GS:OGL:Shaders: Formatting

This commit is contained in:
TellowKrinkle 2023-04-12 22:23:31 -05:00 committed by lightningterror
parent dfbdaa651c
commit e221d31b45
7 changed files with 787 additions and 787 deletions

View File

@ -30,7 +30,7 @@ layout(binding=0, rgba8) uniform writeonly image2D imgDst;
AF3 CasLoad(ASU2 p)
{
return texelFetch(imgSrc, srcOffset + ivec2(p), 0).rgb;
return texelFetch(imgSrc, srcOffset + ivec2(p), 0).rgb;
}
// Lets you transform input from the load into a linear color space between 0 and 1. See ffx_cas.h
@ -42,23 +42,23 @@ void CasInput(inout AF1 r, inout AF1 g, inout AF1 b) {}
layout(local_size_x=64) in;
void main()
{
// Do remapping of local xy in workgroup for a more PS-like swizzle pattern.
AU2 gxy = ARmp8x8(gl_LocalInvocationID.x)+AU2(gl_WorkGroupID.x<<4u,gl_WorkGroupID.y<<4u);
// Do remapping of local xy in workgroup for a more PS-like swizzle pattern.
AU2 gxy = ARmp8x8(gl_LocalInvocationID.x)+AU2(gl_WorkGroupID.x<<4u,gl_WorkGroupID.y<<4u);
// Filter.
AF4 c;
CasFilter(c.r, c.g, c.b, gxy, const0, const1, CAS_SHARPEN_ONLY);
imageStore(imgDst, ASU2(gxy), c);
gxy.x += 8u;
// Filter.
AF4 c;
CasFilter(c.r, c.g, c.b, gxy, const0, const1, CAS_SHARPEN_ONLY);
imageStore(imgDst, ASU2(gxy), c);
gxy.x += 8u;
CasFilter(c.r, c.g, c.b, gxy, const0, const1, CAS_SHARPEN_ONLY);
imageStore(imgDst, ASU2(gxy), c);
gxy.y += 8u;
CasFilter(c.r, c.g, c.b, gxy, const0, const1, CAS_SHARPEN_ONLY);
imageStore(imgDst, ASU2(gxy), c);
gxy.y += 8u;
CasFilter(c.r, c.g, c.b, gxy, const0, const1, CAS_SHARPEN_ONLY);
imageStore(imgDst, ASU2(gxy), c);
gxy.x -= 8u;
CasFilter(c.r, c.g, c.b, gxy, const0, const1, CAS_SHARPEN_ONLY);
imageStore(imgDst, ASU2(gxy), c);
gxy.x -= 8u;
CasFilter(c.r, c.g, c.b, gxy, const0, const1, CAS_SHARPEN_ONLY);
imageStore(imgDst, ASU2(gxy), c);
CasFilter(c.r, c.g, c.b, gxy, const0, const1, CAS_SHARPEN_ONLY);
imageStore(imgDst, ASU2(gxy), c);
}

View File

@ -21,10 +21,10 @@ out vec4 PSin_c;
void vs_main()
{
PSin_p = vec4(POSITION, 0.5f, 1.0f);
PSin_t = TEXCOORD0;
PSin_c = COLOR;
gl_Position = vec4(POSITION, 0.5f, 1.0f); // NOTE I don't know if it is possible to merge POSITION_OUT and gl_Position
PSin_p = vec4(POSITION, 0.5f, 1.0f);
PSin_t = TEXCOORD0;
PSin_c = COLOR;
gl_Position = vec4(POSITION, 0.5f, 1.0f); // NOTE I don't know if it is possible to merge POSITION_OUT and gl_Position
}
#endif
@ -46,13 +46,13 @@ layout(location = 0) out vec4 SV_Target0;
vec4 sample_c()
{
return texture(TextureSampler, PSin_t);
return texture(TextureSampler, PSin_t);
}
#ifdef ps_copy
void ps_copy()
{
SV_Target0 = sample_c();
SV_Target0 = sample_c();
}
#endif
@ -67,20 +67,20 @@ void ps_depth_copy()
// Need to be careful with precision here, it can break games like Spider-Man 3 and Dogs Life
void ps_convert_rgba8_16bits()
{
highp uvec4 i = uvec4(sample_c() * vec4(255.5f, 255.5f, 255.5f, 255.5f));
highp uvec4 i = uvec4(sample_c() * vec4(255.5f, 255.5f, 255.5f, 255.5f));
SV_Target1 = ((i.x & 0x00F8u) >> 3) | ((i.y & 0x00F8u) << 2) | ((i.z & 0x00f8u) << 7) | ((i.w & 0x80u) << 8);
SV_Target1 = ((i.x & 0x00F8u) >> 3) | ((i.y & 0x00F8u) << 2) | ((i.z & 0x00f8u) << 7) | ((i.w & 0x80u) << 8);
}
#endif
#ifdef ps_convert_float32_32bits
void ps_convert_float32_32bits()
{
// Convert a GL_FLOAT32 depth texture into a 32 bits UINT texture
// Convert a GL_FLOAT32 depth texture into a 32 bits UINT texture
#if HAS_CLIP_CONTROL
SV_Target1 = uint(exp2(32.0f) * sample_c().r);
SV_Target1 = uint(exp2(32.0f) * sample_c().r);
#else
SV_Target1 = uint(exp2(24.0f) * sample_c().r);
SV_Target1 = uint(exp2(24.0f) * sample_c().r);
#endif
}
#endif
@ -88,150 +88,150 @@ void ps_convert_float32_32bits()
#ifdef ps_convert_float32_rgba8
void ps_convert_float32_rgba8()
{
// Convert a GL_FLOAT32 depth texture into a RGBA color texture
// Convert a GL_FLOAT32 depth texture into a RGBA color texture
#if HAS_CLIP_CONTROL
uint d = uint(sample_c().r * exp2(32.0f));
uint d = uint(sample_c().r * exp2(32.0f));
#else
uint d = uint(sample_c().r * exp2(24.0f));
uint d = uint(sample_c().r * exp2(24.0f));
#endif
SV_Target0 = vec4(uvec4((d & 0xFFu), ((d >> 8) & 0xFFu), ((d >> 16) & 0xFFu), (d >> 24))) / vec4(255.0);
SV_Target0 = vec4(uvec4((d & 0xFFu), ((d >> 8) & 0xFFu), ((d >> 16) & 0xFFu), (d >> 24))) / vec4(255.0);
}
#endif
#ifdef ps_convert_float16_rgb5a1
void ps_convert_float16_rgb5a1()
{
// Convert a GL_FLOAT32 (only 16 lsb) depth into a RGB5A1 color texture
// Convert a GL_FLOAT32 (only 16 lsb) depth into a RGB5A1 color texture
#if HAS_CLIP_CONTROL
uint d = uint(sample_c().r * exp2(32.0f));
uint d = uint(sample_c().r * exp2(32.0f));
#else
uint d = uint(sample_c().r * exp2(24.0f));
uint d = uint(sample_c().r * exp2(24.0f));
#endif
SV_Target0 = vec4(uvec4((d & 0x1Fu), ((d >> 5) & 0x1Fu), ((d >> 10) & 0x1Fu), (d >> 15) & 0x01u)) / vec4(32.0f, 32.0f, 32.0f, 1.0f);
SV_Target0 = vec4(uvec4((d & 0x1Fu), ((d >> 5) & 0x1Fu), ((d >> 10) & 0x1Fu), (d >> 15) & 0x01u)) / vec4(32.0f, 32.0f, 32.0f, 1.0f);
}
#endif
float rgba8_to_depth32(vec4 unorm)
{
uvec4 c = uvec4(unorm * vec4(255.5f));
uvec4 c = uvec4(unorm * vec4(255.5f));
#if HAS_CLIP_CONTROL
return float(c.r | (c.g << 8) | (c.b << 16) | (c.a << 24)) * exp2(-32.0f);
return float(c.r | (c.g << 8) | (c.b << 16) | (c.a << 24)) * exp2(-32.0f);
#else
return float(c.r | (c.g << 8) | (c.b << 16) | (c.a << 24)) * exp2(-24.0f);
return float(c.r | (c.g << 8) | (c.b << 16) | (c.a << 24)) * exp2(-24.0f);
#endif
}
float rgba8_to_depth24(vec4 unorm)
{
uvec3 c = uvec3(unorm.rgb * vec3(255.5f));
uvec3 c = uvec3(unorm.rgb * vec3(255.5f));
#if HAS_CLIP_CONTROL
return float(c.r | (c.g << 8) | (c.b << 16)) * exp2(-32.0f);
return float(c.r | (c.g << 8) | (c.b << 16)) * exp2(-32.0f);
#else
return float(c.r | (c.g << 8) | (c.b << 16)) * exp2(-24.0f);
return float(c.r | (c.g << 8) | (c.b << 16)) * exp2(-24.0f);
#endif
}
float rgba8_to_depth16(vec4 unorm)
{
uvec2 c = uvec2(unorm.rg * vec2(255.5f));
uvec2 c = uvec2(unorm.rg * vec2(255.5f));
#if HAS_CLIP_CONTROL
return float(c.r | (c.g << 8)) * exp2(-32.0f);
return float(c.r | (c.g << 8)) * exp2(-32.0f);
#else
return float(c.r | (c.g << 8)) * exp2(-24.0f);
return float(c.r | (c.g << 8)) * exp2(-24.0f);
#endif
}
float rgb5a1_to_depth16(vec4 unorm)
{
uvec4 c = uvec4(unorm * vec4(255.5f));
uvec4 c = uvec4(unorm * vec4(255.5f));
#if HAS_CLIP_CONTROL
return float(((c.r & 0xF8u) >> 3) | ((c.g & 0xF8u) << 2) | ((c.b & 0xF8u) << 7) | ((c.a & 0x80u) << 8)) * exp2(-32.0f);
return float(((c.r & 0xF8u) >> 3) | ((c.g & 0xF8u) << 2) | ((c.b & 0xF8u) << 7) | ((c.a & 0x80u) << 8)) * exp2(-32.0f);
#else
return float(((c.r & 0xF8u) >> 3) | ((c.g & 0xF8u) << 2) | ((c.b & 0xF8u) << 7) | ((c.a & 0x80u) << 8)) * exp2(-24.0f);
return float(((c.r & 0xF8u) >> 3) | ((c.g & 0xF8u) << 2) | ((c.b & 0xF8u) << 7) | ((c.a & 0x80u) << 8)) * exp2(-24.0f);
#endif
}
#ifdef ps_convert_rgba8_float32
void ps_convert_rgba8_float32()
{
// Convert an RGBA texture into a float depth texture
gl_FragDepth = rgba8_to_depth32(sample_c());
// Convert an RGBA texture into a float depth texture
gl_FragDepth = rgba8_to_depth32(sample_c());
}
#endif
#ifdef ps_convert_rgba8_float24
void ps_convert_rgba8_float24()
{
// Same as above but without the alpha channel (24 bits Z)
// Same as above but without the alpha channel (24 bits Z)
// Convert an RGBA texture into a float depth texture
gl_FragDepth = rgba8_to_depth24(sample_c());
// Convert an RGBA texture into a float depth texture
gl_FragDepth = rgba8_to_depth24(sample_c());
}
#endif
#ifdef ps_convert_rgba8_float16
void ps_convert_rgba8_float16()
{
// Same as above but without the A/B channels (16 bits Z)
// Same as above but without the A/B channels (16 bits Z)
// Convert an RGBA texture into a float depth texture
gl_FragDepth = rgba8_to_depth16(sample_c());
// Convert an RGBA texture into a float depth texture
gl_FragDepth = rgba8_to_depth16(sample_c());
}
#endif
#ifdef ps_convert_rgb5a1_float16
void ps_convert_rgb5a1_float16()
{
// Convert an RGB5A1 (saved as RGBA8) color to a 16 bit Z
gl_FragDepth = rgb5a1_to_depth16(sample_c());
// Convert an RGB5A1 (saved as RGBA8) color to a 16 bit Z
gl_FragDepth = rgb5a1_to_depth16(sample_c());
}
#endif
#define SAMPLE_RGBA_DEPTH_BILN(CONVERT_FN) \
ivec2 dims = textureSize(TextureSampler, 0); \
vec2 top_left_f = PSin_t * vec2(dims) - 0.5f; \
ivec2 top_left = ivec2(floor(top_left_f)); \
ivec4 coords = clamp(ivec4(top_left, top_left + 1), ivec4(0), dims.xyxy - 1); \
vec2 mix_vals = fract(top_left_f); \
float depthTL = CONVERT_FN(texelFetch(TextureSampler, coords.xy, 0)); \
float depthTR = CONVERT_FN(texelFetch(TextureSampler, coords.zy, 0)); \
float depthBL = CONVERT_FN(texelFetch(TextureSampler, coords.xw, 0)); \
float depthBR = CONVERT_FN(texelFetch(TextureSampler, coords.zw, 0)); \
gl_FragDepth = mix(mix(depthTL, depthTR, mix_vals.x), mix(depthBL, depthBR, mix_vals.x), mix_vals.y);
ivec2 dims = textureSize(TextureSampler, 0); \
vec2 top_left_f = PSin_t * vec2(dims) - 0.5f; \
ivec2 top_left = ivec2(floor(top_left_f)); \
ivec4 coords = clamp(ivec4(top_left, top_left + 1), ivec4(0), dims.xyxy - 1); \
vec2 mix_vals = fract(top_left_f); \
float depthTL = CONVERT_FN(texelFetch(TextureSampler, coords.xy, 0)); \
float depthTR = CONVERT_FN(texelFetch(TextureSampler, coords.zy, 0)); \
float depthBL = CONVERT_FN(texelFetch(TextureSampler, coords.xw, 0)); \
float depthBR = CONVERT_FN(texelFetch(TextureSampler, coords.zw, 0)); \
gl_FragDepth = mix(mix(depthTL, depthTR, mix_vals.x), mix(depthBL, depthBR, mix_vals.x), mix_vals.y);
#ifdef ps_convert_rgba8_float32_biln
void ps_convert_rgba8_float32_biln()
{
// Convert an RGBA texture into a float depth texture
SAMPLE_RGBA_DEPTH_BILN(rgba8_to_depth32);
// Convert an RGBA texture into a float depth texture
SAMPLE_RGBA_DEPTH_BILN(rgba8_to_depth32);
}
#endif
#ifdef ps_convert_rgba8_float24_biln
void ps_convert_rgba8_float24_biln()
{
// Same as above but without the alpha channel (24 bits Z)
// Same as above but without the alpha channel (24 bits Z)
// Convert an RGBA texture into a float depth texture
SAMPLE_RGBA_DEPTH_BILN(rgba8_to_depth24);
// Convert an RGBA texture into a float depth texture
SAMPLE_RGBA_DEPTH_BILN(rgba8_to_depth24);
}
#endif
#ifdef ps_convert_rgba8_float16_biln
void ps_convert_rgba8_float16_biln()
{
// Same as above but without the A/B channels (16 bits Z)
// Same as above but without the A/B channels (16 bits Z)
// Convert an RGBA texture into a float depth texture
SAMPLE_RGBA_DEPTH_BILN(rgba8_to_depth16);
// Convert an RGBA texture into a float depth texture
SAMPLE_RGBA_DEPTH_BILN(rgba8_to_depth16);
}
#endif
#ifdef ps_convert_rgb5a1_float16_biln
void ps_convert_rgb5a1_float16_biln()
{
// Convert an RGB5A1 (saved as RGBA8) color to a 16 bit Z
SAMPLE_RGBA_DEPTH_BILN(rgb5a1_to_depth16);
// Convert an RGB5A1 (saved as RGBA8) color to a 16 bit Z
SAMPLE_RGBA_DEPTH_BILN(rgb5a1_to_depth16);
}
#endif
@ -242,51 +242,51 @@ uniform float ScaleFactor;
void ps_convert_rgba_8i()
{
// Convert a RGBA texture into a 8 bits packed texture
// Input column: 8x2 RGBA pixels
// 0: 8 RGBA
// 1: 8 RGBA
// Output column: 16x4 Index pixels
// 0: 8 R | 8 B
// 1: 8 R | 8 B
// 2: 8 G | 8 A
// 3: 8 G | 8 A
uvec2 pos = uvec2(gl_FragCoord.xy);
// Convert a RGBA texture into a 8 bits packed texture
// Input column: 8x2 RGBA pixels
// 0: 8 RGBA
// 1: 8 RGBA
// Output column: 16x4 Index pixels
// 0: 8 R | 8 B
// 1: 8 R | 8 B
// 2: 8 G | 8 A
// 3: 8 G | 8 A
uvec2 pos = uvec2(gl_FragCoord.xy);
// Collapse separate R G B A areas into their base pixel
uvec2 block = (pos & ~uvec2(15u, 3u)) >> 1;
uvec2 subblock = pos & uvec2(7u, 1u);
uvec2 coord = block | subblock;
// Collapse separate R G B A areas into their base pixel
uvec2 block = (pos & ~uvec2(15u, 3u)) >> 1;
uvec2 subblock = pos & uvec2(7u, 1u);
uvec2 coord = block | subblock;
// Compensate for potentially differing page pitch.
// Compensate for potentially differing page pitch.
uvec2 block_xy = coord / uvec2(64u, 32u);
uint block_num = (block_xy.y * (DBW / 128u)) + block_xy.x;
uvec2 block_offset = uvec2((block_num % (SBW / 64u)) * 64u, (block_num / (SBW / 64u)) * 32u);
coord = (coord % uvec2(64u, 32u)) + block_offset;
// Apply offset to cols 1 and 2
uint is_col23 = pos.y & 4u;
uint is_col13 = pos.y & 2u;
uint is_col12 = is_col23 ^ (is_col13 << 1);
coord.x ^= is_col12; // If cols 1 or 2, flip bit 3 of x
// Apply offset to cols 1 and 2
uint is_col23 = pos.y & 4u;
uint is_col13 = pos.y & 2u;
uint is_col12 = is_col23 ^ (is_col13 << 1);
coord.x ^= is_col12; // If cols 1 or 2, flip bit 3 of x
if (floor(ScaleFactor) != ScaleFactor)
coord = uvec2(vec2(coord) * ScaleFactor);
else
coord *= uvec2(ScaleFactor);
if (floor(ScaleFactor) != ScaleFactor)
coord = uvec2(vec2(coord) * ScaleFactor);
else
coord *= uvec2(ScaleFactor);
vec4 pixel = texelFetch(TextureSampler, ivec2(coord), 0);
vec2 sel0 = (pos.y & 2u) == 0u ? pixel.rb : pixel.ga;
float sel1 = (pos.x & 8u) == 0u ? sel0.x : sel0.y;
SV_Target0 = vec4(sel1);
vec4 pixel = texelFetch(TextureSampler, ivec2(coord), 0);
vec2 sel0 = (pos.y & 2u) == 0u ? pixel.rb : pixel.ga;
float sel1 = (pos.x & 8u) == 0u ? sel0.x : sel0.y;
SV_Target0 = vec4(sel1);
}
#endif
#ifdef ps_filter_transparency
void ps_filter_transparency()
{
vec4 c = sample_c();
SV_Target0 = vec4(c.rgb, 1.0);
vec4 c = sample_c();
SV_Target0 = vec4(c.rgb, 1.0);
}
#endif
@ -295,8 +295,8 @@ void ps_filter_transparency()
#ifdef ps_datm1
void ps_datm1()
{
if(sample_c().a < (127.5f / 255.0f)) // >= 0x80 pass
discard;
if(sample_c().a < (127.5f / 255.0f)) // >= 0x80 pass
discard;
}
#endif
@ -305,24 +305,24 @@ void ps_datm1()
#ifdef ps_datm0
void ps_datm0()
{
if((127.5f / 255.0f) < sample_c().a) // < 0x80 pass (== 0x80 should not pass)
discard;
if((127.5f / 255.0f) < sample_c().a) // < 0x80 pass (== 0x80 should not pass)
discard;
}
#endif
#ifdef ps_hdr_init
void ps_hdr_init()
{
vec4 value = sample_c();
SV_Target0 = vec4(round(value.rgb * 255.0f) / 65535.0f, value.a);
vec4 value = sample_c();
SV_Target0 = vec4(round(value.rgb * 255.0f) / 65535.0f, value.a);
}
#endif
#ifdef ps_hdr_resolve
void ps_hdr_resolve()
{
vec4 value = sample_c();
SV_Target0 = vec4(vec3(uvec3(value.rgb * 65535.0f) & 255u) / 255.0f, value.a);
vec4 value = sample_c();
SV_Target0 = vec4(vec3(uvec3(value.rgb * 65535.0f) & 255u) / 255.0f, value.a);
}
#endif
@ -366,51 +366,51 @@ uniform ivec2 EMOD;
void ps_yuv()
{
vec4 i = sample_c();
vec4 o;
vec4 i = sample_c();
vec4 o;
mat3 rgb2yuv; // Value from GS manual
rgb2yuv[0] = vec3(0.587, -0.311, -0.419);
rgb2yuv[1] = vec3(0.114, 0.500, -0.081);
rgb2yuv[2] = vec3(0.299, -0.169, 0.500);
mat3 rgb2yuv; // Value from GS manual
rgb2yuv[0] = vec3(0.587, -0.311, -0.419);
rgb2yuv[1] = vec3(0.114, 0.500, -0.081);
rgb2yuv[2] = vec3(0.299, -0.169, 0.500);
vec3 yuv = rgb2yuv * i.gbr;
vec3 yuv = rgb2yuv * i.gbr;
float Y = float(0xDB)/255.0f * yuv.x + float(0x10)/255.0f;
float Cr = float(0xE0)/255.0f * yuv.y + float(0x80)/255.0f;
float Cb = float(0xE0)/255.0f * yuv.z + float(0x80)/255.0f;
float Y = float(0xDB)/255.0f * yuv.x + float(0x10)/255.0f;
float Cr = float(0xE0)/255.0f * yuv.y + float(0x80)/255.0f;
float Cb = float(0xE0)/255.0f * yuv.z + float(0x80)/255.0f;
switch(EMOD.x) {
case 0:
o.a = i.a;
break;
case 1:
o.a = Y;
break;
case 2:
o.a = Y/2.0f;
break;
case 3:
o.a = 0.0f;
break;
}
switch(EMOD.x) {
case 0:
o.a = i.a;
break;
case 1:
o.a = Y;
break;
case 2:
o.a = Y/2.0f;
break;
case 3:
o.a = 0.0f;
break;
}
switch(EMOD.y) {
case 0:
o.rgb = i.rgb;
break;
case 1:
o.rgb = vec3(Y);
break;
case 2:
o.rgb = vec3(Y, Cb, Cr);
break;
case 3:
o.rgb = vec3(i.a);
break;
}
switch(EMOD.y) {
case 0:
o.rgb = i.rgb;
break;
case 1:
o.rgb = vec3(Y);
break;
case 2:
o.rgb = vec3(Y, Cb, Cr);
break;
case 3:
o.rgb = vec3(i.a);
break;
}
SV_Target0 = o;
SV_Target0 = o;
}
#endif
@ -418,16 +418,16 @@ void ps_yuv()
void main()
{
SV_Target0 = vec4(0x7FFFFFFF);
SV_Target0 = vec4(0x7FFFFFFF);
#ifdef ps_stencil_image_init_0
if((127.5f / 255.0f) < sample_c().a) // < 0x80 pass (== 0x80 should not pass)
SV_Target0 = vec4(-1);
#endif
#ifdef ps_stencil_image_init_1
if(sample_c().a < (127.5f / 255.0f)) // >= 0x80 pass
SV_Target0 = vec4(-1);
#endif
#ifdef ps_stencil_image_init_0
if((127.5f / 255.0f) < sample_c().a) // < 0x80 pass (== 0x80 should not pass)
SV_Target0 = vec4(-1);
#endif
#ifdef ps_stencil_image_init_1
if(sample_c().a < (127.5f / 255.0f)) // >= 0x80 pass
SV_Target0 = vec4(-1);
#endif
}
#endif

View File

@ -14,17 +14,17 @@ layout(location = 0) out vec4 SV_Target0;
void ps_main0()
{
vec4 c = texture(TextureSampler, PSin_t);
// Note: clamping will be done by fixed unit
c.a *= 2.0f;
SV_Target0 = c;
vec4 c = texture(TextureSampler, PSin_t);
// Note: clamping will be done by fixed unit
c.a *= 2.0f;
SV_Target0 = c;
}
void ps_main1()
{
vec4 c = texture(TextureSampler, PSin_t);
c.a = BGColor.a;
SV_Target0 = c;
vec4 c = texture(TextureSampler, PSin_t);
c.a = BGColor.a;
SV_Target0 = c;
}
#endif

View File

@ -119,10 +119,10 @@ void ps_filter_triangular() // triangular
#ifdef ps_filter_complex
void ps_filter_complex()
{
const float PI = 3.14159265359f;
vec2 texdim = vec2(textureSize(TextureSampler, 0));
float factor = (0.9f - 0.4f * cos(2.0f * PI * PSin_t.y * texdim.y));
vec4 c = factor * texture(TextureSampler, vec2(PSin_t.x, (floor(PSin_t.y * texdim.y) + 0.5f) / texdim.y));
const float PI = 3.14159265359f;
vec2 texdim = vec2(textureSize(TextureSampler, 0));
float factor = (0.9f - 0.4f * cos(2.0f * PI * PSin_t.y * texdim.y));
vec4 c = factor * texture(TextureSampler, vec2(PSin_t.x, (floor(PSin_t.y * texdim.y) + 0.5f) / texdim.y));
SV_Target0 = c;
}

View File

@ -24,33 +24,33 @@ layout(location = 0) out vec4 SV_Target0;
// For all settings: 1.0 = 100% 0.5=50% 1.5 = 150%
vec4 ContrastSaturationBrightness(vec4 color)
{
float brt = params.x;
float con = params.y;
float sat = params.z;
float brt = params.x;
float con = params.y;
float sat = params.z;
// Increase or decrease these values to adjust r, g and b color channels separately
const float AvgLumR = 0.5;
const float AvgLumG = 0.5;
const float AvgLumB = 0.5;
// Increase or decrease these values to adjust r, g and b color channels separately
const float AvgLumR = 0.5;
const float AvgLumG = 0.5;
const float AvgLumB = 0.5;
const vec3 LumCoeff = vec3(0.2125, 0.7154, 0.0721);
const vec3 LumCoeff = vec3(0.2125, 0.7154, 0.0721);
vec3 AvgLumin = vec3(AvgLumR, AvgLumG, AvgLumB);
vec3 brtColor = color.rgb * brt;
float dot_intensity = dot(brtColor, LumCoeff);
vec3 intensity = vec3(dot_intensity, dot_intensity, dot_intensity);
vec3 satColor = mix(intensity, brtColor, sat);
vec3 conColor = mix(AvgLumin, satColor, con);
vec3 AvgLumin = vec3(AvgLumR, AvgLumG, AvgLumB);
vec3 brtColor = color.rgb * brt;
float dot_intensity = dot(brtColor, LumCoeff);
vec3 intensity = vec3(dot_intensity, dot_intensity, dot_intensity);
vec3 satColor = mix(intensity, brtColor, sat);
vec3 conColor = mix(AvgLumin, satColor, con);
color.rgb = conColor;
return color;
color.rgb = conColor;
return color;
}
void ps_main()
{
vec4 c = texture(TextureSampler, PSin_t);
SV_Target0 = ContrastSaturationBrightness(c);
vec4 c = texture(TextureSampler, PSin_t);
SV_Target0 = ContrastSaturationBrightness(c);
}

File diff suppressed because it is too large Load Diff

View File

@ -2,28 +2,28 @@
layout(std140, binding = 1) uniform cb20
{
vec2 VertexScale;
vec2 VertexOffset;
vec2 VertexScale;
vec2 VertexOffset;
vec2 TextureScale;
vec2 TextureOffset;
vec2 TextureScale;
vec2 TextureOffset;
vec2 PointSize;
uint MaxDepth;
uint pad_cb20;
vec2 PointSize;
uint MaxDepth;
uint pad_cb20;
};
#ifdef VERTEX_SHADER
out SHADER
{
vec4 t_float;
vec4 t_int;
#if VS_IIP != 0
vec4 c;
#else
flat vec4 c;
#endif
vec4 t_float;
vec4 t_int;
#if VS_IIP != 0
vec4 c;
#else
flat vec4 c;
#endif
} VSout;
const float exp_min32 = exp2(-32.0f);
@ -40,198 +40,198 @@ layout(location = 7) in vec4 i_f;
void texture_coord()
{
vec2 uv = vec2(i_uv) - TextureOffset;
vec2 st = i_st - TextureOffset;
vec2 uv = vec2(i_uv) - TextureOffset;
vec2 st = i_st - TextureOffset;
// Float coordinate
VSout.t_float.xy = st;
VSout.t_float.w = i_q;
// Float coordinate
VSout.t_float.xy = st;
VSout.t_float.w = i_q;
// Integer coordinate => normalized
VSout.t_int.xy = uv * TextureScale;
// Integer coordinate => normalized
VSout.t_int.xy = uv * TextureScale;
#if VS_FST
// Integer coordinate => integral
VSout.t_int.zw = uv;
// Integer coordinate => integral
VSout.t_int.zw = uv;
#else
// Some games uses float coordinate for post-processing effect
VSout.t_int.zw = st / TextureScale;
// Some games uses float coordinate for post-processing effect
VSout.t_int.zw = st / TextureScale;
#endif
}
void vs_main()
{
// Clamp to max depth, gs doesn't wrap
highp uint z = min(i_z, MaxDepth);
// Clamp to max depth, gs doesn't wrap
highp uint z = min(i_z, MaxDepth);
// pos -= 0.05 (1/320 pixel) helps avoiding rounding problems (integral part of pos is usually 5 digits, 0.05 is about as low as we can go)
// example: ceil(afterseveralvertextransformations(y = 133)) => 134 => line 133 stays empty
// input granularity is 1/16 pixel, anything smaller than that won't step drawing up/left by one pixel
// example: 133.0625 (133 + 1/16) should start from line 134, ceil(133.0625 - 0.05) still above 133
vec4 p;
// pos -= 0.05 (1/320 pixel) helps avoiding rounding problems (integral part of pos is usually 5 digits, 0.05 is about as low as we can go)
// example: ceil(afterseveralvertextransformations(y = 133)) => 134 => line 133 stays empty
// input granularity is 1/16 pixel, anything smaller than that won't step drawing up/left by one pixel
// example: 133.0625 (133 + 1/16) should start from line 134, ceil(133.0625 - 0.05) still above 133
vec4 p;
p.xy = vec2(i_p) - vec2(0.05f, 0.05f);
p.xy = p.xy * VertexScale - VertexOffset;
p.w = 1.0f;
p.xy = vec2(i_p) - vec2(0.05f, 0.05f);
p.xy = p.xy * VertexScale - VertexOffset;
p.w = 1.0f;
#if HAS_CLIP_CONTROL
p.z = float(z) * exp_min32;
p.z = float(z) * exp_min32;
#else
// GLES doesn't support ARB_clip_control, so remap it to -1..1. We also reduce the range from 32 bits
// to 24 bits, which means some games with very large depth ranges will not render correctly. But,
// for most, it's okay, and really, the best we can do.
p.z = min(float(z) * exp2(-23.0f), 2.0f) - 1.0f;
// GLES doesn't support ARB_clip_control, so remap it to -1..1. We also reduce the range from 32 bits
// to 24 bits, which means some games with very large depth ranges will not render correctly. But,
// for most, it's okay, and really, the best we can do.
p.z = min(float(z) * exp2(-23.0f), 2.0f) - 1.0f;
#endif
gl_Position = p;
gl_Position = p;
texture_coord();
texture_coord();
VSout.c = i_c;
VSout.t_float.z = i_f.x; // pack for with texture
VSout.c = i_c;
VSout.t_float.z = i_f.x; // pack for with texture
#if VS_POINT_SIZE
gl_PointSize = PointSize.x;
#endif
#if VS_POINT_SIZE
gl_PointSize = PointSize.x;
#endif
}
#else // VS_EXPAND
struct RawVertex
{
vec2 ST;
uint RGBA;
float Q;
uint XY;
uint Z;
uint UV;
uint FOG;
vec2 ST;
uint RGBA;
float Q;
uint XY;
uint Z;
uint UV;
uint FOG;
};
layout(std140, binding = 2) readonly buffer VertexBuffer {
RawVertex vertex_buffer[];
RawVertex vertex_buffer[];
};
struct ProcessedVertex
{
vec4 p;
vec4 t_float;
vec4 t_int;
vec4 c;
vec4 p;
vec4 t_float;
vec4 t_int;
vec4 c;
};
ProcessedVertex load_vertex(uint index)
{
#if defined(GL_ARB_shader_draw_parameters) && GL_ARB_shader_draw_parameters
RawVertex rvtx = vertex_buffer[index + gl_BaseVertexARB];
RawVertex rvtx = vertex_buffer[index + gl_BaseVertexARB];
#else
RawVertex rvtx = vertex_buffer[index];
RawVertex rvtx = vertex_buffer[index];
#endif
vec2 i_st = rvtx.ST;
vec4 i_c = vec4(uvec4(bitfieldExtract(rvtx.RGBA, 0, 8), bitfieldExtract(rvtx.RGBA, 8, 8),
vec2 i_st = rvtx.ST;
vec4 i_c = vec4(uvec4(bitfieldExtract(rvtx.RGBA, 0, 8), bitfieldExtract(rvtx.RGBA, 8, 8),
bitfieldExtract(rvtx.RGBA, 16, 8), bitfieldExtract(rvtx.RGBA, 24, 8)));
float i_q = rvtx.Q;
uvec2 i_p = uvec2(bitfieldExtract(rvtx.XY, 0, 16), bitfieldExtract(rvtx.XY, 16, 16));
uint i_z = rvtx.Z;
uvec2 i_uv = uvec2(bitfieldExtract(rvtx.UV, 0, 16), bitfieldExtract(rvtx.UV, 16, 16));
vec4 i_f = unpackUnorm4x8(rvtx.FOG);
float i_q = rvtx.Q;
uvec2 i_p = uvec2(bitfieldExtract(rvtx.XY, 0, 16), bitfieldExtract(rvtx.XY, 16, 16));
uint i_z = rvtx.Z;
uvec2 i_uv = uvec2(bitfieldExtract(rvtx.UV, 0, 16), bitfieldExtract(rvtx.UV, 16, 16));
vec4 i_f = unpackUnorm4x8(rvtx.FOG);
ProcessedVertex vtx;
ProcessedVertex vtx;
uint z = min(i_z, MaxDepth);
vtx.p.xy = vec2(i_p) - vec2(0.05f, 0.05f);
vtx.p.xy = vtx.p.xy * VertexScale - VertexOffset;
vtx.p.w = 1.0f;
uint z = min(i_z, MaxDepth);
vtx.p.xy = vec2(i_p) - vec2(0.05f, 0.05f);
vtx.p.xy = vtx.p.xy * VertexScale - VertexOffset;
vtx.p.w = 1.0f;
#if HAS_CLIP_CONTROL
vtx.p.z = float(z) * exp_min32;
vtx.p.z = float(z) * exp_min32;
#else
vtx.p.z = min(float(z) * exp2(-23.0f), 2.0f) - 1.0f;
vtx.p.z = min(float(z) * exp2(-23.0f), 2.0f) - 1.0f;
#endif
vec2 uv = vec2(i_uv) - TextureOffset;
vec2 st = i_st - TextureOffset;
vec2 uv = vec2(i_uv) - TextureOffset;
vec2 st = i_st - TextureOffset;
vtx.t_float.xy = st;
vtx.t_float.w = i_q;
vtx.t_float.xy = st;
vtx.t_float.w = i_q;
vtx.t_int.xy = uv * TextureScale;
vtx.t_int.xy = uv * TextureScale;
#if VS_FST
vtx.t_int.zw = uv;
vtx.t_int.zw = uv;
#else
vtx.t_int.zw = st / TextureScale;
vtx.t_int.zw = st / TextureScale;
#endif
vtx.c = i_c;
vtx.t_float.z = i_f.x;
vtx.c = i_c;
vtx.t_float.z = i_f.x;
return vtx;
return vtx;
}
void main()
{
ProcessedVertex vtx;
ProcessedVertex vtx;
#if defined(GL_ARB_shader_draw_parameters) && GL_ARB_shader_draw_parameters
uint vid = uint(gl_VertexID - gl_BaseVertexARB);
uint vid = uint(gl_VertexID - gl_BaseVertexARB);
#else
uint vid = uint(gl_VertexID);
uint vid = uint(gl_VertexID);
#endif
#if VS_EXPAND == 1 // Point
vtx = load_vertex(vid >> 2);
vtx = load_vertex(vid >> 2);
vtx.p.x += ((vid & 1u) != 0u) ? PointSize.x : 0.0f;
vtx.p.y += ((vid & 2u) != 0u) ? PointSize.y : 0.0f;
vtx.p.x += ((vid & 1u) != 0u) ? PointSize.x : 0.0f;
vtx.p.y += ((vid & 2u) != 0u) ? PointSize.y : 0.0f;
#elif VS_EXPAND == 2 // Line
uint vid_base = vid >> 2;
bool is_bottom = (vid & 2u) != 0u;
bool is_right = (vid & 1u) != 0u;
uint vid_other = is_bottom ? vid_base - 1 : vid_base + 1;
vtx = load_vertex(vid_base);
ProcessedVertex other = load_vertex(vid_other);
uint vid_base = vid >> 2;
bool is_bottom = (vid & 2u) != 0u;
bool is_right = (vid & 1u) != 0u;
uint vid_other = is_bottom ? vid_base - 1 : vid_base + 1;
vtx = load_vertex(vid_base);
ProcessedVertex other = load_vertex(vid_other);
vec2 line_vector = normalize(vtx.p.xy - other.p.xy);
vec2 line_normal = vec2(line_vector.y, -line_vector.x);
vec2 line_width = (line_normal * PointSize) / 2;
// line_normal is inverted for bottom point
vec2 offset = ((uint(is_bottom) ^ uint(is_right)) != 0u) ? line_width : -line_width;
vtx.p.xy += offset;
vec2 line_vector = normalize(vtx.p.xy - other.p.xy);
vec2 line_normal = vec2(line_vector.y, -line_vector.x);
vec2 line_width = (line_normal * PointSize) / 2;
// line_normal is inverted for bottom point
vec2 offset = ((uint(is_bottom) ^ uint(is_right)) != 0u) ? line_width : -line_width;
vtx.p.xy += offset;
// Lines will be run as (0 1 2) (1 2 3)
// This means that both triangles will have a point based off the top line point as their first point
// So we don't have to do anything for !IIP
// Lines will be run as (0 1 2) (1 2 3)
// This means that both triangles will have a point based off the top line point as their first point
// So we don't have to do anything for !IIP
#elif VS_EXPAND == 3 // Sprite
// Sprite points are always in pairs
uint vid_base = vid >> 1;
uint vid_lt = vid_base & ~1u;
uint vid_rb = vid_base | 1u;
// Sprite points are always in pairs
uint vid_base = vid >> 1;
uint vid_lt = vid_base & ~1u;
uint vid_rb = vid_base | 1u;
ProcessedVertex lt = load_vertex(vid_lt);
ProcessedVertex rb = load_vertex(vid_rb);
vtx = rb;
ProcessedVertex lt = load_vertex(vid_lt);
ProcessedVertex rb = load_vertex(vid_rb);
vtx = rb;
bool is_right = ((vid & 1u) != 0u);
vtx.p.x = is_right ? lt.p.x : vtx.p.x;
vtx.t_float.x = is_right ? lt.t_float.x : vtx.t_float.x;
vtx.t_int.xz = is_right ? lt.t_int.xz : vtx.t_int.xz;
bool is_right = ((vid & 1u) != 0u);
vtx.p.x = is_right ? lt.p.x : vtx.p.x;
vtx.t_float.x = is_right ? lt.t_float.x : vtx.t_float.x;
vtx.t_int.xz = is_right ? lt.t_int.xz : vtx.t_int.xz;
bool is_bottom = ((vid & 2u) != 0u);
vtx.p.y = is_bottom ? lt.p.y : vtx.p.y;
vtx.t_float.y = is_bottom ? lt.t_float.y : vtx.t_float.y;
vtx.t_int.yw = is_bottom ? lt.t_int.yw : vtx.t_int.yw;
bool is_bottom = ((vid & 2u) != 0u);
vtx.p.y = is_bottom ? lt.p.y : vtx.p.y;
vtx.t_float.y = is_bottom ? lt.t_float.y : vtx.t_float.y;
vtx.t_int.yw = is_bottom ? lt.t_int.yw : vtx.t_int.yw;
#endif
gl_Position = vtx.p;
VSout.t_float = vtx.t_float;
VSout.t_int = vtx.t_int;
VSout.c = vtx.c;
gl_Position = vtx.p;
VSout.t_float = vtx.t_float;
VSout.t_int = vtx.t_int;
VSout.c = vtx.c;
}
#endif // VS_EXPAND