surge-math 0.2.6-alpha.0

surge synthesizer -- dsp math used in synthesis
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

crate::ix!();

/// y = x - (4/27)*x^3,  x € [-1.5 .. 1.5]
#[inline] 
#[cfg(target_arch = "x86_64")] 
pub fn softclip_ss(x_in: __m128) -> __m128
{
   let a: __m128     = unsafe{ _mm_set_ss(-4.0 / 27.0) };

   let x_min: __m128 = unsafe{ _mm_set_ss(-1.5) };
   let x_max: __m128 = unsafe{ _mm_set_ss(1.5) };

   let x: __m128     = unsafe{ _mm_max_ss(_mm_min_ss(x_in, x_max), x_min) };
   let xx: __m128    = unsafe{ _mm_mul_ss(x, x) };

   let mut t: __m128 = unsafe{ _mm_mul_ss(x, a) };
   t = unsafe{ _mm_mul_ss(t, xx) };
   t = unsafe{ _mm_add_ss(t, x) };
   t
}

/// y = x - (4/27)*x^3,  x € [-1.5 .. 1.5]
#[inline] 
#[cfg(target_arch = "x86_64")] 
pub fn softclip_ps(x_in: __m128) -> __m128
{
    unsafe {
        let a: __m128 = _mm_set1_ps(-4.0 / 27.0);

        let x_min: __m128 = _mm_set1_ps(-1.5);
        let x_max: __m128 = _mm_set1_ps(1.5);

        let x: __m128     = _mm_max_ps(_mm_min_ps(x_in, x_max), x_min);
        let xx: __m128    = _mm_mul_ps(x, x);

        let mut t: __m128 = _mm_mul_ps(x, a);
        t =  _mm_mul_ps(t, xx);
        t =  _mm_add_ps(t, x);
        t
    }
}

#[inline] pub fn softclip_f64(input: f64 ) -> f64
{
    let c0: f64 = 4.0 / 27.0;
    let mut x: f64 = match input > -1.5 { true => input, false => -1.5 };
    x = match x < 1.5  { true => x , false =>  1.5 };
    let ax: f64 = c0 * x;
    let xx: f64 = x * x;
    x - ax * xx
}

#[inline] pub fn softclip4_f64(input: f64 ) -> f64
{
   let mut x: f64 = match input > -6.0 { true => input, false => -6.0 };
   x = match x < 6.0 { true => x, false => 6.0 };
   let ax: f64 = 0.0023148148148 * x; // 1.0 / 27*16
   let xx: f64 = x * x;
   x - ax * xx
}

#[inline] pub fn softclip8_f64(input: f64 ) -> f64
{
   let mut x: f64 = match input > -12.0 { true => input, false => -12.0 };
   x = match x < 12.0 { true => x, false => 12.0 };
   let ax: f64 = 0.00028935185185 * x; // 1.0 / 27*128
   let xx: f64 = x * x;
   x - ax * xx
}

#[inline] pub fn softclip2_f64(input: f64 ) -> f64
{
    let mut x: f64 = match input > -3.0 { true => input, false => -3.0 };

    x = match x < 3.0 { true => x, false => 3.0 };

    let ax: f64 = 
        0.185_185_185_185_185_17
        * x; // 1.0 / 27*2

    let xx: f64 = x * x;
    x - ax * xx
}

#[inline] 
pub fn softclip8_ps(input: __m128 ) -> __m128
{
    unsafe {

        let a: __m128 = _mm_set1_ps(
            -0.000_289_351_85
        );

        let x_min: __m128 = _mm_set1_ps(-12.0);
        let x_max: __m128 = _mm_set1_ps(12.0);

        let x: __m128 = _mm_max_ps(_mm_min_ps(input, x_max), x_min);
        let xx: __m128 = _mm_mul_ps(x, x);
        let mut t: __m128 = _mm_mul_ps(x, a);
        t = _mm_mul_ps(t, xx);
        t = _mm_add_ps(t, x);
        t
    }
}

/**
  |# Safety
  |
  |must be able to access nquads * 4 contiguous
  |values from the valid starting point "input"
  */
pub unsafe fn softclip_block<NQ>(input: *mut f32, nquads: NQ) 
    where 
        <NQ as std::convert::TryInto<usize>>::Error: std::fmt::Debug, 
        NQ: TryInto<usize>
{
    let nquads: usize = nquads.try_into().unwrap();

    // y = x - (4/27)*x^3,  x [-1.5 .. 1.5]
    let a:     __m128 = _mm_set1_ps(-4.0 / 27.0);
    let x_min: __m128 = _mm_set1_ps(-1.50);
    let x_max: __m128 = _mm_set1_ps(1.50);

    for i in (0_usize..nquads).step_by(2) 
    {
        let mut x:  [__m128; 2] = [z128![]; 2];
        let mut xx: [__m128; 2] = [z128![]; 2];
        let mut t:  [__m128; 2] = [z128![]; 2];

        x[0] = _mm_min_ps(_mm_load_ps(input.add(i << 2)), x_max);
        x[1] = _mm_min_ps(_mm_load_ps(input.add((i + 1) << 2)), x_max);

        x[0] = _mm_max_ps(x[0], x_min);
        x[1] = _mm_max_ps(x[1], x_min);

        xx[0] = _mm_mul_ps(x[0], x[0]);
        xx[1] = _mm_mul_ps(x[1], x[1]);

        t[0] = _mm_mul_ps(x[0], a);
        t[1] = _mm_mul_ps(x[1], a);

        t[0] = _mm_mul_ps(t[0], xx[0]);
        t[1] = _mm_mul_ps(t[1], xx[1]);

        t[0] = _mm_add_ps(t[0], x[0]);
        t[1] = _mm_add_ps(t[1], x[1]);

        _mm_store_ps(input.add(i << 2), t[0]);
        _mm_store_ps(input.add((i + 1) << 2), t[1]);
    }
}