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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917
|
/*
* BasicFilters.h - simple but powerful filter-class with most used filters
*
* original file by ???
* modified and enhanced by Tobias Doerffel
*
* Lowpass_SV code originally from Nekobee, Copyright (C) 2004 Sean Bolton and others
* adapted & modified for use in LMMS
*
* Copyright (c) 2004-2009 Tobias Doerffel <tobydox/at/users.sourceforge.net>
*
* This file is part of LMMS - https://lmms.io
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* You should have received a copy of the GNU General Public
* License along with this program (see COPYING); if not, write to the
* Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
* Boston, MA 02110-1301 USA.
*
*/
#ifndef BASIC_FILTERS_H
#define BASIC_FILTERS_H
#ifndef __USE_XOPEN
#define __USE_XOPEN
#endif
#include <math.h>
#include "lmms_basics.h"
#include "templates.h"
#include "lmms_constants.h"
#include "interpolation.h"
#include "MemoryManager.h"
template<ch_cnt_t CHANNELS> class BasicFilters;
template<ch_cnt_t CHANNELS>
class LinkwitzRiley
{
MM_OPERATORS
public:
LinkwitzRiley( float sampleRate )
{
m_sampleRate = sampleRate;
clearHistory();
}
virtual ~LinkwitzRiley() {}
inline void clearHistory()
{
for( int i = 0; i < CHANNELS; ++i )
{
m_z1[i] = m_z2[i] = m_z3[i] = m_z4[i] = 0.0f;
}
}
inline void setSampleRate( float sampleRate )
{
m_sampleRate = sampleRate;
}
inline void setCoeffs( float freq )
{
// wc
const double wc = D_2PI * freq;
const double wc2 = wc * wc;
const double wc3 = wc2 * wc;
m_wc4 = wc2 * wc2;
// k
const double k = wc / tan( D_PI * freq / m_sampleRate );
const double k2 = k * k;
const double k3 = k2 * k;
m_k4 = k2 * k2;
// a
static const double sqrt2 = sqrt( 2.0 );
const double sq_tmp1 = sqrt2 * wc3 * k;
const double sq_tmp2 = sqrt2 * wc * k3;
m_a = 1.0 / ( 4.0 * wc2 * k2 + 2.0 * sq_tmp1 + m_k4 + 2.0 * sq_tmp2 + m_wc4 );
// b
m_b1 = ( 4.0 * ( m_wc4 + sq_tmp1 - m_k4 - sq_tmp2 ) ) * m_a;
m_b2 = ( 6.0 * m_wc4 - 8.0 * wc2 * k2 + 6.0 * m_k4 ) * m_a;
m_b3 = ( 4.0 * ( m_wc4 - sq_tmp1 + sq_tmp2 - m_k4 ) ) * m_a;
m_b4 = ( m_k4 - 2.0 * sq_tmp1 + m_wc4 - 2.0 * sq_tmp2 + 4.0 * wc2 * k2 ) * m_a;
}
inline void setLowpass( float freq )
{
setCoeffs( freq );
m_a0 = m_wc4 * m_a;
m_a1 = 4.0 * m_a0;
m_a2 = 6.0 * m_a0;
}
inline void setHighpass( float freq )
{
setCoeffs( freq );
m_a0 = m_k4 * m_a;
m_a1 = -4.0 * m_a0;
m_a2 = 6.0 * m_a0;
}
inline float update( float in, ch_cnt_t ch )
{
const double x = in - ( m_z1[ch] * m_b1 ) - ( m_z2[ch] * m_b2 ) -
( m_z3[ch] * m_b3 ) - ( m_z4[ch] * m_b4 );
const double y = ( m_a0 * x ) + ( m_z1[ch] * m_a1 ) + ( m_z2[ch] * m_a2 ) +
( m_z3[ch] * m_a1 ) + ( m_z4[ch] * m_a0 );
m_z4[ch] = m_z3[ch];
m_z3[ch] = m_z2[ch];
m_z2[ch] = m_z1[ch];
m_z1[ch] = x;
return y;
}
private:
float m_sampleRate;
double m_wc4;
double m_k4;
double m_a, m_a0, m_a1, m_a2;
double m_b1, m_b2, m_b3, m_b4;
typedef double frame[CHANNELS];
frame m_z1, m_z2, m_z3, m_z4;
};
typedef LinkwitzRiley<2> StereoLinkwitzRiley;
template<ch_cnt_t CHANNELS>
class BiQuad
{
MM_OPERATORS
public:
BiQuad()
{
clearHistory();
}
virtual ~BiQuad() {}
inline void setCoeffs( float a1, float a2, float b0, float b1, float b2 )
{
m_a1 = a1;
m_a2 = a2;
m_b0 = b0;
m_b1 = b1;
m_b2 = b2;
}
inline void clearHistory()
{
for( int i = 0; i < CHANNELS; ++i )
{
m_z1[i] = 0.0f;
m_z2[i] = 0.0f;
}
}
inline float update( float in, ch_cnt_t ch )
{
// biquad filter in transposed form
const float out = m_z1[ch] + m_b0 * in;
m_z1[ch] = m_b1 * in + m_z2[ch] - m_a1 * out;
m_z2[ch] = m_b2 * in - m_a2 * out;
return out;
}
private:
float m_a1, m_a2, m_b0, m_b1, m_b2;
float m_z1 [CHANNELS], m_z2 [CHANNELS];
friend class BasicFilters<CHANNELS>; // needed for subfilter stuff in BasicFilters
};
typedef BiQuad<2> StereoBiQuad;
template<ch_cnt_t CHANNELS>
class OnePole
{
MM_OPERATORS
public:
OnePole()
{
m_a0 = 1.0;
m_b1 = 0.0;
for( int i = 0; i < CHANNELS; ++i )
{
m_z1[i] = 0.0;
}
}
virtual ~OnePole() {}
inline void setCoeffs( float a0, float b1 )
{
m_a0 = a0;
m_b1 = b1;
}
inline float update( float s, ch_cnt_t ch )
{
if( qAbs( s ) < 1.0e-10f && qAbs( m_z1[ch] ) < 1.0e-10f ) return 0.0f;
return m_z1[ch] = s * m_a0 + m_z1[ch] * m_b1;
}
private:
float m_a0, m_b1;
float m_z1 [CHANNELS];
};
typedef OnePole<2> StereoOnePole;
template<ch_cnt_t CHANNELS>
class BasicFilters
{
MM_OPERATORS
public:
enum FilterTypes
{
LowPass,
HiPass,
BandPass_CSG,
BandPass_CZPG,
Notch,
AllPass,
Moog,
DoubleLowPass,
Lowpass_RC12,
Bandpass_RC12,
Highpass_RC12,
Lowpass_RC24,
Bandpass_RC24,
Highpass_RC24,
Formantfilter,
DoubleMoog,
Lowpass_SV,
Bandpass_SV,
Highpass_SV,
Notch_SV,
FastFormant,
Tripole,
NumFilters
};
static inline float minFreq()
{
return( 5.0f );
}
static inline float minQ()
{
return( 0.01f );
}
inline void setFilterType( const int _idx )
{
m_doubleFilter = _idx == DoubleLowPass || _idx == DoubleMoog;
if( !m_doubleFilter )
{
m_type = static_cast<FilterTypes>( _idx );
return;
}
// Double lowpass mode, backwards-compat for the goofy
// Add-NumFilters to signify doubleFilter stuff
m_type = _idx == DoubleLowPass
? LowPass
: Moog;
if( m_subFilter == NULL )
{
m_subFilter = new BasicFilters<CHANNELS>(
static_cast<sample_rate_t>(
m_sampleRate ) );
}
m_subFilter->m_type = m_type;
}
inline BasicFilters( const sample_rate_t _sample_rate ) :
m_doubleFilter( false ),
m_sampleRate( (float) _sample_rate ),
m_sampleRatio( 1.0f / m_sampleRate ),
m_subFilter( NULL )
{
clearHistory();
}
inline ~BasicFilters()
{
delete m_subFilter;
}
inline void clearHistory()
{
// reset in/out history for biquads
m_biQuad.clearHistory();
// reset in/out history
for( ch_cnt_t _chnl = 0; _chnl < CHANNELS; ++_chnl )
{
// reset in/out history for moog-filter
m_y1[_chnl] = m_y2[_chnl] = m_y3[_chnl] = m_y4[_chnl] =
m_oldx[_chnl] = m_oldy1[_chnl] =
m_oldy2[_chnl] = m_oldy3[_chnl] = 0.0f;
// tripole
m_last[_chnl] = 0.0f;
// reset in/out history for RC-filters
m_rclp0[_chnl] = m_rcbp0[_chnl] = m_rchp0[_chnl] = m_rclast0[_chnl] = 0.0f;
m_rclp1[_chnl] = m_rcbp1[_chnl] = m_rchp1[_chnl] = m_rclast1[_chnl] = 0.0f;
for(int i=0; i<6; i++)
m_vfbp[i][_chnl] = m_vfhp[i][_chnl] = m_vflast[i][_chnl] = 0.0f;
// reset in/out history for SV-filters
m_delay1[_chnl] = 0.0f;
m_delay2[_chnl] = 0.0f;
m_delay3[_chnl] = 0.0f;
m_delay4[_chnl] = 0.0f;
}
}
inline sample_t update( sample_t _in0, ch_cnt_t _chnl )
{
sample_t out;
switch( m_type )
{
case Moog:
{
sample_t x = _in0 - m_r*m_y4[_chnl];
// four cascaded onepole filters
// (bilinear transform)
m_y1[_chnl] = qBound( -10.0f,
( x + m_oldx[_chnl] ) * m_p
- m_k * m_y1[_chnl],
10.0f );
m_y2[_chnl] = qBound( -10.0f,
( m_y1[_chnl] + m_oldy1[_chnl] ) * m_p
- m_k * m_y2[_chnl],
10.0f );
m_y3[_chnl] = qBound( -10.0f,
( m_y2[_chnl] + m_oldy2[_chnl] ) * m_p
- m_k * m_y3[_chnl],
10.0f );
m_y4[_chnl] = qBound( -10.0f,
( m_y3[_chnl] + m_oldy3[_chnl] ) * m_p
- m_k * m_y4[_chnl],
10.0f );
m_oldx[_chnl] = x;
m_oldy1[_chnl] = m_y1[_chnl];
m_oldy2[_chnl] = m_y2[_chnl];
m_oldy3[_chnl] = m_y3[_chnl];
out = m_y4[_chnl] - m_y4[_chnl] * m_y4[_chnl] *
m_y4[_chnl] * ( 1.0f / 6.0f );
break;
}
// 3x onepole filters with 4x oversampling and interpolation of oversampled signal:
// input signal is linear-interpolated after oversampling, output signal is averaged from oversampled outputs
case Tripole:
{
out = 0.0f;
float ip = 0.0f;
for( int i = 0; i < 4; ++i )
{
ip += 0.25f;
sample_t x = linearInterpolate( m_last[_chnl], _in0, ip ) - m_r * m_y3[_chnl];
m_y1[_chnl] = qBound( -10.0f,
( x + m_oldx[_chnl] ) * m_p
- m_k * m_y1[_chnl],
10.0f );
m_y2[_chnl] = qBound( -10.0f,
( m_y1[_chnl] + m_oldy1[_chnl] ) * m_p
- m_k * m_y2[_chnl],
10.0f );
m_y3[_chnl] = qBound( -10.0f,
( m_y2[_chnl] + m_oldy2[_chnl] ) * m_p
- m_k * m_y3[_chnl],
10.0f );
m_oldx[_chnl] = x;
m_oldy1[_chnl] = m_y1[_chnl];
m_oldy2[_chnl] = m_y2[_chnl];
out += ( m_y3[_chnl] - m_y3[_chnl] * m_y3[_chnl] * m_y3[_chnl] * ( 1.0f / 6.0f ) );
}
out *= 0.25f;
m_last[_chnl] = _in0;
return out;
}
// 4-pole state-variant lowpass filter, adapted from Nekobee source code
// and extended to other SV filter types
// /* Hal Chamberlin's state variable filter */
case Lowpass_SV:
case Bandpass_SV:
{
float highpass;
for( int i = 0; i < 2; ++i ) // 2x oversample
{
m_delay2[_chnl] = m_delay2[_chnl] + m_svf1 * m_delay1[_chnl]; /* delay2/4 = lowpass output */
highpass = _in0 - m_delay2[_chnl] - m_svq * m_delay1[_chnl];
m_delay1[_chnl] = m_svf1 * highpass + m_delay1[_chnl]; /* delay1/3 = bandpass output */
m_delay4[_chnl] = m_delay4[_chnl] + m_svf2 * m_delay3[_chnl];
highpass = m_delay2[_chnl] - m_delay4[_chnl] - m_svq * m_delay3[_chnl];
m_delay3[_chnl] = m_svf2 * highpass + m_delay3[_chnl];
}
/* mix filter output into output buffer */
return m_type == Lowpass_SV
? m_delay4[_chnl]
: m_delay3[_chnl];
}
case Highpass_SV:
{
float hp;
for( int i = 0; i < 2; ++i ) // 2x oversample
{
m_delay2[_chnl] = m_delay2[_chnl] + m_svf1 * m_delay1[_chnl];
hp = _in0 - m_delay2[_chnl] - m_svq * m_delay1[_chnl];
m_delay1[_chnl] = m_svf1 * hp + m_delay1[_chnl];
}
return hp;
}
case Notch_SV:
{
float hp1, hp2;
for( int i = 0; i < 2; ++i ) // 2x oversample
{
m_delay2[_chnl] = m_delay2[_chnl] + m_svf1 * m_delay1[_chnl]; /* delay2/4 = lowpass output */
hp1 = _in0 - m_delay2[_chnl] - m_svq * m_delay1[_chnl];
m_delay1[_chnl] = m_svf1 * hp1 + m_delay1[_chnl]; /* delay1/3 = bandpass output */
m_delay4[_chnl] = m_delay4[_chnl] + m_svf2 * m_delay3[_chnl];
hp2 = m_delay2[_chnl] - m_delay4[_chnl] - m_svq * m_delay3[_chnl];
m_delay3[_chnl] = m_svf2 * hp2 + m_delay3[_chnl];
}
/* mix filter output into output buffer */
return m_delay4[_chnl] + hp1;
}
// 4-times oversampled simulation of an active RC-Bandpass,-Lowpass,-Highpass-
// Filter-Network as it was used in nearly all modern analog synthesizers. This
// can be driven up to self-oscillation (BTW: do not remove the limits!!!).
// (C) 1998 ... 2009 S.Fendt. Released under the GPL v2.0 or any later version.
case Lowpass_RC12:
{
sample_t lp, bp, hp, in;
for( int n = 4; n != 0; --n )
{
in = _in0 + m_rcbp0[_chnl] * m_rcq;
in = qBound( -1.0f, in, 1.0f );
lp = in * m_rcb + m_rclp0[_chnl] * m_rca;
lp = qBound( -1.0f, lp, 1.0f );
hp = m_rcc * ( m_rchp0[_chnl] + in - m_rclast0[_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_rcb + m_rcbp0[_chnl] * m_rca;
bp = qBound( -1.0f, bp, 1.0f );
m_rclast0[_chnl] = in;
m_rclp0[_chnl] = lp;
m_rchp0[_chnl] = hp;
m_rcbp0[_chnl] = bp;
}
return lp;
}
case Highpass_RC12:
case Bandpass_RC12:
{
sample_t hp, bp, in;
for( int n = 4; n != 0; --n )
{
in = _in0 + m_rcbp0[_chnl] * m_rcq;
in = qBound( -1.0f, in, 1.0f );
hp = m_rcc * ( m_rchp0[_chnl] + in - m_rclast0[_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_rcb + m_rcbp0[_chnl] * m_rca;
bp = qBound( -1.0f, bp, 1.0f );
m_rclast0[_chnl] = in;
m_rchp0[_chnl] = hp;
m_rcbp0[_chnl] = bp;
}
return m_type == Highpass_RC12 ? hp : bp;
}
case Lowpass_RC24:
{
sample_t lp, bp, hp, in;
for( int n = 4; n != 0; --n )
{
// first stage is as for the 12dB case...
in = _in0 + m_rcbp0[_chnl] * m_rcq;
in = qBound( -1.0f, in, 1.0f );
lp = in * m_rcb + m_rclp0[_chnl] * m_rca;
lp = qBound( -1.0f, lp, 1.0f );
hp = m_rcc * ( m_rchp0[_chnl] + in - m_rclast0[_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_rcb + m_rcbp0[_chnl] * m_rca;
bp = qBound( -1.0f, bp, 1.0f );
m_rclast0[_chnl] = in;
m_rclp0[_chnl] = lp;
m_rcbp0[_chnl] = bp;
m_rchp0[_chnl] = hp;
// second stage gets the output of the first stage as input...
in = lp + m_rcbp1[_chnl] * m_rcq;
in = qBound( -1.0f, in, 1.0f );
lp = in * m_rcb + m_rclp1[_chnl] * m_rca;
lp = qBound( -1.0f, lp, 1.0f );
hp = m_rcc * ( m_rchp1[_chnl] + in - m_rclast1[_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_rcb + m_rcbp1[_chnl] * m_rca;
bp = qBound( -1.0f, bp, 1.0f );
m_rclast1[_chnl] = in;
m_rclp1[_chnl] = lp;
m_rcbp1[_chnl] = bp;
m_rchp1[_chnl] = hp;
}
return lp;
}
case Highpass_RC24:
case Bandpass_RC24:
{
sample_t hp, bp, in;
for( int n = 4; n != 0; --n )
{
// first stage is as for the 12dB case...
in = _in0 + m_rcbp0[_chnl] * m_rcq;
in = qBound( -1.0f, in, 1.0f );
hp = m_rcc * ( m_rchp0[_chnl] + in - m_rclast0[_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_rcb + m_rcbp0[_chnl] * m_rca;
bp = qBound( -1.0f, bp, 1.0f );
m_rclast0[_chnl] = in;
m_rchp0[_chnl] = hp;
m_rcbp0[_chnl] = bp;
// second stage gets the output of the first stage as input...
in = m_type == Highpass_RC24
? hp + m_rcbp1[_chnl] * m_rcq
: bp + m_rcbp1[_chnl] * m_rcq;
in = qBound( -1.0f, in, 1.0f );
hp = m_rcc * ( m_rchp1[_chnl] + in - m_rclast1[_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_rcb + m_rcbp1[_chnl] * m_rca;
bp = qBound( -1.0f, bp, 1.0f );
m_rclast1[_chnl] = in;
m_rchp1[_chnl] = hp;
m_rcbp1[_chnl] = bp;
}
return m_type == Highpass_RC24 ? hp : bp;
}
case Formantfilter:
case FastFormant:
{
if( qAbs( _in0 ) < 1.0e-10f && qAbs( m_vflast[0][_chnl] ) < 1.0e-10f ) { return 0.0f; } // performance hack - skip processing when the numbers get too small
sample_t hp, bp, in;
out = 0;
const int os = m_type == FastFormant ? 1 : 4; // no oversampling for fast formant
for( int o = 0; o < os; ++o )
{
// first formant
in = _in0 + m_vfbp[0][_chnl] * m_vfq;
in = qBound( -1.0f, in, 1.0f );
hp = m_vfc[0] * ( m_vfhp[0][_chnl] + in - m_vflast[0][_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_vfb[0] + m_vfbp[0][_chnl] * m_vfa[0];
bp = qBound( -1.0f, bp, 1.0f );
m_vflast[0][_chnl] = in;
m_vfhp[0][_chnl] = hp;
m_vfbp[0][_chnl] = bp;
in = bp + m_vfbp[2][_chnl] * m_vfq;
in = qBound( -1.0f, in, 1.0f );
hp = m_vfc[0] * ( m_vfhp[2][_chnl] + in - m_vflast[2][_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_vfb[0] + m_vfbp[2][_chnl] * m_vfa[0];
bp = qBound( -1.0f, bp, 1.0f );
m_vflast[2][_chnl] = in;
m_vfhp[2][_chnl] = hp;
m_vfbp[2][_chnl] = bp;
in = bp + m_vfbp[4][_chnl] * m_vfq;
in = qBound( -1.0f, in, 1.0f );
hp = m_vfc[0] * ( m_vfhp[4][_chnl] + in - m_vflast[4][_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_vfb[0] + m_vfbp[4][_chnl] * m_vfa[0];
bp = qBound( -1.0f, bp, 1.0f );
m_vflast[4][_chnl] = in;
m_vfhp[4][_chnl] = hp;
m_vfbp[4][_chnl] = bp;
out += bp;
// second formant
in = _in0 + m_vfbp[0][_chnl] * m_vfq;
in = qBound( -1.0f, in, 1.0f );
hp = m_vfc[1] * ( m_vfhp[1][_chnl] + in - m_vflast[1][_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_vfb[1] + m_vfbp[1][_chnl] * m_vfa[1];
bp = qBound( -1.0f, bp, 1.0f );
m_vflast[1][_chnl] = in;
m_vfhp[1][_chnl] = hp;
m_vfbp[1][_chnl] = bp;
in = bp + m_vfbp[3][_chnl] * m_vfq;
in = qBound( -1.0f, in, 1.0f );
hp = m_vfc[1] * ( m_vfhp[3][_chnl] + in - m_vflast[3][_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_vfb[1] + m_vfbp[3][_chnl] * m_vfa[1];
bp = qBound( -1.0f, bp, 1.0f );
m_vflast[3][_chnl] = in;
m_vfhp[3][_chnl] = hp;
m_vfbp[3][_chnl] = bp;
in = bp + m_vfbp[5][_chnl] * m_vfq;
in = qBound( -1.0f, in, 1.0f );
hp = m_vfc[1] * ( m_vfhp[5][_chnl] + in - m_vflast[5][_chnl] );
hp = qBound( -1.0f, hp, 1.0f );
bp = hp * m_vfb[1] + m_vfbp[5][_chnl] * m_vfa[1];
bp = qBound( -1.0f, bp, 1.0f );
m_vflast[5][_chnl] = in;
m_vfhp[5][_chnl] = hp;
m_vfbp[5][_chnl] = bp;
out += bp;
}
return m_type == FastFormant ? out * 2.0f : out * 0.5f;
}
default:
out = m_biQuad.update( _in0, _chnl );
break;
}
if( m_doubleFilter )
{
return m_subFilter->update( out, _chnl );
}
// Clipper band limited sigmoid
return out;
}
inline void calcFilterCoeffs( float _freq, float _q )
{
// temp coef vars
_q = qMax( _q, minQ() );
if( m_type == Lowpass_RC12 ||
m_type == Bandpass_RC12 ||
m_type == Highpass_RC12 ||
m_type == Lowpass_RC24 ||
m_type == Bandpass_RC24 ||
m_type == Highpass_RC24 )
{
_freq = qBound( 50.0f, _freq, 20000.0f );
const float sr = m_sampleRatio * 0.25f;
const float f = 1.0f / ( _freq * F_2PI );
m_rca = 1.0f - sr / ( f + sr );
m_rcb = 1.0f - m_rca;
m_rcc = f / ( f + sr );
// Stretch Q/resonance, as self-oscillation reliably starts at a q of ~2.5 - ~2.6
m_rcq = _q * 0.25f;
return;
}
if( m_type == Formantfilter ||
m_type == FastFormant )
{
_freq = qBound( minFreq(), _freq, 20000.0f ); // limit freq and q for not getting bad noise out of the filter...
// formats for a, e, i, o, u, a
static const float _f[6][2] = { { 1000, 1400 }, { 500, 2300 },
{ 320, 3200 },
{ 500, 1000 },
{ 320, 800 },
{ 1000, 1400 } };
static const float freqRatio = 4.0f / 14000.0f;
// Stretch Q/resonance
m_vfq = _q * 0.25f;
// frequency in lmms ranges from 1Hz to 14000Hz
const float vowelf = _freq * freqRatio;
const int vowel = static_cast<int>( vowelf );
const float fract = vowelf - vowel;
// interpolate between formant frequencies
const float f0 = 1.0f / ( linearInterpolate( _f[vowel+0][0], _f[vowel+1][0], fract ) * F_2PI );
const float f1 = 1.0f / ( linearInterpolate( _f[vowel+0][1], _f[vowel+1][1], fract ) * F_2PI );
// samplerate coeff: depends on oversampling
const float sr = m_type == FastFormant ? m_sampleRatio : m_sampleRatio * 0.25f;
m_vfa[0] = 1.0f - sr / ( f0 + sr );
m_vfb[0] = 1.0f - m_vfa[0];
m_vfc[0] = f0 / ( f0 + sr );
m_vfa[1] = 1.0f - sr / ( f1 + sr );
m_vfb[1] = 1.0f - m_vfa[1];
m_vfc[1] = f1 / ( f1 + sr );
return;
}
if( m_type == Moog ||
m_type == DoubleMoog )
{
// [ 0 - 0.5 ]
const float f = qBound( minFreq(), _freq, 20000.0f ) * m_sampleRatio;
// (Empirical tunning)
m_p = ( 3.6f - 3.2f * f ) * f;
m_k = 2.0f * m_p - 1;
m_r = _q * powf( F_E, ( 1 - m_p ) * 1.386249f );
if( m_doubleFilter )
{
m_subFilter->m_r = m_r;
m_subFilter->m_p = m_p;
m_subFilter->m_k = m_k;
}
return;
}
if( m_type == Tripole )
{
const float f = qBound( 20.0f, _freq, 20000.0f ) * m_sampleRatio * 0.25f;
m_p = ( 3.6f - 3.2f * f ) * f;
m_k = 2.0f * m_p - 1.0f;
m_r = _q * 0.1f * powf( F_E, ( 1 - m_p ) * 1.386249f );
return;
}
if( m_type == Lowpass_SV ||
m_type == Bandpass_SV ||
m_type == Highpass_SV ||
m_type == Notch_SV )
{
const float f = sinf( qMax( minFreq(), _freq ) * m_sampleRatio * F_PI );
m_svf1 = qMin( f, 0.825f );
m_svf2 = qMin( f * 2.0f, 0.825f );
m_svq = qMax( 0.0001f, 2.0f - ( _q * 0.1995f ) );
return;
}
// other filters
_freq = qBound( minFreq(), _freq, 20000.0f );
const float omega = F_2PI * _freq * m_sampleRatio;
const float tsin = sinf( omega ) * 0.5f;
const float tcos = cosf( omega );
const float alpha = tsin / _q;
const float a0 = 1.0f / ( 1.0f + alpha );
const float a1 = -2.0f * tcos * a0;
const float a2 = ( 1.0f - alpha ) * a0;
switch( m_type )
{
case LowPass:
{
const float b1 = ( 1.0f - tcos ) * a0;
const float b0 = b1 * 0.5f;
m_biQuad.setCoeffs( a1, a2, b0, b1, b0 );
break;
}
case HiPass:
{
const float b1 = ( -1.0f - tcos ) * a0;
const float b0 = b1 * -0.5f;
m_biQuad.setCoeffs( a1, a2, b0, b1, b0 );
break;
}
case BandPass_CSG:
{
const float b0 = tsin * a0;
m_biQuad.setCoeffs( a1, a2, b0, 0.0f, -b0 );
break;
}
case BandPass_CZPG:
{
const float b0 = alpha * a0;
m_biQuad.setCoeffs( a1, a2, b0, 0.0f, -b0 );
break;
}
case Notch:
{
m_biQuad.setCoeffs( a1, a2, a0, a1, a0 );
break;
}
case AllPass:
{
m_biQuad.setCoeffs( a1, a2, a2, a1, 1.0f );
break;
}
default:
break;
}
if( m_doubleFilter )
{
m_subFilter->m_biQuad.setCoeffs( m_biQuad.m_a1, m_biQuad.m_a2, m_biQuad.m_b0, m_biQuad.m_b1, m_biQuad.m_b2 );
}
}
private:
// biquad filter
BiQuad<CHANNELS> m_biQuad;
// coeffs for moog-filter
float m_r, m_p, m_k;
// coeffs for RC-type-filters
float m_rca, m_rcb, m_rcc, m_rcq;
// coeffs for formant-filters
float m_vfa[4], m_vfb[4], m_vfc[4], m_vfq;
// coeffs for Lowpass_SV (state-variant lowpass)
float m_svf1, m_svf2, m_svq;
typedef sample_t frame[CHANNELS];
// in/out history for moog-filter
frame m_y1, m_y2, m_y3, m_y4, m_oldx, m_oldy1, m_oldy2, m_oldy3;
// additional one for Tripole filter
frame m_last;
// in/out history for RC-type-filters
frame m_rcbp0, m_rclp0, m_rchp0, m_rclast0;
frame m_rcbp1, m_rclp1, m_rchp1, m_rclast1;
// in/out history for Formant-filters
frame m_vfbp[6], m_vfhp[6], m_vflast[6];
// in/out history for Lowpass_SV (state-variant lowpass)
frame m_delay1, m_delay2, m_delay3, m_delay4;
FilterTypes m_type;
bool m_doubleFilter;
float m_sampleRate;
float m_sampleRatio;
BasicFilters<CHANNELS> * m_subFilter;
} ;
#endif
|