invcot (example 3.9)

Average Error: 59.9 → 0.3

Time: 12.1s

Precision: 64

\[-0.0259999999999999988 \lt x \land x \lt 0.0259999999999999988\]

Math

\[\frac{1}{x} - \frac{1}{\tan x}\]

↓

\[\mathsf{fma}\left(0.0222222222222222231, {x}^{3}, \mathsf{fma}\left(0.00211640211640211654, {x}^{5}, 0.333333333333333315 \cdot x\right)\right)\]

C

double code(double x) {
	return ((1.0 / x) - (1.0 / tan(x)));
}

↓

double code(double x) {
	return fma(0.022222222222222223, pow(x, 3.0), fma(0.0021164021164021165, pow(x, 5.0), (0.3333333333333333 * x)));
}

Error

Bits error versus x

Target

Original	59.9
Target	0.1
Herbie	0.3

\[\begin{array}{l} \mathbf{if}\;\left|x\right| \lt 0.0259999999999999988:\\ \;\;\;\;\frac{x}{3} \cdot \left(1 + \frac{x \cdot x}{15}\right)\\ \mathbf{else}:\\ \;\;\;\;\frac{1}{x} - \frac{1}{\tan x}\\ \end{array}\]

Derivation

1. Initial program 59.9

   \[\frac{1}{x} - \frac{1}{\tan x}\]

2. Taylor expanded around 0 0.3

   \[\leadsto \color{blue}{0.0222222222222222231 \cdot {x}^{3} + \left(0.00211640211640211654 \cdot {x}^{5} + 0.333333333333333315 \cdot x\right)}\]

3. Simplified 0.3

   \[\leadsto \color{blue}{\mathsf{fma}\left(0.0222222222222222231, {x}^{3}, \mathsf{fma}\left(0.00211640211640211654, {x}^{5}, 0.333333333333333315 \cdot x\right)\right)}\]

4. Final simplification 0.3

   \[\leadsto \mathsf{fma}\left(0.0222222222222222231, {x}^{3}, \mathsf{fma}\left(0.00211640211640211654, {x}^{5}, 0.333333333333333315 \cdot x\right)\right)\]

Reproduce

herbie shell --seed 2020106 +o rules:numerics
(FPCore (x)
  :name "invcot (example 3.9)"
  :precision binary64
  :pre (and (< -0.026 x) (< x 0.026))

  :herbie-target
  (if (< (fabs x) 0.026) (* (/ x 3) (+ 1 (/ (* x x) 15))) (- (/ 1 x) (/ 1 (tan x))))

  (- (/ 1 x) (/ 1 (tan x))))