y)/ resScale); -R0f.z = R1f.x + (intBitsToFloat(uf_remappedPS[0].z)/ resScale); -R0f.w = R1f.y + (intBitsToFloat(uf_remappedPS[0].w)/ resScale); -// 1 + 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; + // 0er + R2f.x = R1f.x + (intBitsToFloat(uf_remappedPS[0].x) / resScale); + R2f.y = R1f.y + (intBitsToFloat(uf_remappedPS[0].y) / resScale); + 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)/ resScale,intBitsToFloat(uf_remappedPS[1].y)/ resScale,intBitsToFloat(uf_remappedPS[1].z)/ resScale); -R1f.w = backupReg1f + intBitsToFloat(uf_remappedPS[1].w)/ resScale; +R1f.xyz = vec3(backupReg0f,backupReg1f,backupReg0f) + vec3(intBitsToFloat(uf_remappedPS[1].x) / resScale,intBitsToFloat(uf_remappedPS[1].y) / resScale,intBitsToFloat(uf_remappedPS[1].z)); +R1f.w = backupReg1f + intBitsToFloat(uf_remappedPS[1].w) ; +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); -vec2 texsizetest = vec2(1280.0, 720.0); -//R4f = blur9(textureUnitPS0, R1f.xy, texsizetest, vec2(1.0, 1.0)); -//R3f = blur9(textureUnitPS0, R0f.zw, texsizetest, vec2(1.0, 1.0)); - -//R3f.xyz = (texture(textureUnitPS1, R2f.xy).xyz); -R3f = (blur9(textureUnitPS1, R2f.xy, texsizetest, vec2(1.0, 1.0))); // remove, redundant, blur on next step -R4f = (blur9(textureUnitPS1, R0f.zw, texsizetest, vec2(1.0, 1.0))); -R5f = (blur9(textureUnitPS1, R1f.xy, texsizetest, vec2(1.0, 1.0))); -R6f = (blur9(textureUnitPS1, R1f.zw, texsizetest, vec2(1.0, 1.0))); -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); +//R6f = circleBlur(textureUnitPS0, R1f.xy, textureSize(textureUnitPS0, 0)); +R2f.xyz = (circleBlur(textureUnitPS0, R2f.xy, textureSize(textureUnitPS0, 0)).xyz); +R7f.xyz = (circleBlur(textureUnitPS0, R0f.zw, textureSize(textureUnitPS0, 0)).xyz); +R8f.xyz = (circleBlur(textureUnitPS0, R1f.xy, textureSize(textureUnitPS0, 0)).xyz); +R1f.xyz = (circleBlur(textureUnitPS0, R1f.zw, textureSize(textureUnitPS0, 0)).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; diff --git a/Source/XenobladeX/8c1e55fd967b0496_0000000000000079_ps.txt b/Source/XenobladeX/8c1e55fd967b0496_0000000000000079_ps.txt index 0138d664..bf3daaf9 100644 --- a/Source/XenobladeX/8c1e55fd967b0496_0000000000000079_ps.txt +++ b/Source/XenobladeX/8c1e55fd967b0496_0000000000000079_ps.txt @@ -11,39 +11,27 @@ $scaleFactorY = always_decimal_format($fullHeight / 720.0); // shader 8c1e55fd967b0496 // ver/horz alignment tweak, just a magic value // To-do. Go back and re-center all passes instead. - const float resScale = ; -//const float resScale = 3.0; // +//const float resScale = 4.0; +const float repositionBloom = 1.00125; + uniform ivec4 uf_remappedPS[5]; layout(binding = 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf4e76000 res 320x180x1 dim 1 tm: 4 format 0816 compSel: 0 1 2 5 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler0 ClampX/Y/Z: 6 6 6 border: 0 layout(location = 0) in vec4 passParameterSem0; layout(location = 0) out vec4 passPixelColor0; uniform vec2 uf_fragCoordScale; -const float repositionBloom = 1.00125; -//mattdesl MIT - https://github.com/Jam3/glsl-fast-gaussian-blur -vec4 blur9(sampler2D image, vec2 uv, vec2 resolution, vec2 direction) { - vec4 color = vec4(0.0); - vec2 off1 = vec2(1.3846153846) * direction; - vec2 off2 = vec2(3.2307692308) * direction; - color += texture2D(image, uv) * 0.2270270270; - color += texture2D(image, uv + (off1 / resolution)) * 0.3162162162; - color += texture2D(image, uv - (off1 / resolution)) * 0.3162162162; - color += texture2D(image, uv + (off2 / resolution)) * 0.0702702703; - color += texture2D(image, uv - (off2 / resolution)) * 0.0702702703; - return color; -} int clampFI32(int v) { - if( v == 0x7FFFFFFF ) - return floatBitsToInt(1.0); - else if( v == 0xFFFFFFFF ) - return floatBitsToInt(0.0); + 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)); } +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); @@ -68,13 +56,13 @@ void main() R0f = passParameterSem0; vec2 scaleFactor = vec2(repositionBloom, repositionBloom); //Added R0f.xy = R0f.xy * scaleFactor;// re-align bloom.. hack until issue source is found. - - // 0 + + // 0 R1f.x = R0f.x + (intBitsToFloat(uf_remappedPS[0].x) / resScale); R1f.y = R0f.y + (intBitsToFloat(uf_remappedPS[0].y) / resScale); R0f.z = R0f.x + (intBitsToFloat(uf_remappedPS[1].x) / resScale); R0f.w = R0f.y + (intBitsToFloat(uf_remappedPS[1].y) / resScale); - R5f.w = 1.0 ; + R5f.w = 1.0; PS0f = R5f.w; // 1 R2f.x = R0f.x + (intBitsToFloat(uf_remappedPS[2].x) / resScale); @@ -87,21 +75,13 @@ void main() R0f.x = backupReg0f + (intBitsToFloat(uf_remappedPS[4].x) / resScale); R0f.y = backupReg1f + (intBitsToFloat(uf_remappedPS[4].y) / resScale); - //hack aproximate 720 blur - //vec2 texsizetest = vec2(2560.0,1440.0 ); -R4f = blur9(textureUnitPS0, R1f.xy, textureSize(textureUnitPS0, 0), vec2(1.0, 1.0)); -R3f = blur9(textureUnitPS0, R0f.zw, textureSize(textureUnitPS0, 0), vec2(1.0, 1.0)); -R2f = blur9(textureUnitPS0, R2f.xy, textureSize(textureUnitPS0, 0), vec2(1.0, 1.0)); -R1f = blur9(textureUnitPS0, R1f.zw, textureSize(textureUnitPS0, 0), vec2(1.0, 1.0)); -R0f = blur9(textureUnitPS0, R0f.xy, textureSize(textureUnitPS0, 0), vec2(1.0, 1.0)); - - /* + R4f.xyz = (texture(textureUnitPS0, R1f.xy).xyz); R3f.xyz = (texture(textureUnitPS0, R0f.zw).xyz); R2f.xyz = (texture(textureUnitPS0, R2f.xy).xyz); R1f.xyz = (texture(textureUnitPS0, R1f.zw).xyz); R0f.xyz = (texture(textureUnitPS0, R0f.xy).xyz); - */ + // 0 PV0f.x = R4f.y * intBitsToFloat(uf_remappedPS[0].z); PV0f.y = R4f.x * intBitsToFloat(uf_remappedPS[0].z); @@ -110,7 +90,7 @@ R0f = blur9(textureUnitPS0, R0f.xy, textureSize(textureUnitPS0, 0), vec2(1.0, 1. R127f.z = max(R3f.y, R4f.y); PS0f = R127f.z; // 1 - R123f.x = (R3f.z * intBitsToFloat(uf_remappedPS[1].z) + PV0f.z); + R123f.x = (R3f.z * intBitsToFloat(uf_remappedPS[1].z) + PV0f.z); PV1f.x = R123f.x; PV1f.y = max(R3f.z, R4f.z); R123f.z = (R3f.y * intBitsToFloat(uf_remappedPS[1].z) + PV0f.x); @@ -157,11 +137,11 @@ R0f = blur9(textureUnitPS0, R0f.xy, textureSize(textureUnitPS0, 0), vec2(1.0, 1. PV1f.w = R127f.w; // 6 PV0f.x = PV1f.z + -(PV1f.w); - R5f.y = (PV1f.y * 0.25 + R127f.x);//degree of bloom - R5f.x = (PV1f.x * 0.25 + R127f.z); + R5f.y = (PV1f.y * 0.30 + R127f.x);//degree of bloom + R5f.x = (PV1f.x * 0.30 + R127f.z); PS0f = R5f.x; // 7 - R5f.z = (PV0f.x * 0.25 + R127f.w); + R5f.z = (PV0f.x * 0.30 + R127f.w); // export passPixelColor0 = vec4(R5f.x, R5f.y, R5f.z, R5f.w); } diff --git a/Source/XenobladeX/d8e69e8df8c227f5_00000000000003c9_ps.txt b/Source/XenobladeX/d8e69e8df8c227f5_00000000000003c9_ps.txt index a27415e5..a6ba3374 100644 --- a/Source/XenobladeX/d8e69e8df8c227f5_00000000000003c9_ps.txt +++ b/Source/XenobladeX/d8e69e8df8c227f5_00000000000003c9_ps.txt @@ -5,33 +5,167 @@ $fullHeight = $argv[2]; $scaleFactorX = always_decimal_format($fullWidth / 1280.0); $scaleFactorY = always_decimal_format($fullHeight / 720.0); ?> + #version 420 #extension GL_ARB_texture_gather : enable // shader d8e69e8df8c227f5 -// Inital blur step pixelation fix, naive blur9 hack to aproximate 720 blur. -// To-do review performance hit of multi pass and depth for bokeh blur -const float resScale = ; -//const float resScale = 3.0; // - +// Bloom/blur 1st step, 1280->640->1280 +// To do, go back and remove all test blurs.. uniform ivec4 uf_remappedPS[3]; - layout(binding = 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf470a000 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 = 1) uniform sampler2D textureUnitPS1;// Tex1 addr 0xf4386000 res 1280x720x1 dim 1 tm: 4 format 001a compSel: 0 1 2 3 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler1 ClampX/Y/Z: 2 2 2 border: 0 layout(location = 0) in vec4 passParameterSem0; layout(location = 0) out vec4 passPixelColor0; uniform vec2 uf_fragCoordScale; -const float repositionBlur = 1.00425; -//mattdesl MIT - https://github.com/Jam3/glsl-fast-gaussian-blur -vec4 blur9(sampler2D image, vec2 uv, vec2 resolution, vec2 direction) { +const float resScale = ; +//const float resScale = 3.5; //3.5 looks nicer + +highp float lineRand(vec2 co) +{ + highp float a = 12.9898; + highp float b = 78.233; + highp float c = 43758.5453; + highp float dt = dot(co.xy, vec2(a, b)); + highp float sn = mod(dt, 3.14); + return fract(sin(sn) * c); +} + +vec4 blur21Vert(sampler2D image, vec2 vuv, vec2 res) { + // Note, must use gl_FragCord not emulated coordinates to sample between textels + // Two pass implementation + + //iq CC , http://rastergrid.com/blog/2010/09/efficient-gaussian-blur-with-linear-sampling/ + vec3 blr = vec3(0.0); + + blr += 0.026109*texture(image, (vuv + vec2(0.0, -15.5)) / res.xy).xyz; + blr += 0.034202*texture(image, (vuv + vec2(0.0, -13.5)) / res.xy).xyz; + blr += 0.043219*texture(image, (vuv + vec2(0.0, -11.5)) / res.xy).xyz; + blr += 0.052683*texture(image, (vuv + vec2(0.0, -9.5)) / res.xy).xyz; + blr += 0.061948*texture(image, (vuv + vec2(0.0, -7.5)) / res.xy).xyz; + blr += 0.070266*texture(image, (vuv + vec2(0.0, -5.5)) / res.xy).xyz; + blr += 0.076883*texture(image, (vuv + vec2(0.0, -3.5)) / res.xy).xyz; + blr += 0.081149*texture(image, (vuv + vec2(0.0, -1.5)) / res.xy).xyz; + blr += 0.041312*texture(image, (vuv + vec2(0.0, 0.0)) / res.xy).xyz; + blr += 0.081149*texture(image, (vuv + vec2(0.0, 1.5)) / res.xy).xyz; + blr += 0.076883*texture(image, (vuv + vec2(0.0, 3.5)) / res.xy).xyz; + blr += 0.070266*texture(image, (vuv + vec2(0.0, 5.5)) / res.xy).xyz; + blr += 0.061948*texture(image, (vuv + vec2(0.0, 7.5)) / res.xy).xyz; + blr += 0.052683*texture(image, (vuv + vec2(0.0, 9.5)) / res.xy).xyz; + blr += 0.043219*texture(image, (vuv + vec2(0.0, 11.5)) / res.xy).xyz; + blr += 0.034202*texture(image, (vuv + vec2(0.0, 13.5)) / res.xy).xyz; + blr += 0.026109*texture(image, (vuv + vec2(0.0, 15.5)) / res.xy).xyz; + + blr /= 0.93423; // renormalize to compensate for the 4 taps I skipped + return vec4(blr, 1.0); +} + + +vec4 blur21Horiz(sampler2D image, vec2 vuv, vec2 res) { + //iq CC + vec3 blr = vec3(0.0); + + blr += 0.026109*texture(image, (vuv + vec2(-15.5, 0.0)) / res.xy).xyz; + blr += 0.034202*texture(image, (vuv + vec2(-13.5, 0.0)) / res.xy).xyz; + blr += 0.043219*texture(image, (vuv + vec2(-11.5, 0.0)) / res.xy).xyz; + blr += 0.052683*texture(image, (vuv + vec2(-9.5, 0.0)) / res.xy).xyz; + blr += 0.061948*texture(image, (vuv + vec2(-7.5, 0.0)) / res.xy).xyz; + blr += 0.070266*texture(image, (vuv + vec2(-5.5, 0.0)) / res.xy).xyz; + blr += 0.076883*texture(image, (vuv + vec2(-3.5, 0.0)) / res.xy).xyz; + blr += 0.081149*texture(image, (vuv + vec2(-1.5, 0.0)) / res.xy).xyz; + blr += 0.041312*texture(image, (vuv + vec2(0.0, 0.0)) / res.xy).xyz; + blr += 0.081149*texture(image, (vuv + vec2(1.5, 0.0)) / res.xy).xyz; + blr += 0.076883*texture(image, (vuv + vec2(3.5, 0.0)) / res.xy).xyz; + blr += 0.070266*texture(image, (vuv + vec2(5.5, 0.0)) / res.xy).xyz; + blr += 0.061948*texture(image, (vuv + vec2(7.5, 0.0)) / res.xy).xyz; + blr += 0.052683*texture(image, (vuv + vec2(9.5, 0.0)) / res.xy).xyz; + blr += 0.043219*texture(image, (vuv + vec2(11.5, 0.0)) / res.xy).xyz; + blr += 0.034202*texture(image, (vuv + vec2(13.5, 0.0)) / res.xy).xyz; + blr += 0.026109*texture(image, (vuv + vec2(15.5, 0.0)) / res.xy).xyz; + + blr /= 0.93423; // renormalize to compensate for the 4 taps I skipped + return vec4(blr, 1.0); +} + +vec2 Circle(float Start, float Points, float Point) +{ + float Rad = (3.141592 * 2.0 * (1.0 / Points)) * (Point + Start); + return vec2(sin(Rad), cos(Rad)); +} + +vec4 circleBlur(sampler2D image, vec2 vuv, vec2 res){ + //original by jcant0n ? + //vec2 uv = gl_FragCoord.xy / res.xy; + vec2 uv = vuv; + vec2 PixelOffset = 1.9 / res.xy; //1.0 + + float Start = 2.0 / 14.0; + vec2 Scale = 0.66 * 4.0 * 2.0 * PixelOffset.xy; + + vec3 N0 = texture(image, uv + Circle(Start, 14.0, 0.0) * Scale).rgb; + vec3 N1 = texture(image, uv + Circle(Start, 14.0, 1.0) * Scale).rgb; + vec3 N2 = texture(image, uv + Circle(Start, 14.0, 2.0) * Scale).rgb; + vec3 N3 = texture(image, uv + Circle(Start, 14.0, 3.0) * Scale).rgb; + vec3 N4 = texture(image, uv + Circle(Start, 14.0, 4.0) * Scale).rgb; + vec3 N5 = texture(image, uv + Circle(Start, 14.0, 5.0) * Scale).rgb; + vec3 N6 = texture(image, uv + Circle(Start, 14.0, 6.0) * Scale).rgb; + vec3 N7 = texture(image, uv + Circle(Start, 14.0, 7.0) * Scale).rgb; + vec3 N8 = texture(image, uv + Circle(Start, 14.0, 8.0) * Scale).rgb; + vec3 N9 = texture(image, uv + Circle(Start, 14.0, 9.0) * Scale).rgb; + vec3 N10 = texture(image, uv + Circle(Start, 14.0, 10.0) * Scale).rgb; + vec3 N11 = texture(image, uv + Circle(Start, 14.0, 11.0) * Scale).rgb; + vec3 N12 = texture(image, uv + Circle(Start, 14.0, 12.0) * Scale).rgb; + vec3 N13 = texture(image, uv + Circle(Start, 14.0, 13.0) * Scale).rgb; + vec3 N14 = texture(image, uv).rgb; + + vec3 color = vec3(0, 0, 0); + color.rgb = + (N0 * 0.01478636) + + (N1 * 0.03696590) + + (N2 * 0.07393179) + + (N3 * 0.07393179) + + (N4 * 0.12013917) + + (N5 * 0.16018555) + + (N6 * 0.17620411) + + (N7 * 0.17620411) + + (N8 * 0.16018555) + + (N9 * 0.12013917) + + (N10 * 0.07393179) + + (N11 * 0.07393179) + + (N12 * 0.03696590) + + (N13 * 0.03696590) + + (N14 * 0.01478636); + + return vec4(color.rgb, 1.0); +} + + + +vec4 blur25(sampler2D image, vec2 uv, vec2 resolution, vec2 direction) { + //curent implentation doesn't work in cemu.. + //mattdesl MIT - https://github.com/Jam3/glsl-fast-gaussian-blur vec4 color = vec4(0.0); - vec2 off1 = vec2(1.3846153846) * direction; - vec2 off2 = vec2(3.2307692308) * direction; - color += texture2D(image, uv) * 0.2270270270; - color += texture2D(image, uv + (off1 / resolution)) * 0.3162162162; - color += texture2D(image, uv - (off1 / resolution)) * 0.3162162162; - color += texture2D(image, uv + (off2 / resolution)) * 0.0702702703; - color += texture2D(image, uv - (off2 / resolution)) * 0.0702702703; - return color; + vec2 off1 = vec2(1.44827586) * direction; + vec2 off2 = vec2(3.37931034) * direction; + vec2 off3 = vec2(5.31034483) * direction; + vec2 off4 = vec2(7.24137931) * direction; + vec2 off5 = vec2(9.17241379) * direction; + vec2 off6 = vec2(11.10344828) * direction; + + //color += texture2D(image, uv - (off6 / resolution)) * 0.00001361; + color += texture2D(image, uv - (off5 / resolution)) * 0.00044240; + color += texture2D(image, uv - (off4 / resolution)) * 0.00581436; + color += texture2D(image, uv - (off3 / resolution)) * 0.03730881; + color += texture2D(image, uv - (off2 / resolution)) * 0.12888498; + color += texture2D(image, uv - (off1 / resolution)) * 0.25281284; + color += texture2D(image, uv) * 0.14944601; + color += texture2D(image, uv + (off1 / resolution)) * 0.25281284; + color += texture2D(image, uv + (off2 / resolution)) * 0.12888498; + color += texture2D(image, uv + (off3 / resolution)) * 0.03730881; + color += texture2D(image, uv + (off4 / resolution)) * 0.00581436; + color += texture2D(image, uv + (off5 / resolution)) * 0.00044240; + //color += texture2D(image, uv + (off6 / resolution)) * 0.00001361; + ///= 0.93423 + return color /= 0.93423; } int clampFI32(int v) @@ -66,25 +200,22 @@ bool predResult = true; vec3 cubeMapSTM; int cubeMapFaceId; R0f = passParameterSem0; -vec2 scaleFactor = vec2(repositionBlur, repositionBlur); //Added -R0f.xy = R0f.xy * scaleFactor;// re-align bloom.. hack until issue source is found. +R0f.xy = vec2((passParameterSem0.x + passParameterSem0.z) , (passParameterSem0.y + passParameterSem0.w) ); //center point + // 0 -R1f.x = R0f.x + (intBitsToFloat(uf_remappedPS[0].x)/ resScale); -R1f.y = R0f.y + (intBitsToFloat(uf_remappedPS[0].y)/ resScale); -R0f.z = R0f.x + (intBitsToFloat(uf_remappedPS[0].z)/ resScale); -R0f.w = R0f.y + (intBitsToFloat(uf_remappedPS[0].w)/ resScale); +R1f.x = R0f.x + intBitsToFloat(uf_remappedPS[0].x)/resScale; +R1f.y = R0f.y + intBitsToFloat(uf_remappedPS[0].y)/resScale; +R0f.z = R0f.x + intBitsToFloat(uf_remappedPS[0].z)/resScale; +R0f.w = R0f.y + intBitsToFloat(uf_remappedPS[0].w)/resScale; // 1 - backupReg0f = R0f.x; backupReg1f = R0f.y; backupReg0f = R0f.x; backupReg1f = R0f.y; -R0f.x = backupReg0f + (intBitsToFloat(uf_remappedPS[1].x)/ resScale); -R0f.y = backupReg1f + (intBitsToFloat(uf_remappedPS[1].y)/ resScale); -R1f.z = backupReg0f + (intBitsToFloat(uf_remappedPS[1].z)/ resScale); -R1f.w = backupReg1f + (intBitsToFloat(uf_remappedPS[1].w)/ resScale); - - +R0f.x = backupReg0f + intBitsToFloat(uf_remappedPS[1].x)/resScale; +R0f.y = backupReg1f + intBitsToFloat(uf_remappedPS[1].y)/resScale; +R1f.z = backupReg0f + intBitsToFloat(uf_remappedPS[1].z)/resScale; +R1f.w = backupReg1f + intBitsToFloat(uf_remappedPS[1].w)/resScale; R2f.xyz = (texture(textureUnitPS1, R1f.xy).xyz); @@ -92,20 +223,22 @@ R3f.xyz = (texture(textureUnitPS1, R0f.zw).xyz); R4f.xyz = (texture(textureUnitPS1, R0f.xy).xyz); R5f.xyz = (texture(textureUnitPS1, R1f.zw).xyz); -//hack aproximate 720 blur -R6f = blur9(textureUnitPS0, R1f.xy, textureSize(textureUnitPS0, 0), vec2(0.0, 1.0)); // note, check what goes wrong with textureSize(textureUnitPS0, 0), -R7f = blur9(textureUnitPS0, R0f.zw, textureSize(textureUnitPS0, 0), vec2(1.0, 0.0)); -R0f = blur9(textureUnitPS0, R0f.xy, textureSize(textureUnitPS0, 0), vec2(0.0, 1.0)); -R1f = blur9(textureUnitPS0, R1f.zw, textureSize(textureUnitPS0, 0), vec2(1.0, 0.0)); +//vec2 iResolution = textureSize(textureUnitPS0, 0); + +R6f = circleBlur(textureUnitPS0, R1f.xy, textureSize(textureUnitPS0, 0)); +R7f = circleBlur(textureUnitPS0, R0f.zw, textureSize(textureUnitPS0, 0)); +R0f = circleBlur(textureUnitPS0, R0f.xy, textureSize(textureUnitPS0, 0)); +R1f = circleBlur(textureUnitPS0, R1f.zw, textureSize(textureUnitPS0, 0)); + + + +// Sample distance fails :( +//R6f = mix(blur25(textureUnitPS0, R1f.xy, vec2(2560, 1440), vec2(0.0, 1.0)), blur13(textureUnitPS0, R1f.xy, textureSize(textureUnitPS0, 0), vec2(1.0,0.0)), 0.5); +//R7f = mix(blur25(textureUnitPS0, R0f.zw, textureSize(textureUnitPS0, 0), vec2(2.0, -3.0)), blur13(textureUnitPS0, R0f.zw, textureSize(textureUnitPS0, 0), vec2(4.0, 6.0)), 0.5); +//R0f = mix(blur25(textureUnitPS0, R0f.xy, textureSize(textureUnitPS0, 0), vec2(4.0, -6.0)), blur13(textureUnitPS0, R0f.xy, textureSize(textureUnitPS0, 0), vec2(2.0, 3.0)), 0.5); +//R1f = mix(blur25(textureUnitPS0, R1f.zw, textureSize(textureUnitPS0, 0), vec2(8.0, -9.0)), blur13(textureUnitPS0, R1f.zw, textureSize(textureUnitPS0, 0), vec2(1.1, 1.5)), 0.5); -/* -R6f.xyz = (texture(textureUnitPS0, R1f.xy).xyz); -R7f.xyz = (texture(textureUnitPS0, R0f.zw).xyz); -R0f.xyz = (texture(textureUnitPS0, R0f.xy).xyz); -R1f.xyz = (texture(textureUnitPS0, R1f.zw).xyz); -*/ -// 0// 0.125, 31.875, 8128.125 tempf.x = dot(vec4(R2f.x,R2f.y,R2f.z,-0.0),vec4(intBitsToFloat(0x3e000000),intBitsToFloat(0x41ff0000),intBitsToFloat(0x45fe0100),0.0)); PV0f.x = tempf.x; PV0f.y = tempf.x; @@ -175,8 +308,6 @@ PS1f = R1f.x; PS0f = exp2(PV1f.x); // 11 R1f.w = PS0f * intBitsToFloat(uf_remappedPS[2].z); - -//R1f = blur9(textureUnitPS0, R1f.xy, vec2(2560.0, 1440.0), vec2(1.0, 0.0)); // export passPixelColor0 = vec4(R1f.x, R1f.y, R1f.z, R1f.w); } diff --git a/Source/XenobladeX/rules.txt b/Source/XenobladeX/rules.txt index 88e67a60..004b91c8 100644 --- a/Source/XenobladeX/rules.txt +++ b/Source/XenobladeX/rules.txt @@ -31,6 +31,12 @@ tileModesExcluded = 0x001 overwriteWidth = overwriteHeight = +[TextureRedefine] # Do not remove, must be scaled to TV or probe view breaks +width = 854 +height = 480 +overwriteWidth = +overwriteHeight = + [TextureRedefine] # half-res alpha width = 640 height = 360 From fa05f6e2dcf0795a781a290b9ad3805d426b9ca1 Mon Sep 17 00:00:00 2001 From: Michael Date: Mon, 20 Nov 2017 10:13:10 -0800 Subject: [PATCH 04/10] Fix Lego not appearing & XCX gamepad scaling --- Source/XenobladeX/rules.txt | 4 ++-- build.sh | 1 + 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/Source/XenobladeX/rules.txt b/Source/XenobladeX/rules.txt index 004b91c8..6c3446b9 100644 --- a/Source/XenobladeX/rules.txt +++ b/Source/XenobladeX/rules.txt @@ -34,8 +34,8 @@ overwriteHeight = [TextureRedefine] # Do not remove, must be scaled to TV or probe view breaks width = 854 height = 480 -overwriteWidth = -overwriteHeight = +overwriteWidth = +overwriteHeight = [TextureRedefine] # half-res alpha width = 640 diff --git a/build.sh b/build.sh index 71f0aefb..80e7c70d 100644 --- a/build.sh +++ b/build.sh @@ -124,6 +124,7 @@ std_respack "KamenRider" "${res16by9[@]/$just720p}" std_respack "KickandFennick" "${res16by9[@]/$just720p}" std_respack "KirbyRainbowCurse" "${res16by9[@]/$just720p}" std_respack "KungFuPanda" "${res16by9[@]/$just720p}" +std_respack "LEGOCityUndercover" "${res16by9[@]/$just720p}" std_respack "MarioKart8" "${res16by9[@]/$just720p}" "${res21by9[@]}" std_respack "MarioSonicRio" "${res16by9[@]/$just720p}" std_respack "MarioSonicSochi" "${res16by9[@]/$just720p}" From 9eb95d1bf549c9f4ca3566e69359c82b710f32ef Mon Sep 17 00:00:00 2001 From: getdls Date: Mon, 20 Nov 2017 20:35:03 +0100 Subject: [PATCH 05/10] [XCX] - Brightness balance Tweak - Lowers default brigtness to better match new blur implementation weights --- Source/XenobladeX/3cc7e98f78c258b4_00000000000003ca_ps.txt | 2 +- Source/XenobladeX/d936195db0dd8e7d_0000000000001e52_ps.txt | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/Source/XenobladeX/3cc7e98f78c258b4_00000000000003ca_ps.txt b/Source/XenobladeX/3cc7e98f78c258b4_00000000000003ca_ps.txt index 5be3f01a..c4dfeb33 100644 --- a/Source/XenobladeX/3cc7e98f78c258b4_00000000000003ca_ps.txt +++ b/Source/XenobladeX/3cc7e98f78c258b4_00000000000003ca_ps.txt @@ -4,7 +4,7 @@ // To-do, .5 is daylight and 1.0 night is wiiu "correct" for nvidia // changes here in turn "breaks" bloom as they over or under expose depending on day/night -const float preExposure = 0.85; // old brigntess tweak. Truncates at around .45+ +const float preExposure = 0.75; // old brigntess tweak. Truncates at around .45+ const float gammaPostExposure = 0.9; // compensate pre exposure, but loss of contrast when positive. uniform ivec4 uf_remappedPS[1]; layout(binding = 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf470a000 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 diff --git a/Source/XenobladeX/d936195db0dd8e7d_0000000000001e52_ps.txt b/Source/XenobladeX/d936195db0dd8e7d_0000000000001e52_ps.txt index 8f060a63..3e53c688 100644 --- a/Source/XenobladeX/d936195db0dd8e7d_0000000000001e52_ps.txt +++ b/Source/XenobladeX/d936195db0dd8e7d_0000000000001e52_ps.txt @@ -4,7 +4,7 @@ // cross fade brightness // To-do, .5 is daylight and 1.0 night is wiiu "correct" for nvidia // changes here in turn "breaks" bloom as they over or under expose depending on day/night -const float preExposure = 0.85; // old brigntess tweak. Truncates at around .45+ +const float preExposure = 0.75; // old brigntess tweak. Truncates at around .45+ const float gammaPostExposure = 0.9; // compensate pre exposure, but loss of contrast when positive. uniform ivec4 uf_remappedPS[1]; From e10fdfaa3e428ad00471517d553a24b4f5c48af2 Mon Sep 17 00:00:00 2001 From: Alex Chirila Date: Tue, 21 Nov 2017 08:39:07 +0200 Subject: [PATCH 06/10] [BotW] Removed obsolete code in AMD Light Circle fix --- .../cf2e9b7aab93e342_000000000001fe49_ps.txt | 31 +------------------ .../e95d54571be1c0a9_000000000001fe49_ps.txt | 30 +----------------- 2 files changed, 2 insertions(+), 59 deletions(-) diff --git a/Workaround/BreathOfTheWild_!AMDLights/cf2e9b7aab93e342_000000000001fe49_ps.txt b/Workaround/BreathOfTheWild_!AMDLights/cf2e9b7aab93e342_000000000001fe49_ps.txt index 699569ae..884ebab9 100644 --- a/Workaround/BreathOfTheWild_!AMDLights/cf2e9b7aab93e342_000000000001fe49_ps.txt +++ b/Workaround/BreathOfTheWild_!AMDLights/cf2e9b7aab93e342_000000000001fe49_ps.txt @@ -1,16 +1,8 @@ #version 420 #extension GL_ARB_texture_gather : enable // shader cf2e9b7aab93e342 -// Sheikah Tower - inactive, reflection pass -// Glow Lights Color Mod -// Credit for hsv functions http://lolengine.net/blog/2013/07/27/rgb-to-hsv-in-glsl +// AMD Light Circle fix - World lights reflection pass -#define HUE_ROTATION 0 //[0, 360] where 0 and 360 is unchanged Hue and 180 is completely opposite Hue. Check http://dba.med.sc.edu/price/irf/Adobe_tg/models/images/hsl_top.JPG -#define SATURATION_FACTOR 0.9 //[0.0, 1.0] 1.0 means unchanged Saturation, 0.0 means completely desaturated. Values above 1.0 are accepted, but they may cause clipping -#define VALUE_FACTOR 0.65 //same as above; applies to Value -#define ALPHA_FACTOR 1.0 //same as above; applies to Transparency - -const float hueRotation = HUE_ROTATION / 360.0; uniform ivec4 uf_remappedPS[10]; layout(binding = 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf5054000 res 640x360x1 dim 1 tm: 4 format 080e compSel: 0 4 4 5 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 0xf494a800 res 1280x720x1 dim 1 tm: 4 format 001a compSel: 0 1 2 3 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler1 ClampX/Y/Z: 2 2 2 border: 1 @@ -28,20 +20,6 @@ else if( v == 0xFFFFFFFF ) 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; } -vec3 rgb2hsv(vec3 c) { - vec4 K = vec4(0.0, -1.0 / 3.0, 2.0 / 3.0, -1.0); - vec4 p = mix(vec4(c.bg, K.wz), vec4(c.gb, K.xy), step(c.b, c.g)); - vec4 q = mix(vec4(p.xyw, c.r), vec4(c.r, p.yzx), step(p.x, c.r)); - - float d = q.x - min(q.w, q.y); - float e = 1.0e-10; - return vec3(abs(q.z + (q.w - q.y) / (6.0 * d + e)), d / (q.x + e), q.x); -} -vec3 hsv2rgb(vec3 c) { - vec4 K = vec4(1.0, 2.0 / 3.0, 1.0 / 3.0, 3.0); - vec3 p = abs(fract(c.xxx + K.xyz) * 6.0 - K.www); - return c.z * mix(K.xxx, clamp(p - K.xxx, 0.0, 1.0), c.y); -} void main() { ivec4 R0i = ivec4(0); @@ -513,11 +491,4 @@ R1i.w = R4i.w; // export passPixelColor0 = vec4(intBitsToFloat(R0i.x), intBitsToFloat(R0i.y), intBitsToFloat(R0i.z), intBitsToFloat(R0i.w)); passPixelColor1 = vec4(intBitsToFloat(R1i.x), intBitsToFloat(R1i.y), intBitsToFloat(R1i.z), intBitsToFloat(R1i.w)); -vec3 colhsv = rgb2hsv(passPixelColor0.rgb); -passPixelColor0.rgb = hsv2rgb(vec3(mod(colhsv.x + hueRotation, 1.0), colhsv.y*SATURATION_FACTOR, colhsv.z*VALUE_FACTOR)); -passPixelColor0.a *= ALPHA_FACTOR; -colhsv = rgb2hsv(passPixelColor1.rgb); -colhsv = rgb2hsv(passPixelColor1.rgb); -passPixelColor1.rgb = hsv2rgb(vec3(mod(colhsv.x + hueRotation, 1.0), colhsv.y*SATURATION_FACTOR, colhsv.z*VALUE_FACTOR)); -passPixelColor1.a *= ALPHA_FACTOR; } diff --git a/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt b/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt index 7002600e..9d3d90e8 100644 --- a/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt +++ b/Workaround/BreathOfTheWild_!AMDLights/e95d54571be1c0a9_000000000001fe49_ps.txt @@ -1,16 +1,8 @@ #version 420 #extension GL_ARB_texture_gather : enable // shader e95d54571be1c0a9 -// Shrine - balls, reflection pass -// Glow Lights Color Mod -// Credit for hsv functions http://lolengine.net/blog/2013/07/27/rgb-to-hsv-in-glsl +// AMD Light Circle fix - Shrine lights reflection pass -#define HUE_ROTATION 0 //[0, 360] where 0 and 360 is unchanged Hue and 180 is completely opposite Hue. Check http://dba.med.sc.edu/price/irf/Adobe_tg/models/images/hsl_top.JPG -#define SATURATION_FACTOR 0.9 //[0.0, 1.0] 1.0 means unchanged Saturation, 0.0 means completely desaturated. Values above 1.0 are accepted, but they may cause clipping -#define VALUE_FACTOR 0.65 //same as above; applies to Value -#define ALPHA_FACTOR 1.0 //same as above; applies to Transparency - -const float hueRotation = HUE_ROTATION / 360.0; uniform ivec4 uf_remappedPS[8]; layout(binding = 0) uniform sampler2D textureUnitPS0;// Tex0 addr 0xf5054000 res 640x360x1 dim 1 tm: 4 format 080e compSel: 0 4 4 5 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 0xf494a800 res 1280x720x1 dim 1 tm: 4 format 001a compSel: 0 1 2 3 mipView: 0x0 (num 0x1) sliceView: 0x0 (num 0x1) Sampler1 ClampX/Y/Z: 2 2 2 border: 1 @@ -28,20 +20,6 @@ else if( v == 0xFFFFFFFF ) 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; } -vec3 rgb2hsv(vec3 c) { - vec4 K = vec4(0.0, -1.0 / 3.0, 2.0 / 3.0, -1.0); - vec4 p = mix(vec4(c.bg, K.wz), vec4(c.gb, K.xy), step(c.b, c.g)); - vec4 q = mix(vec4(p.xyw, c.r), vec4(c.r, p.yzx), step(p.x, c.r)); - - float d = q.x - min(q.w, q.y); - float e = 1.0e-10; - return vec3(abs(q.z + (q.w - q.y) / (6.0 * d + e)), d / (q.x + e), q.x); -} -vec3 hsv2rgb(vec3 c) { - vec4 K = vec4(1.0, 2.0 / 3.0, 1.0 / 3.0, 3.0); - vec3 p = abs(fract(c.xxx + K.xyz) * 6.0 - K.www); - return c.z * mix(K.xxx, clamp(p - K.xxx, 0.0, 1.0), c.y); -} void main() { ivec4 R0i = ivec4(0); @@ -416,11 +394,5 @@ R1i.xyz = ivec3(R7i.x,R7i.y,R7i.z); R1i.w = R7i.w; // export passPixelColor0 = vec4(intBitsToFloat(R0i.x), intBitsToFloat(R0i.y), intBitsToFloat(R0i.z), intBitsToFloat(R0i.w)); -vec3 colhsv = rgb2hsv(passPixelColor0.rgb); -passPixelColor0.rgb = hsv2rgb(vec3(mod(colhsv.x + hueRotation, 1.0), colhsv.y*SATURATION_FACTOR, colhsv.z*VALUE_FACTOR)); -passPixelColor0.a *= ALPHA_FACTOR; passPixelColor1 = vec4(intBitsToFloat(R1i.x), intBitsToFloat(R1i.y), intBitsToFloat(R1i.z), intBitsToFloat(R1i.w)); -colhsv = rgb2hsv(passPixelColor1.rgb); -passPixelColor1.rgb = hsv2rgb(vec3(mod(colhsv.x + hueRotation, 1.0), colhsv.y*SATURATION_FACTOR, colhsv.z*VALUE_FACTOR)); -passPixelColor1.a *= ALPHA_FACTOR; } From c130421f7aa46174b36422f756b1684ab6847366 Mon Sep 17 00:00:00 2001 From: getdls Date: Tue, 21 Nov 2017 21:25:15 +0100 Subject: [PATCH 07/10] [XCX] Aligment tweak [fix] Rule aligment fix. Game uses 16:9 logic for aligment, but had to scale smallest buffer to even. So upscaling from even multiplied the error. --- Source/XenobladeX/rules.txt | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/Source/XenobladeX/rules.txt b/Source/XenobladeX/rules.txt index 6c3446b9..c3e25e76 100644 --- a/Source/XenobladeX/rules.txt +++ b/Source/XenobladeX/rules.txt @@ -31,7 +31,7 @@ tileModesExcluded = 0x001 overwriteWidth = overwriteHeight = -[TextureRedefine] # Do not remove, must be scaled to TV or probe view breaks +[TextureRedefine] # Do not remove, must be upscaled with same factor as TV or probe view breaks width = 854 height = 480 overwriteWidth = @@ -85,9 +85,9 @@ height = 90 overwriteWidth = overwriteHeight = -[TextureRedefine]# Probe Beacon glow. +[TextureRedefine]# should up-scale from 35 to keep aligment with other 16:9 buffers width = 80 height = 46 #formatsExcluded = 0x816 #fixed in shader overwriteWidth = -overwriteHeight = +overwriteHeight = From 972fcc6aa8de3ba01b48c8c4d475a6204863e695 Mon Sep 17 00:00:00 2001 From: Alex Kiri Date: Tue, 21 Nov 2017 22:39:57 +0200 Subject: [PATCH 08/10] [BotW] Fixed self shadowing issue --- .../93f16bf1d083933b_0000000000000000_vs.txt | 102 ++++++++++++++++++ 1 file changed, 102 insertions(+) create mode 100644 Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt diff --git a/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt b/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt new file mode 100644 index 00000000..e25b12b6 --- /dev/null +++ b/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt @@ -0,0 +1,102 @@ + +#version 420 +#extension GL_ARB_texture_gather : enable +#extension GL_ARB_shading_language_packing : enable +// shader 93f16bf1d083933b +// Self shadowing fix +// original shader dumped using cemu 1.11.0, BotW 1.3.1 +const float resXScale = ; +const float resYScale = ; + +uniform ivec4 uf_remappedVS[1]; +uniform vec2 uf_windowSpaceToClipSpaceTransform; +layout(location = 0) in uvec4 attrDataSem0; +layout(location = 1) in uvec4 attrDataSem1; +out gl_PerVertex +{ + vec4 gl_Position; + float gl_PointSize; +}; +layout(location = 1) out vec4 passParameterSem1; +layout(location = 0) out vec4 passParameterSem0; +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 R127f = vec4(0.0); +uvec4 attrDecoder; +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 = floatBitsToInt(ivec4(gl_VertexID, 0, 0, gl_InstanceID)); +attrDecoder.xyz = attrDataSem0.xyz; +attrDecoder.xyz = (attrDecoder.xyz>>24)|((attrDecoder.xyz>>8)&0xFF00)|((attrDecoder.xyz<<8)&0xFF0000)|((attrDecoder.xyz<<24)); +attrDecoder.w = 0; +R1f = vec4(intBitsToFloat(int(attrDecoder.x)), intBitsToFloat(int(attrDecoder.y)), intBitsToFloat(int(attrDecoder.z)), intBitsToFloat(floatBitsToInt(1.0))); +attrDecoder.xy = attrDataSem1.xy; +attrDecoder.xy = (attrDecoder.xy>>24)|((attrDecoder.xy>>8)&0xFF00)|((attrDecoder.xy<<8)&0xFF0000)|((attrDecoder.xy<<24)); +attrDecoder.z = 0; +attrDecoder.w = 0; +R2f = vec4(intBitsToFloat(int(attrDecoder.x)), intBitsToFloat(int(attrDecoder.y)), intBitsToFloat(floatBitsToInt(0.0)), intBitsToFloat(floatBitsToInt(1.0))); +// 0 +backupReg0f = R1f.x; +backupReg1f = R1f.y; +R1f.x = backupReg0f; +R1f.x *= 2.0; +R1f.y = backupReg1f; +R1f.y *= 2.0; +R1f.z = intBitsToFloat(0xbf5fdfe0); +R1f.w = 1.0; +!!! R127f.y = intBitsToFloat(uf_remappedVS[0].x) * intBitsToFloat(0x3fb13a93)/resXScale; +PS0f = R127f.y; +// 1 +!!! R127f.x = intBitsToFloat(uf_remappedVS[0].x) * intBitsToFloat(0x404ec4f0)/resXScale; +PV1f.x = R127f.x; +R4f.y = R2f.y; +R4f.z = R2f.x; +R3f.w = R2f.x; +R3f.x = R2f.y; +PS1f = R3f.x; +// 2 +R0f.x = PS1f; +R3f.y = R2f.x + -(R127f.y); +R3f.z = R2f.x + R127f.y; +R0f.w = R2f.x; +R0f.y = R2f.x + -(PV1f.x); +PS0f = R0f.y; +// 3 +R0f.z = R2f.x + R127f.x; +// export +gl_Position = vec4(R1f.x, R1f.y, R1f.z, R1f.w); +// export +passParameterSem1 = vec4(R0f.x, R0f.y, R0f.z, R0f.w); +// export +passParameterSem0 = vec4(R3f.x, R3f.y, R3f.z, R3f.w); +// export +// skipped export to semanticId 255 +} From 937e5b074a1d2d93d1ecf17b082dede75110f840 Mon Sep 17 00:00:00 2001 From: Alex Kiri Date: Tue, 21 Nov 2017 22:55:07 +0200 Subject: [PATCH 09/10] [BotW] Self shadowing issue - compile fix --- .../BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt b/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt index e25b12b6..e6b219ae 100644 --- a/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt +++ b/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt @@ -72,10 +72,10 @@ R1f.y = backupReg1f; R1f.y *= 2.0; R1f.z = intBitsToFloat(0xbf5fdfe0); R1f.w = 1.0; -!!! R127f.y = intBitsToFloat(uf_remappedVS[0].x) * intBitsToFloat(0x3fb13a93)/resXScale; +R127f.y = intBitsToFloat(uf_remappedVS[0].x) * intBitsToFloat(0x3fb13a93)/resXScale; PS0f = R127f.y; // 1 -!!! R127f.x = intBitsToFloat(uf_remappedVS[0].x) * intBitsToFloat(0x404ec4f0)/resXScale; +R127f.x = intBitsToFloat(uf_remappedVS[0].x) * intBitsToFloat(0x404ec4f0)/resXScale; PV1f.x = R127f.x; R4f.y = R2f.y; R4f.z = R2f.x; From 9e0c51ec0290dd79c403b915cdf9e0df27eccd96 Mon Sep 17 00:00:00 2001 From: Alex Kiri Date: Tue, 21 Nov 2017 23:30:35 +0200 Subject: [PATCH 10/10] [BotW] Self shadowing fix - part 2 --- .../44b73ce02e05c2e6_0000000000000000_vs.txt | 105 ++++++++++++++++++ .../93f16bf1d083933b_0000000000000000_vs.txt | 6 +- 2 files changed, 108 insertions(+), 3 deletions(-) create mode 100644 Source/BreathOfTheWild/44b73ce02e05c2e6_0000000000000000_vs.txt diff --git a/Source/BreathOfTheWild/44b73ce02e05c2e6_0000000000000000_vs.txt b/Source/BreathOfTheWild/44b73ce02e05c2e6_0000000000000000_vs.txt new file mode 100644 index 00000000..7095485e --- /dev/null +++ b/Source/BreathOfTheWild/44b73ce02e05c2e6_0000000000000000_vs.txt @@ -0,0 +1,105 @@ + +#version 420 +#extension GL_ARB_texture_gather : enable +#extension GL_ARB_shading_language_packing : enable +// shader 44b73ce02e05c2e6 +// Self shadowing mask fix - v blur +const float resXScale = ; +const float resYScale = ; + +uniform ivec4 uf_remappedVS[1]; +uniform vec2 uf_windowSpaceToClipSpaceTransform; +layout(location = 0) in uvec4 attrDataSem0; +layout(location = 1) in uvec4 attrDataSem1; +out gl_PerVertex +{ + vec4 gl_Position; + float gl_PointSize; +}; +layout(location = 0) out vec4 passParameterSem0; +layout(location = 1) out vec4 passParameterSem1; +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 R126f = vec4(0.0); +vec4 R127f = vec4(0.0); +uvec4 attrDecoder; +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 = floatBitsToInt(ivec4(gl_VertexID, 0, 0, gl_InstanceID)); +attrDecoder.xyz = attrDataSem0.xyz; +attrDecoder.xyz = (attrDecoder.xyz>>24)|((attrDecoder.xyz>>8)&0xFF00)|((attrDecoder.xyz<<8)&0xFF0000)|((attrDecoder.xyz<<24)); +attrDecoder.w = 0; +R1f = vec4(intBitsToFloat(int(attrDecoder.x)), intBitsToFloat(int(attrDecoder.y)), intBitsToFloat(int(attrDecoder.z)), intBitsToFloat(floatBitsToInt(1.0))); +attrDecoder.xy = attrDataSem1.xy; +attrDecoder.xy = (attrDecoder.xy>>24)|((attrDecoder.xy>>8)&0xFF00)|((attrDecoder.xy<<8)&0xFF0000)|((attrDecoder.xy<<24)); +attrDecoder.z = 0; +attrDecoder.w = 0; +R2f = vec4(intBitsToFloat(int(attrDecoder.x)), intBitsToFloat(int(attrDecoder.y)), intBitsToFloat(floatBitsToInt(0.0)), intBitsToFloat(floatBitsToInt(1.0))); +// 0 +backupReg0f = R1f.x; +backupReg1f = R1f.y; +R1f.x = backupReg0f; +R1f.x *= 2.0; +R1f.y = backupReg1f; +R1f.y *= 2.0; +R1f.z = intBitsToFloat(0xbf5fdfe0); +R1f.w = 1.0; +R127f.x = intBitsToFloat(uf_remappedVS[0].y) * intBitsToFloat(0x3fb13a93) / resYScale; +PS0f = R127f.x; +// 1 +R126f.x = intBitsToFloat(uf_remappedVS[0].y) * intBitsToFloat(0x404ec4f0) / resYScale; +PV1f.x = R126f.x; +R3f.y = R2f.y; +R3f.z = R2f.x; +R2f.w = R2f.y; +PV1f.w = R2f.w; +R2f.z = R2f.y + PS0f; +PS1f = R2f.z; +// 2 +R0f.x = R2f.x; +R0f.y = R2f.y + -(R127f.x); +R0f.z = PS1f; +R0f.w = PV1f.w; +R2f.z = R2f.y + PV1f.x; +PS0f = R2f.z; +// 3 +backupReg0f = R2f.y; +backupReg0f = R2f.y; +R2f.y = backupReg0f + -(R126f.x); +R2f.w = backupReg0f; +// export +gl_Position = vec4(R1f.x, R1f.y, R1f.z, R1f.w); +// export +passParameterSem0 = vec4(R0f.x, R0f.y, R0f.z, R0f.w); +// export +passParameterSem1 = vec4(R2f.x, R2f.y, R2f.z, R2f.w); +// export +// skipped export to semanticId 255 +} diff --git a/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt b/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt index e6b219ae..7570f81d 100644 --- a/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt +++ b/Source/BreathOfTheWild/93f16bf1d083933b_0000000000000000_vs.txt @@ -9,7 +9,7 @@ $scaleFactorY = always_decimal_format($fullHeight / 720.0); #extension GL_ARB_texture_gather : enable #extension GL_ARB_shading_language_packing : enable // shader 93f16bf1d083933b -// Self shadowing fix +// Self shadowing mask fix - h blur // original shader dumped using cemu 1.11.0, BotW 1.3.1 const float resXScale = ; const float resYScale = ; @@ -72,10 +72,10 @@ R1f.y = backupReg1f; R1f.y *= 2.0; R1f.z = intBitsToFloat(0xbf5fdfe0); R1f.w = 1.0; -R127f.y = intBitsToFloat(uf_remappedVS[0].x) * intBitsToFloat(0x3fb13a93)/resXScale; +R127f.y = intBitsToFloat(uf_remappedVS[0].x) * intBitsToFloat(0x3fb13a93) / resXScale; PS0f = R127f.y; // 1 -R127f.x = intBitsToFloat(uf_remappedVS[0].x) * intBitsToFloat(0x404ec4f0)/resXScale; +R127f.x = intBitsToFloat(uf_remappedVS[0].x) * intBitsToFloat(0x404ec4f0) / resXScale; PV1f.x = R127f.x; R4f.y = R2f.y; R4f.z = R2f.x;