2018-10-31 11:03:04 +01:00
#version 420
#extension GL_ARB_texture_gather : enable
Update BotW packs for Vulkan (#411)
But now done properly! Basically, a bunch of improvements were made to the script. The previous attempt at this conversion was quickly followed by a rollback since I realized that the script was overlooking certain things that made most of the packs hit or miss whether they would work. A few things missing were:
- It only tested the values from 1 preset. Now, each shader gets compiled per each preset, like what Cemu would do. It also merges the changes done for each preset into one. This should solve cases where one shader would define things separately or repeatedly from preset to preset.
- All* of the shaders are tested to see if they use the converter used the right values for the locations for Vulkan.
Both of these *should* mean that they should both compile and be linkable in Vulkan, which means that I don't have to test each individual shader to see if they work. I will release the two scripts (one used for converting, one used for checking the right values for the locations) tomorrow so that other people might be able to help, if they want. It's fairly straightforward now at least.
* Organize workaround graphic packs
Pretty hard to organize these correctly, but according to our discord discussion, this was the best layout from a bunch I proposed, together with some suggestions.
* Add V4 converter script and instructions on how to use it
Now everyone BotW is done and all of the bugs have been kinked out using it (hopefully...), here's the release of the converter script in all of it's very badly coded glory. I hope I didn't leave too much debug glory in there...
Also, I hope that I didn't make too many grammatical mistakes in the instructions, but hopefully it's easy enough to follow.
2019-12-28 05:55:52 +01:00
#ifdef VULKAN
#define ATTR_LAYOUT(__vkSet, __location) layout(set = __vkSet, location = __location)
#define UNIFORM_BUFFER_LAYOUT(__glLocation, __vkSet, __vkLocation) layout(set = __vkSet, binding = __vkLocation, std140)
#define TEXTURE_LAYOUT(__glLocation, __vkSet, __vkLocation) layout(set = __vkSet, binding = __vkLocation)
#define SET_POSITION(_v) gl_Position = _v; gl_Position.z = (gl_Position.z + gl_Position.w) / 2.0
#define GET_FRAGCOORD() vec4(gl_FragCoord.xy*uf_fragCoordScale.xy,gl_FragCoord.zw)
#define gl_VertexID gl_VertexIndex
#define gl_InstanceID gl_InstanceIndex
#else
#define ATTR_LAYOUT(__vkSet, __location) layout(location = __location)
#define UNIFORM_BUFFER_LAYOUT(__glLocation, __vkSet, __vkLocation) layout(binding = __glLocation, std140)
#define TEXTURE_LAYOUT(__glLocation, __vkSet, __vkLocation) layout(binding = __glLocation)
#define SET_POSITION(_v) gl_Position = _v
#define GET_FRAGCOORD() vec4(gl_FragCoord.xy*uf_fragCoordScale,gl_FragCoord.zw)
#endif
2020-01-05 04:16:42 +01:00
// This shader was automatically converted to be cross-compatible with Vulkan and OpenGL.
2018-12-03 06:05:15 +01:00
Update BotW packs for Vulkan (#411)
But now done properly! Basically, a bunch of improvements were made to the script. The previous attempt at this conversion was quickly followed by a rollback since I realized that the script was overlooking certain things that made most of the packs hit or miss whether they would work. A few things missing were:
- It only tested the values from 1 preset. Now, each shader gets compiled per each preset, like what Cemu would do. It also merges the changes done for each preset into one. This should solve cases where one shader would define things separately or repeatedly from preset to preset.
- All* of the shaders are tested to see if they use the converter used the right values for the locations for Vulkan.
Both of these *should* mean that they should both compile and be linkable in Vulkan, which means that I don't have to test each individual shader to see if they work. I will release the two scripts (one used for converting, one used for checking the right values for the locations) tomorrow so that other people might be able to help, if they want. It's fairly straightforward now at least.
* Organize workaround graphic packs
Pretty hard to organize these correctly, but according to our discord discussion, this was the best layout from a bunch I proposed, together with some suggestions.
* Add V4 converter script and instructions on how to use it
Now everyone BotW is done and all of the bugs have been kinked out using it (hopefully...), here's the release of the converter script in all of it's very badly coded glory. I hope I didn't leave too much debug glory in there...
Also, I hope that I didn't make too many grammatical mistakes in the instructions, but hopefully it's easy enough to follow.
2019-12-28 05:55:52 +01:00
#ifdef VULKAN
layout(set = 1, binding = 2) uniform ufBlock
{
uniform ivec4 uf_remappedPS[2];
2020-01-05 04:16:42 +01:00
uniform vec4 uf_fragCoordScale;
Update BotW packs for Vulkan (#411)
But now done properly! Basically, a bunch of improvements were made to the script. The previous attempt at this conversion was quickly followed by a rollback since I realized that the script was overlooking certain things that made most of the packs hit or miss whether they would work. A few things missing were:
- It only tested the values from 1 preset. Now, each shader gets compiled per each preset, like what Cemu would do. It also merges the changes done for each preset into one. This should solve cases where one shader would define things separately or repeatedly from preset to preset.
- All* of the shaders are tested to see if they use the converter used the right values for the locations for Vulkan.
Both of these *should* mean that they should both compile and be linkable in Vulkan, which means that I don't have to test each individual shader to see if they work. I will release the two scripts (one used for converting, one used for checking the right values for the locations) tomorrow so that other people might be able to help, if they want. It's fairly straightforward now at least.
* Organize workaround graphic packs
Pretty hard to organize these correctly, but according to our discord discussion, this was the best layout from a bunch I proposed, together with some suggestions.
* Add V4 converter script and instructions on how to use it
Now everyone BotW is done and all of the bugs have been kinked out using it (hopefully...), here's the release of the converter script in all of it's very badly coded glory. I hope I didn't leave too much debug glory in there...
Also, I hope that I didn't make too many grammatical mistakes in the instructions, but hopefully it's easy enough to follow.
2019-12-28 05:55:52 +01:00
};
#else
uniform ivec4 uf_remappedPS[2];
2020-01-05 04:16:42 +01:00
uniform vec2 uf_fragCoordScale;
Update BotW packs for Vulkan (#411)
But now done properly! Basically, a bunch of improvements were made to the script. The previous attempt at this conversion was quickly followed by a rollback since I realized that the script was overlooking certain things that made most of the packs hit or miss whether they would work. A few things missing were:
- It only tested the values from 1 preset. Now, each shader gets compiled per each preset, like what Cemu would do. It also merges the changes done for each preset into one. This should solve cases where one shader would define things separately or repeatedly from preset to preset.
- All* of the shaders are tested to see if they use the converter used the right values for the locations for Vulkan.
Both of these *should* mean that they should both compile and be linkable in Vulkan, which means that I don't have to test each individual shader to see if they work. I will release the two scripts (one used for converting, one used for checking the right values for the locations) tomorrow so that other people might be able to help, if they want. It's fairly straightforward now at least.
* Organize workaround graphic packs
Pretty hard to organize these correctly, but according to our discord discussion, this was the best layout from a bunch I proposed, together with some suggestions.
* Add V4 converter script and instructions on how to use it
Now everyone BotW is done and all of the bugs have been kinked out using it (hopefully...), here's the release of the converter script in all of it's very badly coded glory. I hope I didn't leave too much debug glory in there...
Also, I hope that I didn't make too many grammatical mistakes in the instructions, but hopefully it's easy enough to follow.
2019-12-28 05:55:52 +01:00
#endif
2018-12-03 06:05:15 +01:00
// shader 0f2b9ee517917425 - dumped 1.15
// Used for: Removing/Restoring the native BotW Anti-Aliasing implementation to link in inventory screen
2018-10-31 11:03:04 +01:00
#define preset $preset
2018-12-12 02:07:26 +01:00
#define iSharper $inventorySharper
#define iBlurrier $inventoryBlurrier
2018-10-31 11:03:04 +01:00
2018-12-05 01:18:00 +01:00
#if (preset == 0) // Native AA Disabled
Update BotW packs for Vulkan (#411)
But now done properly! Basically, a bunch of improvements were made to the script. The previous attempt at this conversion was quickly followed by a rollback since I realized that the script was overlooking certain things that made most of the packs hit or miss whether they would work. A few things missing were:
- It only tested the values from 1 preset. Now, each shader gets compiled per each preset, like what Cemu would do. It also merges the changes done for each preset into one. This should solve cases where one shader would define things separately or repeatedly from preset to preset.
- All* of the shaders are tested to see if they use the converter used the right values for the locations for Vulkan.
Both of these *should* mean that they should both compile and be linkable in Vulkan, which means that I don't have to test each individual shader to see if they work. I will release the two scripts (one used for converting, one used for checking the right values for the locations) tomorrow so that other people might be able to help, if they want. It's fairly straightforward now at least.
* Organize workaround graphic packs
Pretty hard to organize these correctly, but according to our discord discussion, this was the best layout from a bunch I proposed, together with some suggestions.
* Add V4 converter script and instructions on how to use it
Now everyone BotW is done and all of the bugs have been kinked out using it (hopefully...), here's the release of the converter script in all of it's very badly coded glory. I hope I didn't leave too much debug glory in there...
Also, I hope that I didn't make too many grammatical mistakes in the instructions, but hopefully it's easy enough to follow.
2019-12-28 05:55:52 +01:00
TEXTURE_LAYOUT(0, 1, 0) uniform sampler2D textureUnitPS0;
2018-10-31 11:03:04 +01:00
layout(location = 0) in vec4 passParameterSem2;
layout(location = 0) out vec4 passPixelColor0;
void main()
{
passPixelColor0 = texture(textureUnitPS0, passParameterSem2.xy);
}
#endif
2018-12-05 01:18:00 +01:00
#if (preset == 1) // Native AA Enabled
Update BotW packs for Vulkan (#411)
But now done properly! Basically, a bunch of improvements were made to the script. The previous attempt at this conversion was quickly followed by a rollback since I realized that the script was overlooking certain things that made most of the packs hit or miss whether they would work. A few things missing were:
- It only tested the values from 1 preset. Now, each shader gets compiled per each preset, like what Cemu would do. It also merges the changes done for each preset into one. This should solve cases where one shader would define things separately or repeatedly from preset to preset.
- All* of the shaders are tested to see if they use the converter used the right values for the locations for Vulkan.
Both of these *should* mean that they should both compile and be linkable in Vulkan, which means that I don't have to test each individual shader to see if they work. I will release the two scripts (one used for converting, one used for checking the right values for the locations) tomorrow so that other people might be able to help, if they want. It's fairly straightforward now at least.
* Organize workaround graphic packs
Pretty hard to organize these correctly, but according to our discord discussion, this was the best layout from a bunch I proposed, together with some suggestions.
* Add V4 converter script and instructions on how to use it
Now everyone BotW is done and all of the bugs have been kinked out using it (hopefully...), here's the release of the converter script in all of it's very badly coded glory. I hope I didn't leave too much debug glory in there...
Also, I hope that I didn't make too many grammatical mistakes in the instructions, but hopefully it's easy enough to follow.
2019-12-28 05:55:52 +01:00
TEXTURE_LAYOUT(0, 1, 0) uniform sampler2D textureUnitPS0;
TEXTURE_LAYOUT(1, 1, 1) uniform sampler2D textureUnitPS1;
2018-10-31 11:03:04 +01:00
layout(location = 0) in vec4 passParameterSem2;
layout(location = 0) out vec4 passPixelColor0;
Update BotW packs for Vulkan (#411)
But now done properly! Basically, a bunch of improvements were made to the script. The previous attempt at this conversion was quickly followed by a rollback since I realized that the script was overlooking certain things that made most of the packs hit or miss whether they would work. A few things missing were:
- It only tested the values from 1 preset. Now, each shader gets compiled per each preset, like what Cemu would do. It also merges the changes done for each preset into one. This should solve cases where one shader would define things separately or repeatedly from preset to preset.
- All* of the shaders are tested to see if they use the converter used the right values for the locations for Vulkan.
Both of these *should* mean that they should both compile and be linkable in Vulkan, which means that I don't have to test each individual shader to see if they work. I will release the two scripts (one used for converting, one used for checking the right values for the locations) tomorrow so that other people might be able to help, if they want. It's fairly straightforward now at least.
* Organize workaround graphic packs
Pretty hard to organize these correctly, but according to our discord discussion, this was the best layout from a bunch I proposed, together with some suggestions.
* Add V4 converter script and instructions on how to use it
Now everyone BotW is done and all of the bugs have been kinked out using it (hopefully...), here's the release of the converter script in all of it's very badly coded glory. I hope I didn't leave too much debug glory in there...
Also, I hope that I didn't make too many grammatical mistakes in the instructions, but hopefully it's easy enough to follow.
2019-12-28 05:55:52 +01:00
// uf_fragCoordScale was moved to the ufBlock
2018-12-03 06:05:15 +01:00
2019-01-06 23:46:59 +01:00
ivec2 resDim = textureSize(textureUnitPS0,0); // Retrieve texture dimensions vector holds data-type-float
float iresX = ( (float(resDim.x)/float(1280)) + iSharper ) - iBlurrier; // 1st comes aaSharper needs to be added to the direct result of resolution ratio to make it more sharper
float iresY = ( (float(resDim.y)/float(720)) + iSharper ) - iBlurrier; // 2nd comes aablurier needs to be subtracted from final result to make it more blurrier
2018-12-03 06:05:15 +01:00
2018-10-31 11:03:04 +01:00
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));
}
2018-12-03 06:05:15 +01:00
float mul_nonIEEE(float a, float b){return mix(0.0, a*b, (a != 0.0) && (b != 0.0));}
2018-10-31 11:03:04 +01:00
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 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;
2018-12-03 06:05:15 +01:00
ivec4 PV0i = ivec4(0), PV1i = ivec4(0); // These variables make the difference
2018-10-31 11:03:04 +01:00
int PS0i = 0, PS1i = 0;
ivec4 tempi = ivec4(0);
float tempResultf;
int tempResulti;
ivec4 ARi = ivec4(0);
bool predResult = true;
bool activeMaskStack[2];
bool activeMaskStackC[3];
activeMaskStack[0] = false;
activeMaskStackC[0] = false;
activeMaskStackC[1] = false;
activeMaskStack[0] = true;
activeMaskStackC[0] = true;
activeMaskStackC[1] = true;
vec3 cubeMapSTM;
int cubeMapFaceId;
R0i = floatBitsToInt(passParameterSem2);
if( activeMaskStackC[1] == true ) {
R2i.xzw = floatBitsToInt(textureGather(textureUnitPS1, intBitsToFloat(R0i.xy)).xzw);
R1i.xz = floatBitsToInt(textureGather(textureUnitPS1, intBitsToFloat(R0i.zw)).xz);
R3i.xyzw = floatBitsToInt(texture(textureUnitPS0, intBitsToFloat(R0i.xy)).xyzw);
R0i.w = floatBitsToInt(textureOffset(textureUnitPS1, intBitsToFloat(R0i.xy),ivec2(1,-1)).x);
R1i.y = floatBitsToInt(textureOffset(textureUnitPS1, intBitsToFloat(R0i.xy),ivec2(-1,1)).x);
}
if( activeMaskStackC[1] == true ) {
activeMaskStack[1] = activeMaskStack[0];
activeMaskStackC[2] = activeMaskStackC[1];
2018-12-03 06:05:15 +01:00
// 0 --- Point of Interest 1
2018-12-12 02:07:26 +01:00
PV0i.x = floatBitsToInt(min(intBitsToFloat(R1i.x), intBitsToFloat(R1i.z)) / iresX ); // Divide looks better for minimum - Must place the varaibles in that location of the round brackets to use floats correctly
PV0i.y = floatBitsToInt(max(intBitsToFloat(R2i.x), intBitsToFloat(R2i.z)) * iresY ); // Multiply looks beeter for max - Must place the varaibles in that location of the round brackets to use floats correctly
PV0i.z = floatBitsToInt(max(intBitsToFloat(R1i.x), intBitsToFloat(R1i.z)) * iresX ); // Multiply looks better for max - Must place the varaibles in that location of the round brackets to use floats correctly
2018-10-31 11:03:04 +01:00
PV0i.w = floatBitsToInt(min(intBitsToFloat(R2i.x), intBitsToFloat(R2i.z)));
2018-12-03 06:05:15 +01:00
// 1 ---Point of Interest 2
2018-10-31 11:03:04 +01:00
PV1i.x = floatBitsToInt(max(intBitsToFloat(PV0i.z), intBitsToFloat(PV0i.y)));
PV1i.y = floatBitsToInt(min(intBitsToFloat(PV0i.x), intBitsToFloat(PV0i.w)));
// 2
PV0i.z = floatBitsToInt(min(intBitsToFloat(R2i.w), intBitsToFloat(PV1i.y)));
PV0i.w = floatBitsToInt(max(intBitsToFloat(R2i.w), intBitsToFloat(PV1i.x)));
2018-12-03 06:05:15 +01:00
// 3
2019-01-04 05:39:34 +01:00
PV1i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.w), intBitsToFloat(uf_remappedPS[0].x) / iresX ));
2018-10-31 11:03:04 +01:00
R1i.w = floatBitsToInt(intBitsToFloat(PV0i.w) + -(intBitsToFloat(PV0i.z)));
2018-12-03 06:05:15 +01:00
// 4
2018-10-31 11:03:04 +01:00
R2i.y = floatBitsToInt(max(intBitsToFloat(PV1i.x), intBitsToFloat(uf_remappedPS[0].y)));
// 5
predResult = (intBitsToFloat(R1i.w) >= intBitsToFloat(R2i.y));
activeMaskStack[1] = predResult;
activeMaskStackC[2] = predResult == true && activeMaskStackC[1] == true;
}
else {
activeMaskStack[1] = false;
activeMaskStackC[2] = false;
}
if( activeMaskStackC[2] == true ) {
// 0
R126i.xyz = floatBitsToInt(vec3(intBitsToFloat(R2i.z),intBitsToFloat(R1i.x),intBitsToFloat(R2i.z)) + vec3(intBitsToFloat(R0i.w),intBitsToFloat(R1i.y),intBitsToFloat(R1i.x)));
PV0i.z = R126i.z;
R127i.w = floatBitsToInt(intBitsToFloat(R2i.x) + intBitsToFloat(R1i.z));
PV0i.w = R127i.w;
R127i.y = R1i.z;
R127i.y = floatBitsToInt(intBitsToFloat(R127i.y) * 2.0);
PS0i = R127i.y;
// 1
PV1i.x = R2i.x;
PV1i.x = floatBitsToInt(intBitsToFloat(PV1i.x) * 2.0);
PV1i.y = floatBitsToInt(intBitsToFloat(R1i.x) + intBitsToFloat(R0i.w));
R127i.z = floatBitsToInt((-(intBitsToFloat(R2i.w)) * 2.0 + intBitsToFloat(PV0i.z)));
PV1i.w = PV0i.w;
PS1i = floatBitsToInt(intBitsToFloat(R2i.z) + intBitsToFloat(R1i.y));
// 2
R127i.x = floatBitsToInt((-(intBitsToFloat(R2i.w)) * 2.0 + intBitsToFloat(PV1i.w)));
R1i.y = R2i.z;
PV0i.y = R1i.y;
PV0i.z = floatBitsToInt(intBitsToFloat(PS1i) + -(intBitsToFloat(PV1i.x)));
PV0i.w = floatBitsToInt(intBitsToFloat(PV1i.y) + -(intBitsToFloat(R127i.y)));
PS0i = R126i.x;
// 3
backupReg0i = R127i.z;
backupReg0i = R127i.z;
R123i.x = floatBitsToInt((-(intBitsToFloat(R1i.x)) * 2.0 + intBitsToFloat(R126i.y)));
PV1i.x = R123i.x;
PV1i.y = floatBitsToInt(max(intBitsToFloat(PV0i.w), -(intBitsToFloat(PV0i.w))));
R127i.z = floatBitsToInt(max(intBitsToFloat(PV0i.z), -(intBitsToFloat(PV0i.z))));
R123i.w = floatBitsToInt((-(intBitsToFloat(PV0i.y)) * 2.0 + intBitsToFloat(PS0i)));
PV1i.w = R123i.w;
PS1i = floatBitsToInt(max(intBitsToFloat(backupReg0i), -(intBitsToFloat(backupReg0i))));
PS1i = floatBitsToInt(intBitsToFloat(PS1i) * 2.0);
// 4
backupReg0i = R126i.y;
PV0i.x = floatBitsToInt(max(intBitsToFloat(PV1i.x), -(intBitsToFloat(PV1i.x))));
R126i.y = floatBitsToInt(max(intBitsToFloat(PV1i.w), -(intBitsToFloat(PV1i.w))));
PV0i.z = floatBitsToInt(max(intBitsToFloat(R127i.x), -(intBitsToFloat(R127i.x))));
PV0i.z = floatBitsToInt(intBitsToFloat(PV0i.z) * 2.0);
PV0i.w = floatBitsToInt(intBitsToFloat(PV1i.y) + intBitsToFloat(PS1i));
R126i.w = floatBitsToInt(intBitsToFloat(R126i.x) + intBitsToFloat(backupReg0i));
PS0i = R126i.w;
// 5
backupReg0i = R127i.z;
PV1i.x = floatBitsToInt(intBitsToFloat(PV0i.x) + intBitsToFloat(PV0i.z));
PV1i.y = floatBitsToInt(intBitsToFloat(R127i.w) + intBitsToFloat(R126i.z));
PV1i.y = floatBitsToInt(intBitsToFloat(PV1i.y) * 2.0);
R127i.z = floatBitsToInt(intBitsToFloat(backupReg0i) + intBitsToFloat(PV0i.w));
2018-12-03 06:05:15 +01:00
R127i.y = floatBitsToInt(1.0 / intBitsToFloat(R1i.w)); // Rli.w * 2 has the same affect as line 85------------------------------------------------------
2018-10-31 11:03:04 +01:00
PS1i = R127i.y;
// 6
PV0i.x = floatBitsToInt(intBitsToFloat(R126i.w) + intBitsToFloat(PV1i.y));
PV0i.y = floatBitsToInt(intBitsToFloat(R126i.y) + intBitsToFloat(PV1i.x));
// 7
PV1i.x = ((intBitsToFloat(PV0i.y) >= intBitsToFloat(R127i.z))?int(0xFFFFFFFF):int(0x0));
PV1i.y = floatBitsToInt(intBitsToFloat(PV0i.x) * intBitsToFloat(0x3daaaaab));
// 8
PV0i.x = floatBitsToInt(intBitsToFloat(R2i.w) + -(intBitsToFloat(PV1i.y)));
R4i.z = ((PV1i.x == 0)?(0x3f800000):(0));
PV0i.z = R4i.z;
R5i.w = ((PV1i.x == 0)?(0):(0x3f800000));
PV0i.w = R5i.w;
2018-12-03 06:05:15 +01:00
// 9 --- Point fo Interest
2019-01-04 05:39:34 +01:00
R5i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.w), intBitsToFloat(uf_remappedPS[1].x) / iresX )); // Default implementation division took place here
PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.z), intBitsToFloat(uf_remappedPS[1].x) / iresX )); // Default implementation division took place here
2018-10-31 11:03:04 +01:00
PV1i.z = floatBitsToInt(max(intBitsToFloat(PV0i.x), -(intBitsToFloat(PV0i.x))));
2019-01-04 05:39:34 +01:00
R3i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(PV0i.z), intBitsToFloat(uf_remappedPS[1].y) / iresY )); // Default implementation division took place here
2018-10-31 11:03:04 +01:00
PS1i = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R1i.x), intBitsToFloat(PV0i.z)));
2018-12-03 06:05:15 +01:00
// 10 --- Point of Interest
R127i.x = floatBitsToInt(intBitsToFloat(PV1i.z) * intBitsToFloat(R127i.y)); // Divide looks good same as below line ----------------------------------------------------------------------
R127i.x = clampFI32(R127i.x); // Divide looks good same as above line----------------------------------------------------------------------------------------------------
PV0i.x = R127i.x;
2019-01-04 05:39:34 +01:00
R127i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.w),intBitsToFloat(uf_remappedPS[1].y) / iresY ) + intBitsToFloat(PV1i.y))); // Default implementation division took place here
2018-10-31 11:03:04 +01:00
R127i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R1i.z),intBitsToFloat(R5i.w)) + intBitsToFloat(PS1i)));
PV0i.z = R127i.z;
PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R2i.z), intBitsToFloat(R4i.z)));
// 11
R124i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R2i.x),intBitsToFloat(R5i.w)) + intBitsToFloat(PV0i.w)));
PV1i.x = R124i.x;
PV1i.y = floatBitsToInt(-(intBitsToFloat(R2i.w)) + intBitsToFloat(PV0i.z));
R123i.w = floatBitsToInt((intBitsToFloat(PV0i.x) * intBitsToFloat(0x40c00000) + intBitsToFloat(0xc1700000)));
PV1i.w = R123i.w;
// 12
R125i.x = floatBitsToInt(max(intBitsToFloat(PV1i.y), -(intBitsToFloat(PV1i.y))));
PV0i.y = floatBitsToInt(-(intBitsToFloat(R2i.w)) + intBitsToFloat(PV1i.x));
R123i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R127i.x),intBitsToFloat(PV1i.w)) + intBitsToFloat(0x41200000)));
PV0i.z = R123i.z;
// 13
R126i.x = floatBitsToInt(max(intBitsToFloat(PV0i.y), -(intBitsToFloat(PV0i.y))));
PV1i.x = R126i.x;
PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PV0i.z)));
// 14
PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PV1i.y)));
R126i.w = ((intBitsToFloat(R125i.x) >= intBitsToFloat(PV1i.x))?int(0xFFFFFFFF):int(0x0));
PV0i.w = R126i.w;
// 15
R6i.x = floatBitsToInt(((PV0i.w == 0)?(intBitsToFloat(R127i.y)):(-(intBitsToFloat(R127i.y)))));
PV1i.x = R6i.x;
R123i.z = ((PV0i.w == 0)?(R126i.x):(R125i.x));
PV1i.z = R123i.z;
PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R127i.x), intBitsToFloat(PV0i.x)));
2018-12-03 06:05:15 +01:00
// 16 --- Point of Interest
2018-10-31 11:03:04 +01:00
PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R4i.z), intBitsToFloat(PV1i.x)));
2018-12-03 06:05:15 +01:00
PV0i.x = floatBitsToInt(intBitsToFloat(PV0i.x) / 2.0); // Important Doubling improves curves and clarity
2018-10-31 11:03:04 +01:00
R123i.y = ((R126i.w == 0)?(R124i.x):(R127i.z));
PV0i.y = R123i.y;
R3i.z = floatBitsToInt(intBitsToFloat(PV1i.w) * intBitsToFloat(0x3f400000));
PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(PV1i.x)));
PV0i.w = floatBitsToInt(intBitsToFloat(PV0i.w) / 2.0);
R4i.x = floatBitsToInt(intBitsToFloat(PV1i.z) * 0.25);
PS0i = R4i.x;
// 17
backupReg0i = R0i.x;
PV1i.x = floatBitsToInt(intBitsToFloat(backupReg0i) + intBitsToFloat(PV0i.x));
PV1i.z = floatBitsToInt(intBitsToFloat(R0i.y) + intBitsToFloat(PV0i.w));
R4i.w = floatBitsToInt(intBitsToFloat(R2i.w) + intBitsToFloat(PV0i.y));
R4i.w = floatBitsToInt(intBitsToFloat(R4i.w) / 2.0);
PV1i.w = R4i.w;
// 18
R3i.x = floatBitsToInt(-(intBitsToFloat(R5i.x)) + intBitsToFloat(PV1i.x));
R3i.y = floatBitsToInt(-(intBitsToFloat(R3i.w)) + intBitsToFloat(PV1i.z));
R1i.z = floatBitsToInt(intBitsToFloat(R5i.x) + intBitsToFloat(PV1i.x));
R1i.w = floatBitsToInt(intBitsToFloat(R3i.w) + intBitsToFloat(PV1i.z));
R2i.x = floatBitsToInt(intBitsToFloat(R2i.w) + -(intBitsToFloat(PV1i.w)));
PS0i = R2i.x;
}
if( activeMaskStackC[2] == true ) {
R1i.y = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R3i.xy)).x);
R1i.x = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
backupReg0i = R2i.x;
R2i.x = ((0.0 > intBitsToFloat(backupReg0i))?int(0xFFFFFFFF):int(0x0));
PV0i.z = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.x));
PV0i.w = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.y));
// 1
PV1i.y = floatBitsToInt(max(intBitsToFloat(PV0i.w), -(intBitsToFloat(PV0i.w))));
PV1i.z = floatBitsToInt(max(intBitsToFloat(PV0i.z), -(intBitsToFloat(PV0i.z))));
// 2
PV0i.x = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.y)));
PV0i.y = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.z)));
// 3
R123i.z = ((intBitsToFloat(PV0i.y) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.z = R123i.z;
R123i.w = ((intBitsToFloat(PV0i.x) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.w = R123i.w;
// 4
backupReg0i = R3i.x;
backupReg1i = R3i.y;
R3i.x = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R5i.x)),intBitsToFloat(PV1i.w)) + intBitsToFloat(backupReg0i)));
R3i.y = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R3i.w)),intBitsToFloat(PV1i.w)) + intBitsToFloat(backupReg1i)));
R2i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.x),intBitsToFloat(PV1i.z)) + intBitsToFloat(R1i.z)));
R2i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R3i.w),intBitsToFloat(PV1i.z)) + intBitsToFloat(R1i.w)));
}
if( activeMaskStackC[2] == true ) {
R1i.w = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R3i.xy)).x);
R1i.z = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R2i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.x = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.w));
PV0i.y = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.z));
// 1
PV1i.x = floatBitsToInt(max(intBitsToFloat(PV0i.y), -(intBitsToFloat(PV0i.y))));
PV1i.w = floatBitsToInt(max(intBitsToFloat(PV0i.x), -(intBitsToFloat(PV0i.x))));
// 2
PV0i.z = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.w)));
PV0i.w = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.x)));
// 3
R123i.z = ((intBitsToFloat(PV0i.w) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.z = R123i.z;
R123i.w = ((intBitsToFloat(PV0i.z) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.w = R123i.w;
// 4
backupReg0i = R2i.z;
backupReg1i = R2i.w;
R1i.x = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R5i.x)),intBitsToFloat(PV1i.w)) + intBitsToFloat(R3i.x)));
R1i.y = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R3i.w)),intBitsToFloat(PV1i.w)) + intBitsToFloat(R3i.y)));
R2i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.x),intBitsToFloat(PV1i.z)) + intBitsToFloat(backupReg0i)));
R2i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R3i.w),intBitsToFloat(PV1i.z)) + intBitsToFloat(backupReg1i)));
}
if( activeMaskStackC[2] == true ) {
R3i.y = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.xy)).x);
R3i.x = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R2i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.x = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R3i.y));
PV0i.w = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R3i.x));
// 1
PV1i.z = floatBitsToInt(max(intBitsToFloat(PV0i.w), -(intBitsToFloat(PV0i.w))));
PV1i.w = floatBitsToInt(max(intBitsToFloat(PV0i.x), -(intBitsToFloat(PV0i.x))));
// 2
PV0i.y = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.z)));
PV0i.z = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.w)));
// 3
R123i.y = ((intBitsToFloat(PV0i.z) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.y = R123i.y;
R123i.z = ((intBitsToFloat(PV0i.y) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.z = R123i.z;
// 4
backupReg0i = R1i.x;
backupReg1i = R1i.y;
R1i.x = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R5i.x)),intBitsToFloat(PV1i.y)) + intBitsToFloat(backupReg0i)));
R1i.y = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R3i.w)),intBitsToFloat(PV1i.y)) + intBitsToFloat(backupReg1i)));
R1i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.x),intBitsToFloat(PV1i.z)) + intBitsToFloat(R2i.z)));
R1i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R3i.w),intBitsToFloat(PV1i.z)) + intBitsToFloat(R2i.w)));
}
if( activeMaskStackC[2] == true ) {
R2i.w = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.xy)).x);
R2i.z = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.x = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R2i.w));
PV0i.y = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R2i.z));
// 1
PV1i.x = floatBitsToInt(max(intBitsToFloat(PV0i.y), -(intBitsToFloat(PV0i.y))));
PV1i.w = floatBitsToInt(max(intBitsToFloat(PV0i.x), -(intBitsToFloat(PV0i.x))));
// 2
PV0i.z = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.w)));
PV0i.w = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.x)));
// 3
R123i.z = ((intBitsToFloat(PV0i.w) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.z = R123i.z;
R123i.w = ((intBitsToFloat(PV0i.z) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.w = R123i.w;
// 4
backupReg0i = R1i.z;
backupReg1i = R1i.w;
R3i.x = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R5i.x)),intBitsToFloat(PV1i.w)) + intBitsToFloat(R1i.x)));
R3i.y = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R3i.w)),intBitsToFloat(PV1i.w)) + intBitsToFloat(R1i.y)));
R1i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.x),intBitsToFloat(PV1i.z)) + intBitsToFloat(backupReg0i)));
R1i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R3i.w),intBitsToFloat(PV1i.z)) + intBitsToFloat(backupReg1i)));
}
if( activeMaskStackC[2] == true ) {
R1i.y = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R3i.xy)).x);
R1i.x = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.x = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.y));
PV0i.w = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.x));
// 1
PV1i.z = floatBitsToInt(max(intBitsToFloat(PV0i.w), -(intBitsToFloat(PV0i.w))));
PV1i.w = floatBitsToInt(max(intBitsToFloat(PV0i.x), -(intBitsToFloat(PV0i.x))));
// 2
PV0i.y = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.z)));
PV0i.z = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.w)));
// 3
R123i.y = ((intBitsToFloat(PV0i.z) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.y = R123i.y;
R123i.z = ((intBitsToFloat(PV0i.y) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.z = R123i.z;
// 4
backupReg0i = R3i.x;
backupReg1i = R3i.y;
R3i.x = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R5i.x)),intBitsToFloat(PV1i.y)) + intBitsToFloat(backupReg0i)));
R3i.y = floatBitsToInt((mul_nonIEEE(-(intBitsToFloat(R3i.w)),intBitsToFloat(PV1i.y)) + intBitsToFloat(backupReg1i)));
R2i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.x),intBitsToFloat(PV1i.z)) + intBitsToFloat(R1i.z)));
R2i.w = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R3i.w),intBitsToFloat(PV1i.z)) + intBitsToFloat(R1i.w)));
}
if( activeMaskStackC[2] == true ) {
R1i.w = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R3i.xy)).x);
R1i.z = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R2i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.x = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.w));
PV0i.y = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.z));
// 1
PV1i.x = floatBitsToInt(max(intBitsToFloat(PV0i.y), -(intBitsToFloat(PV0i.y))));
PV1i.w = floatBitsToInt(max(intBitsToFloat(PV0i.x), -(intBitsToFloat(PV0i.x))));
// 2
PV0i.z = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.w)));
PV0i.w = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.x)));
// 3
R123i.z = ((intBitsToFloat(PV0i.w) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.z = R123i.z;
R123i.w = ((intBitsToFloat(PV0i.z) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.w = R123i.w;
// 4
PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.w)));
PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.z)));
PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.z)));
PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.w)));
// 5
backupReg0i = R2i.z;
backupReg1i = R2i.w;
R1i.x = floatBitsToInt((-(intBitsToFloat(PV0i.x)) * 1.5 + intBitsToFloat(R3i.x)));
R1i.y = floatBitsToInt((-(intBitsToFloat(PV0i.w)) * 1.5 + intBitsToFloat(R3i.y)));
R2i.z = floatBitsToInt((intBitsToFloat(PV0i.z) * 1.5 + intBitsToFloat(backupReg0i)));
R2i.w = floatBitsToInt((intBitsToFloat(PV0i.y) * 1.5 + intBitsToFloat(backupReg1i)));
}
if( activeMaskStackC[2] == true ) {
R3i.y = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.xy)).x);
R3i.x = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R2i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.z = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R3i.y));
PV0i.w = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R3i.x));
// 1
PV1i.y = floatBitsToInt(max(intBitsToFloat(PV0i.z), -(intBitsToFloat(PV0i.z))));
PV1i.z = floatBitsToInt(max(intBitsToFloat(PV0i.w), -(intBitsToFloat(PV0i.w))));
// 2
PV0i.x = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.y)));
PV0i.y = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.z)));
// 3
R123i.x = ((intBitsToFloat(PV0i.y) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.x = R123i.x;
R123i.w = ((intBitsToFloat(PV0i.x) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.w = R123i.w;
// 4
PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.x)));
PV0i.x = floatBitsToInt(intBitsToFloat(PV0i.x) * 2.0);
PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.x)));
PV0i.y = floatBitsToInt(intBitsToFloat(PV0i.y) * 2.0);
PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.w)));
PV0i.z = floatBitsToInt(intBitsToFloat(PV0i.z) * 2.0);
PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.w)));
PV0i.w = floatBitsToInt(intBitsToFloat(PV0i.w) * 2.0);
// 5
backupReg0i = R1i.x;
backupReg1i = R1i.y;
R1i.xyz = floatBitsToInt(vec3(intBitsToFloat(backupReg0i),intBitsToFloat(backupReg1i),intBitsToFloat(R2i.z)) + vec3(-(intBitsToFloat(PV0i.w)),-(intBitsToFloat(PV0i.z)),intBitsToFloat(PV0i.x)));
R1i.w = floatBitsToInt(intBitsToFloat(R2i.w) + intBitsToFloat(PV0i.y));
}
if( activeMaskStackC[2] == true ) {
R2i.w = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.xy)).x);
R2i.z = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.y = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R2i.z));
PV0i.z = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R2i.w));
// 1
PV1i.x = floatBitsToInt(max(intBitsToFloat(PV0i.y), -(intBitsToFloat(PV0i.y))));
PV1i.y = floatBitsToInt(max(intBitsToFloat(PV0i.z), -(intBitsToFloat(PV0i.z))));
// 2
PV0i.x = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.y)));
PV0i.w = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.x)));
// 3
R123i.z = ((intBitsToFloat(PV0i.w) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.z = R123i.z;
R123i.w = ((intBitsToFloat(PV0i.x) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.w = R123i.w;
// 4
PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.w)));
PV0i.x = floatBitsToInt(intBitsToFloat(PV0i.x) * 2.0);
PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.z)));
PV0i.y = floatBitsToInt(intBitsToFloat(PV0i.y) * 2.0);
PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.z)));
PV0i.z = floatBitsToInt(intBitsToFloat(PV0i.z) * 2.0);
PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.w)));
PV0i.w = floatBitsToInt(intBitsToFloat(PV0i.w) * 2.0);
// 5
backupReg0i = R1i.z;
backupReg1i = R1i.w;
R3i.x = floatBitsToInt(intBitsToFloat(R1i.x) + -(intBitsToFloat(PV0i.x)));
R3i.y = floatBitsToInt(intBitsToFloat(R1i.y) + -(intBitsToFloat(PV0i.w)));
R1i.z = floatBitsToInt(intBitsToFloat(backupReg0i) + intBitsToFloat(PV0i.z));
R1i.w = floatBitsToInt(intBitsToFloat(backupReg1i) + intBitsToFloat(PV0i.y));
}
if( activeMaskStackC[2] == true ) {
R1i.y = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R3i.xy)).x);
R1i.x = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.z = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.y));
PV0i.w = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.x));
// 1
PV1i.y = floatBitsToInt(max(intBitsToFloat(PV0i.z), -(intBitsToFloat(PV0i.z))));
PV1i.z = floatBitsToInt(max(intBitsToFloat(PV0i.w), -(intBitsToFloat(PV0i.w))));
// 2
PV0i.x = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.y)));
PV0i.y = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.z)));
// 3
R123i.x = ((intBitsToFloat(PV0i.y) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.x = R123i.x;
R123i.w = ((intBitsToFloat(PV0i.x) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.w = R123i.w;
// 4
PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.x)));
PV0i.x = floatBitsToInt(intBitsToFloat(PV0i.x) * 2.0);
PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.x)));
PV0i.y = floatBitsToInt(intBitsToFloat(PV0i.y) * 2.0);
PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.w)));
PV0i.z = floatBitsToInt(intBitsToFloat(PV0i.z) * 2.0);
PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.w)));
PV0i.w = floatBitsToInt(intBitsToFloat(PV0i.w) * 2.0);
// 5
backupReg0i = R3i.x;
backupReg1i = R3i.y;
R3i.x = floatBitsToInt(intBitsToFloat(backupReg0i) + -(intBitsToFloat(PV0i.w)));
R3i.y = floatBitsToInt(intBitsToFloat(backupReg1i) + -(intBitsToFloat(PV0i.z)));
R2i.z = floatBitsToInt(intBitsToFloat(R1i.z) + intBitsToFloat(PV0i.x));
R2i.w = floatBitsToInt(intBitsToFloat(R1i.w) + intBitsToFloat(PV0i.y));
}
if( activeMaskStackC[2] == true ) {
R1i.w = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R3i.xy)).x);
R1i.z = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R2i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.y = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.z));
PV0i.z = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.w));
// 1
PV1i.x = floatBitsToInt(max(intBitsToFloat(PV0i.y), -(intBitsToFloat(PV0i.y))));
PV1i.y = floatBitsToInt(max(intBitsToFloat(PV0i.z), -(intBitsToFloat(PV0i.z))));
// 2
PV0i.x = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.y)));
PV0i.w = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.x)));
// 3
R123i.z = ((intBitsToFloat(PV0i.w) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.z = R123i.z;
R123i.w = ((intBitsToFloat(PV0i.x) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.w = R123i.w;
// 4
PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.w)));
PV0i.x = floatBitsToInt(intBitsToFloat(PV0i.x) * 2.0);
PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.z)));
PV0i.y = floatBitsToInt(intBitsToFloat(PV0i.y) * 2.0);
PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.z)));
PV0i.z = floatBitsToInt(intBitsToFloat(PV0i.z) * 2.0);
PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.w)));
PV0i.w = floatBitsToInt(intBitsToFloat(PV0i.w) * 2.0);
// 5
backupReg0i = R2i.z;
backupReg1i = R2i.w;
R1i.x = floatBitsToInt(intBitsToFloat(R3i.x) + -(intBitsToFloat(PV0i.x)));
R1i.y = floatBitsToInt(intBitsToFloat(R3i.y) + -(intBitsToFloat(PV0i.w)));
R2i.z = floatBitsToInt(intBitsToFloat(backupReg0i) + intBitsToFloat(PV0i.z));
R2i.w = floatBitsToInt(intBitsToFloat(backupReg1i) + intBitsToFloat(PV0i.y));
}
if( activeMaskStackC[2] == true ) {
R3i.y = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.xy)).x);
R3i.x = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R2i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.z = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R3i.y));
PV0i.w = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R3i.x));
// 1
PV1i.y = floatBitsToInt(max(intBitsToFloat(PV0i.z), -(intBitsToFloat(PV0i.z))));
PV1i.z = floatBitsToInt(max(intBitsToFloat(PV0i.w), -(intBitsToFloat(PV0i.w))));
// 2
PV0i.x = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.y)));
PV0i.y = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.z)));
// 3
R123i.x = ((intBitsToFloat(PV0i.y) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.x = R123i.x;
R123i.w = ((intBitsToFloat(PV0i.x) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.w = R123i.w;
// 4
PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.x)));
PV0i.x = floatBitsToInt(intBitsToFloat(PV0i.x) * 4.0);
PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.x)));
PV0i.y = floatBitsToInt(intBitsToFloat(PV0i.y) * 4.0);
PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.w)));
PV0i.z = floatBitsToInt(intBitsToFloat(PV0i.z) * 4.0);
PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.w)));
PV0i.w = floatBitsToInt(intBitsToFloat(PV0i.w) * 4.0);
// 5
backupReg0i = R1i.x;
backupReg1i = R1i.y;
R1i.xyz = floatBitsToInt(vec3(intBitsToFloat(backupReg0i),intBitsToFloat(backupReg1i),intBitsToFloat(R2i.z)) + vec3(-(intBitsToFloat(PV0i.w)),-(intBitsToFloat(PV0i.z)),intBitsToFloat(PV0i.x)));
R1i.w = floatBitsToInt(intBitsToFloat(R2i.w) + intBitsToFloat(PV0i.y));
}
if( activeMaskStackC[2] == true ) {
R2i.w = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.xy)).x);
R2i.z = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
PV0i.y = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R2i.z));
PV0i.z = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R2i.w));
// 1
PV1i.x = floatBitsToInt(max(intBitsToFloat(PV0i.y), -(intBitsToFloat(PV0i.y))));
PV1i.y = floatBitsToInt(max(intBitsToFloat(PV0i.z), -(intBitsToFloat(PV0i.z))));
// 2
PV0i.x = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.y)));
PV0i.w = floatBitsToInt(intBitsToFloat(R4i.x) + -(intBitsToFloat(PV1i.x)));
// 3
R123i.z = ((intBitsToFloat(PV0i.w) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.z = R123i.z;
R123i.w = ((intBitsToFloat(PV0i.x) >= 0.0)?(floatBitsToInt(1.0)):(0));
PV1i.w = R123i.w;
// 4
PV0i.x = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.w)));
PV0i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.x), intBitsToFloat(PV1i.z)));
PV0i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.z)));
PV0i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R3i.w), intBitsToFloat(PV1i.w)));
// 5
backupReg0i = R1i.z;
backupReg1i = R1i.w;
R5i.x = floatBitsToInt((-(intBitsToFloat(PV0i.x)) * intBitsToFloat(0x41000000) + intBitsToFloat(R1i.x)));
R5i.y = floatBitsToInt((-(intBitsToFloat(PV0i.w)) * intBitsToFloat(0x41000000) + intBitsToFloat(R1i.y)));
R1i.z = floatBitsToInt((intBitsToFloat(PV0i.y) * intBitsToFloat(0x41000000) + intBitsToFloat(backupReg0i)));
R1i.w = floatBitsToInt((intBitsToFloat(PV0i.z) * intBitsToFloat(0x41000000) + intBitsToFloat(backupReg1i)));
}
if( activeMaskStackC[2] == true ) {
R1i.y = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R5i.xy)).x);
R1i.x = floatBitsToInt(texture(textureUnitPS1, intBitsToFloat(R1i.zw)).x);
}
if( activeMaskStackC[2] == true ) {
// 0
backupReg0i = R0i.x;
backupReg1i = R0i.y;
backupReg0i = R0i.x;
PV0i.x = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.x));
PV0i.y = floatBitsToInt(intBitsToFloat(backupReg0i) + -(intBitsToFloat(R5i.x)));
PV0i.z = floatBitsToInt(-(intBitsToFloat(R4i.w)) + intBitsToFloat(R1i.y));
R126i.w = floatBitsToInt(intBitsToFloat(backupReg1i) + -(intBitsToFloat(R5i.y)));
PS0i = floatBitsToInt(-(intBitsToFloat(backupReg0i)) + intBitsToFloat(R1i.z));
// 1
PV1i.x = floatBitsToInt(-(intBitsToFloat(R0i.y)) + intBitsToFloat(R1i.w));
PV1i.y = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(PS0i)));
PV1i.z = ((0.0 > intBitsToFloat(PV0i.z))?int(0xFFFFFFFF):int(0x0));
PV1i.w = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R5i.w), intBitsToFloat(PV0i.y)));
PS1i = ((0.0 > intBitsToFloat(PV0i.x))?int(0xFFFFFFFF):int(0x0));
// 2
R123i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R4i.z),intBitsToFloat(PV1i.x)) + intBitsToFloat(PV1i.y)));
PV0i.x = R123i.x;
R127i.y = (PV1i.z != R2i.x)?int(0xFFFFFFFF):int(0x0);
R123i.z = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R4i.z),intBitsToFloat(R126i.w)) + intBitsToFloat(PV1i.w)));
PV0i.z = R123i.z;
R126i.w = (PS1i != R2i.x)?int(0xFFFFFFFF):int(0x0);
// 3
PV1i.x = floatBitsToInt(intBitsToFloat(PV0i.z) + intBitsToFloat(PV0i.x));
R126i.y = floatBitsToInt(min(intBitsToFloat(PV0i.z), intBitsToFloat(PV0i.x)));
PV1i.z = ((intBitsToFloat(PV0i.x) > intBitsToFloat(PV0i.z))?int(0xFFFFFFFF):int(0x0));
// 4
backupReg0i = R127i.y;
R127i.y = ((PV1i.z == 0)?(R126i.w):(backupReg0i));
PS0i = floatBitsToInt(1.0 / intBitsToFloat(PV1i.x));
// 5
PV1i.z = floatBitsToInt(intBitsToFloat(R126i.y) * intBitsToFloat(PS0i));
// 6
PV0i.y = floatBitsToInt(-(intBitsToFloat(PV1i.z)) + 0.5);
// 7
R123i.x = ((R127i.y == 0)?(0):(PV0i.y));
PV1i.x = R123i.x;
// 8
PV0i.w = floatBitsToInt(max(intBitsToFloat(R3i.z), intBitsToFloat(PV1i.x)));
// 9
PV1i.z = floatBitsToInt(mul_nonIEEE(intBitsToFloat(R6i.x), intBitsToFloat(PV0i.w)));
// 10
backupReg0i = R0i.x;
backupReg1i = R0i.y;
R0i.x = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R4i.z),intBitsToFloat(PV1i.z)) + intBitsToFloat(backupReg0i)));
R0i.y = floatBitsToInt((mul_nonIEEE(intBitsToFloat(R5i.w),intBitsToFloat(PV1i.z)) + intBitsToFloat(backupReg1i)));
}
if( activeMaskStackC[2] == true ) {
R3i.xyzw = floatBitsToInt(texture(textureUnitPS0, intBitsToFloat(R0i.xy)).xyzw);
}
activeMaskStackC[1] = activeMaskStack[0] == true && activeMaskStackC[0] == true;
// export
passPixelColor0 = vec4(intBitsToFloat(R3i.x), intBitsToFloat(R3i.y), intBitsToFloat(R3i.z), intBitsToFloat(R3i.w));
}
2018-12-03 06:05:15 +01:00
2018-10-31 11:03:04 +01:00
#endif
2018-12-03 08:10:30 +01:00
#if (preset == 2) // FXAA
2019-01-06 23:46:59 +01:00
//-----------------------------settings-------------------------------------//
2018-10-31 11:03:04 +01:00
2018-12-03 06:05:15 +01:00
#define Subpix $subPix //[0.000 to 1.000] Choose the amount of sub-pixel aliasing removal.
#define EdgeThreshold $edgeThreshold //[0.000 to 1.000] Edge detection threshold. The minimum amount of local contrast required to apply algorithm.
#define EdgeThresholdMin $edgeThresholdMin //[0.000 to 1.000] Darkness threshold. Trims the algorithm from processing darks.
2018-10-31 11:03:04 +01:00
2019-01-06 23:46:59 +01:00
//--------------------------------------------------------------------------//
2018-10-31 11:03:04 +01:00
Update BotW packs for Vulkan (#411)
But now done properly! Basically, a bunch of improvements were made to the script. The previous attempt at this conversion was quickly followed by a rollback since I realized that the script was overlooking certain things that made most of the packs hit or miss whether they would work. A few things missing were:
- It only tested the values from 1 preset. Now, each shader gets compiled per each preset, like what Cemu would do. It also merges the changes done for each preset into one. This should solve cases where one shader would define things separately or repeatedly from preset to preset.
- All* of the shaders are tested to see if they use the converter used the right values for the locations for Vulkan.
Both of these *should* mean that they should both compile and be linkable in Vulkan, which means that I don't have to test each individual shader to see if they work. I will release the two scripts (one used for converting, one used for checking the right values for the locations) tomorrow so that other people might be able to help, if they want. It's fairly straightforward now at least.
* Organize workaround graphic packs
Pretty hard to organize these correctly, but according to our discord discussion, this was the best layout from a bunch I proposed, together with some suggestions.
* Add V4 converter script and instructions on how to use it
Now everyone BotW is done and all of the bugs have been kinked out using it (hopefully...), here's the release of the converter script in all of it's very badly coded glory. I hope I didn't leave too much debug glory in there...
Also, I hope that I didn't make too many grammatical mistakes in the instructions, but hopefully it's easy enough to follow.
2019-12-28 05:55:52 +01:00
TEXTURE_LAYOUT(0, 1, 0) uniform sampler2D textureUnitPS0;
TEXTURE_LAYOUT(1, 1, 1) uniform sampler2D textureUnitPS1;
2018-10-31 11:03:04 +01:00
layout(location = 0) in vec4 passParameterSem2;
layout(location = 0) out vec4 passPixelColor0;
Update BotW packs for Vulkan (#411)
But now done properly! Basically, a bunch of improvements were made to the script. The previous attempt at this conversion was quickly followed by a rollback since I realized that the script was overlooking certain things that made most of the packs hit or miss whether they would work. A few things missing were:
- It only tested the values from 1 preset. Now, each shader gets compiled per each preset, like what Cemu would do. It also merges the changes done for each preset into one. This should solve cases where one shader would define things separately or repeatedly from preset to preset.
- All* of the shaders are tested to see if they use the converter used the right values for the locations for Vulkan.
Both of these *should* mean that they should both compile and be linkable in Vulkan, which means that I don't have to test each individual shader to see if they work. I will release the two scripts (one used for converting, one used for checking the right values for the locations) tomorrow so that other people might be able to help, if they want. It's fairly straightforward now at least.
* Organize workaround graphic packs
Pretty hard to organize these correctly, but according to our discord discussion, this was the best layout from a bunch I proposed, together with some suggestions.
* Add V4 converter script and instructions on how to use it
Now everyone BotW is done and all of the bugs have been kinked out using it (hopefully...), here's the release of the converter script in all of it's very badly coded glory. I hope I didn't leave too much debug glory in there...
Also, I hope that I didn't make too many grammatical mistakes in the instructions, but hopefully it's easy enough to follow.
2019-12-28 05:55:52 +01:00
2018-10-31 11:03:04 +01:00
2019-01-06 23:46:59 +01:00
#define FXAA_QUALITY_PS 12
#define FXAA_QUALITY_P0 1.0
#define FXAA_QUALITY_P1 1.0
#define FXAA_QUALITY_P2 1.0
#define FXAA_QUALITY_P3 1.0
#define FXAA_QUALITY_P4 1.0
#define FXAA_QUALITY_P5 1.5
#define FXAA_QUALITY_P6 2.0
#define FXAA_QUALITY_P7 2.0
#define FXAA_QUALITY_P8 2.0
#define FXAA_QUALITY_P9 2.0
#define FXAA_QUALITY_P10 4.0
#define FXAA_QUALITY_P11 8.0
2018-10-31 11:03:04 +01:00
#define FxaaBool bool
#define FxaaDiscard discard
#define FxaaFloat float
#define FxaaFloat2 vec2
#define FxaaFloat3 vec3
#define FxaaFloat4 vec4
#define FxaaHalf float
#define FxaaHalf2 vec2
#define FxaaHalf3 vec3
#define FxaaHalf4 vec4
#define FxaaInt2 ivec2
#define FxaaSat(x) clamp(x, 0.0, 1.0)
#define FxaaTex sampler2D
#define FxaaTexTop(t, p) textureLod(t, p, 0.0)
#define FxaaTexOff(t, p, o, r) textureLodOffset(t, p, 0.0, o)
#define FxaaTexAlpha4(t, p) textureGather(t, p, 3)
#define FxaaTexOffAlpha4(t, p, o) textureGatherOffset(t, p, o, 3)
#define FxaaTexGreen4(t, p) textureGather(t, p, 1)
#define FxaaTexOffGreen4(t, p, o) textureGatherOffset(t, p, o, 1)
FxaaFloat4 FxaaPixelShader(
FxaaFloat2 pos,
FxaaTex tex,
FxaaTex lum,
FxaaFloat2 fxaaQualityRcpFrame,
FxaaFloat fxaaQualitySubpix,
FxaaFloat fxaaQualityEdgeThreshold,
FxaaFloat fxaaQualityEdgeThresholdMin
) {
FxaaFloat2 posM;
posM.x = pos.x;
posM.y = pos.y;
FxaaFloat4 rgbyM = vec4(FxaaTexTop(tex, posM).xyz, FxaaTexTop(lum, posM).x);
#define lumaM rgbyM.w
FxaaFloat4 luma4A = textureGather(lum, posM);
FxaaFloat4 luma4B = textureGatherOffset(lum, posM, FxaaInt2(-1, -1));
#define lumaE luma4A.z
#define lumaS luma4A.x
#define lumaSE luma4A.y
#define lumaNW luma4B.w
#define lumaN luma4B.z
#define lumaW luma4B.x
FxaaFloat maxSM = max(lumaS, lumaM);
FxaaFloat minSM = min(lumaS, lumaM);
FxaaFloat maxESM = max(lumaE, maxSM);
FxaaFloat minESM = min(lumaE, minSM);
FxaaFloat maxWN = max(lumaN, lumaW);
FxaaFloat minWN = min(lumaN, lumaW);
FxaaFloat rangeMax = max(maxWN, maxESM);
FxaaFloat rangeMin = min(minWN, minESM);
FxaaFloat rangeMaxScaled = rangeMax * fxaaQualityEdgeThreshold;
FxaaFloat range = rangeMax - rangeMin;
FxaaFloat rangeMaxClamped = max(fxaaQualityEdgeThresholdMin, rangeMaxScaled);
FxaaBool earlyExit = range < rangeMaxClamped;
if(earlyExit)
return rgbyM;
FxaaFloat lumaNE = FxaaTexOff(lum, posM, FxaaInt2(1, -1), fxaaQualityRcpFrame.xy).x;
FxaaFloat lumaSW = FxaaTexOff(lum, posM, FxaaInt2(-1, 1), fxaaQualityRcpFrame.xy).x;
FxaaFloat lumaNS = lumaN + lumaS;
FxaaFloat lumaWE = lumaW + lumaE;
FxaaFloat subpixRcpRange = 1.0/range;
FxaaFloat subpixNSWE = lumaNS + lumaWE;
FxaaFloat edgeHorz1 = (-2.0 * lumaM) + lumaNS;
FxaaFloat edgeVert1 = (-2.0 * lumaM) + lumaWE;
FxaaFloat lumaNESE = lumaNE + lumaSE;
FxaaFloat lumaNWNE = lumaNW + lumaNE;
FxaaFloat edgeHorz2 = (-2.0 * lumaE) + lumaNESE;
FxaaFloat edgeVert2 = (-2.0 * lumaN) + lumaNWNE;
FxaaFloat lumaNWSW = lumaNW + lumaSW;
FxaaFloat lumaSWSE = lumaSW + lumaSE;
FxaaFloat edgeHorz4 = (abs(edgeHorz1) * 2.0) + abs(edgeHorz2);
FxaaFloat edgeVert4 = (abs(edgeVert1) * 2.0) + abs(edgeVert2);
FxaaFloat edgeHorz3 = (-2.0 * lumaW) + lumaNWSW;
FxaaFloat edgeVert3 = (-2.0 * lumaS) + lumaSWSE;
FxaaFloat edgeHorz = abs(edgeHorz3) + edgeHorz4;
FxaaFloat edgeVert = abs(edgeVert3) + edgeVert4;
FxaaFloat subpixNWSWNESE = lumaNWSW + lumaNESE;
FxaaFloat lengthSign = fxaaQualityRcpFrame.x;
FxaaBool horzSpan = edgeHorz >= edgeVert;
FxaaFloat subpixA = subpixNSWE * 2.0 + subpixNWSWNESE;
if(!horzSpan) lumaN = lumaW;
if(!horzSpan) lumaS = lumaE;
if(horzSpan) lengthSign = fxaaQualityRcpFrame.y;
FxaaFloat subpixB = (subpixA * (1.0/12.0)) - lumaM;
FxaaFloat gradientN = lumaN - lumaM;
FxaaFloat gradientS = lumaS - lumaM;
FxaaFloat lumaNN = lumaN + lumaM;
FxaaFloat lumaSS = lumaS + lumaM;
FxaaBool pairN = abs(gradientN) >= abs(gradientS);
FxaaFloat gradient = max(abs(gradientN), abs(gradientS));
if(pairN) lengthSign = -lengthSign;
FxaaFloat subpixC = FxaaSat(abs(subpixB) * subpixRcpRange);
FxaaFloat2 posB;
posB.x = posM.x;
posB.y = posM.y;
FxaaFloat2 offNP;
offNP.x = (!horzSpan) ? 0.0 : fxaaQualityRcpFrame.x;
offNP.y = ( horzSpan) ? 0.0 : fxaaQualityRcpFrame.y;
if(!horzSpan) posB.x += lengthSign * 0.5;
if( horzSpan) posB.y += lengthSign * 0.5;
FxaaFloat2 posN;
2019-01-06 23:46:59 +01:00
posN.x = posB.x - offNP.x * FXAA_QUALITY_P0;
posN.y = posB.y - offNP.y * FXAA_QUALITY_P0;
2018-10-31 11:03:04 +01:00
FxaaFloat2 posP;
2019-01-06 23:46:59 +01:00
posP.x = posB.x + offNP.x * FXAA_QUALITY_P0;
posP.y = posB.y + offNP.y * FXAA_QUALITY_P0;
2018-10-31 11:03:04 +01:00
FxaaFloat subpixD = ((-2.0)*subpixC) + 3.0;
FxaaFloat lumaEndN = FxaaTexTop(lum, posN).x;
FxaaFloat subpixE = subpixC * subpixC;
FxaaFloat lumaEndP = FxaaTexTop(lum, posP).x;
if(!pairN) lumaNN = lumaSS;
FxaaFloat gradientScaled = gradient * 1.0/4.0;
FxaaFloat lumaMM = lumaM - lumaNN * 0.5;
FxaaFloat subpixF = subpixD * subpixE;
FxaaBool lumaMLTZero = lumaMM < 0.0;
lumaEndN -= lumaNN * 0.5;
lumaEndP -= lumaNN * 0.5;
FxaaBool doneN = abs(lumaEndN) >= gradientScaled;
FxaaBool doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P1;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P1;
2018-10-31 11:03:04 +01:00
FxaaBool doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P1;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P1;
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P2;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P2;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P2;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P2;
#if (FXAA_QUALITY_PS > 3)
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P3;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P3;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P3;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P3;
#if (FXAA_QUALITY_PS > 4)
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P4;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P4;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P4;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P4;
#if (FXAA_QUALITY_PS > 5)
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P5;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P5;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P5;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P5;
#if (FXAA_QUALITY_PS > 6)
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P6;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P6;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P6;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P6;
#if (FXAA_QUALITY_PS > 7)
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P7;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P7;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P7;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P7;
#if (FXAA_QUALITY_PS > 8)
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P8;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P8;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P8;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P8;
#if (FXAA_QUALITY_PS > 9)
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P9;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P9;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P9;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P9;
#if (FXAA_QUALITY_PS > 10)
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P10;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P10;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P10;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P10;
#if (FXAA_QUALITY_PS > 11)
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P11;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P11;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P11;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P11;
#if (FXAA_QUALITY_PS > 12)
2018-10-31 11:03:04 +01:00
if(doneNP) {
if(!doneN) lumaEndN = FxaaTexTop(lum, posN.xy).x;
if(!doneP) lumaEndP = FxaaTexTop(lum, posP.xy).x;
if(!doneN) lumaEndN = lumaEndN - lumaNN * 0.5;
if(!doneP) lumaEndP = lumaEndP - lumaNN * 0.5;
doneN = abs(lumaEndN) >= gradientScaled;
doneP = abs(lumaEndP) >= gradientScaled;
2019-01-06 23:46:59 +01:00
if(!doneN) posN.x -= offNP.x * FXAA_QUALITY_P12;
if(!doneN) posN.y -= offNP.y * FXAA_QUALITY_P12;
2018-10-31 11:03:04 +01:00
doneNP = (!doneN) || (!doneP);
2019-01-06 23:46:59 +01:00
if(!doneP) posP.x += offNP.x * FXAA_QUALITY_P12;
if(!doneP) posP.y += offNP.y * FXAA_QUALITY_P12;
2018-10-31 11:03:04 +01:00
}
#endif
}
#endif
}
#endif
}
#endif
}
#endif
}
#endif
}
#endif
}
#endif
}
#endif
}
#endif
}
FxaaFloat dstN = posM.x - posN.x;
FxaaFloat dstP = posP.x - posM.x;
if(!horzSpan) dstN = posM.y - posN.y;
if(!horzSpan) dstP = posP.y - posM.y;
FxaaBool goodSpanN = (lumaEndN < 0.0) != lumaMLTZero;
FxaaFloat spanLength = (dstP + dstN);
FxaaBool goodSpanP = (lumaEndP < 0.0) != lumaMLTZero;
FxaaFloat spanLengthRcp = 1.0/spanLength;
FxaaBool directionN = dstN < dstP;
FxaaFloat dst = min(dstN, dstP);
FxaaBool goodSpan = directionN ? goodSpanN : goodSpanP;
FxaaFloat subpixG = subpixF * subpixF;
FxaaFloat pixelOffset = (dst * (-spanLengthRcp)) + 0.5;
FxaaFloat subpixH = subpixG * fxaaQualitySubpix;
FxaaFloat pixelOffsetGood = goodSpan ? pixelOffset : 0.0;
FxaaFloat pixelOffsetSubpix = max(pixelOffsetGood, subpixH);
if(!horzSpan) posM.x += pixelOffsetSubpix * lengthSign;
if( horzSpan) posM.y += pixelOffsetSubpix * lengthSign;
return FxaaFloat4(FxaaTexTop(tex, posM).xyz, lumaM);
}
2019-01-06 23:46:59 +01:00
ivec2 resolution = textureSize(textureUnitPS0,0);
2018-12-12 02:07:26 +01:00
vec2 RcpFrame = vec2(1.0 / float(resolution.x), 1.0 / float(resolution.y));
2018-10-31 11:03:04 +01:00
void main()
{
passPixelColor0 = FxaaPixelShader(passParameterSem2.xy, textureUnitPS0, textureUnitPS1, RcpFrame, Subpix, EdgeThreshold, EdgeThresholdMin);
}
2018-12-01 04:19:14 +01:00
#endif
