Average Error: 32.1 → 7.1
Time: 3.5s
Precision: binary64
\[[re, im]=\mathsf{sort}([re, im])\]
\[\log \left(\sqrt{re \cdot re + im \cdot im}\right)\]
↓
\[\begin{array}{l} \mathbf{if}\;re \leq -8.383593090521001 \cdot 10^{+110}:\\ \;\;\;\;\log \left(-re\right)\\ \mathbf{elif}\;re \leq -8.647414243699812 \cdot 10^{-165}:\\ \;\;\;\;\log \left(\sqrt{re \cdot re + im \cdot im}\right)\\ \mathbf{else}:\\ \;\;\;\;\log im\\ \end{array}\]
\log \left(\sqrt{re \cdot re + im \cdot im}\right)
↓
\begin{array}{l}
\mathbf{if}\;re \leq -8.383593090521001 \cdot 10^{+110}:\\
\;\;\;\;\log \left(-re\right)\\

\mathbf{elif}\;re \leq -8.647414243699812 \cdot 10^{-165}:\\
\;\;\;\;\log \left(\sqrt{re \cdot re + im \cdot im}\right)\\

\mathbf{else}:\\
\;\;\;\;\log im\\

\end{array}
(FPCore (re im) :precision binary64 (log (sqrt (+ (* re re) (* im im)))))
↓
(FPCore (re im)
 :precision binary64
 (if (<= re -8.383593090521001e+110)
   (log (- re))
   (if (<= re -8.647414243699812e-165)
     (log (sqrt (+ (* re re) (* im im))))
     (log im))))
double code(double re, double im) {
	return log(sqrt((re * re) + (im * im)));
}
↓
double code(double re, double im) {
	double tmp;
	if (re <= -8.383593090521001e+110) {
		tmp = log(-re);
	} else if (re <= -8.647414243699812e-165) {
		tmp = log(sqrt((re * re) + (im * im)));
	} else {
		tmp = log(im);
	}
	return tmp;
}

Error

Bits error versus re

Bits error versus im

Try it out

Your Program's Arguments

Results

Enter valid numbers for all inputs

Derivation

  1. Split input into 3 regimes

  2. if re < -8.3835930905210013e110

    1. Initial program 53.1

      \[\log \left(\sqrt{re \cdot re + im \cdot im}\right)\]

    2. Taylor expanded around -inf 5.4

      \[\leadsto \log \color{blue}{\left(-1 \cdot re\right)}\]

    if -8.3835930905210013e110 < re < -8.64741424369981227e-165

    1. Initial program 11.7

      \[\log \left(\sqrt{re \cdot re + im \cdot im}\right)\]

    if -8.64741424369981227e-165 < re

    1. Initial program 32.7

      \[\log \left(\sqrt{re \cdot re + im \cdot im}\right)\]

    2. Taylor expanded around 0 7.0

      \[\leadsto \log \color{blue}{\left(0.5 \cdot \frac{{re}^{2}}{im} + im\right)}\]

    3. Simplified7.0

      \[\leadsto \log \color{blue}{\left(im + 0.5 \cdot \frac{re \cdot re}{im}\right)}\]

    4. Taylor expanded around 0 4.6

      \[\leadsto \color{blue}{\log im}\]
  3. Recombined 3 regimes into one program.

  4. Final simplification7.1

    \[\leadsto \begin{array}{l} \mathbf{if}\;re \leq -8.383593090521001 \cdot 10^{+110}:\\ \;\;\;\;\log \left(-re\right)\\ \mathbf{elif}\;re \leq -8.647414243699812 \cdot 10^{-165}:\\ \;\;\;\;\log \left(\sqrt{re \cdot re + im \cdot im}\right)\\ \mathbf{else}:\\ \;\;\;\;\log im\\ \end{array}\]

Reproduce

herbie shell --seed 2021098 
(FPCore (re im)
  :name "math.log/1 on complex, real part"
  :precision binary64
  (log (sqrt (+ (* re re) (* im im)))))