Untitled
unknown
plain_text
2 years ago
7.0 kB
4
Indexable
- "long"
- "float"
- "point2D"
- "color"
- "image"
- "audio"
- "audioFFT"
2) Functions:
- IMG_NORM_PIXEL() -> get a pixel from input with normalized coordinates
- IMG_PIXEL() -> get a pixel from input with screen space coordinates
3) Predefined variables:
- RENDERSIZE (resolution of the shader)
- TIME (run time)
- gl_FragCoord.xy (screen space coordinates of current fragment)
- isf_FragNormCoord.xy (normalized coordinates)
To learn more see:
https://github.com/mrRay/ISF_Spec/
*/
#define PI 3.14159265359
float animTime;
vec4 fft, ffts;
void compressFft(){ //compress sound in iChannel0 to simple frequency-range amplitude estimations
fft = vec4(0), ffts = vec4(0);
//IMG_NORM_PIXEL(inputImage, uvS2);
// // Sound (assume sound texture with 44.1kHz in 512 texels, cf. shadertoy.com/view/Xds3Rr)
// for (int n=1;n<3;n++) fft.x += texelFetch( iChannel0, ivec2(n,0), 0 ).x; //bass, 0-517Hz, reduced to 86-258Hz
// for (int n=6;n<8;n++) ffts.x += texelFetch( iChannel0, ivec2(n,0), 0 ).x; //speech I, 517-689Hz
// for (int n=8;n<14;n+=2) ffts.y += texelFetch( iChannel0, ivec2(n,0), 0 ).x; //speech II, 689-1206Hz
// for (int n=14;n<24;n+=4) ffts.z += texelFetch( iChannel0, ivec2(n,0), 0 ).x; //speech III, 1206-2067Hz
// for (int n=24;n<95;n+=10) fft.z += texelFetch( iChannel0, ivec2(n,0), 0 ).x; //presence, 2067-8183Hz, tenth sample
// for (int n=95;n<512;n+=100) fft.w += texelFetch( iChannel0, ivec2(n,0), 0 ).x; //brilliance, 8183-44100Hz, tenth2 sample
float fftdivider = 512.0;
// Sound (assume sound texture with 44.1kHz in 512 texels, cf. shadertoy.com/view/Xds3Rr)
for (int n=1;n<3;n++) fft.x += IMG_NORM_PIXEL( audio, vec2( float(n)/fftdivider ,0)).x; //bass, 0-517Hz, reduced to 86-258Hz
for (int n=6;n<8;n++) ffts.x += IMG_NORM_PIXEL( audio, vec2( float(n)/fftdivider ,0)).x; //speech I, 517-689Hz
for (int n=8;n<14;n+=2) ffts.y += IMG_NORM_PIXEL( audio, vec2( float(n)/fftdivider ,0)).x; //speech II, 689-1206Hz
for (int n=14;n<24;n+=4) ffts.z += IMG_NORM_PIXEL( audio, vec2( float(n)/fftdivider ,0)).x; //speech III, 1206-2067Hz
for (int n=24;n<95;n+=10) fft.z += IMG_NORM_PIXEL( audio, vec2( float(n)/fftdivider ,0)).x; //presence, 2067-8183Hz, tenth sample
for (int n=95;n<512;n+=100) fft.w += IMG_NORM_PIXEL( audio, vec2( float(n)/fftdivider ,0)).x; //brilliance, 8183-44100Hz, tenth2 sample
fft.y = dot(ffts.xyz,vec3(1)); //speech I-III, 517-2067Hz
ffts.w = dot(fft.xyzw,vec4(1)); //overall loudness
fft /= vec4(2,8,7,4); ffts /= vec4(2,3,3,21); //normalize
fft.x = step(.91,fft.x); //weaken weaker sounds, hard limit
}
float hash21(vec2 p){ //pseudorandom generator, cf. The Art of Code on youtu.be/rvDo9LvfoVE
p = fract(p*vec2(13.81, 741.76));
p += dot(p, p+42.23);
return fract(p.x*p.y);
}
mat2 rotM(float deg){
deg /= 180./PI;
return mat2(cos(deg),-sin(deg),sin(deg),cos(deg));
}
float particle(vec2 p){
return smoothstep(.1,.0,length(p)) * smoothstep(.1,.06,length(p-vec2(0.,.02)));
}
float dustLayer(vec2 p){
float id = hash21(floor(p));
return smoothstep(0.,1.,id)*particle((fract(p)-vec2(.5+.4*cos(id*animTime),.5+.4*sin(.8*id*animTime)))*rotM(id*360.)/vec2(cos(.5*id*animTime),1)); //...is there performance gain in doing this rather than via sub steps?
}
vec3 Field(vec3 Pos) //forked
{
Pos *= 0.1;
float f = 0.1;
for (int i = 0; i < 5; ++i)
{
Pos = (Pos.yzx * mat3(0.8, 0.6, 0, -0.6, 0.8, 0.0, 0.0, 0.0, 1.0)) + vec3(0.123, 0.456, 0.789) * float(i);
// Pos = (Pos.yzx * mat3(0.7071, 0.4071, 0, -0.4071, 0.7071, 0.0, 0.0, 0.0, 1.0));// + vec3(0.123, 0.456, 0.789) * float(i);
Pos = (abs(fract(Pos) - 0.5)) * 2.0;
f *= 2.0;
}
Pos *= Pos;
return sqrt(Pos + Pos.yzx) / f + 0.0001;
}
out vec4 fragColor;
//void mainImage( out vec4 fragColor, in vec2 fragCoord )
void main()
{
// vec2 uv = isf_FragNormCoord.xy;
vec2 fragCoord = gl_FragCoord.xy;
vec2 uv = (2.*fragCoord-RENDERSIZE.xy) / max(RENDERSIZE.x, RENDERSIZE.y); // viewport max -1..1
float d2 = uv.x*uv.x+uv.y*uv.y; //polar distance squared
animTime = 2.133333*TIME;
vec3 col;
float aFrac, amp = 0.;
compressFft(); //initializes fft, ffts
vec3 Position = vec3(0.5, 0.8, .5*animTime+.4*fft.w*fft.w*fft.w);
vec3 Direction = vec3((1.-.5*cos(animTime/16.))*uv,1.);
// Dust layers
for (float n=0.;n<4.;n++){
aFrac = fract(-.05*animTime+.25*n)-.03*fft.w*fft.w*fft.w;
amp += 1.4*(.2+.8*fft.z)*dustLayer((uv+n*vec2(.1,.05))*25.*aFrac)*smoothstep(1.,.33,aFrac);
}
amp *= (.7+.5*length(uv)); //anti-vignette
// Original geometry field, forked
for (int i = 0; i < 50; ++i){
vec3 f2 = Field(Position);
Position += Direction * min(min(f2.x, f2.y), f2.z);
col += float(50 - i) / (f2 + 0.005);
}
col = vec3(1.0 - 1.0 / (1.0 + col * (-0.06 / 2500.0)));
//col = vec3(1.0 - 1.0 / (1.0 + col * (-0.06 / 2500.0))); //!Variable
col *= col;
// Misc
col = (.5*amp+3.*col.r)*vec3(ffts.x<=ffts.y,ffts.y<=ffts.z,ffts.z<=ffts.x); //colors
// col = vec3(0.0);
col += .1*vec3(clamp(.015/abs(d2-.04*(.5+ffts.w))*(.1+.9*fft.x),0.,1.))/(length(col)); //center eye
float divider = length(col);
// divider = 0.1;
// col += .1*vec3(clamp(.015/abs(d2-.04*(.5+ffts.w))*(.1+.9*fft.x),0.,1.))/(divider); //center eye
col = vec3(col.r + col.g + col.b*0.0);
vec3 col1 = vec3(0.0);
col1 = mix(col1, vec3(fft.x), step(uv.x, -0.5));
col1 = mix(col1, vec3(fft.y), step(-0.5, uv.x) * step(uv.x, 0.) );
col1 = mix(col1, vec3(fft.z), step(0., uv.x) * step(uv.x, 0.5) );
col1 = mix(col1, vec3(fft.w), step(0.5, uv.x) * step(uv.x, 1.0) );
vec3 col2 = vec3(0.0);
col2 = mix(col2, vec3(ffts.x), step(uv.x, -0.5));
col2 = mix(col2, vec3(ffts.y), step(-0.5, uv.x) * step(uv.x, 0.) );
col2 = mix(col2, vec3(ffts.z), step(0., uv.x) * step(uv.x, 0.5) );
col2 = mix(col2, vec3(ffts.w), step(0.5, uv.x) * step(uv.x, 1.0) );
//col = mix(col1, col2, step(uv.y, 0.0));
// col = vec3(step(-0.5, uv.x) * step(uv.x, 0.));
gl_FragColor = vec4(col,1.0);
} Editor is loading...