1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
|
///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
// Copyright (c) 2016, Intel Corporation
// Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated
// documentation files (the "Software"), to deal in the Software without restriction, including without limitation
// the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to
// permit persons to whom the Software is furnished to do so, subject to the following conditions:
// The above copyright notice and this permission notice shall be included in all copies or substantial portions of
// the Software.
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO
// THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
// TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
// SOFTWARE.
///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
// File changes (yyyy-mm-dd)
// 2016-09-07: filip.strugar@intel.com: first commit
// 2020-12-05: clayjohn: convert to Vulkan and Godot
///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
#[compute]
#version 450
#VERSION_DEFINES
layout(local_size_x = 8, local_size_y = 8, local_size_z = 1) in;
layout(set = 0, binding = 0) uniform sampler2D source_ssao;
layout(rg8, set = 1, binding = 0) uniform restrict writeonly image2D dest_image;
layout(push_constant, std430) uniform Params {
float edge_sharpness;
float pad;
vec2 half_screen_pixel_size;
}
params;
vec4 unpack_edges(float p_packed_val) {
uint packed_val = uint(p_packed_val * 255.5);
vec4 edgesLRTB;
edgesLRTB.x = float((packed_val >> 6) & 0x03) / 3.0;
edgesLRTB.y = float((packed_val >> 4) & 0x03) / 3.0;
edgesLRTB.z = float((packed_val >> 2) & 0x03) / 3.0;
edgesLRTB.w = float((packed_val >> 0) & 0x03) / 3.0;
return clamp(edgesLRTB + params.edge_sharpness, 0.0, 1.0);
}
void add_sample(float p_ssao_value, float p_edge_value, inout float r_sum, inout float r_sum_weight) {
float weight = p_edge_value;
r_sum += (weight * p_ssao_value);
r_sum_weight += weight;
}
#ifdef MODE_WIDE
vec2 sample_blurred_wide(vec2 p_coord) {
vec2 vC = textureLodOffset(source_ssao, vec2(p_coord), 0.0, ivec2(0, 0)).xy;
vec2 vL = textureLodOffset(source_ssao, vec2(p_coord), 0.0, ivec2(-2, 0)).xy;
vec2 vT = textureLodOffset(source_ssao, vec2(p_coord), 0.0, ivec2(0, -2)).xy;
vec2 vR = textureLodOffset(source_ssao, vec2(p_coord), 0.0, ivec2(2, 0)).xy;
vec2 vB = textureLodOffset(source_ssao, vec2(p_coord), 0.0, ivec2(0, 2)).xy;
float packed_edges = vC.y;
vec4 edgesLRTB = unpack_edges(packed_edges);
edgesLRTB.x *= unpack_edges(vL.y).y;
edgesLRTB.z *= unpack_edges(vT.y).w;
edgesLRTB.y *= unpack_edges(vR.y).x;
edgesLRTB.w *= unpack_edges(vB.y).z;
float ssao_value = vC.x;
float ssao_valueL = vL.x;
float ssao_valueT = vT.x;
float ssao_valueR = vR.x;
float ssao_valueB = vB.x;
float sum_weight = 0.8f;
float sum = ssao_value * sum_weight;
add_sample(ssao_valueL, edgesLRTB.x, sum, sum_weight);
add_sample(ssao_valueR, edgesLRTB.y, sum, sum_weight);
add_sample(ssao_valueT, edgesLRTB.z, sum, sum_weight);
add_sample(ssao_valueB, edgesLRTB.w, sum, sum_weight);
float ssao_avg = sum / sum_weight;
ssao_value = ssao_avg;
return vec2(ssao_value, packed_edges);
}
#endif
#ifdef MODE_SMART
vec2 sample_blurred(vec3 p_pos, vec2 p_coord) {
float packed_edges = texelFetch(source_ssao, ivec2(p_pos.xy), 0).y;
vec4 edgesLRTB = unpack_edges(packed_edges);
vec4 valuesUL = textureGather(source_ssao, vec2(p_coord - params.half_screen_pixel_size * 0.5));
vec4 valuesBR = textureGather(source_ssao, vec2(p_coord + params.half_screen_pixel_size * 0.5));
float ssao_value = valuesUL.y;
float ssao_valueL = valuesUL.x;
float ssao_valueT = valuesUL.z;
float ssao_valueR = valuesBR.z;
float ssao_valueB = valuesBR.x;
float sum_weight = 0.5;
float sum = ssao_value * sum_weight;
add_sample(ssao_valueL, edgesLRTB.x, sum, sum_weight);
add_sample(ssao_valueR, edgesLRTB.y, sum, sum_weight);
add_sample(ssao_valueT, edgesLRTB.z, sum, sum_weight);
add_sample(ssao_valueB, edgesLRTB.w, sum, sum_weight);
float ssao_avg = sum / sum_weight;
ssao_value = ssao_avg;
return vec2(ssao_value, packed_edges);
}
#endif
void main() {
// Pixel being shaded
ivec2 ssC = ivec2(gl_GlobalInvocationID.xy);
#ifdef MODE_NON_SMART
vec2 half_pixel = params.half_screen_pixel_size * 0.5;
vec2 uv = (vec2(gl_GlobalInvocationID.xy) + vec2(0.5, 0.5)) * params.half_screen_pixel_size;
vec2 center = textureLod(source_ssao, vec2(uv), 0.0).xy;
vec4 vals;
vals.x = textureLod(source_ssao, vec2(uv + vec2(-half_pixel.x * 3, -half_pixel.y)), 0.0).x;
vals.y = textureLod(source_ssao, vec2(uv + vec2(+half_pixel.x, -half_pixel.y * 3)), 0.0).x;
vals.z = textureLod(source_ssao, vec2(uv + vec2(-half_pixel.x, +half_pixel.y * 3)), 0.0).x;
vals.w = textureLod(source_ssao, vec2(uv + vec2(+half_pixel.x * 3, +half_pixel.y)), 0.0).x;
vec2 sampled = vec2(dot(vals, vec4(0.2)) + center.x * 0.2, center.y);
#else
#ifdef MODE_SMART
vec2 sampled = sample_blurred(vec3(gl_GlobalInvocationID), (vec2(gl_GlobalInvocationID.xy) + vec2(0.5, 0.5)) * params.half_screen_pixel_size);
#else // MODE_WIDE
vec2 sampled = sample_blurred_wide((vec2(gl_GlobalInvocationID.xy) + vec2(0.5, 0.5)) * params.half_screen_pixel_size);
#endif
#endif
imageStore(dest_image, ivec2(ssC), vec4(sampled, 0.0, 0.0));
}
|