zrythm/inc/utils/math.h

279 lines
5.6 KiB
C

// SPDX-FileCopyrightText: © 2018-2023 Alexandros Theodotou <alex@zrythm.org>
// SPDX-License-Identifier: LicenseRef-ZrythmLicense
/*
* This file incorporates work covered by the following copyright and
* permission notices:
*
* ---
*
* Copyright (C) 2017-2019 Robin Gareus <robin@gareus.org>
*
* 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; if not, write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
*
* ---
*/
/**
* \file
*
* Math utils.
*
* For more, look at libs/pbd/pbd/control_math.h in
* ardour.
*/
#ifndef __UTILS_MATH_H__
#define __UTILS_MATH_H__
#include <math.h>
#include <stdbool.h>
#include <stdint.h>
#include "utils/types.h"
#include <float.h>
/**
* @addtogroup utils
*
* @{
*/
/**
* Frames to skip when calculating the RMS.
*
* The lower the more CPU intensive.
*/
#define MATH_RMS_FRAMES 1
/** Tiny number to be used for denormaml prevention
* (-140dB). */
#define MATH_TINY_NUMBER (0.0000001)
#define MATH_MINUS_INFINITY (-HUGE_VAL)
/**
* Checks if 2 doubles are equal.
*
* @param e The allowed difference (epsilon).
*/
#define math_floats_equal_epsilon(a, b, e) \
((a) > (b) ? (a) - (b) < e : (b) - (a) < e)
#define math_doubles_equal_epsilon math_floats_equal_epsilon
/**
* Checks if 2 doubles are equal.
*/
#define math_floats_equal(a, b) \
((a) > (b) \
? (a) - (b) < FLT_EPSILON \
: (b) - (a) < FLT_EPSILON)
#define math_doubles_equal(a, b) \
((a) > (b) \
? (a) - (b) < DBL_EPSILON \
: (b) - (a) < DBL_EPSILON)
/**
* Rounds a double to a (minimum) signed 32-bit
* integer.
*/
#define math_round_double_to_signed_32(x) (lround (x))
/**
* Rounds a double to a (minimum) signed 64-bit
* integer.
*/
#define math_round_double_to_signed_64(x) (llround (x))
#define math_round_double_to_signed_frame_t(x) \
math_round_double_to_signed_64 (x)
/**
* Rounds a float to a (minimum) signed 32-bit
* integer.
*/
#define math_round_float_to_signed_32(x) (lroundf (x))
/**
* Rounds a float to a (minimum) signed 64-bit
* integer.
*/
#define math_round_float_to_signed_64(x) (llroundf (x))
#define math_round_float_to_signed_frame_t(x) \
math_round_float_to_signed_64 (x)
/**
* Fast log calculation to be used where precision
* is not required (like log curves).
*
* Taken from ardour from code in the public domain.
*/
CONST
static inline float
math_fast_log2 (float val)
{
union
{
float f;
int i;
} t;
t.f = val;
int * const exp_ptr = &t.i;
int x = *exp_ptr;
const float log_2 = (float) (((x >> 23) & 255) - 128);
x &= ~(255 << 23);
x += 127 << 23;
*exp_ptr = x;
val = ((-1.0f / 3) * t.f + 2) * t.f - 2.0f / 3;
return (val + log_2);
}
CONST
static inline float
math_fast_log (const float val)
{
return (math_fast_log2 (val) * 0.69314718f);
}
CONST
static inline float
math_fast_log10 (const float val)
{
return math_fast_log2 (val) / 3.312500f;
}
/**
* Returns fader value 0.0 to 1.0 from amp value
* 0.0 to 2.0 (+6 dbFS).
*/
CONST
static inline sample_t
math_get_fader_val_from_amp (sample_t amp)
{
static const float fader_coefficient1 =
/*192.f * logf (2.f);*/
133.084258667509499408f;
static const float fader_coefficient2 =
/*powf (logf (2.f), 8.f) * powf (198.f, 8.f);*/
1.25870863180257576e17f;
/* to prevent weird values when amp is very
* small */
if (amp <= 0.00001f)
{
return 1e-20f;
}
else
{
if (math_floats_equal (amp, 1.f))
{
amp = 1.f + 1e-20f;
}
sample_t fader =
powf (
6.f * math_fast_log (amp) + fader_coefficient1, 8.f)
/ fader_coefficient2;
return (sample_t) fader;
}
}
/**
* Returns amp value 0.0 to 2.0 (+6 dbFS) from
* fader value 0.0 to 1.0.
*/
CONST
static inline sample_t
math_get_amp_val_from_fader (sample_t fader)
{
static const float val1 = 1.f / 6.f;
return powf (
2.f, (val1) * (-192.f + 198.f * powf (fader, 1.f / 8.f)));
}
/**
* Convert from amplitude 0.0 to 2.0 to dbFS.
*/
CONST
static inline sample_t
math_amp_to_dbfs (sample_t amp)
{
return 20.f * log10f (amp);
}
sample_t
math_calculate_rms_amp (
sample_t * buf,
const nframes_t nframes);
/**
* Calculate db using RMS method.
*
* @param buf Buffer containing the samples.
* @param nframes Number of samples.
*/
sample_t
math_calculate_rms_db (sample_t * buf, const nframes_t nframes);
/**
* Convert form dbFS to amplitude 0.0 to 2.0.
*/
CONST
static inline sample_t
math_dbfs_to_amp (sample_t dbfs)
{
return powf (10.f, (dbfs / 20.f));
}
/**
* Convert form dbFS to fader val 0.0 to 1.0.
*/
CONST
static inline sample_t
math_dbfs_to_fader_val (sample_t dbfs)
{
return math_get_fader_val_from_amp (math_dbfs_to_amp (dbfs));
}
/**
* Asserts that the value is non-nan.
*
* Not real-time safe.
*
* @return Whether the value is valid (nonnan).
*/
bool
math_assert_nonnann (float x);
/**
* Returns whether the given string is a valid float.
*
* @param ret If non-NULL, the result will be placed here.
*/
NONNULL_ARGS (1)
bool math_is_string_valid_float (const char * str, float * ret);
/**
* @}
*/
#endif