mirror of
https://github.com/airwindows/airwindows.git
synced 2026-05-15 14:16:00 -06:00
152 lines
5.4 KiB
C++
Executable file
152 lines
5.4 KiB
C++
Executable file
/* ========================================
|
|
* SubTight - SubTight.h
|
|
* Copyright (c) airwindows, Airwindows uses the MIT license
|
|
* ======================================== */
|
|
|
|
#ifndef __SubTight_H
|
|
#include "SubTight.h"
|
|
#endif
|
|
|
|
void SubTight::processReplacing(float **inputs, float **outputs, VstInt32 sampleFrames)
|
|
{
|
|
float* in1 = inputs[0];
|
|
float* in2 = inputs[1];
|
|
float* out1 = outputs[0];
|
|
float* out2 = outputs[1];
|
|
|
|
double overallscale = 1.0;
|
|
overallscale /= 44100.0;
|
|
overallscale *= getSampleRate();
|
|
int subStages = pow(B,2)*16.0;
|
|
if (subStages < 1) subStages = 1;
|
|
double subTrim = pow((A*0.3)+(pow(B,2)*0.2),subStages)/overallscale;
|
|
//to use this as an analog modeler for restricting digital lows, find set values that still show bass
|
|
//Note that this is best used sparingly, on the 'not enough subtraction' side of the node.
|
|
|
|
while (--sampleFrames >= 0)
|
|
{
|
|
double inputSampleL = *in1;
|
|
double inputSampleR = *in2;
|
|
if (fabs(inputSampleL)<1.18e-23) inputSampleL = fpdL * 1.18e-17;
|
|
if (fabs(inputSampleR)<1.18e-23) inputSampleR = fpdR * 1.18e-17;
|
|
|
|
//you want subStages and subTrim to be hardcoded values when embedding this into something else
|
|
//then it only needs the sub[] array, and to have it initialized to 0.0
|
|
|
|
//begin SubTight section
|
|
double subSampleL = inputSampleL * subTrim;
|
|
double subSampleR = inputSampleR * subTrim;
|
|
for (int x = 0; x < subStages; x++) {
|
|
double scale = 0.5+fabs(subSampleL*0.5);
|
|
subSampleL = (subL[x]+(sin(subL[x]-subSampleL)*scale));
|
|
subL[x] = subSampleL*scale;
|
|
scale = 0.5+fabs(subSampleR*0.5);
|
|
subSampleR = (subR[x]+(sin(subR[x]-subSampleR)*scale));
|
|
subR[x] = subSampleR*scale;
|
|
}
|
|
if (subStages % 2 > 0) {
|
|
subSampleL = -subSampleL;
|
|
subSampleR = -subSampleR;
|
|
}
|
|
if (subSampleL > 0.25) subSampleL = 0.25;
|
|
if (subSampleL < -0.25) subSampleL = -0.25;
|
|
if (subSampleR > 0.25) subSampleR = 0.25;
|
|
if (subSampleR < -0.25) subSampleR = -0.25;
|
|
inputSampleL -= (subSampleL*16.0);
|
|
inputSampleR -= (subSampleR*16.0);
|
|
//end SubTight section
|
|
|
|
//cut the level WAY down, then the modified Creature code blows up subs.
|
|
//the adjustment of scale destabilizes the routine to blow up more DC.
|
|
//this is boosted by 24dB and subtracted from the dry signal
|
|
|
|
//begin 32 bit stereo floating point dither
|
|
int expon; frexpf((float)inputSampleL, &expon);
|
|
fpdL ^= fpdL << 13; fpdL ^= fpdL >> 17; fpdL ^= fpdL << 5;
|
|
inputSampleL += ((double(fpdL)-uint32_t(0x7fffffff)) * 5.5e-36l * pow(2,expon+62));
|
|
frexpf((float)inputSampleR, &expon);
|
|
fpdR ^= fpdR << 13; fpdR ^= fpdR >> 17; fpdR ^= fpdR << 5;
|
|
inputSampleR += ((double(fpdR)-uint32_t(0x7fffffff)) * 5.5e-36l * pow(2,expon+62));
|
|
//end 32 bit stereo floating point dither
|
|
|
|
*out1 = inputSampleL;
|
|
*out2 = inputSampleR;
|
|
|
|
in1++;
|
|
in2++;
|
|
out1++;
|
|
out2++;
|
|
}
|
|
}
|
|
|
|
void SubTight::processDoubleReplacing(double **inputs, double **outputs, VstInt32 sampleFrames)
|
|
{
|
|
double* in1 = inputs[0];
|
|
double* in2 = inputs[1];
|
|
double* out1 = outputs[0];
|
|
double* out2 = outputs[1];
|
|
|
|
double overallscale = 1.0;
|
|
overallscale /= 44100.0;
|
|
overallscale *= getSampleRate();
|
|
int subStages = pow(B,2)*16.0;
|
|
if (subStages < 1) subStages = 1;
|
|
double subTrim = pow((A*0.3)+(pow(B,2)*0.2),subStages)/overallscale;
|
|
//to use this as an analog modeler for restricting digital lows, find set values that still show bass
|
|
//Note that this is best used sparingly, on the 'not enough subtraction' side of the node.
|
|
|
|
while (--sampleFrames >= 0)
|
|
{
|
|
double inputSampleL = *in1;
|
|
double inputSampleR = *in2;
|
|
if (fabs(inputSampleL)<1.18e-23) inputSampleL = fpdL * 1.18e-17;
|
|
if (fabs(inputSampleR)<1.18e-23) inputSampleR = fpdR * 1.18e-17;
|
|
|
|
//you want subStages and subTrim to be hardcoded values when embedding this into something else
|
|
//then it only needs the sub[] array, and to have it initialized to 0.0
|
|
|
|
//begin SubTight section
|
|
double subSampleL = inputSampleL * subTrim;
|
|
double subSampleR = inputSampleR * subTrim;
|
|
for (int x = 0; x < subStages; x++) {
|
|
double scale = 0.5+fabs(subSampleL*0.5);
|
|
subSampleL = (subL[x]+(sin(subL[x]-subSampleL)*scale));
|
|
subL[x] = subSampleL*scale;
|
|
scale = 0.5+fabs(subSampleR*0.5);
|
|
subSampleR = (subR[x]+(sin(subR[x]-subSampleR)*scale));
|
|
subR[x] = subSampleR*scale;
|
|
}
|
|
if (subStages % 2 > 0) {
|
|
subSampleL = -subSampleL;
|
|
subSampleR = -subSampleR;
|
|
}
|
|
if (subSampleL > 0.25) subSampleL = 0.25;
|
|
if (subSampleL < -0.25) subSampleL = -0.25;
|
|
if (subSampleR > 0.25) subSampleR = 0.25;
|
|
if (subSampleR < -0.25) subSampleR = -0.25;
|
|
inputSampleL -= (subSampleL*16.0);
|
|
inputSampleR -= (subSampleR*16.0);
|
|
//end SubTight section
|
|
|
|
//cut the level WAY down, then the modified Creature code blows up subs.
|
|
//the adjustment of scale destabilizes the routine to blow up more DC.
|
|
//this is boosted by 24dB and subtracted from the dry signal
|
|
|
|
//begin 64 bit stereo floating point dither
|
|
//int expon; frexp((double)inputSampleL, &expon);
|
|
fpdL ^= fpdL << 13; fpdL ^= fpdL >> 17; fpdL ^= fpdL << 5;
|
|
//inputSampleL += ((double(fpdL)-uint32_t(0x7fffffff)) * 1.1e-44l * pow(2,expon+62));
|
|
//frexp((double)inputSampleR, &expon);
|
|
fpdR ^= fpdR << 13; fpdR ^= fpdR >> 17; fpdR ^= fpdR << 5;
|
|
//inputSampleR += ((double(fpdR)-uint32_t(0x7fffffff)) * 1.1e-44l * pow(2,expon+62));
|
|
//end 64 bit stereo floating point dither
|
|
|
|
*out1 = inputSampleL;
|
|
*out2 = inputSampleR;
|
|
|
|
in1++;
|
|
in2++;
|
|
out1++;
|
|
out2++;
|
|
}
|
|
}
|