Initial multi-band EQ support for software codec platforms. Now go start

making that user interface!


git-svn-id: svn://svn.rockbox.org/rockbox/trunk@8478 a1c6a512-1295-4272-9138-f99709370657
This commit is contained in:
Thom Johansen 2006-01-29 02:10:28 +00:00
parent b0302f0cbb
commit a8cc6a7454
4 changed files with 305 additions and 0 deletions

View file

@ -76,4 +76,8 @@ playback.c
metadata.c metadata.c
codecs.c codecs.c
dsp.c dsp.c
eq.c
#if defined(CPU_COLDFIRE) && !defined(SIMULATOR)
eq_cf.S
#endif
#endif #endif

View file

@ -19,6 +19,7 @@
#include <inttypes.h> #include <inttypes.h>
#include <string.h> #include <string.h>
#include "dsp.h" #include "dsp.h"
#include "eq.h"
#include "kernel.h" #include "kernel.h"
#include "playback.h" #include "playback.h"
#include "system.h" #include "system.h"
@ -166,10 +167,21 @@ struct crossfeed_data
int index; int index;
}; };
/* Current setup is one lowshelf filters, three peaking filters and one
highshelf filter. Varying the number of shelving filters make no sense,
but adding peaking filters are possible. */
struct eq_state {
char enabled[5]; /* Flags for active filters */
struct eqfilter ls;
struct eqfilter pk[3];
struct eqfilter hs;
};
static struct dsp_config dsp_conf[2] IBSS_ATTR; static struct dsp_config dsp_conf[2] IBSS_ATTR;
static struct dither_data dither_data[2] IBSS_ATTR; static struct dither_data dither_data[2] IBSS_ATTR;
static struct resample_data resample_data[2] IBSS_ATTR; static struct resample_data resample_data[2] IBSS_ATTR;
struct crossfeed_data crossfeed_data IBSS_ATTR; struct crossfeed_data crossfeed_data IBSS_ATTR;
static struct eq_state eq_data;
static int pitch_ratio = 1000; static int pitch_ratio = 1000;
@ -608,6 +620,25 @@ static void apply_crossfeed(long* src[], int count)
} }
#endif #endif
/* Apply EQ filters to those bands that have got it switched on. */
void eq_process(long **x, unsigned num)
{
int i;
unsigned int channels = dsp->stereo_mode != STEREO_MONO ? 2 : 1;
/* filter configuration currently is 1 low shelf filter, 3 band peaking
filters and 1 high shelf filter, in that order.
*/
if (eq_data.enabled[0])
eq_filter(x, &eq_data.ls, num, channels, EQ_SHELF_SHIFT);
for (i = 0; i < 3; i++) {
if (eq_data.enabled[1 + i])
eq_filter(x, &eq_data.pk[i], num, channels, EQ_PEAK_SHIFT);
}
if (eq_data.enabled[4])
eq_filter(x, &eq_data.hs, num, channels, EQ_SHELF_SHIFT);
}
/* Apply a constant gain to the samples (e.g., for ReplayGain). May update /* Apply a constant gain to the samples (e.g., for ReplayGain). May update
* the src array if gain was applied. * the src array if gain was applied.
* Note that this must be called before the resampler. * Note that this must be called before the resampler.
@ -713,6 +744,9 @@ long dsp_process(char* dst, char* src[], long size)
samples = resample(tmp, samples); samples = resample(tmp, samples);
if (dsp->crossfeed_enabled && dsp->stereo_mode != STEREO_MONO) if (dsp->crossfeed_enabled && dsp->stereo_mode != STEREO_MONO)
apply_crossfeed(tmp, samples); apply_crossfeed(tmp, samples);
/* TODO: Might want to wrap this with a generic eq_enabled when the
settings are in place */
eq_process(tmp, samples);
write_samples((short*) dst, tmp, samples); write_samples((short*) dst, tmp, samples);
written += samples; written += samples;
dst += samples * sizeof(short) * 2; dst += samples * sizeof(short) * 2;

226
apps/eq.c Normal file
View file

@ -0,0 +1,226 @@
/***************************************************************************
* __________ __ ___.
* Open \______ \ ____ ____ | | _\_ |__ _______ ___
* Source | _// _ \_/ ___\| |/ /| __ \ / _ \ \/ /
* Jukebox | | ( <_> ) \___| < | \_\ ( <_> > < <
* Firmware |____|_ /\____/ \___ >__|_ \|___ /\____/__/\_ \
* \/ \/ \/ \/ \/
* $Id$
*
* Copyright (C) 2006 Thom Johansen
*
* All files in this archive are subject to the GNU General Public License.
* See the file COPYING in the source tree root for full license agreement.
*
* This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
* KIND, either express or implied.
*
****************************************************************************/
#include "config.h"
#include "eq.h"
/* Coef calculation taken from Audio-EQ-Cookbook.txt by Robert Bristow-Johnson.
Slightly faster calculation can be done by deriving forms which use tan()
instead of cos() and sin(), but the latter are far easier to use when doing
fixed point math, and performance is not a big point in the calculation part.
We realise the filters as a second order direct form 1 structure. Direct
form 1 was chosen because of better numerical properties for fixed point
implementations.
*/
#define DIV64(x, y, z) (long)(((long long)(x) << (z))/(y))
/* TODO: This macro requires the EMAC unit to be in fractional mode
when the coef generator routines are called. If this can be guaranteeed,
then remove the "&& 0" below for faster coef calculation on Coldfire.
*/
#if defined(CPU_COLDFIRE) && !defined(SIMULATOR) && 0
#define FRACMUL(x, y) \
({ \
long t; \
asm volatile ("mac.l %[a], %[b], %%acc0\n\t" \
"movclr.l %%acc0, %[t]\n\t" \
: [t] "=r" (t) : [a] "r" (x), [b] "r" (y)); \
t; \
})
#else
#define FRACMUL(x, y) ((long)(((((long long) (x)) * ((long long) (y))) >> 31)))
#endif
/* TODO: replaygain.c has some fixed point routines. perhaps we could reuse
them? */
/* 128 sixteen bit sine samples + guard point */
short sinetab[] = {
0, 1607, 3211, 4807, 6392, 7961, 9511, 11038, 12539, 14009, 15446, 16845,
18204, 19519, 20787, 22004, 23169, 24278, 25329, 26318, 27244, 28105, 28897,
29621, 30272, 30851, 31356, 31785, 32137, 32412, 32609,32727, 32767, 32727,
32609, 32412, 32137, 31785, 31356, 30851, 30272, 29621, 28897, 28105, 27244,
26318, 25329, 24278, 23169, 22004, 20787, 19519, 18204, 16845, 15446, 14009,
12539, 11038, 9511, 7961, 6392, 4807, 3211, 1607, 0, -1607, -3211, -4807,
-6392, -7961, -9511, -11038, -12539, -14009, -15446, -16845, -18204, -19519,
-20787, -22004, -23169, -24278, -25329, -26318, -27244, -28105, -28897,
-29621, -30272, -30851, -31356, -31785, -32137, -32412, -32609, -32727,
-32767, -32727, -32609, -32412, -32137, -31785, -31356, -30851, -30272,
-29621, -28897, -28105, -27244, -26318, -25329, -24278, -23169, -22004,
-20787, -19519, -18204, -16845, -15446, -14009, -12539, -11038, -9511,
-7961, -6392, -4807, -3211, -1607, 0
};
/* Good quality sine calculated by linearly interpolating
* a 128 sample sine table. First harmonic has amplitude of about -84 dB.
* phase has range from 0 to 0xffffffff, representing 0 and
* 2*pi respectively.
* Return value is a signed value from LONG_MIN to LONG_MAX, representing
* -1 and 1 respectively.
*/
static long fsin(unsigned long phase)
{
unsigned int pos = phase >> 25;
unsigned short frac = (phase & 0x01ffffff) >> 9;
short diff = sinetab[pos + 1] - sinetab[pos];
return (sinetab[pos] << 16) + frac*diff;
}
static inline long fcos(unsigned long phase)
{
return fsin(phase + 0xffffffff/4);
}
/* Fixed point square root via Newton-Raphson.
* Output is in same fixed point format as input.
* fracbits specifies number of fractional bits in argument.
*/
static long fsqrt(long a, unsigned int fracbits)
{
long b = a/2 + (1 << fracbits); /* initial approximation */
unsigned n;
const unsigned iterations = 4;
for (n = 0; n < iterations; ++n)
b = (b + DIV64(a, b, fracbits))/2;
return b;
}
short dbtoatab[49] = {
2058, 2180, 2309, 2446, 2591, 2744, 2907, 3079, 3261, 3455, 3659, 3876,
4106, 4349, 4607, 4880, 5169, 5475, 5799, 6143, 6507, 6893, 7301, 7734,
8192, 8677, 9192, 9736, 10313, 10924, 11572, 12257, 12983, 13753, 14568,
15431, 16345, 17314, 18340, 19426, 20577, 21797, 23088, 24456, 25905, 27440,
29066, 30789, 32613
};
/* Function for converting dB to squared amplitude factor (A = 10^(dB/40)).
Range is -24 to 24 dB. If gain values outside of this is needed, the above
table needs to be extended.
Parameter format is s15.16 fixed point. Return format is s2.29 fixed point.
*/
static long dbtoA(long db)
{
const unsigned long bias = 24 << 16;
unsigned short frac = (db + bias) & 0x0000ffff;
unsigned short pos = (db + bias) >> 16;
short diff = dbtoatab[pos + 1] - dbtoatab[pos];
return (dbtoatab[pos] << 16) + frac*diff;
}
/* Calculate second order section peaking filter coefficients.
cutoff is a value from 0 to 0xffffffff, where 0 represents 0 hz and
0xffffffff represents nyquist (samplerate/2).
Q is an unsigned 6.26 fixed point number, lower bound is artificially set
at 0.5.
db is s15.16 fixed point and describes gain/attenuation at peak freq.
c is a pointer where the coefs will be stored.
*/
void eq_pk_coefs(unsigned long cutoff, unsigned long Q, long db, long *c)
{
const long one = 1 << 28; /* s3.28 */
const long A = dbtoA(db);
const long alpha = DIV64(fsin(cutoff), 2*Q, 25); /* s1.30 */
long a0, a1, a2; /* these are all s3.28 format */
long b0, b1, b2;
/* possible numerical ranges listed after each coef */
b0 = one + FRACMUL(alpha, A); /* [1.25..5] */
b1 = a1 = -2*(fcos(cutoff) >> 3); /* [-2..2] */
b2 = one - FRACMUL(alpha, A); /* [-3..0.75] */
a0 = one + DIV64(alpha, A, 27); /* [1.25..5] */
a2 = one - DIV64(alpha, A, 27); /* [-3..0.75] */
c[0] = DIV64(b0, a0, 28);
c[1] = DIV64(b1, a0, 28);
c[2] = DIV64(b2, a0, 28);
c[3] = DIV64(a1, a0, 28);
c[4] = DIV64(a2, a0, 28);
}
/* Calculate coefficients for lowshelf filter */
void eq_ls_coefs(unsigned long cutoff, unsigned long Q, long db, long *c)
{
const long one = 1 << 24; /* s7.24 */
const long A = dbtoA(db);
const long alpha = DIV64(fsin(cutoff), 2*Q, 25); /* s1.30 */
const long ap1 = (A >> 5) + one;
const long am1 = (A >> 5) - one;
const long twosqrtalpha = 2*(FRACMUL(fsqrt(A >> 5, 24), alpha) << 1);
long a0, a1, a2; /* these are all s7.24 format */
long b0, b1, b2;
long cs = fcos(cutoff);
b0 = FRACMUL(A, ap1 - FRACMUL(am1, cs) + twosqrtalpha) << 2;
b1 = FRACMUL(A, am1 - FRACMUL(ap1, cs)) << 3;
b2 = FRACMUL(A, ap1 - FRACMUL(am1, cs) - twosqrtalpha) << 2;
a0 = ap1 + FRACMUL(am1, cs) + twosqrtalpha;
a1 = -2*((am1 + FRACMUL(ap1, cs)));
a2 = ap1 + FRACMUL(am1, cs) - twosqrtalpha;
c[0] = DIV64(b0, a0, 24);
c[1] = DIV64(b1, a0, 24);
c[2] = DIV64(b2, a0, 24);
c[3] = DIV64(a1, a0, 24);
c[4] = DIV64(a2, a0, 24);
}
/* Calculate coefficients for highshelf filter */
void eq_hs_coefs(unsigned long cutoff, unsigned long Q, long db, long *c)
{
const long one = 1 << 24; /* s7.24 */
const long A = dbtoA(db);
const long alpha = DIV64(fsin(cutoff), 2*Q, 25); /* s1.30 */
const long ap1 = (A >> 5) + one;
const long am1 = (A >> 5) - one;
const long twosqrtalpha = 2*(FRACMUL(fsqrt(A >> 5, 24), alpha) << 1);
long a0, a1, a2; /* these are all s7.24 format */
long b0, b1, b2;
long cs = fcos(cutoff);
b0 = FRACMUL(A, ap1 + FRACMUL(am1, cs) + twosqrtalpha) << 2;
b1 = -FRACMUL(A, am1 + FRACMUL(ap1, cs)) << 3;
b2 = FRACMUL(A, ap1 + FRACMUL(am1, cs) - twosqrtalpha) << 2;
a0 = ap1 - FRACMUL(am1, cs) + twosqrtalpha;
a1 = 2*((am1 - FRACMUL(ap1, cs)));
a2 = ap1 - FRACMUL(am1, cs) - twosqrtalpha;
c[0] = DIV64(b0, a0, 24);
c[1] = DIV64(b1, a0, 24);
c[2] = DIV64(b2, a0, 24);
c[3] = DIV64(a1, a0, 24);
c[4] = DIV64(a2, a0, 24);
}
#if !defined(CPU_COLDFIRE) || defined(SIMULATOR)
void eq_filter(long **x, struct eqfilter *f, unsigned num,
unsigned channels, unsigned shift)
{
/* TODO: Implement generic filtering routine. */
(void)x;
(void)f;
(void)num;
(void)channels;
(void)shift;
}
#endif

41
apps/eq.h Normal file
View file

@ -0,0 +1,41 @@
/***************************************************************************
* __________ __ ___.
* Open \______ \ ____ ____ | | _\_ |__ _______ ___
* Source | _// _ \_/ ___\| |/ /| __ \ / _ \ \/ /
* Jukebox | | ( <_> ) \___| < | \_\ ( <_> > < <
* Firmware |____|_ /\____/ \___ >__|_ \|___ /\____/__/\_ \
* \/ \/ \/ \/ \/
* $Id$
*
* Copyright (C) 2006 Thom Johansen
*
* All files in this archive are subject to the GNU General Public License.
* See the file COPYING in the source tree root for full license agreement.
*
* This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
* KIND, either express or implied.
*
****************************************************************************/
#ifndef _EQ_H
#define _EQ_H
/* These depend on the fixed point formats used by the different filter types
and need to be changed when they change.
*/
#define EQ_PEAK_SHIFT 3
#define EQ_SHELF_SHIFT 7
struct eqfilter {
long coefs[5]; /* Order is b0, b1, b2, a1, a2 */
long history[2][4];
};
void eq_pk_coefs(unsigned long cutoff, unsigned long Q, long db, long *c);
void eq_ls_coefs(unsigned long cutoff, unsigned long Q, long db, long *c);
void eq_hs_coefs(unsigned long cutoff, unsigned long Q, long db, long *c);
void eq_filter(long **x, struct eqfilter *f, unsigned num,
unsigned samples, unsigned shift);
#endif