Migdal et al, Equation (51)

Percentage Accurate: 99.4% → 99.4%
Time: 11.0s
Alternatives: 14
Speedup: 1.1×

Specification

?
\[\begin{array}{l} \\ \frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \end{array} \]
(FPCore (k n)
 :precision binary64
 (* (/ 1.0 (sqrt k)) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0))))
\begin{array}{l}

\\
\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}
\end{array}

Sampling outcomes in binary64 precision:

Local Percentage Accuracy vs ?

The average percentage accuracy by input value. Horizontal axis shows value of an input variable; the variable is choosen in the title. Vertical axis is accuracy; higher is better. Red represent the original program, while blue represents Herbie's suggestion. These can be toggled with buttons below the plot. The line is an average while dots represent individual samples.

Accuracy vs Speed?

Herbie found 14 alternatives:

AlternativeAccuracySpeedup
The accuracy (vertical axis) and speed (horizontal axis) of each alternatives. Up and to the right is better. The red square shows the initial program, and each blue circle shows an alternative.The line shows the best available speed-accuracy tradeoffs.

Initial Program: 99.4% accurate, 1.0× speedup?

\[\begin{array}{l} \\ \frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \end{array} \]
(FPCore (k n)
 :precision binary64
 (* (/ 1.0 (sqrt k)) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0))))
\begin{array}{l}

\\
\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}
\end{array}

Alternative 1: 99.4% accurate, 0.9× speedup?

\[\begin{array}{l} \\ \frac{\sqrt{\frac{2}{k}} \cdot \sqrt{n \cdot \mathsf{PI}\left(\right)}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(k \cdot 0.5\right)}} \end{array} \]
(FPCore (k n)
 :precision binary64
 (/ (* (sqrt (/ 2.0 k)) (sqrt (* n (PI)))) (pow (* (* 2.0 n) (PI)) (* k 0.5))))
\begin{array}{l}

\\
\frac{\sqrt{\frac{2}{k}} \cdot \sqrt{n \cdot \mathsf{PI}\left(\right)}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(k \cdot 0.5\right)}}
\end{array}
Derivation
  1. Initial program 99.5%

    \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
  2. Add Preprocessing
  3. Step-by-step derivation
    1. lift-/.f64N/A

      \[\leadsto \color{blue}{\frac{1}{\sqrt{k}}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
    2. inv-powN/A

      \[\leadsto \color{blue}{{\left(\sqrt{k}\right)}^{-1}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
    3. lift-sqrt.f64N/A

      \[\leadsto {\color{blue}{\left(\sqrt{k}\right)}}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
    4. pow1/2N/A

      \[\leadsto {\color{blue}{\left({k}^{\frac{1}{2}}\right)}}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
    5. pow-powN/A

      \[\leadsto \color{blue}{{k}^{\left(\frac{1}{2} \cdot -1\right)}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
    6. lower-pow.f64N/A

      \[\leadsto \color{blue}{{k}^{\left(\frac{1}{2} \cdot -1\right)}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
    7. metadata-eval99.5

      \[\leadsto {k}^{\color{blue}{-0.5}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
  4. Applied rewrites99.5%

    \[\leadsto \color{blue}{{k}^{-0.5}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
  5. Step-by-step derivation
    1. lift-*.f64N/A

      \[\leadsto \color{blue}{{k}^{\frac{-1}{2}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}} \]
    2. *-commutativeN/A

      \[\leadsto \color{blue}{{\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}}} \]
    3. lift-pow.f64N/A

      \[\leadsto \color{blue}{{\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}} \cdot {k}^{\frac{-1}{2}} \]
    4. lift-*.f64N/A

      \[\leadsto {\color{blue}{\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
    5. lift-*.f64N/A

      \[\leadsto {\left(\color{blue}{\left(2 \cdot \mathsf{PI}\left(\right)\right)} \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
    6. associate-*l*N/A

      \[\leadsto {\color{blue}{\left(2 \cdot \left(\mathsf{PI}\left(\right) \cdot n\right)\right)}}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
    7. *-commutativeN/A

      \[\leadsto {\left(2 \cdot \color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right)}\right)}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
    8. associate-*l*N/A

      \[\leadsto {\color{blue}{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
    9. lift-*.f64N/A

      \[\leadsto {\left(\color{blue}{\left(2 \cdot n\right)} \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
    10. lift-*.f64N/A

      \[\leadsto {\color{blue}{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
    11. lift-/.f64N/A

      \[\leadsto {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\color{blue}{\left(\frac{1 - k}{2}\right)}} \cdot {k}^{\frac{-1}{2}} \]
    12. lift--.f64N/A

      \[\leadsto {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{\color{blue}{1 - k}}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
    13. div-subN/A

      \[\leadsto {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\color{blue}{\left(\frac{1}{2} - \frac{k}{2}\right)}} \cdot {k}^{\frac{-1}{2}} \]
    14. metadata-evalN/A

      \[\leadsto {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\color{blue}{\frac{1}{2}} - \frac{k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
    15. pow-subN/A

      \[\leadsto \color{blue}{\frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}}} \cdot {k}^{\frac{-1}{2}} \]
    16. lift-pow.f64N/A

      \[\leadsto \frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot \color{blue}{{k}^{\frac{-1}{2}}} \]
    17. metadata-evalN/A

      \[\leadsto \frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot {k}^{\color{blue}{\left(\mathsf{neg}\left(\frac{1}{2}\right)\right)}} \]
    18. pow-flipN/A

      \[\leadsto \frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot \color{blue}{\frac{1}{{k}^{\frac{1}{2}}}} \]
    19. pow1/2N/A

      \[\leadsto \frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot \frac{1}{\color{blue}{\sqrt{k}}} \]
    20. lift-sqrt.f64N/A

      \[\leadsto \frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot \frac{1}{\color{blue}{\sqrt{k}}} \]
  6. Applied rewrites99.7%

    \[\leadsto \color{blue}{\frac{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)} \cdot 1}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}}} \]
  7. Step-by-step derivation
    1. lift-*.f64N/A

      \[\leadsto \frac{\color{blue}{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)} \cdot 1}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    2. *-rgt-identity99.7

      \[\leadsto \frac{\color{blue}{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    3. lift-sqrt.f64N/A

      \[\leadsto \frac{\color{blue}{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    4. lift-*.f64N/A

      \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    5. lift-*.f64N/A

      \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot 2\right)} \cdot \mathsf{PI}\left(\right)}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    6. associate-*l*N/A

      \[\leadsto \frac{\sqrt{\color{blue}{n \cdot \left(2 \cdot \mathsf{PI}\left(\right)\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    7. *-commutativeN/A

      \[\leadsto \frac{\sqrt{n \cdot \color{blue}{\left(\mathsf{PI}\left(\right) \cdot 2\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    8. associate-*r*N/A

      \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    9. lift-*.f64N/A

      \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot 2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    10. sqrt-prodN/A

      \[\leadsto \frac{\color{blue}{\sqrt{n \cdot \mathsf{PI}\left(\right)} \cdot \sqrt{2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    11. lift-sqrt.f64N/A

      \[\leadsto \frac{\color{blue}{\sqrt{n \cdot \mathsf{PI}\left(\right)}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    12. lift-sqrt.f64N/A

      \[\leadsto \frac{\sqrt{n \cdot \mathsf{PI}\left(\right)} \cdot \color{blue}{\sqrt{2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    13. lower-*.f6499.5

      \[\leadsto \frac{\color{blue}{\sqrt{n \cdot \mathsf{PI}\left(\right)} \cdot \sqrt{2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    14. lift-sqrt.f64N/A

      \[\leadsto \frac{\color{blue}{\sqrt{n \cdot \mathsf{PI}\left(\right)}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    15. pow1/2N/A

      \[\leadsto \frac{\color{blue}{{\left(n \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    16. lift-*.f64N/A

      \[\leadsto \frac{{\color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right)}}^{\frac{1}{2}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    17. *-commutativeN/A

      \[\leadsto \frac{{\color{blue}{\left(\mathsf{PI}\left(\right) \cdot n\right)}}^{\frac{1}{2}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    18. lift-*.f64N/A

      \[\leadsto \frac{{\color{blue}{\left(\mathsf{PI}\left(\right) \cdot n\right)}}^{\frac{1}{2}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    19. pow1/2N/A

      \[\leadsto \frac{\color{blue}{\sqrt{\mathsf{PI}\left(\right) \cdot n}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    20. lift-sqrt.f6499.5

      \[\leadsto \frac{\color{blue}{\sqrt{\mathsf{PI}\left(\right) \cdot n}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
  8. Applied rewrites99.5%

    \[\leadsto \frac{\color{blue}{\sqrt{\mathsf{PI}\left(\right) \cdot n} \cdot \sqrt{2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
  9. Step-by-step derivation
    1. lift-/.f64N/A

      \[\leadsto \color{blue}{\frac{\sqrt{\mathsf{PI}\left(\right) \cdot n} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}}} \]
    2. lift-*.f64N/A

      \[\leadsto \frac{\color{blue}{\sqrt{\mathsf{PI}\left(\right) \cdot n} \cdot \sqrt{2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
    3. lift-*.f64N/A

      \[\leadsto \frac{\sqrt{\mathsf{PI}\left(\right) \cdot n} \cdot \sqrt{2}}{\color{blue}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}}} \]
    4. times-fracN/A

      \[\leadsto \color{blue}{\frac{\sqrt{\mathsf{PI}\left(\right) \cdot n}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot \frac{\sqrt{2}}{\sqrt{k}}} \]
    5. *-commutativeN/A

      \[\leadsto \color{blue}{\frac{\sqrt{2}}{\sqrt{k}} \cdot \frac{\sqrt{\mathsf{PI}\left(\right) \cdot n}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}}} \]
    6. associate-*r/N/A

      \[\leadsto \color{blue}{\frac{\frac{\sqrt{2}}{\sqrt{k}} \cdot \sqrt{\mathsf{PI}\left(\right) \cdot n}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}}} \]
    7. lower-/.f64N/A

      \[\leadsto \color{blue}{\frac{\frac{\sqrt{2}}{\sqrt{k}} \cdot \sqrt{\mathsf{PI}\left(\right) \cdot n}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}}} \]
  10. Applied rewrites99.7%

    \[\leadsto \color{blue}{\frac{\sqrt{\frac{2}{k}} \cdot \sqrt{n \cdot \mathsf{PI}\left(\right)}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(k \cdot 0.5\right)}}} \]
  11. Add Preprocessing

Alternative 2: 42.3% accurate, 0.3× speedup?

\[\begin{array}{l} \\ \begin{array}{l} t_0 := {\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}\\ t_1 := \mathsf{fma}\left(k, n, n \cdot k\right)\\ \mathbf{if}\;t\_0 \leq 0:\\ \;\;\;\;\sqrt{\frac{t\_1}{k \cdot k}}\\ \mathbf{elif}\;t\_0 \leq 2 \cdot 10^{+154}:\\ \;\;\;\;\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}\\ \mathbf{else}:\\ \;\;\;\;\frac{\sqrt{t\_1}}{\left|k\right|}\\ \end{array} \end{array} \]
(FPCore (k n)
 :precision binary64
 (let* ((t_0
         (* (pow (sqrt k) -1.0) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0))))
        (t_1 (fma k n (* n k))))
   (if (<= t_0 0.0)
     (sqrt (/ t_1 (* k k)))
     (if (<= t_0 2e+154)
       (sqrt (* (* 2.0 (/ n k)) (PI)))
       (/ (sqrt t_1) (fabs k))))))
\begin{array}{l}

\\
\begin{array}{l}
t_0 := {\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}\\
t_1 := \mathsf{fma}\left(k, n, n \cdot k\right)\\
\mathbf{if}\;t\_0 \leq 0:\\
\;\;\;\;\sqrt{\frac{t\_1}{k \cdot k}}\\

\mathbf{elif}\;t\_0 \leq 2 \cdot 10^{+154}:\\
\;\;\;\;\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}\\

\mathbf{else}:\\
\;\;\;\;\frac{\sqrt{t\_1}}{\left|k\right|}\\


\end{array}
\end{array}
Derivation
  1. Split input into 3 regimes
  2. if (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 0.0

    1. Initial program 100.0%

      \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
    2. Add Preprocessing
    3. Taylor expanded in k around 0

      \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
    4. Step-by-step derivation
      1. *-commutativeN/A

        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
      2. lower-*.f64N/A

        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
      3. lower-sqrt.f64N/A

        \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
      4. lower-sqrt.f64N/A

        \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
      5. lower-/.f64N/A

        \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
      6. *-commutativeN/A

        \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
      7. lower-*.f64N/A

        \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
      8. lower-PI.f643.2

        \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
    5. Applied rewrites3.2%

      \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
    6. Step-by-step derivation
      1. Applied rewrites3.2%

        \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
      2. Applied rewrites3.2%

        \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
      3. Step-by-step derivation
        1. Applied rewrites10.2%

          \[\leadsto \sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}} \]

        if 0.0 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 2.00000000000000007e154

        1. Initial program 98.9%

          \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
        2. Add Preprocessing
        3. Taylor expanded in k around 0

          \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
        4. Step-by-step derivation
          1. *-commutativeN/A

            \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
          2. lower-*.f64N/A

            \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
          3. lower-sqrt.f64N/A

            \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
          4. lower-sqrt.f64N/A

            \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
          5. lower-/.f64N/A

            \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
          6. *-commutativeN/A

            \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
          7. lower-*.f64N/A

            \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
          8. lower-PI.f6494.8

            \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
        5. Applied rewrites94.8%

          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
        6. Step-by-step derivation
          1. Applied rewrites95.2%

            \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
          2. Applied rewrites95.2%

            \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]

          if 2.00000000000000007e154 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64))))

          1. Initial program 99.8%

            \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
          2. Add Preprocessing
          3. Taylor expanded in k around 0

            \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
          4. Step-by-step derivation
            1. *-commutativeN/A

              \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
            2. lower-*.f64N/A

              \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
            3. lower-sqrt.f64N/A

              \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
            4. lower-sqrt.f64N/A

              \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
            5. lower-/.f64N/A

              \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
            6. *-commutativeN/A

              \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
            7. lower-*.f64N/A

              \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
            8. lower-PI.f643.5

              \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
          5. Applied rewrites3.5%

            \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
          6. Step-by-step derivation
            1. Applied rewrites3.5%

              \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
            2. Applied rewrites3.5%

              \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
            3. Step-by-step derivation
              1. Applied rewrites8.2%

                \[\leadsto \frac{\sqrt{\mathsf{fma}\left(k, n, n \cdot k\right)}}{\color{blue}{\left|k\right|}} \]
            4. Recombined 3 regimes into one program.
            5. Final simplification40.7%

              \[\leadsto \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;\sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}}\\ \mathbf{elif}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 2 \cdot 10^{+154}:\\ \;\;\;\;\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}\\ \mathbf{else}:\\ \;\;\;\;\frac{\sqrt{\mathsf{fma}\left(k, n, n \cdot k\right)}}{\left|k\right|}\\ \end{array} \]
            6. Add Preprocessing

            Alternative 3: 74.8% accurate, 0.5× speedup?

            \[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;\sqrt{0}\\ \mathbf{else}:\\ \;\;\;\;\frac{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}{\sqrt{k}}\\ \end{array} \end{array} \]
            (FPCore (k n)
             :precision binary64
             (if (<=
                  (* (pow (sqrt k) -1.0) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0)))
                  0.0)
               (sqrt 0.0)
               (/ (sqrt (* (* n 2.0) (PI))) (sqrt k))))
            \begin{array}{l}
            
            \\
            \begin{array}{l}
            \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\
            \;\;\;\;\sqrt{0}\\
            
            \mathbf{else}:\\
            \;\;\;\;\frac{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}{\sqrt{k}}\\
            
            
            \end{array}
            \end{array}
            
            Derivation
            1. Split input into 2 regimes
            2. if (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 0.0

              1. Initial program 100.0%

                \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
              2. Add Preprocessing
              3. Taylor expanded in k around 0

                \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
              4. Step-by-step derivation
                1. *-commutativeN/A

                  \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                2. lower-*.f64N/A

                  \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                3. lower-sqrt.f64N/A

                  \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                4. lower-sqrt.f64N/A

                  \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                5. lower-/.f64N/A

                  \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                6. *-commutativeN/A

                  \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                7. lower-*.f64N/A

                  \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                8. lower-PI.f643.2

                  \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
              5. Applied rewrites3.2%

                \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
              6. Step-by-step derivation
                1. Applied rewrites3.2%

                  \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                2. Step-by-step derivation
                  1. Applied rewrites100.0%

                    \[\leadsto \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k} - \frac{\mathsf{PI}\left(\right) \cdot n}{k}} \]

                  if 0.0 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64))))

                  1. Initial program 99.3%

                    \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                  2. Add Preprocessing
                  3. Taylor expanded in k around 0

                    \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                  4. Step-by-step derivation
                    1. *-commutativeN/A

                      \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                    2. lower-*.f64N/A

                      \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                    3. lower-sqrt.f64N/A

                      \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                    4. lower-sqrt.f64N/A

                      \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                    5. lower-/.f64N/A

                      \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                    6. *-commutativeN/A

                      \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                    7. lower-*.f64N/A

                      \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                    8. lower-PI.f6448.7

                      \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                  5. Applied rewrites48.7%

                    \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                  6. Step-by-step derivation
                    1. Applied rewrites65.5%

                      \[\leadsto \frac{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}{\color{blue}{\sqrt{k}}} \]
                  7. Recombined 2 regimes into one program.
                  8. Final simplification74.4%

                    \[\leadsto \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 0:\\ \;\;\;\;\sqrt{0}\\ \mathbf{else}:\\ \;\;\;\;\frac{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}{\sqrt{k}}\\ \end{array} \]
                  9. Add Preprocessing

                  Alternative 4: 39.5% accurate, 0.5× speedup?

                  \[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 2 \cdot 10^{+154}:\\ \;\;\;\;\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}\\ \mathbf{else}:\\ \;\;\;\;\frac{\sqrt{\mathsf{fma}\left(k, n, n \cdot k\right)}}{\left|k\right|}\\ \end{array} \end{array} \]
                  (FPCore (k n)
                   :precision binary64
                   (if (<=
                        (* (pow (sqrt k) -1.0) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0)))
                        2e+154)
                     (sqrt (* (* 2.0 (/ n k)) (PI)))
                     (/ (sqrt (fma k n (* n k))) (fabs k))))
                  \begin{array}{l}
                  
                  \\
                  \begin{array}{l}
                  \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 2 \cdot 10^{+154}:\\
                  \;\;\;\;\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}\\
                  
                  \mathbf{else}:\\
                  \;\;\;\;\frac{\sqrt{\mathsf{fma}\left(k, n, n \cdot k\right)}}{\left|k\right|}\\
                  
                  
                  \end{array}
                  \end{array}
                  
                  Derivation
                  1. Split input into 2 regimes
                  2. if (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64)))) < 2.00000000000000007e154

                    1. Initial program 99.3%

                      \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                    2. Add Preprocessing
                    3. Taylor expanded in k around 0

                      \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                    4. Step-by-step derivation
                      1. *-commutativeN/A

                        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                      2. lower-*.f64N/A

                        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                      3. lower-sqrt.f64N/A

                        \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                      4. lower-sqrt.f64N/A

                        \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                      5. lower-/.f64N/A

                        \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                      6. *-commutativeN/A

                        \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                      7. lower-*.f64N/A

                        \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                      8. lower-PI.f6457.0

                        \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                    5. Applied rewrites57.0%

                      \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                    6. Step-by-step derivation
                      1. Applied rewrites57.3%

                        \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                      2. Applied rewrites57.3%

                        \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]

                      if 2.00000000000000007e154 < (*.f64 (/.f64 #s(literal 1 binary64) (sqrt.f64 k)) (pow.f64 (*.f64 (*.f64 #s(literal 2 binary64) (PI.f64)) n) (/.f64 (-.f64 #s(literal 1 binary64) k) #s(literal 2 binary64))))

                      1. Initial program 99.8%

                        \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                      2. Add Preprocessing
                      3. Taylor expanded in k around 0

                        \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                      4. Step-by-step derivation
                        1. *-commutativeN/A

                          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                        2. lower-*.f64N/A

                          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                        3. lower-sqrt.f64N/A

                          \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                        4. lower-sqrt.f64N/A

                          \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                        5. lower-/.f64N/A

                          \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                        6. *-commutativeN/A

                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                        7. lower-*.f64N/A

                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                        8. lower-PI.f643.5

                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                      5. Applied rewrites3.5%

                        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                      6. Step-by-step derivation
                        1. Applied rewrites3.5%

                          \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                        2. Applied rewrites3.5%

                          \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
                        3. Step-by-step derivation
                          1. Applied rewrites8.2%

                            \[\leadsto \frac{\sqrt{\mathsf{fma}\left(k, n, n \cdot k\right)}}{\color{blue}{\left|k\right|}} \]
                        4. Recombined 2 regimes into one program.
                        5. Final simplification38.9%

                          \[\leadsto \begin{array}{l} \mathbf{if}\;{\left(\sqrt{k}\right)}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \leq 2 \cdot 10^{+154}:\\ \;\;\;\;\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}\\ \mathbf{else}:\\ \;\;\;\;\frac{\sqrt{\mathsf{fma}\left(k, n, n \cdot k\right)}}{\left|k\right|}\\ \end{array} \]
                        6. Add Preprocessing

                        Alternative 5: 99.5% accurate, 0.9× speedup?

                        \[\begin{array}{l} \\ \begin{array}{l} t_0 := \left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\\ \frac{\sqrt{t\_0}}{\sqrt{k} \cdot {t\_0}^{\left(k \cdot 0.5\right)}} \end{array} \end{array} \]
                        (FPCore (k n)
                         :precision binary64
                         (let* ((t_0 (* (* 2.0 n) (PI))))
                           (/ (sqrt t_0) (* (sqrt k) (pow t_0 (* k 0.5))))))
                        \begin{array}{l}
                        
                        \\
                        \begin{array}{l}
                        t_0 := \left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\\
                        \frac{\sqrt{t\_0}}{\sqrt{k} \cdot {t\_0}^{\left(k \cdot 0.5\right)}}
                        \end{array}
                        \end{array}
                        
                        Derivation
                        1. Initial program 99.5%

                          \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                        2. Add Preprocessing
                        3. Step-by-step derivation
                          1. lift-/.f64N/A

                            \[\leadsto \color{blue}{\frac{1}{\sqrt{k}}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                          2. inv-powN/A

                            \[\leadsto \color{blue}{{\left(\sqrt{k}\right)}^{-1}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                          3. lift-sqrt.f64N/A

                            \[\leadsto {\color{blue}{\left(\sqrt{k}\right)}}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                          4. pow1/2N/A

                            \[\leadsto {\color{blue}{\left({k}^{\frac{1}{2}}\right)}}^{-1} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                          5. pow-powN/A

                            \[\leadsto \color{blue}{{k}^{\left(\frac{1}{2} \cdot -1\right)}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                          6. lower-pow.f64N/A

                            \[\leadsto \color{blue}{{k}^{\left(\frac{1}{2} \cdot -1\right)}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                          7. metadata-eval99.5

                            \[\leadsto {k}^{\color{blue}{-0.5}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                        4. Applied rewrites99.5%

                          \[\leadsto \color{blue}{{k}^{-0.5}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                        5. Step-by-step derivation
                          1. lift-*.f64N/A

                            \[\leadsto \color{blue}{{k}^{\frac{-1}{2}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}} \]
                          2. *-commutativeN/A

                            \[\leadsto \color{blue}{{\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}}} \]
                          3. lift-pow.f64N/A

                            \[\leadsto \color{blue}{{\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)}} \cdot {k}^{\frac{-1}{2}} \]
                          4. lift-*.f64N/A

                            \[\leadsto {\color{blue}{\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
                          5. lift-*.f64N/A

                            \[\leadsto {\left(\color{blue}{\left(2 \cdot \mathsf{PI}\left(\right)\right)} \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
                          6. associate-*l*N/A

                            \[\leadsto {\color{blue}{\left(2 \cdot \left(\mathsf{PI}\left(\right) \cdot n\right)\right)}}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
                          7. *-commutativeN/A

                            \[\leadsto {\left(2 \cdot \color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right)}\right)}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
                          8. associate-*l*N/A

                            \[\leadsto {\color{blue}{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
                          9. lift-*.f64N/A

                            \[\leadsto {\left(\color{blue}{\left(2 \cdot n\right)} \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
                          10. lift-*.f64N/A

                            \[\leadsto {\color{blue}{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}}^{\left(\frac{1 - k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
                          11. lift-/.f64N/A

                            \[\leadsto {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\color{blue}{\left(\frac{1 - k}{2}\right)}} \cdot {k}^{\frac{-1}{2}} \]
                          12. lift--.f64N/A

                            \[\leadsto {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{\color{blue}{1 - k}}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
                          13. div-subN/A

                            \[\leadsto {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\color{blue}{\left(\frac{1}{2} - \frac{k}{2}\right)}} \cdot {k}^{\frac{-1}{2}} \]
                          14. metadata-evalN/A

                            \[\leadsto {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\color{blue}{\frac{1}{2}} - \frac{k}{2}\right)} \cdot {k}^{\frac{-1}{2}} \]
                          15. pow-subN/A

                            \[\leadsto \color{blue}{\frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}}} \cdot {k}^{\frac{-1}{2}} \]
                          16. lift-pow.f64N/A

                            \[\leadsto \frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot \color{blue}{{k}^{\frac{-1}{2}}} \]
                          17. metadata-evalN/A

                            \[\leadsto \frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot {k}^{\color{blue}{\left(\mathsf{neg}\left(\frac{1}{2}\right)\right)}} \]
                          18. pow-flipN/A

                            \[\leadsto \frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot \color{blue}{\frac{1}{{k}^{\frac{1}{2}}}} \]
                          19. pow1/2N/A

                            \[\leadsto \frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot \frac{1}{\color{blue}{\sqrt{k}}} \]
                          20. lift-sqrt.f64N/A

                            \[\leadsto \frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}} \cdot \frac{1}{\color{blue}{\sqrt{k}}} \]
                        6. Applied rewrites99.7%

                          \[\leadsto \color{blue}{\frac{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)} \cdot 1}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}}} \]
                        7. Step-by-step derivation
                          1. lift-*.f64N/A

                            \[\leadsto \frac{\color{blue}{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)} \cdot 1}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          2. *-rgt-identity99.7

                            \[\leadsto \frac{\color{blue}{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          3. lift-sqrt.f64N/A

                            \[\leadsto \frac{\color{blue}{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          4. lift-*.f64N/A

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          5. lift-*.f64N/A

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot 2\right)} \cdot \mathsf{PI}\left(\right)}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          6. associate-*l*N/A

                            \[\leadsto \frac{\sqrt{\color{blue}{n \cdot \left(2 \cdot \mathsf{PI}\left(\right)\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          7. *-commutativeN/A

                            \[\leadsto \frac{\sqrt{n \cdot \color{blue}{\left(\mathsf{PI}\left(\right) \cdot 2\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          8. associate-*r*N/A

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right) \cdot 2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          9. lift-*.f64N/A

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right)} \cdot 2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          10. sqrt-prodN/A

                            \[\leadsto \frac{\color{blue}{\sqrt{n \cdot \mathsf{PI}\left(\right)} \cdot \sqrt{2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          11. lift-sqrt.f64N/A

                            \[\leadsto \frac{\color{blue}{\sqrt{n \cdot \mathsf{PI}\left(\right)}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          12. lift-sqrt.f64N/A

                            \[\leadsto \frac{\sqrt{n \cdot \mathsf{PI}\left(\right)} \cdot \color{blue}{\sqrt{2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          13. lower-*.f6499.5

                            \[\leadsto \frac{\color{blue}{\sqrt{n \cdot \mathsf{PI}\left(\right)} \cdot \sqrt{2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          14. lift-sqrt.f64N/A

                            \[\leadsto \frac{\color{blue}{\sqrt{n \cdot \mathsf{PI}\left(\right)}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          15. pow1/2N/A

                            \[\leadsto \frac{\color{blue}{{\left(n \cdot \mathsf{PI}\left(\right)\right)}^{\frac{1}{2}}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          16. lift-*.f64N/A

                            \[\leadsto \frac{{\color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right)}}^{\frac{1}{2}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          17. *-commutativeN/A

                            \[\leadsto \frac{{\color{blue}{\left(\mathsf{PI}\left(\right) \cdot n\right)}}^{\frac{1}{2}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          18. lift-*.f64N/A

                            \[\leadsto \frac{{\color{blue}{\left(\mathsf{PI}\left(\right) \cdot n\right)}}^{\frac{1}{2}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          19. pow1/2N/A

                            \[\leadsto \frac{\color{blue}{\sqrt{\mathsf{PI}\left(\right) \cdot n}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          20. lift-sqrt.f6499.5

                            \[\leadsto \frac{\color{blue}{\sqrt{\mathsf{PI}\left(\right) \cdot n}} \cdot \sqrt{2}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                        8. Applied rewrites99.5%

                          \[\leadsto \frac{\color{blue}{\sqrt{\mathsf{PI}\left(\right) \cdot n} \cdot \sqrt{2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                        9. Step-by-step derivation
                          1. lift-*.f64N/A

                            \[\leadsto \frac{\color{blue}{\sqrt{\mathsf{PI}\left(\right) \cdot n} \cdot \sqrt{2}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          2. *-commutativeN/A

                            \[\leadsto \frac{\color{blue}{\sqrt{2} \cdot \sqrt{\mathsf{PI}\left(\right) \cdot n}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          3. lift-sqrt.f64N/A

                            \[\leadsto \frac{\color{blue}{\sqrt{2}} \cdot \sqrt{\mathsf{PI}\left(\right) \cdot n}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          4. lift-sqrt.f64N/A

                            \[\leadsto \frac{\sqrt{2} \cdot \color{blue}{\sqrt{\mathsf{PI}\left(\right) \cdot n}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          5. sqrt-unprodN/A

                            \[\leadsto \frac{\color{blue}{\sqrt{2 \cdot \left(\mathsf{PI}\left(\right) \cdot n\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          6. lift-*.f64N/A

                            \[\leadsto \frac{\sqrt{2 \cdot \color{blue}{\left(\mathsf{PI}\left(\right) \cdot n\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          7. *-commutativeN/A

                            \[\leadsto \frac{\sqrt{2 \cdot \color{blue}{\left(n \cdot \mathsf{PI}\left(\right)\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          8. associate-*r*N/A

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          9. *-commutativeN/A

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot 2\right)} \cdot \mathsf{PI}\left(\right)}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          10. lift-*.f64N/A

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot 2\right)} \cdot \mathsf{PI}\left(\right)}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          11. lift-*.f64N/A

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          12. lower-sqrt.f6499.7

                            \[\leadsto \frac{\color{blue}{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          13. lift-*.f64N/A

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(n \cdot 2\right)} \cdot \mathsf{PI}\left(\right)}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          14. *-commutativeN/A

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(2 \cdot n\right)} \cdot \mathsf{PI}\left(\right)}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          15. lower-*.f6499.7

                            \[\leadsto \frac{\sqrt{\color{blue}{\left(2 \cdot n\right)} \cdot \mathsf{PI}\left(\right)}}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}} \]
                          16. lift-*.f64N/A

                            \[\leadsto \frac{\sqrt{\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)}}{\color{blue}{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)} \cdot \sqrt{k}}} \]
                          17. *-commutativeN/A

                            \[\leadsto \frac{\sqrt{\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)}}{\color{blue}{\sqrt{k} \cdot {\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}}} \]
                          18. lower-*.f6499.7

                            \[\leadsto \frac{\sqrt{\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)}}{\color{blue}{\sqrt{k} \cdot {\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\frac{k}{2}\right)}}} \]
                        10. Applied rewrites99.7%

                          \[\leadsto \color{blue}{\frac{\sqrt{\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)}}{\sqrt{k} \cdot {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(k \cdot 0.5\right)}}} \]
                        11. Add Preprocessing

                        Alternative 6: 99.5% accurate, 1.1× speedup?

                        \[\begin{array}{l} \\ \frac{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)}}{\sqrt{k}} \end{array} \]
                        (FPCore (k n)
                         :precision binary64
                         (/ (pow (* (* n 2.0) (PI)) (fma -0.5 k 0.5)) (sqrt k)))
                        \begin{array}{l}
                        
                        \\
                        \frac{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)}}{\sqrt{k}}
                        \end{array}
                        
                        Derivation
                        1. Initial program 99.5%

                          \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                        2. Add Preprocessing
                        3. Taylor expanded in k around inf

                          \[\leadsto \frac{1}{\sqrt{k}} \cdot \color{blue}{e^{\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)}} \]
                        4. Step-by-step derivation
                          1. sinh-+-cosh-revN/A

                            \[\leadsto \frac{1}{\sqrt{k}} \cdot \color{blue}{\left(\cosh \left(\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)\right) + \sinh \left(\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)\right)\right)} \]
                          2. +-commutativeN/A

                            \[\leadsto \frac{1}{\sqrt{k}} \cdot \color{blue}{\left(\sinh \left(\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)\right) + \cosh \left(\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)\right)\right)} \]
                          3. distribute-rgt-out--N/A

                            \[\leadsto \frac{1}{\sqrt{k}} \cdot \left(\sinh \left(\frac{1}{2} \cdot \color{blue}{\left(1 \cdot \log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) - k \cdot \log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right)\right)}\right) + \cosh \left(\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)\right)\right) \]
                          4. fp-cancel-sub-signN/A

                            \[\leadsto \frac{1}{\sqrt{k}} \cdot \left(\sinh \left(\frac{1}{2} \cdot \color{blue}{\left(1 \cdot \log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) + \left(\mathsf{neg}\left(k\right)\right) \cdot \log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right)\right)}\right) + \cosh \left(\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)\right)\right) \]
                          5. mul-1-negN/A

                            \[\leadsto \frac{1}{\sqrt{k}} \cdot \left(\sinh \left(\frac{1}{2} \cdot \left(1 \cdot \log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) + \color{blue}{\left(-1 \cdot k\right)} \cdot \log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right)\right)\right) + \cosh \left(\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)\right)\right) \]
                          6. distribute-rgt-inN/A

                            \[\leadsto \frac{1}{\sqrt{k}} \cdot \left(\sinh \left(\frac{1}{2} \cdot \color{blue}{\left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 + -1 \cdot k\right)\right)}\right) + \cosh \left(\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 - k\right)\right)\right)\right) \]
                          7. distribute-rgt-out--N/A

                            \[\leadsto \frac{1}{\sqrt{k}} \cdot \left(\sinh \left(\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 + -1 \cdot k\right)\right)\right) + \cosh \left(\frac{1}{2} \cdot \color{blue}{\left(1 \cdot \log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) - k \cdot \log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right)\right)}\right)\right) \]
                          8. fp-cancel-sub-signN/A

                            \[\leadsto \frac{1}{\sqrt{k}} \cdot \left(\sinh \left(\frac{1}{2} \cdot \left(\log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) \cdot \left(1 + -1 \cdot k\right)\right)\right) + \cosh \left(\frac{1}{2} \cdot \color{blue}{\left(1 \cdot \log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right) + \left(\mathsf{neg}\left(k\right)\right) \cdot \log \left(2 \cdot \left(n \cdot \mathsf{PI}\left(\right)\right)\right)\right)}\right)\right) \]
                        5. Applied rewrites99.5%

                          \[\leadsto \frac{1}{\sqrt{k}} \cdot \color{blue}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)}} \]
                        6. Step-by-step derivation
                          1. lift-*.f64N/A

                            \[\leadsto \color{blue}{\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(\frac{-1}{2}, k, \frac{1}{2}\right)\right)}} \]
                          2. lift-/.f64N/A

                            \[\leadsto \color{blue}{\frac{1}{\sqrt{k}}} \cdot {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(\frac{-1}{2}, k, \frac{1}{2}\right)\right)} \]
                          3. associate-*l/N/A

                            \[\leadsto \color{blue}{\frac{1 \cdot {\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(\frac{-1}{2}, k, \frac{1}{2}\right)\right)}}{\sqrt{k}}} \]
                          4. *-lft-identityN/A

                            \[\leadsto \frac{\color{blue}{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(\frac{-1}{2}, k, \frac{1}{2}\right)\right)}}}{\sqrt{k}} \]
                          5. lower-/.f6499.5

                            \[\leadsto \color{blue}{\frac{{\left(\left(2 \cdot n\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)}}{\sqrt{k}}} \]
                        7. Applied rewrites99.5%

                          \[\leadsto \color{blue}{\frac{{\left(\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)\right)}^{\left(\mathsf{fma}\left(-0.5, k, 0.5\right)\right)}}{\sqrt{k}}} \]
                        8. Add Preprocessing

                        Alternative 7: 52.3% accurate, 3.2× speedup?

                        \[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;k \leq 3.4 \cdot 10^{+163}:\\ \;\;\;\;\frac{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}{\sqrt{k}}\\ \mathbf{else}:\\ \;\;\;\;\sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}}\\ \end{array} \end{array} \]
                        (FPCore (k n)
                         :precision binary64
                         (if (<= k 3.4e+163)
                           (/ (sqrt (* (* n 2.0) (PI))) (sqrt k))
                           (sqrt (/ (fma k n (* n k)) (* k k)))))
                        \begin{array}{l}
                        
                        \\
                        \begin{array}{l}
                        \mathbf{if}\;k \leq 3.4 \cdot 10^{+163}:\\
                        \;\;\;\;\frac{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}{\sqrt{k}}\\
                        
                        \mathbf{else}:\\
                        \;\;\;\;\sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}}\\
                        
                        
                        \end{array}
                        \end{array}
                        
                        Derivation
                        1. Split input into 2 regimes
                        2. if k < 3.4000000000000001e163

                          1. Initial program 99.3%

                            \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                          2. Add Preprocessing
                          3. Taylor expanded in k around 0

                            \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                          4. Step-by-step derivation
                            1. *-commutativeN/A

                              \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                            2. lower-*.f64N/A

                              \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                            3. lower-sqrt.f64N/A

                              \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                            4. lower-sqrt.f64N/A

                              \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                            5. lower-/.f64N/A

                              \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                            6. *-commutativeN/A

                              \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                            7. lower-*.f64N/A

                              \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                            8. lower-PI.f6448.1

                              \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                          5. Applied rewrites48.1%

                            \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                          6. Step-by-step derivation
                            1. Applied rewrites64.6%

                              \[\leadsto \frac{\sqrt{\left(n \cdot 2\right) \cdot \mathsf{PI}\left(\right)}}{\color{blue}{\sqrt{k}}} \]

                            if 3.4000000000000001e163 < k

                            1. Initial program 100.0%

                              \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                            2. Add Preprocessing
                            3. Taylor expanded in k around 0

                              \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                            4. Step-by-step derivation
                              1. *-commutativeN/A

                                \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                              2. lower-*.f64N/A

                                \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                              3. lower-sqrt.f64N/A

                                \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                              4. lower-sqrt.f64N/A

                                \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                              5. lower-/.f64N/A

                                \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                              6. *-commutativeN/A

                                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                              7. lower-*.f64N/A

                                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                              8. lower-PI.f642.7

                                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                            5. Applied rewrites2.7%

                              \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                            6. Step-by-step derivation
                              1. Applied rewrites2.7%

                                \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                              2. Applied rewrites2.7%

                                \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
                              3. Step-by-step derivation
                                1. Applied rewrites10.2%

                                  \[\leadsto \sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}} \]
                              4. Recombined 2 regimes into one program.
                              5. Add Preprocessing

                              Alternative 8: 52.3% accurate, 3.2× speedup?

                              \[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;k \leq 3.4 \cdot 10^{+163}:\\ \;\;\;\;\sqrt{\mathsf{PI}\left(\right) \cdot n} \cdot \sqrt{\frac{2}{k}}\\ \mathbf{else}:\\ \;\;\;\;\sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}}\\ \end{array} \end{array} \]
                              (FPCore (k n)
                               :precision binary64
                               (if (<= k 3.4e+163)
                                 (* (sqrt (* (PI) n)) (sqrt (/ 2.0 k)))
                                 (sqrt (/ (fma k n (* n k)) (* k k)))))
                              \begin{array}{l}
                              
                              \\
                              \begin{array}{l}
                              \mathbf{if}\;k \leq 3.4 \cdot 10^{+163}:\\
                              \;\;\;\;\sqrt{\mathsf{PI}\left(\right) \cdot n} \cdot \sqrt{\frac{2}{k}}\\
                              
                              \mathbf{else}:\\
                              \;\;\;\;\sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}}\\
                              
                              
                              \end{array}
                              \end{array}
                              
                              Derivation
                              1. Split input into 2 regimes
                              2. if k < 3.4000000000000001e163

                                1. Initial program 99.3%

                                  \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                2. Add Preprocessing
                                3. Taylor expanded in k around 0

                                  \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                4. Step-by-step derivation
                                  1. *-commutativeN/A

                                    \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                  2. lower-*.f64N/A

                                    \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                  3. lower-sqrt.f64N/A

                                    \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                  4. lower-sqrt.f64N/A

                                    \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                  5. lower-/.f64N/A

                                    \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                  6. *-commutativeN/A

                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                  7. lower-*.f64N/A

                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                  8. lower-PI.f6448.1

                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                5. Applied rewrites48.1%

                                  \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                6. Step-by-step derivation
                                  1. Applied rewrites48.3%

                                    \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                                  2. Applied rewrites64.6%

                                    \[\leadsto \sqrt{\mathsf{PI}\left(\right) \cdot n} \cdot \color{blue}{\sqrt{\frac{2}{k}}} \]

                                  if 3.4000000000000001e163 < k

                                  1. Initial program 100.0%

                                    \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                  2. Add Preprocessing
                                  3. Taylor expanded in k around 0

                                    \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                  4. Step-by-step derivation
                                    1. *-commutativeN/A

                                      \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                    2. lower-*.f64N/A

                                      \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                    3. lower-sqrt.f64N/A

                                      \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                    4. lower-sqrt.f64N/A

                                      \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                    5. lower-/.f64N/A

                                      \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                    6. *-commutativeN/A

                                      \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                    7. lower-*.f64N/A

                                      \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                    8. lower-PI.f642.7

                                      \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                  5. Applied rewrites2.7%

                                    \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                  6. Step-by-step derivation
                                    1. Applied rewrites2.7%

                                      \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                                    2. Applied rewrites2.7%

                                      \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
                                    3. Step-by-step derivation
                                      1. Applied rewrites10.2%

                                        \[\leadsto \sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}} \]
                                    4. Recombined 2 regimes into one program.
                                    5. Add Preprocessing

                                    Alternative 9: 52.3% accurate, 3.2× speedup?

                                    \[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;k \leq 3.4 \cdot 10^{+163}:\\ \;\;\;\;\sqrt{n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k} \cdot 2}\\ \mathbf{else}:\\ \;\;\;\;\sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}}\\ \end{array} \end{array} \]
                                    (FPCore (k n)
                                     :precision binary64
                                     (if (<= k 3.4e+163)
                                       (* (sqrt n) (sqrt (* (/ (PI) k) 2.0)))
                                       (sqrt (/ (fma k n (* n k)) (* k k)))))
                                    \begin{array}{l}
                                    
                                    \\
                                    \begin{array}{l}
                                    \mathbf{if}\;k \leq 3.4 \cdot 10^{+163}:\\
                                    \;\;\;\;\sqrt{n} \cdot \sqrt{\frac{\mathsf{PI}\left(\right)}{k} \cdot 2}\\
                                    
                                    \mathbf{else}:\\
                                    \;\;\;\;\sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}}\\
                                    
                                    
                                    \end{array}
                                    \end{array}
                                    
                                    Derivation
                                    1. Split input into 2 regimes
                                    2. if k < 3.4000000000000001e163

                                      1. Initial program 99.3%

                                        \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                      2. Add Preprocessing
                                      3. Taylor expanded in k around 0

                                        \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                      4. Step-by-step derivation
                                        1. *-commutativeN/A

                                          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                        2. lower-*.f64N/A

                                          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                        3. lower-sqrt.f64N/A

                                          \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                        4. lower-sqrt.f64N/A

                                          \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                        5. lower-/.f64N/A

                                          \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                        6. *-commutativeN/A

                                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                        7. lower-*.f64N/A

                                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                        8. lower-PI.f6448.1

                                          \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                      5. Applied rewrites48.1%

                                        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                      6. Step-by-step derivation
                                        1. Applied rewrites48.3%

                                          \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                                        2. Applied rewrites64.6%

                                          \[\leadsto \sqrt{n} \cdot \color{blue}{\sqrt{\frac{\mathsf{PI}\left(\right)}{k} \cdot 2}} \]

                                        if 3.4000000000000001e163 < k

                                        1. Initial program 100.0%

                                          \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                        2. Add Preprocessing
                                        3. Taylor expanded in k around 0

                                          \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                        4. Step-by-step derivation
                                          1. *-commutativeN/A

                                            \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                          2. lower-*.f64N/A

                                            \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                          3. lower-sqrt.f64N/A

                                            \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                          4. lower-sqrt.f64N/A

                                            \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                          5. lower-/.f64N/A

                                            \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                          6. *-commutativeN/A

                                            \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                          7. lower-*.f64N/A

                                            \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                          8. lower-PI.f642.7

                                            \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                        5. Applied rewrites2.7%

                                          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                        6. Step-by-step derivation
                                          1. Applied rewrites2.7%

                                            \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                                          2. Applied rewrites2.7%

                                            \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
                                          3. Step-by-step derivation
                                            1. Applied rewrites10.2%

                                              \[\leadsto \sqrt{\frac{\mathsf{fma}\left(k, n, n \cdot k\right)}{k \cdot k}} \]
                                          4. Recombined 2 regimes into one program.
                                          5. Add Preprocessing

                                          Alternative 10: 37.7% accurate, 4.8× speedup?

                                          \[\begin{array}{l} \\ \sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)} \end{array} \]
                                          (FPCore (k n) :precision binary64 (sqrt (* (* 2.0 (/ n k)) (PI))))
                                          \begin{array}{l}
                                          
                                          \\
                                          \sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}
                                          \end{array}
                                          
                                          Derivation
                                          1. Initial program 99.5%

                                            \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                          2. Add Preprocessing
                                          3. Taylor expanded in k around 0

                                            \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                          4. Step-by-step derivation
                                            1. *-commutativeN/A

                                              \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                            2. lower-*.f64N/A

                                              \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                            3. lower-sqrt.f64N/A

                                              \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                            4. lower-sqrt.f64N/A

                                              \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                            5. lower-/.f64N/A

                                              \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                            6. *-commutativeN/A

                                              \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                            7. lower-*.f64N/A

                                              \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                            8. lower-PI.f6437.0

                                              \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                          5. Applied rewrites37.0%

                                            \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                          6. Step-by-step derivation
                                            1. Applied rewrites37.1%

                                              \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                                            2. Applied rewrites37.1%

                                              \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
                                            3. Add Preprocessing

                                            Alternative 11: 37.6% accurate, 4.8× speedup?

                                            \[\begin{array}{l} \\ \sqrt{n \cdot \left(\frac{2}{k} \cdot \mathsf{PI}\left(\right)\right)} \end{array} \]
                                            (FPCore (k n) :precision binary64 (sqrt (* n (* (/ 2.0 k) (PI)))))
                                            \begin{array}{l}
                                            
                                            \\
                                            \sqrt{n \cdot \left(\frac{2}{k} \cdot \mathsf{PI}\left(\right)\right)}
                                            \end{array}
                                            
                                            Derivation
                                            1. Initial program 99.5%

                                              \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                            2. Add Preprocessing
                                            3. Taylor expanded in k around 0

                                              \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                            4. Step-by-step derivation
                                              1. *-commutativeN/A

                                                \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                              2. lower-*.f64N/A

                                                \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                              3. lower-sqrt.f64N/A

                                                \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                              4. lower-sqrt.f64N/A

                                                \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                              5. lower-/.f64N/A

                                                \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                              6. *-commutativeN/A

                                                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                              7. lower-*.f64N/A

                                                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                              8. lower-PI.f6437.0

                                                \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                            5. Applied rewrites37.0%

                                              \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                            6. Step-by-step derivation
                                              1. Applied rewrites37.1%

                                                \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                                              2. Applied rewrites37.1%

                                                \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
                                              3. Step-by-step derivation
                                                1. Applied rewrites37.1%

                                                  \[\leadsto \sqrt{n \cdot \left(\frac{2}{k} \cdot \mathsf{PI}\left(\right)\right)} \]
                                                2. Add Preprocessing

                                                Alternative 12: 9.1% accurate, 5.6× speedup?

                                                \[\begin{array}{l} \\ \sqrt{n \cdot \frac{2}{k}} \end{array} \]
                                                (FPCore (k n) :precision binary64 (sqrt (* n (/ 2.0 k))))
                                                double code(double k, double n) {
                                                	return sqrt((n * (2.0 / k)));
                                                }
                                                
                                                module fmin_fmax_functions
                                                    implicit none
                                                    private
                                                    public fmax
                                                    public fmin
                                                
                                                    interface fmax
                                                        module procedure fmax88
                                                        module procedure fmax44
                                                        module procedure fmax84
                                                        module procedure fmax48
                                                    end interface
                                                    interface fmin
                                                        module procedure fmin88
                                                        module procedure fmin44
                                                        module procedure fmin84
                                                        module procedure fmin48
                                                    end interface
                                                contains
                                                    real(8) function fmax88(x, y) result (res)
                                                        real(8), intent (in) :: x
                                                        real(8), intent (in) :: y
                                                        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
                                                    end function
                                                    real(4) function fmax44(x, y) result (res)
                                                        real(4), intent (in) :: x
                                                        real(4), intent (in) :: y
                                                        res = merge(y, merge(x, max(x, y), y /= y), x /= x)
                                                    end function
                                                    real(8) function fmax84(x, y) result(res)
                                                        real(8), intent (in) :: x
                                                        real(4), intent (in) :: y
                                                        res = merge(dble(y), merge(x, max(x, dble(y)), y /= y), x /= x)
                                                    end function
                                                    real(8) function fmax48(x, y) result(res)
                                                        real(4), intent (in) :: x
                                                        real(8), intent (in) :: y
                                                        res = merge(y, merge(dble(x), max(dble(x), y), y /= y), x /= x)
                                                    end function
                                                    real(8) function fmin88(x, y) result (res)
                                                        real(8), intent (in) :: x
                                                        real(8), intent (in) :: y
                                                        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
                                                    end function
                                                    real(4) function fmin44(x, y) result (res)
                                                        real(4), intent (in) :: x
                                                        real(4), intent (in) :: y
                                                        res = merge(y, merge(x, min(x, y), y /= y), x /= x)
                                                    end function
                                                    real(8) function fmin84(x, y) result(res)
                                                        real(8), intent (in) :: x
                                                        real(4), intent (in) :: y
                                                        res = merge(dble(y), merge(x, min(x, dble(y)), y /= y), x /= x)
                                                    end function
                                                    real(8) function fmin48(x, y) result(res)
                                                        real(4), intent (in) :: x
                                                        real(8), intent (in) :: y
                                                        res = merge(y, merge(dble(x), min(dble(x), y), y /= y), x /= x)
                                                    end function
                                                end module
                                                
                                                real(8) function code(k, n)
                                                use fmin_fmax_functions
                                                    real(8), intent (in) :: k
                                                    real(8), intent (in) :: n
                                                    code = sqrt((n * (2.0d0 / k)))
                                                end function
                                                
                                                public static double code(double k, double n) {
                                                	return Math.sqrt((n * (2.0 / k)));
                                                }
                                                
                                                def code(k, n):
                                                	return math.sqrt((n * (2.0 / k)))
                                                
                                                function code(k, n)
                                                	return sqrt(Float64(n * Float64(2.0 / k)))
                                                end
                                                
                                                function tmp = code(k, n)
                                                	tmp = sqrt((n * (2.0 / k)));
                                                end
                                                
                                                code[k_, n_] := N[Sqrt[N[(n * N[(2.0 / k), $MachinePrecision]), $MachinePrecision]], $MachinePrecision]
                                                
                                                \begin{array}{l}
                                                
                                                \\
                                                \sqrt{n \cdot \frac{2}{k}}
                                                \end{array}
                                                
                                                Derivation
                                                1. Initial program 99.5%

                                                  \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                                2. Add Preprocessing
                                                3. Taylor expanded in k around 0

                                                  \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                                4. Step-by-step derivation
                                                  1. *-commutativeN/A

                                                    \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                  2. lower-*.f64N/A

                                                    \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                  3. lower-sqrt.f64N/A

                                                    \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                                  4. lower-sqrt.f64N/A

                                                    \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                  5. lower-/.f64N/A

                                                    \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                  6. *-commutativeN/A

                                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                                  7. lower-*.f64N/A

                                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                                  8. lower-PI.f6437.0

                                                    \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                                5. Applied rewrites37.0%

                                                  \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                                6. Step-by-step derivation
                                                  1. Applied rewrites37.1%

                                                    \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                                                  2. Applied rewrites37.1%

                                                    \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
                                                  3. Step-by-step derivation
                                                    1. Applied rewrites9.1%

                                                      \[\leadsto \sqrt{n \cdot \frac{2}{k}} \]
                                                    2. Add Preprocessing

                                                    Alternative 13: 4.8% accurate, 7.2× speedup?

                                                    \[\begin{array}{l} \\ \sqrt{\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot \mathsf{PI}\left(\right)} \end{array} \]
                                                    (FPCore (k n) :precision binary64 (sqrt (* (* 2.0 (PI)) (PI))))
                                                    \begin{array}{l}
                                                    
                                                    \\
                                                    \sqrt{\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot \mathsf{PI}\left(\right)}
                                                    \end{array}
                                                    
                                                    Derivation
                                                    1. Initial program 99.5%

                                                      \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                                    2. Add Preprocessing
                                                    3. Taylor expanded in k around 0

                                                      \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                                    4. Step-by-step derivation
                                                      1. *-commutativeN/A

                                                        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                      2. lower-*.f64N/A

                                                        \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                      3. lower-sqrt.f64N/A

                                                        \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                                      4. lower-sqrt.f64N/A

                                                        \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                      5. lower-/.f64N/A

                                                        \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                      6. *-commutativeN/A

                                                        \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                                      7. lower-*.f64N/A

                                                        \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                                      8. lower-PI.f6437.0

                                                        \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                                    5. Applied rewrites37.0%

                                                      \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                                    6. Step-by-step derivation
                                                      1. Applied rewrites37.1%

                                                        \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                                                      2. Applied rewrites37.1%

                                                        \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
                                                      3. Step-by-step derivation
                                                        1. Applied rewrites4.6%

                                                          \[\leadsto \sqrt{\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot \mathsf{PI}\left(\right)} \]
                                                        2. Add Preprocessing

                                                        Alternative 14: 4.8% accurate, 10.9× speedup?

                                                        \[\begin{array}{l} \\ \sqrt{\mathsf{PI}\left(\right) + \mathsf{PI}\left(\right)} \end{array} \]
                                                        (FPCore (k n) :precision binary64 (sqrt (+ (PI) (PI))))
                                                        \begin{array}{l}
                                                        
                                                        \\
                                                        \sqrt{\mathsf{PI}\left(\right) + \mathsf{PI}\left(\right)}
                                                        \end{array}
                                                        
                                                        Derivation
                                                        1. Initial program 99.5%

                                                          \[\frac{1}{\sqrt{k}} \cdot {\left(\left(2 \cdot \mathsf{PI}\left(\right)\right) \cdot n\right)}^{\left(\frac{1 - k}{2}\right)} \]
                                                        2. Add Preprocessing
                                                        3. Taylor expanded in k around 0

                                                          \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \cdot \sqrt{2}} \]
                                                        4. Step-by-step derivation
                                                          1. *-commutativeN/A

                                                            \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                          2. lower-*.f64N/A

                                                            \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                          3. lower-sqrt.f64N/A

                                                            \[\leadsto \color{blue}{\sqrt{2}} \cdot \sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}} \]
                                                          4. lower-sqrt.f64N/A

                                                            \[\leadsto \sqrt{2} \cdot \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                          5. lower-/.f64N/A

                                                            \[\leadsto \sqrt{2} \cdot \sqrt{\color{blue}{\frac{n \cdot \mathsf{PI}\left(\right)}{k}}} \]
                                                          6. *-commutativeN/A

                                                            \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                                          7. lower-*.f64N/A

                                                            \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right) \cdot n}}{k}} \]
                                                          8. lower-PI.f6437.0

                                                            \[\leadsto \sqrt{2} \cdot \sqrt{\frac{\color{blue}{\mathsf{PI}\left(\right)} \cdot n}{k}} \]
                                                        5. Applied rewrites37.0%

                                                          \[\leadsto \color{blue}{\sqrt{2} \cdot \sqrt{\frac{\mathsf{PI}\left(\right) \cdot n}{k}}} \]
                                                        6. Step-by-step derivation
                                                          1. Applied rewrites37.1%

                                                            \[\leadsto \color{blue}{\sqrt{\frac{n \cdot \mathsf{PI}\left(\right)}{k} \cdot 2}} \]
                                                          2. Applied rewrites37.1%

                                                            \[\leadsto \color{blue}{\sqrt{\left(2 \cdot \frac{n}{k}\right) \cdot \mathsf{PI}\left(\right)}} \]
                                                          3. Step-by-step derivation
                                                            1. Applied rewrites4.6%

                                                              \[\leadsto \sqrt{\mathsf{PI}\left(\right) + \mathsf{PI}\left(\right)} \]
                                                            2. Add Preprocessing

                                                            Reproduce

                                                            ?
                                                            herbie shell --seed 2024353 
                                                            (FPCore (k n)
                                                              :name "Migdal et al, Equation (51)"
                                                              :precision binary64
                                                              (* (/ 1.0 (sqrt k)) (pow (* (* 2.0 (PI)) n) (/ (- 1.0 k) 2.0))))