Henrywood and Agarwal, Equation (13)

Percentage Accurate: 23.7% → 47.9%

Time: 8.7s

Alternatives: 10

Speedup: 7.8×

• could not determine a ground truth

  1. c0 = 5.1896707557269823e+303
  2. w = 1.0683847532637158e-174
  3. h = -2.0180283166588474e-244
  4. D = 5.889844121659402e-302
  5. d = 8.654747841328284e+306
  6. M = -2.016361705051164e-206

Specification

Math

\[\begin{array}{l} \\ \begin{array}{l} t_0 := \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)}\\ \frac{c0}{2 \cdot w} \cdot \left(t\_0 + \sqrt{t\_0 \cdot t\_0 - M \cdot M}\right) \end{array} \end{array} \]

FPCore

(FPCore (c0 w h D d M)
 :precision binary64
 (let* ((t_0 (/ (* c0 (* d d)) (* (* w h) (* D D)))))
   (* (/ c0 (* 2.0 w)) (+ t_0 (sqrt (- (* t_0 t_0) (* M M)))))))

C

double code(double c0, double w, double h, double D, double d, double M) {
	double t_0 = (c0 * (d * d)) / ((w * h) * (D * D));
	return (c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))));
}

Fortran

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(c0, w, h, d, d_1, m)
use fmin_fmax_functions
    real(8), intent (in) :: c0
    real(8), intent (in) :: w
    real(8), intent (in) :: h
    real(8), intent (in) :: d
    real(8), intent (in) :: d_1
    real(8), intent (in) :: m
    real(8) :: t_0
    t_0 = (c0 * (d_1 * d_1)) / ((w * h) * (d * d))
    code = (c0 / (2.0d0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (m * m))))
end function

Java

public static double code(double c0, double w, double h, double D, double d, double M) {
	double t_0 = (c0 * (d * d)) / ((w * h) * (D * D));
	return (c0 / (2.0 * w)) * (t_0 + Math.sqrt(((t_0 * t_0) - (M * M))));
}

Python

def code(c0, w, h, D, d, M):
	t_0 = (c0 * (d * d)) / ((w * h) * (D * D))
	return (c0 / (2.0 * w)) * (t_0 + math.sqrt(((t_0 * t_0) - (M * M))))

Julia

function code(c0, w, h, D, d, M)
	t_0 = Float64(Float64(c0 * Float64(d * d)) / Float64(Float64(w * h) * Float64(D * D)))
	return Float64(Float64(c0 / Float64(2.0 * w)) * Float64(t_0 + sqrt(Float64(Float64(t_0 * t_0) - Float64(M * M)))))
end

MATLAB

function tmp = code(c0, w, h, D, d, M)
	t_0 = (c0 * (d * d)) / ((w * h) * (D * D));
	tmp = (c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))));
end

Wolfram

code[c0_, w_, h_, D_, d_, M_] := Block[{t$95$0 = N[(N[(c0 * N[(d * d), $MachinePrecision]), $MachinePrecision] / N[(N[(w * h), $MachinePrecision] * N[(D * D), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, N[(N[(c0 / N[(2.0 * w), $MachinePrecision]), $MachinePrecision] * N[(t$95$0 + N[Sqrt[N[(N[(t$95$0 * t$95$0), $MachinePrecision] - N[(M * M), $MachinePrecision]), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]), $MachinePrecision]]

Initial Program: 23.7% accurate, 1.0× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} t_0 := \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)}\\ \frac{c0}{2 \cdot w} \cdot \left(t\_0 + \sqrt{t\_0 \cdot t\_0 - M \cdot M}\right) \end{array} \end{array} \]

FPCore

(FPCore (c0 w h D d M)
 :precision binary64
 (let* ((t_0 (/ (* c0 (* d d)) (* (* w h) (* D D)))))
   (* (/ c0 (* 2.0 w)) (+ t_0 (sqrt (- (* t_0 t_0) (* M M)))))))

C

double code(double c0, double w, double h, double D, double d, double M) {
	double t_0 = (c0 * (d * d)) / ((w * h) * (D * D));
	return (c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))));
}

Fortran

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(c0, w, h, d, d_1, m)
use fmin_fmax_functions
    real(8), intent (in) :: c0
    real(8), intent (in) :: w
    real(8), intent (in) :: h
    real(8), intent (in) :: d
    real(8), intent (in) :: d_1
    real(8), intent (in) :: m
    real(8) :: t_0
    t_0 = (c0 * (d_1 * d_1)) / ((w * h) * (d * d))
    code = (c0 / (2.0d0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (m * m))))
end function

Java

public static double code(double c0, double w, double h, double D, double d, double M) {
	double t_0 = (c0 * (d * d)) / ((w * h) * (D * D));
	return (c0 / (2.0 * w)) * (t_0 + Math.sqrt(((t_0 * t_0) - (M * M))));
}

Python

def code(c0, w, h, D, d, M):
	t_0 = (c0 * (d * d)) / ((w * h) * (D * D))
	return (c0 / (2.0 * w)) * (t_0 + math.sqrt(((t_0 * t_0) - (M * M))))

Julia

function code(c0, w, h, D, d, M)
	t_0 = Float64(Float64(c0 * Float64(d * d)) / Float64(Float64(w * h) * Float64(D * D)))
	return Float64(Float64(c0 / Float64(2.0 * w)) * Float64(t_0 + sqrt(Float64(Float64(t_0 * t_0) - Float64(M * M)))))
end

MATLAB

function tmp = code(c0, w, h, D, d, M)
	t_0 = (c0 * (d * d)) / ((w * h) * (D * D));
	tmp = (c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))));
end

Wolfram

code[c0_, w_, h_, D_, d_, M_] := Block[{t$95$0 = N[(N[(c0 * N[(d * d), $MachinePrecision]), $MachinePrecision] / N[(N[(w * h), $MachinePrecision] * N[(D * D), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, N[(N[(c0 / N[(2.0 * w), $MachinePrecision]), $MachinePrecision] * N[(t$95$0 + N[Sqrt[N[(N[(t$95$0 * t$95$0), $MachinePrecision] - N[(M * M), $MachinePrecision]), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]), $MachinePrecision]]

Alternative 1: 47.9% accurate, 2.0× speedup

Math

\[\begin{array}{l} d_m = \left|d\right| \\ \begin{array}{l} t_0 := \frac{c0}{D} \cdot \frac{d\_m}{w}\\ \mathbf{if}\;d\_m \leq 2.85 \cdot 10^{+256}:\\ \;\;\;\;\frac{t\_0 \cdot t\_0}{h}\\ \mathbf{else}:\\ \;\;\;\;\frac{c0 \cdot 0}{w \cdot 2}\\ \end{array} \end{array} \]

FPCore

d_m = (fabs.f64 d)
(FPCore (c0 w h D d_m M)
 :precision binary64
 (let* ((t_0 (* (/ c0 D) (/ d_m w))))
   (if (<= d_m 2.85e+256) (/ (* t_0 t_0) h) (/ (* c0 0.0) (* w 2.0)))))

C

d_m = fabs(d);
double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 / D) * (d_m / w);
	double tmp;
	if (d_m <= 2.85e+256) {
		tmp = (t_0 * t_0) / h;
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Fortran

d_m =     private
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(c0, w, h, d, d_m, m)
use fmin_fmax_functions
    real(8), intent (in) :: c0
    real(8), intent (in) :: w
    real(8), intent (in) :: h
    real(8), intent (in) :: d
    real(8), intent (in) :: d_m
    real(8), intent (in) :: m
    real(8) :: t_0
    real(8) :: tmp
    t_0 = (c0 / d) * (d_m / w)
    if (d_m <= 2.85d+256) then
        tmp = (t_0 * t_0) / h
    else
        tmp = (c0 * 0.0d0) / (w * 2.0d0)
    end if
    code = tmp
end function

Java

d_m = Math.abs(d);
public static double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 / D) * (d_m / w);
	double tmp;
	if (d_m <= 2.85e+256) {
		tmp = (t_0 * t_0) / h;
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Python

d_m = math.fabs(d)
def code(c0, w, h, D, d_m, M):
	t_0 = (c0 / D) * (d_m / w)
	tmp = 0
	if d_m <= 2.85e+256:
		tmp = (t_0 * t_0) / h
	else:
		tmp = (c0 * 0.0) / (w * 2.0)
	return tmp

Julia

d_m = abs(d)
function code(c0, w, h, D, d_m, M)
	t_0 = Float64(Float64(c0 / D) * Float64(d_m / w))
	tmp = 0.0
	if (d_m <= 2.85e+256)
		tmp = Float64(Float64(t_0 * t_0) / h);
	else
		tmp = Float64(Float64(c0 * 0.0) / Float64(w * 2.0));
	end
	return tmp
end

MATLAB

d_m = abs(d);
function tmp_2 = code(c0, w, h, D, d_m, M)
	t_0 = (c0 / D) * (d_m / w);
	tmp = 0.0;
	if (d_m <= 2.85e+256)
		tmp = (t_0 * t_0) / h;
	else
		tmp = (c0 * 0.0) / (w * 2.0);
	end
	tmp_2 = tmp;
end

Wolfram

d_m = N[Abs[d], $MachinePrecision]
code[c0_, w_, h_, D_, d$95$m_, M_] := Block[{t$95$0 = N[(N[(c0 / D), $MachinePrecision] * N[(d$95$m / w), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[d$95$m, 2.85e+256], N[(N[(t$95$0 * t$95$0), $MachinePrecision] / h), $MachinePrecision], N[(N[(c0 * 0.0), $MachinePrecision] / N[(w * 2.0), $MachinePrecision]), $MachinePrecision]]]

Derivation

1. Split input into 2 regimes

2. if d < 2.8499999999999999e256

   1. Initial program 24.2%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in h around 0

      \[\leadsto \color{blue}{\frac{\frac{-1}{4} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot {h}^{2}\right)}{{d}^{2}} + \frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h}} \]
   4. Step-by-step derivation

      1. lower-/.f64 N/A

         \[\leadsto \frac{\frac{-1}{4} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot {h}^{2}\right)}{{d}^{2}} + \frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{\color{blue}{h}} \]

   5. Applied rewrites 15.6%

      \[\leadsto \color{blue}{\frac{\mathsf{fma}\left(\frac{c0 \cdot c0}{D \cdot D}, \frac{d \cdot d}{w \cdot w}, \frac{\left(\left(M \cdot M\right) \cdot \left(h \cdot h\right)\right) \cdot \left(D \cdot D\right)}{d \cdot d} \cdot -0.25\right)}{h}} \]

   6. Taylor expanded in c0 around inf

      \[\leadsto \frac{\frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h} \]
   7. Step-by-step derivation

      1. frac-times N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{{d}^{2}}{{w}^{2}}}{h} \]

      2. pow2 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{{w}^{2}}}{h} \]

      3. pow2 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      5. pow2 N/A

         \[\leadsto \frac{\frac{c0 \cdot c0}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      6. pow2 N/A

         \[\leadsto \frac{\frac{c0 \cdot c0}{D \cdot D} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      7. times-frac N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      8. lower-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      9. lower-/.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      10. lower-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      11. times-frac N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      12. lower-*.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      13. lower-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      14. lower-/.f64 41.7

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

   8. Applied rewrites 41.7%

      \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      2. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      3. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      4. unswap-sqr N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

      5. lower-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

      6. lower-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

      7. lower-*.f64 49.0

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

   10. Applied rewrites 49.0%

       \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

   if 2.8499999999999999e256 < d

   1. Initial program 20.5%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in c0 around -inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-1 \cdot \left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right)} \]
   4. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\mathsf{neg}\left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right) \]

      2. lower-neg.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right) \]

      3. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

   5. Applied rewrites 0.0%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-\left(0 \cdot \frac{d \cdot d}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot c0\right)} \]

   6. Taylor expanded in c0 around 0

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   7. Step-by-step derivation

   8. Applied rewrites 36.2%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \color{blue}{\frac{c0}{2 \cdot w} \cdot 0} \]

   10. Applied rewrites 41.3%

       \[\leadsto \color{blue}{\frac{c0 \cdot 0}{w \cdot 2}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 2: 55.6% accurate, 0.7× speedup

Math

\[\begin{array}{l} d_m = \left|d\right| \\ \begin{array}{l} t_0 := \frac{c0}{2 \cdot w}\\ t_1 := \frac{c0 \cdot \left(d\_m \cdot d\_m\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)}\\ \mathbf{if}\;t\_0 \cdot \left(t\_1 + \sqrt{t\_1 \cdot t\_1 - M \cdot M}\right) \leq \infty:\\ \;\;\;\;t\_0 \cdot \left(\left(\left(\frac{c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot 2\right) \cdot d\_m\right) \cdot d\_m\right)\\ \mathbf{else}:\\ \;\;\;\;\frac{c0 \cdot 0}{w \cdot 2}\\ \end{array} \end{array} \]

FPCore

d_m = (fabs.f64 d)
(FPCore (c0 w h D d_m M)
 :precision binary64
 (let* ((t_0 (/ c0 (* 2.0 w)))
        (t_1 (/ (* c0 (* d_m d_m)) (* (* w h) (* D D)))))
   (if (<= (* t_0 (+ t_1 (sqrt (- (* t_1 t_1) (* M M))))) INFINITY)
     (* t_0 (* (* (* (/ c0 (* (* (* h w) D) D)) 2.0) d_m) d_m))
     (/ (* c0 0.0) (* w 2.0)))))

C

d_m = fabs(d);
double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = c0 / (2.0 * w);
	double t_1 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if ((t_0 * (t_1 + sqrt(((t_1 * t_1) - (M * M))))) <= ((double) INFINITY)) {
		tmp = t_0 * ((((c0 / (((h * w) * D) * D)) * 2.0) * d_m) * d_m);
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Java

d_m = Math.abs(d);
public static double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = c0 / (2.0 * w);
	double t_1 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if ((t_0 * (t_1 + Math.sqrt(((t_1 * t_1) - (M * M))))) <= Double.POSITIVE_INFINITY) {
		tmp = t_0 * ((((c0 / (((h * w) * D) * D)) * 2.0) * d_m) * d_m);
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Python

d_m = math.fabs(d)
def code(c0, w, h, D, d_m, M):
	t_0 = c0 / (2.0 * w)
	t_1 = (c0 * (d_m * d_m)) / ((w * h) * (D * D))
	tmp = 0
	if (t_0 * (t_1 + math.sqrt(((t_1 * t_1) - (M * M))))) <= math.inf:
		tmp = t_0 * ((((c0 / (((h * w) * D) * D)) * 2.0) * d_m) * d_m)
	else:
		tmp = (c0 * 0.0) / (w * 2.0)
	return tmp

Julia

d_m = abs(d)
function code(c0, w, h, D, d_m, M)
	t_0 = Float64(c0 / Float64(2.0 * w))
	t_1 = Float64(Float64(c0 * Float64(d_m * d_m)) / Float64(Float64(w * h) * Float64(D * D)))
	tmp = 0.0
	if (Float64(t_0 * Float64(t_1 + sqrt(Float64(Float64(t_1 * t_1) - Float64(M * M))))) <= Inf)
		tmp = Float64(t_0 * Float64(Float64(Float64(Float64(c0 / Float64(Float64(Float64(h * w) * D) * D)) * 2.0) * d_m) * d_m));
	else
		tmp = Float64(Float64(c0 * 0.0) / Float64(w * 2.0));
	end
	return tmp
end

MATLAB

d_m = abs(d);
function tmp_2 = code(c0, w, h, D, d_m, M)
	t_0 = c0 / (2.0 * w);
	t_1 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	tmp = 0.0;
	if ((t_0 * (t_1 + sqrt(((t_1 * t_1) - (M * M))))) <= Inf)
		tmp = t_0 * ((((c0 / (((h * w) * D) * D)) * 2.0) * d_m) * d_m);
	else
		tmp = (c0 * 0.0) / (w * 2.0);
	end
	tmp_2 = tmp;
end

Wolfram

d_m = N[Abs[d], $MachinePrecision]
code[c0_, w_, h_, D_, d$95$m_, M_] := Block[{t$95$0 = N[(c0 / N[(2.0 * w), $MachinePrecision]), $MachinePrecision]}, Block[{t$95$1 = N[(N[(c0 * N[(d$95$m * d$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(w * h), $MachinePrecision] * N[(D * D), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[N[(t$95$0 * N[(t$95$1 + N[Sqrt[N[(N[(t$95$1 * t$95$1), $MachinePrecision] - N[(M * M), $MachinePrecision]), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]), $MachinePrecision], Infinity], N[(t$95$0 * N[(N[(N[(N[(c0 / N[(N[(N[(h * w), $MachinePrecision] * D), $MachinePrecision] * D), $MachinePrecision]), $MachinePrecision] * 2.0), $MachinePrecision] * d$95$m), $MachinePrecision] * d$95$m), $MachinePrecision]), $MachinePrecision], N[(N[(c0 * 0.0), $MachinePrecision] / N[(w * 2.0), $MachinePrecision]), $MachinePrecision]]]]

Derivation

1. Split input into 2 regimes

2. if (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M))))) < +inf.0

   1. Initial program 74.4%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in d around inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left({d}^{2} \cdot \left(\frac{-1}{2} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot \left(h \cdot w\right)\right)}{c0 \cdot {d}^{4}} + 2 \cdot \frac{c0}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{-1}{2} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot \left(h \cdot w\right)\right)}{c0 \cdot {d}^{4}} + 2 \cdot \frac{c0}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot \color{blue}{{d}^{2}}\right) \]

      2. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{-1}{2} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot \left(h \cdot w\right)\right)}{c0 \cdot {d}^{4}} + 2 \cdot \frac{c0}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot \color{blue}{{d}^{2}}\right) \]

   5. Applied rewrites 61.5%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(\mathsf{fma}\left(\frac{D \cdot D}{c0} \cdot \frac{\left(\left(M \cdot M\right) \cdot h\right) \cdot w}{\left(d \cdot d\right) \cdot \left(d \cdot d\right)}, -0.5, \frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot \left(d \cdot d\right)\right)} \]

   6. Taylor expanded in c0 around inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(2 \cdot \frac{c0}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot \left(\color{blue}{d} \cdot d\right)\right) \]
   7. Step-by-step derivation

      1. associate-*r/ N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\frac{2 \cdot c0}{{D}^{2} \cdot \left(h \cdot w\right)} \cdot \left(d \cdot d\right)\right) \]

      2. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\frac{2 \cdot c0}{\left(h \cdot w\right) \cdot {D}^{2}} \cdot \left(d \cdot d\right)\right) \]

      3. pow2 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\frac{2 \cdot c0}{\left(h \cdot w\right) \cdot \left(D \cdot D\right)} \cdot \left(d \cdot d\right)\right) \]

      4. associate-*l* N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

      5. times-frac N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      6. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      7. lower-/.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      8. lift-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      9. lift-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      10. lower-/.f64 75.5

          \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

   8. Applied rewrites 75.5%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(\color{blue}{d} \cdot d\right)\right) \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot \color{blue}{d}\right)\right) \]

      2. lift-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \color{blue}{\left(d \cdot d\right)}\right) \]

      3. associate-*r* N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot d\right) \cdot \color{blue}{d}\right) \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot d\right) \cdot \color{blue}{d}\right) \]

   10. Applied rewrites 77.1%

       \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\left(\frac{c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot 2\right) \cdot d\right) \cdot \color{blue}{d}\right) \]

   if +inf.0 < (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M)))))

   1. Initial program 0.0%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in c0 around -inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-1 \cdot \left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right)} \]
   4. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\mathsf{neg}\left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right) \]

      2. lower-neg.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right) \]

      3. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

   5. Applied rewrites 3.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-\left(0 \cdot \frac{d \cdot d}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot c0\right)} \]

   6. Taylor expanded in c0 around 0

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   7. Step-by-step derivation

   8. Applied rewrites 38.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \color{blue}{\frac{c0}{2 \cdot w} \cdot 0} \]

   10. Applied rewrites 45.5%

       \[\leadsto \color{blue}{\frac{c0 \cdot 0}{w \cdot 2}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 3: 55.0% accurate, 0.7× speedup

Math

\[\begin{array}{l} d_m = \left|d\right| \\ \begin{array}{l} t_0 := \frac{c0 \cdot \left(d\_m \cdot d\_m\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)}\\ \mathbf{if}\;\frac{c0}{2 \cdot w} \cdot \left(t\_0 + \sqrt{t\_0 \cdot t\_0 - M \cdot M}\right) \leq \infty:\\ \;\;\;\;\frac{c0}{w + w} \cdot \frac{2 \cdot \left(\left(d\_m \cdot d\_m\right) \cdot c0\right)}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\\ \mathbf{else}:\\ \;\;\;\;\frac{c0 \cdot 0}{w \cdot 2}\\ \end{array} \end{array} \]

FPCore

d_m = (fabs.f64 d)
(FPCore (c0 w h D d_m M)
 :precision binary64
 (let* ((t_0 (/ (* c0 (* d_m d_m)) (* (* w h) (* D D)))))
   (if (<=
        (* (/ c0 (* 2.0 w)) (+ t_0 (sqrt (- (* t_0 t_0) (* M M)))))
        INFINITY)
     (* (/ c0 (+ w w)) (/ (* 2.0 (* (* d_m d_m) c0)) (* (* (* h w) D) D)))
     (/ (* c0 0.0) (* w 2.0)))))

C

d_m = fabs(d);
double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if (((c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))))) <= ((double) INFINITY)) {
		tmp = (c0 / (w + w)) * ((2.0 * ((d_m * d_m) * c0)) / (((h * w) * D) * D));
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Java

d_m = Math.abs(d);
public static double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if (((c0 / (2.0 * w)) * (t_0 + Math.sqrt(((t_0 * t_0) - (M * M))))) <= Double.POSITIVE_INFINITY) {
		tmp = (c0 / (w + w)) * ((2.0 * ((d_m * d_m) * c0)) / (((h * w) * D) * D));
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Python

d_m = math.fabs(d)
def code(c0, w, h, D, d_m, M):
	t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D))
	tmp = 0
	if ((c0 / (2.0 * w)) * (t_0 + math.sqrt(((t_0 * t_0) - (M * M))))) <= math.inf:
		tmp = (c0 / (w + w)) * ((2.0 * ((d_m * d_m) * c0)) / (((h * w) * D) * D))
	else:
		tmp = (c0 * 0.0) / (w * 2.0)
	return tmp

Julia

d_m = abs(d)
function code(c0, w, h, D, d_m, M)
	t_0 = Float64(Float64(c0 * Float64(d_m * d_m)) / Float64(Float64(w * h) * Float64(D * D)))
	tmp = 0.0
	if (Float64(Float64(c0 / Float64(2.0 * w)) * Float64(t_0 + sqrt(Float64(Float64(t_0 * t_0) - Float64(M * M))))) <= Inf)
		tmp = Float64(Float64(c0 / Float64(w + w)) * Float64(Float64(2.0 * Float64(Float64(d_m * d_m) * c0)) / Float64(Float64(Float64(h * w) * D) * D)));
	else
		tmp = Float64(Float64(c0 * 0.0) / Float64(w * 2.0));
	end
	return tmp
end

MATLAB

d_m = abs(d);
function tmp_2 = code(c0, w, h, D, d_m, M)
	t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	tmp = 0.0;
	if (((c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))))) <= Inf)
		tmp = (c0 / (w + w)) * ((2.0 * ((d_m * d_m) * c0)) / (((h * w) * D) * D));
	else
		tmp = (c0 * 0.0) / (w * 2.0);
	end
	tmp_2 = tmp;
end

Wolfram

d_m = N[Abs[d], $MachinePrecision]
code[c0_, w_, h_, D_, d$95$m_, M_] := Block[{t$95$0 = N[(N[(c0 * N[(d$95$m * d$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(w * h), $MachinePrecision] * N[(D * D), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[N[(N[(c0 / N[(2.0 * w), $MachinePrecision]), $MachinePrecision] * N[(t$95$0 + N[Sqrt[N[(N[(t$95$0 * t$95$0), $MachinePrecision] - N[(M * M), $MachinePrecision]), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]), $MachinePrecision], Infinity], N[(N[(c0 / N[(w + w), $MachinePrecision]), $MachinePrecision] * N[(N[(2.0 * N[(N[(d$95$m * d$95$m), $MachinePrecision] * c0), $MachinePrecision]), $MachinePrecision] / N[(N[(N[(h * w), $MachinePrecision] * D), $MachinePrecision] * D), $MachinePrecision]), $MachinePrecision]), $MachinePrecision], N[(N[(c0 * 0.0), $MachinePrecision] / N[(w * 2.0), $MachinePrecision]), $MachinePrecision]]]

Derivation

1. Split input into 2 regimes

2. if (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M))))) < +inf.0

   1. Initial program 74.4%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in c0 around inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(2 \cdot \frac{c0 \cdot {d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)} \]
   4. Step-by-step derivation

      1. associate-*r/ N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(c0 \cdot {d}^{2}\right)}{\color{blue}{{D}^{2} \cdot \left(h \cdot w\right)}} \]

      2. lower-/.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(c0 \cdot {d}^{2}\right)}{\color{blue}{{D}^{2} \cdot \left(h \cdot w\right)}} \]

      3. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(c0 \cdot {d}^{2}\right)}{\color{blue}{{D}^{2}} \cdot \left(h \cdot w\right)} \]

      4. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left({d}^{2} \cdot c0\right)}{{D}^{\color{blue}{2}} \cdot \left(h \cdot w\right)} \]

      5. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left({d}^{2} \cdot c0\right)}{{D}^{\color{blue}{2}} \cdot \left(h \cdot w\right)} \]

      6. pow2 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{{D}^{2} \cdot \left(h \cdot w\right)} \]

      7. lift-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{{D}^{2} \cdot \left(h \cdot w\right)} \]

      8. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(h \cdot w\right) \cdot \color{blue}{{D}^{2}}} \]

      9. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(w \cdot h\right) \cdot {\color{blue}{D}}^{2}} \]

      10. pow2 N/A

          \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(w \cdot h\right) \cdot \left(D \cdot \color{blue}{D}\right)} \]

      11. associate-*r* N/A

          \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(\left(w \cdot h\right) \cdot D\right) \cdot \color{blue}{D}} \]

      12. lower-*.f64 N/A

          \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(\left(w \cdot h\right) \cdot D\right) \cdot \color{blue}{D}} \]

      13. *-commutative N/A

          \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \]

      14. lower-*.f64 N/A

          \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \]

      15. lower-*.f64 75.2

          \[\leadsto \frac{c0}{2 \cdot w} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \]

   5. Applied rewrites 75.2%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}} \]
   6. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \frac{c0}{\color{blue}{2 \cdot w}} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \]

      2. count-2-rev N/A

         \[\leadsto \frac{c0}{\color{blue}{w + w}} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \]

      3. lower-+.f64 75.2

         \[\leadsto \frac{c0}{\color{blue}{w + w}} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \]

   7. Applied rewrites 75.2%

      \[\leadsto \frac{c0}{\color{blue}{w + w}} \cdot \frac{2 \cdot \left(\left(d \cdot d\right) \cdot c0\right)}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \]

   if +inf.0 < (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M)))))

   1. Initial program 0.0%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in c0 around -inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-1 \cdot \left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right)} \]
   4. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\mathsf{neg}\left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right) \]

      2. lower-neg.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right) \]

      3. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

   5. Applied rewrites 3.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-\left(0 \cdot \frac{d \cdot d}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot c0\right)} \]

   6. Taylor expanded in c0 around 0

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   7. Step-by-step derivation

   8. Applied rewrites 38.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \color{blue}{\frac{c0}{2 \cdot w} \cdot 0} \]

   10. Applied rewrites 45.5%

       \[\leadsto \color{blue}{\frac{c0 \cdot 0}{w \cdot 2}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 4: 54.8% accurate, 0.7× speedup

Math

\[\begin{array}{l} d_m = \left|d\right| \\ \begin{array}{l} t_0 := \frac{c0 \cdot \left(d\_m \cdot d\_m\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)}\\ \mathbf{if}\;\frac{c0}{2 \cdot w} \cdot \left(t\_0 + \sqrt{t\_0 \cdot t\_0 - M \cdot M}\right) \leq \infty:\\ \;\;\;\;\frac{c0}{w + w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d\_m \cdot d\_m\right)\right)\\ \mathbf{else}:\\ \;\;\;\;\frac{c0 \cdot 0}{w \cdot 2}\\ \end{array} \end{array} \]

FPCore

d_m = (fabs.f64 d)
(FPCore (c0 w h D d_m M)
 :precision binary64
 (let* ((t_0 (/ (* c0 (* d_m d_m)) (* (* w h) (* D D)))))
   (if (<=
        (* (/ c0 (* 2.0 w)) (+ t_0 (sqrt (- (* t_0 t_0) (* M M)))))
        INFINITY)
     (* (/ c0 (+ w w)) (* (/ (* 2.0 c0) (* (* (* h w) D) D)) (* d_m d_m)))
     (/ (* c0 0.0) (* w 2.0)))))

C

d_m = fabs(d);
double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if (((c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))))) <= ((double) INFINITY)) {
		tmp = (c0 / (w + w)) * (((2.0 * c0) / (((h * w) * D) * D)) * (d_m * d_m));
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Java

d_m = Math.abs(d);
public static double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if (((c0 / (2.0 * w)) * (t_0 + Math.sqrt(((t_0 * t_0) - (M * M))))) <= Double.POSITIVE_INFINITY) {
		tmp = (c0 / (w + w)) * (((2.0 * c0) / (((h * w) * D) * D)) * (d_m * d_m));
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Python

d_m = math.fabs(d)
def code(c0, w, h, D, d_m, M):
	t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D))
	tmp = 0
	if ((c0 / (2.0 * w)) * (t_0 + math.sqrt(((t_0 * t_0) - (M * M))))) <= math.inf:
		tmp = (c0 / (w + w)) * (((2.0 * c0) / (((h * w) * D) * D)) * (d_m * d_m))
	else:
		tmp = (c0 * 0.0) / (w * 2.0)
	return tmp

Julia

d_m = abs(d)
function code(c0, w, h, D, d_m, M)
	t_0 = Float64(Float64(c0 * Float64(d_m * d_m)) / Float64(Float64(w * h) * Float64(D * D)))
	tmp = 0.0
	if (Float64(Float64(c0 / Float64(2.0 * w)) * Float64(t_0 + sqrt(Float64(Float64(t_0 * t_0) - Float64(M * M))))) <= Inf)
		tmp = Float64(Float64(c0 / Float64(w + w)) * Float64(Float64(Float64(2.0 * c0) / Float64(Float64(Float64(h * w) * D) * D)) * Float64(d_m * d_m)));
	else
		tmp = Float64(Float64(c0 * 0.0) / Float64(w * 2.0));
	end
	return tmp
end

MATLAB

d_m = abs(d);
function tmp_2 = code(c0, w, h, D, d_m, M)
	t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	tmp = 0.0;
	if (((c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))))) <= Inf)
		tmp = (c0 / (w + w)) * (((2.0 * c0) / (((h * w) * D) * D)) * (d_m * d_m));
	else
		tmp = (c0 * 0.0) / (w * 2.0);
	end
	tmp_2 = tmp;
end

Wolfram

d_m = N[Abs[d], $MachinePrecision]
code[c0_, w_, h_, D_, d$95$m_, M_] := Block[{t$95$0 = N[(N[(c0 * N[(d$95$m * d$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(w * h), $MachinePrecision] * N[(D * D), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[N[(N[(c0 / N[(2.0 * w), $MachinePrecision]), $MachinePrecision] * N[(t$95$0 + N[Sqrt[N[(N[(t$95$0 * t$95$0), $MachinePrecision] - N[(M * M), $MachinePrecision]), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]), $MachinePrecision], Infinity], N[(N[(c0 / N[(w + w), $MachinePrecision]), $MachinePrecision] * N[(N[(N[(2.0 * c0), $MachinePrecision] / N[(N[(N[(h * w), $MachinePrecision] * D), $MachinePrecision] * D), $MachinePrecision]), $MachinePrecision] * N[(d$95$m * d$95$m), $MachinePrecision]), $MachinePrecision]), $MachinePrecision], N[(N[(c0 * 0.0), $MachinePrecision] / N[(w * 2.0), $MachinePrecision]), $MachinePrecision]]]

Derivation

1. Split input into 2 regimes

2. if (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M))))) < +inf.0

   1. Initial program 74.4%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in d around inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left({d}^{2} \cdot \left(\frac{-1}{2} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot \left(h \cdot w\right)\right)}{c0 \cdot {d}^{4}} + 2 \cdot \frac{c0}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{-1}{2} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot \left(h \cdot w\right)\right)}{c0 \cdot {d}^{4}} + 2 \cdot \frac{c0}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot \color{blue}{{d}^{2}}\right) \]

      2. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{-1}{2} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot \left(h \cdot w\right)\right)}{c0 \cdot {d}^{4}} + 2 \cdot \frac{c0}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot \color{blue}{{d}^{2}}\right) \]

   5. Applied rewrites 61.5%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(\mathsf{fma}\left(\frac{D \cdot D}{c0} \cdot \frac{\left(\left(M \cdot M\right) \cdot h\right) \cdot w}{\left(d \cdot d\right) \cdot \left(d \cdot d\right)}, -0.5, \frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot \left(d \cdot d\right)\right)} \]

   6. Taylor expanded in c0 around inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(2 \cdot \frac{c0}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot \left(\color{blue}{d} \cdot d\right)\right) \]
   7. Step-by-step derivation

      1. associate-*r/ N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\frac{2 \cdot c0}{{D}^{2} \cdot \left(h \cdot w\right)} \cdot \left(d \cdot d\right)\right) \]

      2. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\frac{2 \cdot c0}{\left(h \cdot w\right) \cdot {D}^{2}} \cdot \left(d \cdot d\right)\right) \]

      3. pow2 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\frac{2 \cdot c0}{\left(h \cdot w\right) \cdot \left(D \cdot D\right)} \cdot \left(d \cdot d\right)\right) \]

      4. associate-*l* N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

      5. times-frac N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      6. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      7. lower-/.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      8. lift-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      9. lift-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      10. lower-/.f64 75.5

          \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

   8. Applied rewrites 75.5%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(\color{blue}{d} \cdot d\right)\right) \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \frac{c0}{\color{blue}{2 \cdot w}} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      2. count-2-rev N/A

         \[\leadsto \frac{c0}{\color{blue}{w + w}} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

      3. lower-+.f64 75.5

         \[\leadsto \frac{c0}{\color{blue}{w + w}} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

   10. Applied rewrites 75.5%

       \[\leadsto \frac{c0}{\color{blue}{w + w}} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]
   11. Step-by-step derivation

       1. lift-*.f64 N/A

          \[\leadsto \frac{c0}{w + w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

       2. lift-/.f64 N/A

          \[\leadsto \frac{c0}{w + w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

       3. lift-/.f64 N/A

          \[\leadsto \frac{c0}{w + w} \cdot \left(\left(\frac{2}{\left(h \cdot w\right) \cdot D} \cdot \frac{c0}{D}\right) \cdot \left(d \cdot d\right)\right) \]

       4. frac-times N/A

          \[\leadsto \frac{c0}{w + w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

       5. lift-*.f64 N/A

          \[\leadsto \frac{c0}{w + w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

       6. lift-*.f64 N/A

          \[\leadsto \frac{c0}{w + w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

       7. lower-/.f64 N/A

          \[\leadsto \frac{c0}{w + w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

       8. lower-*.f64 N/A

          \[\leadsto \frac{c0}{w + w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

       9. lift-*.f64 N/A

          \[\leadsto \frac{c0}{w + w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

       10. lift-*.f64 N/A

           \[\leadsto \frac{c0}{w + w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

       11. lift-*.f64 74.9

           \[\leadsto \frac{c0}{w + w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

   12. Applied rewrites 74.9%

       \[\leadsto \frac{c0}{w + w} \cdot \left(\frac{2 \cdot c0}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D} \cdot \left(d \cdot d\right)\right) \]

   if +inf.0 < (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M)))))

   1. Initial program 0.0%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in c0 around -inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-1 \cdot \left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right)} \]
   4. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\mathsf{neg}\left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right) \]

      2. lower-neg.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right) \]

      3. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

   5. Applied rewrites 3.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-\left(0 \cdot \frac{d \cdot d}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot c0\right)} \]

   6. Taylor expanded in c0 around 0

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   7. Step-by-step derivation

   8. Applied rewrites 38.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \color{blue}{\frac{c0}{2 \cdot w} \cdot 0} \]

   10. Applied rewrites 45.5%

       \[\leadsto \color{blue}{\frac{c0 \cdot 0}{w \cdot 2}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 5: 49.0% accurate, 0.7× speedup

Math

\[\begin{array}{l} d_m = \left|d\right| \\ \begin{array}{l} t_0 := \frac{c0 \cdot \left(d\_m \cdot d\_m\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)}\\ \mathbf{if}\;\frac{c0}{2 \cdot w} \cdot \left(t\_0 + \sqrt{t\_0 \cdot t\_0 - M \cdot M}\right) \leq \infty:\\ \;\;\;\;\frac{\frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d\_m \cdot d\_m\right)}{w \cdot w}}{D \cdot D}}{h}\\ \mathbf{else}:\\ \;\;\;\;\frac{c0 \cdot 0}{w \cdot 2}\\ \end{array} \end{array} \]

FPCore

d_m = (fabs.f64 d)
(FPCore (c0 w h D d_m M)
 :precision binary64
 (let* ((t_0 (/ (* c0 (* d_m d_m)) (* (* w h) (* D D)))))
   (if (<=
        (* (/ c0 (* 2.0 w)) (+ t_0 (sqrt (- (* t_0 t_0) (* M M)))))
        INFINITY)
     (/ (/ (/ (* (* c0 c0) (* d_m d_m)) (* w w)) (* D D)) h)
     (/ (* c0 0.0) (* w 2.0)))))

C

d_m = fabs(d);
double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if (((c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))))) <= ((double) INFINITY)) {
		tmp = ((((c0 * c0) * (d_m * d_m)) / (w * w)) / (D * D)) / h;
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Java

d_m = Math.abs(d);
public static double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if (((c0 / (2.0 * w)) * (t_0 + Math.sqrt(((t_0 * t_0) - (M * M))))) <= Double.POSITIVE_INFINITY) {
		tmp = ((((c0 * c0) * (d_m * d_m)) / (w * w)) / (D * D)) / h;
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Python

d_m = math.fabs(d)
def code(c0, w, h, D, d_m, M):
	t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D))
	tmp = 0
	if ((c0 / (2.0 * w)) * (t_0 + math.sqrt(((t_0 * t_0) - (M * M))))) <= math.inf:
		tmp = ((((c0 * c0) * (d_m * d_m)) / (w * w)) / (D * D)) / h
	else:
		tmp = (c0 * 0.0) / (w * 2.0)
	return tmp

Julia

d_m = abs(d)
function code(c0, w, h, D, d_m, M)
	t_0 = Float64(Float64(c0 * Float64(d_m * d_m)) / Float64(Float64(w * h) * Float64(D * D)))
	tmp = 0.0
	if (Float64(Float64(c0 / Float64(2.0 * w)) * Float64(t_0 + sqrt(Float64(Float64(t_0 * t_0) - Float64(M * M))))) <= Inf)
		tmp = Float64(Float64(Float64(Float64(Float64(c0 * c0) * Float64(d_m * d_m)) / Float64(w * w)) / Float64(D * D)) / h);
	else
		tmp = Float64(Float64(c0 * 0.0) / Float64(w * 2.0));
	end
	return tmp
end

MATLAB

d_m = abs(d);
function tmp_2 = code(c0, w, h, D, d_m, M)
	t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	tmp = 0.0;
	if (((c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))))) <= Inf)
		tmp = ((((c0 * c0) * (d_m * d_m)) / (w * w)) / (D * D)) / h;
	else
		tmp = (c0 * 0.0) / (w * 2.0);
	end
	tmp_2 = tmp;
end

Wolfram

d_m = N[Abs[d], $MachinePrecision]
code[c0_, w_, h_, D_, d$95$m_, M_] := Block[{t$95$0 = N[(N[(c0 * N[(d$95$m * d$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(w * h), $MachinePrecision] * N[(D * D), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[N[(N[(c0 / N[(2.0 * w), $MachinePrecision]), $MachinePrecision] * N[(t$95$0 + N[Sqrt[N[(N[(t$95$0 * t$95$0), $MachinePrecision] - N[(M * M), $MachinePrecision]), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]), $MachinePrecision], Infinity], N[(N[(N[(N[(N[(c0 * c0), $MachinePrecision] * N[(d$95$m * d$95$m), $MachinePrecision]), $MachinePrecision] / N[(w * w), $MachinePrecision]), $MachinePrecision] / N[(D * D), $MachinePrecision]), $MachinePrecision] / h), $MachinePrecision], N[(N[(c0 * 0.0), $MachinePrecision] / N[(w * 2.0), $MachinePrecision]), $MachinePrecision]]]

Derivation

1. Split input into 2 regimes

2. if (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M))))) < +inf.0

   1. Initial program 74.4%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in h around 0

      \[\leadsto \color{blue}{\frac{\frac{-1}{4} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot {h}^{2}\right)}{{d}^{2}} + \frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h}} \]
   4. Step-by-step derivation

      1. lower-/.f64 N/A

         \[\leadsto \frac{\frac{-1}{4} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot {h}^{2}\right)}{{d}^{2}} + \frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{\color{blue}{h}} \]

   5. Applied rewrites 44.9%

      \[\leadsto \color{blue}{\frac{\mathsf{fma}\left(\frac{c0 \cdot c0}{D \cdot D}, \frac{d \cdot d}{w \cdot w}, \frac{\left(\left(M \cdot M\right) \cdot \left(h \cdot h\right)\right) \cdot \left(D \cdot D\right)}{d \cdot d} \cdot -0.25\right)}{h}} \]

   6. Taylor expanded in c0 around inf

      \[\leadsto \frac{\frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h} \]
   7. Step-by-step derivation

      1. frac-times N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{{d}^{2}}{{w}^{2}}}{h} \]

      2. pow2 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{{w}^{2}}}{h} \]

      3. pow2 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      5. pow2 N/A

         \[\leadsto \frac{\frac{c0 \cdot c0}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      6. pow2 N/A

         \[\leadsto \frac{\frac{c0 \cdot c0}{D \cdot D} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      7. times-frac N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      8. lower-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      9. lower-/.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      10. lower-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      11. times-frac N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      12. lower-*.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      13. lower-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      14. lower-/.f64 68.7

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

   8. Applied rewrites 68.7%

      \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      2. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      3. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      4. unswap-sqr N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

      5. lower-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

      6. lower-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

      7. lower-*.f64 79.6

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

   10. Applied rewrites 79.6%

       \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]
   11. Step-by-step derivation

       1. lift-*.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

       2. lift-*.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

       3. lift-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

       4. lift-*.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

       5. lift-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

       6. swap-sqr N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

       7. lift-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

       8. lift-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

       9. times-frac N/A

          \[\leadsto \frac{\frac{c0 \cdot c0}{D \cdot D} \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

       10. times-frac N/A

           \[\leadsto \frac{\frac{c0 \cdot c0}{D \cdot D} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

       11. pow2 N/A

           \[\leadsto \frac{\frac{{c0}^{2}}{D \cdot D} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

       12. pow2 N/A

           \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

       13. associate-*l/ N/A

           \[\leadsto \frac{\frac{{c0}^{2} \cdot \frac{d \cdot d}{w \cdot w}}{{D}^{2}}}{h} \]

       14. pow2 N/A

           \[\leadsto \frac{\frac{{c0}^{2} \cdot \frac{{d}^{2}}{w \cdot w}}{{D}^{2}}}{h} \]

       15. pow2 N/A

           \[\leadsto \frac{\frac{{c0}^{2} \cdot \frac{{d}^{2}}{{w}^{2}}}{{D}^{2}}}{h} \]

       16. associate-/l* N/A

           \[\leadsto \frac{\frac{\frac{{c0}^{2} \cdot {d}^{2}}{{w}^{2}}}{{D}^{2}}}{h} \]

       17. lower-/.f64 N/A

           \[\leadsto \frac{\frac{\frac{{c0}^{2} \cdot {d}^{2}}{{w}^{2}}}{{D}^{2}}}{h} \]

   12. Applied rewrites 56.7%

       \[\leadsto \frac{\frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d \cdot d\right)}{w \cdot w}}{D \cdot D}}{h} \]

   if +inf.0 < (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M)))))

   1. Initial program 0.0%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in c0 around -inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-1 \cdot \left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right)} \]
   4. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\mathsf{neg}\left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right) \]

      2. lower-neg.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right) \]

      3. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

   5. Applied rewrites 3.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-\left(0 \cdot \frac{d \cdot d}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot c0\right)} \]

   6. Taylor expanded in c0 around 0

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   7. Step-by-step derivation

   8. Applied rewrites 38.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \color{blue}{\frac{c0}{2 \cdot w} \cdot 0} \]

   10. Applied rewrites 45.5%

       \[\leadsto \color{blue}{\frac{c0 \cdot 0}{w \cdot 2}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 6: 48.8% accurate, 0.7× speedup

Math

\[\begin{array}{l} d_m = \left|d\right| \\ \begin{array}{l} t_0 := \frac{c0 \cdot \left(d\_m \cdot d\_m\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)}\\ \mathbf{if}\;\frac{c0}{2 \cdot w} \cdot \left(t\_0 + \sqrt{t\_0 \cdot t\_0 - M \cdot M}\right) \leq \infty:\\ \;\;\;\;\frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d\_m \cdot d\_m\right)}{\left(w \cdot w\right) \cdot \left(D \cdot D\right)}}{h}\\ \mathbf{else}:\\ \;\;\;\;\frac{c0 \cdot 0}{w \cdot 2}\\ \end{array} \end{array} \]

FPCore

d_m = (fabs.f64 d)
(FPCore (c0 w h D d_m M)
 :precision binary64
 (let* ((t_0 (/ (* c0 (* d_m d_m)) (* (* w h) (* D D)))))
   (if (<=
        (* (/ c0 (* 2.0 w)) (+ t_0 (sqrt (- (* t_0 t_0) (* M M)))))
        INFINITY)
     (/ (/ (* (* c0 c0) (* d_m d_m)) (* (* w w) (* D D))) h)
     (/ (* c0 0.0) (* w 2.0)))))

C

d_m = fabs(d);
double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if (((c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))))) <= ((double) INFINITY)) {
		tmp = (((c0 * c0) * (d_m * d_m)) / ((w * w) * (D * D))) / h;
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Java

d_m = Math.abs(d);
public static double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if (((c0 / (2.0 * w)) * (t_0 + Math.sqrt(((t_0 * t_0) - (M * M))))) <= Double.POSITIVE_INFINITY) {
		tmp = (((c0 * c0) * (d_m * d_m)) / ((w * w) * (D * D))) / h;
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Python

d_m = math.fabs(d)
def code(c0, w, h, D, d_m, M):
	t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D))
	tmp = 0
	if ((c0 / (2.0 * w)) * (t_0 + math.sqrt(((t_0 * t_0) - (M * M))))) <= math.inf:
		tmp = (((c0 * c0) * (d_m * d_m)) / ((w * w) * (D * D))) / h
	else:
		tmp = (c0 * 0.0) / (w * 2.0)
	return tmp

Julia

d_m = abs(d)
function code(c0, w, h, D, d_m, M)
	t_0 = Float64(Float64(c0 * Float64(d_m * d_m)) / Float64(Float64(w * h) * Float64(D * D)))
	tmp = 0.0
	if (Float64(Float64(c0 / Float64(2.0 * w)) * Float64(t_0 + sqrt(Float64(Float64(t_0 * t_0) - Float64(M * M))))) <= Inf)
		tmp = Float64(Float64(Float64(Float64(c0 * c0) * Float64(d_m * d_m)) / Float64(Float64(w * w) * Float64(D * D))) / h);
	else
		tmp = Float64(Float64(c0 * 0.0) / Float64(w * 2.0));
	end
	return tmp
end

MATLAB

d_m = abs(d);
function tmp_2 = code(c0, w, h, D, d_m, M)
	t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	tmp = 0.0;
	if (((c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))))) <= Inf)
		tmp = (((c0 * c0) * (d_m * d_m)) / ((w * w) * (D * D))) / h;
	else
		tmp = (c0 * 0.0) / (w * 2.0);
	end
	tmp_2 = tmp;
end

Wolfram

d_m = N[Abs[d], $MachinePrecision]
code[c0_, w_, h_, D_, d$95$m_, M_] := Block[{t$95$0 = N[(N[(c0 * N[(d$95$m * d$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(w * h), $MachinePrecision] * N[(D * D), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[N[(N[(c0 / N[(2.0 * w), $MachinePrecision]), $MachinePrecision] * N[(t$95$0 + N[Sqrt[N[(N[(t$95$0 * t$95$0), $MachinePrecision] - N[(M * M), $MachinePrecision]), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]), $MachinePrecision], Infinity], N[(N[(N[(N[(c0 * c0), $MachinePrecision] * N[(d$95$m * d$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(w * w), $MachinePrecision] * N[(D * D), $MachinePrecision]), $MachinePrecision]), $MachinePrecision] / h), $MachinePrecision], N[(N[(c0 * 0.0), $MachinePrecision] / N[(w * 2.0), $MachinePrecision]), $MachinePrecision]]]

Derivation

1. Split input into 2 regimes

2. if (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M))))) < +inf.0

   1. Initial program 74.4%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in h around 0

      \[\leadsto \color{blue}{\frac{\frac{-1}{4} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot {h}^{2}\right)}{{d}^{2}} + \frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h}} \]
   4. Step-by-step derivation

      1. lower-/.f64 N/A

         \[\leadsto \frac{\frac{-1}{4} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot {h}^{2}\right)}{{d}^{2}} + \frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{\color{blue}{h}} \]

   5. Applied rewrites 44.9%

      \[\leadsto \color{blue}{\frac{\mathsf{fma}\left(\frac{c0 \cdot c0}{D \cdot D}, \frac{d \cdot d}{w \cdot w}, \frac{\left(\left(M \cdot M\right) \cdot \left(h \cdot h\right)\right) \cdot \left(D \cdot D\right)}{d \cdot d} \cdot -0.25\right)}{h}} \]

   6. Taylor expanded in c0 around inf

      \[\leadsto \frac{\frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h} \]
   7. Step-by-step derivation

      1. frac-times N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{{d}^{2}}{{w}^{2}}}{h} \]

      2. pow2 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{{w}^{2}}}{h} \]

      3. pow2 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      5. pow2 N/A

         \[\leadsto \frac{\frac{c0 \cdot c0}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      6. pow2 N/A

         \[\leadsto \frac{\frac{c0 \cdot c0}{D \cdot D} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      7. times-frac N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      8. lower-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      9. lower-/.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      10. lower-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      11. times-frac N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      12. lower-*.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      13. lower-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      14. lower-/.f64 68.7

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

   8. Applied rewrites 68.7%

      \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      2. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      3. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      4. lift-/.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      5. lift-/.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      6. frac-times N/A

         \[\leadsto \frac{\frac{c0 \cdot c0}{D \cdot D} \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      7. pow2 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{D \cdot D} \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      8. pow2 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      9. lift-/.f64 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      10. lift-/.f64 N/A

          \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      11. frac-times N/A

          \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      12. pow2 N/A

          \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{{d}^{2}}{w \cdot w}}{h} \]

      13. pow2 N/A

          \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{{d}^{2}}{{w}^{2}}}{h} \]

      14. frac-times N/A

          \[\leadsto \frac{\frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h} \]

      15. lower-/.f64 N/A

          \[\leadsto \frac{\frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h} \]

      16. lower-*.f64 N/A

          \[\leadsto \frac{\frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h} \]

      17. pow2 N/A

          \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h} \]

      18. lift-*.f64 N/A

          \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h} \]

      19. pow2 N/A

          \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d \cdot d\right)}{{D}^{2} \cdot {w}^{2}}}{h} \]

      20. lift-*.f64 N/A

          \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d \cdot d\right)}{{D}^{2} \cdot {w}^{2}}}{h} \]

      21. *-commutative N/A

          \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d \cdot d\right)}{{w}^{2} \cdot {D}^{2}}}{h} \]

      22. lower-*.f64 N/A

          \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d \cdot d\right)}{{w}^{2} \cdot {D}^{2}}}{h} \]

      23. pow2 N/A

          \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d \cdot d\right)}{\left(w \cdot w\right) \cdot {D}^{2}}}{h} \]

      24. lower-*.f64 N/A

          \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d \cdot d\right)}{\left(w \cdot w\right) \cdot {D}^{2}}}{h} \]

      25. pow2 N/A

          \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d \cdot d\right)}{\left(w \cdot w\right) \cdot \left(D \cdot D\right)}}{h} \]

      26. lift-*.f64 55.9

          \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d \cdot d\right)}{\left(w \cdot w\right) \cdot \left(D \cdot D\right)}}{h} \]

   10. Applied rewrites 55.9%

       \[\leadsto \frac{\frac{\left(c0 \cdot c0\right) \cdot \left(d \cdot d\right)}{\left(w \cdot w\right) \cdot \left(D \cdot D\right)}}{h} \]

   if +inf.0 < (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M)))))

   1. Initial program 0.0%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in c0 around -inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-1 \cdot \left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right)} \]
   4. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\mathsf{neg}\left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right) \]

      2. lower-neg.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right) \]

      3. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

   5. Applied rewrites 3.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-\left(0 \cdot \frac{d \cdot d}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot c0\right)} \]

   6. Taylor expanded in c0 around 0

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   7. Step-by-step derivation

   8. Applied rewrites 38.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \color{blue}{\frac{c0}{2 \cdot w} \cdot 0} \]

   10. Applied rewrites 45.5%

       \[\leadsto \color{blue}{\frac{c0 \cdot 0}{w \cdot 2}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 7: 48.5% accurate, 0.7× speedup

Math

\[\begin{array}{l} d_m = \left|d\right| \\ \begin{array}{l} t_0 := \frac{c0 \cdot \left(d\_m \cdot d\_m\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)}\\ \mathbf{if}\;\frac{c0}{2 \cdot w} \cdot \left(t\_0 + \sqrt{t\_0 \cdot t\_0 - M \cdot M}\right) \leq \infty:\\ \;\;\;\;\left(c0 \cdot c0\right) \cdot \frac{d\_m \cdot d\_m}{\left(\left(D \cdot D\right) \cdot h\right) \cdot \left(w \cdot w\right)}\\ \mathbf{else}:\\ \;\;\;\;\frac{c0 \cdot 0}{w \cdot 2}\\ \end{array} \end{array} \]

FPCore

d_m = (fabs.f64 d)
(FPCore (c0 w h D d_m M)
 :precision binary64
 (let* ((t_0 (/ (* c0 (* d_m d_m)) (* (* w h) (* D D)))))
   (if (<=
        (* (/ c0 (* 2.0 w)) (+ t_0 (sqrt (- (* t_0 t_0) (* M M)))))
        INFINITY)
     (* (* c0 c0) (/ (* d_m d_m) (* (* (* D D) h) (* w w))))
     (/ (* c0 0.0) (* w 2.0)))))

C

d_m = fabs(d);
double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if (((c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))))) <= ((double) INFINITY)) {
		tmp = (c0 * c0) * ((d_m * d_m) / (((D * D) * h) * (w * w)));
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Java

d_m = Math.abs(d);
public static double code(double c0, double w, double h, double D, double d_m, double M) {
	double t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	double tmp;
	if (((c0 / (2.0 * w)) * (t_0 + Math.sqrt(((t_0 * t_0) - (M * M))))) <= Double.POSITIVE_INFINITY) {
		tmp = (c0 * c0) * ((d_m * d_m) / (((D * D) * h) * (w * w)));
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Python

d_m = math.fabs(d)
def code(c0, w, h, D, d_m, M):
	t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D))
	tmp = 0
	if ((c0 / (2.0 * w)) * (t_0 + math.sqrt(((t_0 * t_0) - (M * M))))) <= math.inf:
		tmp = (c0 * c0) * ((d_m * d_m) / (((D * D) * h) * (w * w)))
	else:
		tmp = (c0 * 0.0) / (w * 2.0)
	return tmp

Julia

d_m = abs(d)
function code(c0, w, h, D, d_m, M)
	t_0 = Float64(Float64(c0 * Float64(d_m * d_m)) / Float64(Float64(w * h) * Float64(D * D)))
	tmp = 0.0
	if (Float64(Float64(c0 / Float64(2.0 * w)) * Float64(t_0 + sqrt(Float64(Float64(t_0 * t_0) - Float64(M * M))))) <= Inf)
		tmp = Float64(Float64(c0 * c0) * Float64(Float64(d_m * d_m) / Float64(Float64(Float64(D * D) * h) * Float64(w * w))));
	else
		tmp = Float64(Float64(c0 * 0.0) / Float64(w * 2.0));
	end
	return tmp
end

MATLAB

d_m = abs(d);
function tmp_2 = code(c0, w, h, D, d_m, M)
	t_0 = (c0 * (d_m * d_m)) / ((w * h) * (D * D));
	tmp = 0.0;
	if (((c0 / (2.0 * w)) * (t_0 + sqrt(((t_0 * t_0) - (M * M))))) <= Inf)
		tmp = (c0 * c0) * ((d_m * d_m) / (((D * D) * h) * (w * w)));
	else
		tmp = (c0 * 0.0) / (w * 2.0);
	end
	tmp_2 = tmp;
end

Wolfram

d_m = N[Abs[d], $MachinePrecision]
code[c0_, w_, h_, D_, d$95$m_, M_] := Block[{t$95$0 = N[(N[(c0 * N[(d$95$m * d$95$m), $MachinePrecision]), $MachinePrecision] / N[(N[(w * h), $MachinePrecision] * N[(D * D), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]}, If[LessEqual[N[(N[(c0 / N[(2.0 * w), $MachinePrecision]), $MachinePrecision] * N[(t$95$0 + N[Sqrt[N[(N[(t$95$0 * t$95$0), $MachinePrecision] - N[(M * M), $MachinePrecision]), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]), $MachinePrecision], Infinity], N[(N[(c0 * c0), $MachinePrecision] * N[(N[(d$95$m * d$95$m), $MachinePrecision] / N[(N[(N[(D * D), $MachinePrecision] * h), $MachinePrecision] * N[(w * w), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]), $MachinePrecision], N[(N[(c0 * 0.0), $MachinePrecision] / N[(w * 2.0), $MachinePrecision]), $MachinePrecision]]]

Derivation

1. Split input into 2 regimes

2. if (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M))))) < +inf.0

   1. Initial program 74.4%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in c0 around inf

      \[\leadsto \color{blue}{\frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot \left(h \cdot {w}^{2}\right)}} \]
   4. Step-by-step derivation

      1. associate-/l* N/A

         \[\leadsto {c0}^{2} \cdot \color{blue}{\frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot {w}^{2}\right)}} \]

      2. lower-*.f64 N/A

         \[\leadsto {c0}^{2} \cdot \color{blue}{\frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot {w}^{2}\right)}} \]

      3. unpow2 N/A

         \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{\color{blue}{{d}^{2}}}{{D}^{2} \cdot \left(h \cdot {w}^{2}\right)} \]

      4. lower-*.f64 N/A

         \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{\color{blue}{{d}^{2}}}{{D}^{2} \cdot \left(h \cdot {w}^{2}\right)} \]

      5. lower-/.f64 N/A

         \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{{d}^{2}}{\color{blue}{{D}^{2} \cdot \left(h \cdot {w}^{2}\right)}} \]

      6. pow2 N/A

         \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{d \cdot d}{\color{blue}{{D}^{2}} \cdot \left(h \cdot {w}^{2}\right)} \]

      7. lift-*.f64 N/A

         \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{d \cdot d}{\color{blue}{{D}^{2}} \cdot \left(h \cdot {w}^{2}\right)} \]

      8. associate-*r* N/A

         \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{d \cdot d}{\left({D}^{2} \cdot h\right) \cdot \color{blue}{{w}^{2}}} \]

      9. lower-*.f64 N/A

         \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{d \cdot d}{\left({D}^{2} \cdot h\right) \cdot \color{blue}{{w}^{2}}} \]

      10. lower-*.f64 N/A

          \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{d \cdot d}{\left({D}^{2} \cdot h\right) \cdot {\color{blue}{w}}^{2}} \]

      11. pow2 N/A

          \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{d \cdot d}{\left(\left(D \cdot D\right) \cdot h\right) \cdot {w}^{2}} \]

      12. lift-*.f64 N/A

          \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{d \cdot d}{\left(\left(D \cdot D\right) \cdot h\right) \cdot {w}^{2}} \]

      13. unpow2 N/A

          \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{d \cdot d}{\left(\left(D \cdot D\right) \cdot h\right) \cdot \left(w \cdot \color{blue}{w}\right)} \]

      14. lower-*.f64 54.9

          \[\leadsto \left(c0 \cdot c0\right) \cdot \frac{d \cdot d}{\left(\left(D \cdot D\right) \cdot h\right) \cdot \left(w \cdot \color{blue}{w}\right)} \]

   5. Applied rewrites 54.9%

      \[\leadsto \color{blue}{\left(c0 \cdot c0\right) \cdot \frac{d \cdot d}{\left(\left(D \cdot D\right) \cdot h\right) \cdot \left(w \cdot w\right)}} \]

   if +inf.0 < (*.f64 (/.f64 c0 (*.f64 #s(literal 2 binary64) w)) (+.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (sqrt.f64 (-.f64 (*.f64 (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D))) (/.f64 (*.f64 c0 (*.f64 d d)) (*.f64 (*.f64 w h) (*.f64 D D)))) (*.f64 M M)))))

   1. Initial program 0.0%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in c0 around -inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-1 \cdot \left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right)} \]
   4. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\mathsf{neg}\left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right) \]

      2. lower-neg.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right) \]

      3. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

   5. Applied rewrites 3.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-\left(0 \cdot \frac{d \cdot d}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot c0\right)} \]

   6. Taylor expanded in c0 around 0

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   7. Step-by-step derivation

   8. Applied rewrites 38.7%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \color{blue}{\frac{c0}{2 \cdot w} \cdot 0} \]

   10. Applied rewrites 45.5%

       \[\leadsto \color{blue}{\frac{c0 \cdot 0}{w \cdot 2}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 8: 46.9% accurate, 2.2× speedup

Math

\[\begin{array}{l} d_m = \left|d\right| \\ \begin{array}{l} \mathbf{if}\;d\_m \leq 2.85 \cdot 10^{+256}:\\ \;\;\;\;\frac{\left(\frac{c0}{D} \cdot \frac{d\_m}{w}\right) \cdot \left(c0 \cdot \frac{d\_m}{D \cdot w}\right)}{h}\\ \mathbf{else}:\\ \;\;\;\;\frac{c0 \cdot 0}{w \cdot 2}\\ \end{array} \end{array} \]

FPCore

d_m = (fabs.f64 d)
(FPCore (c0 w h D d_m M)
 :precision binary64
 (if (<= d_m 2.85e+256)
   (/ (* (* (/ c0 D) (/ d_m w)) (* c0 (/ d_m (* D w)))) h)
   (/ (* c0 0.0) (* w 2.0))))

C

d_m = fabs(d);
double code(double c0, double w, double h, double D, double d_m, double M) {
	double tmp;
	if (d_m <= 2.85e+256) {
		tmp = (((c0 / D) * (d_m / w)) * (c0 * (d_m / (D * w)))) / h;
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Fortran

d_m =     private
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(c0, w, h, d, d_m, m)
use fmin_fmax_functions
    real(8), intent (in) :: c0
    real(8), intent (in) :: w
    real(8), intent (in) :: h
    real(8), intent (in) :: d
    real(8), intent (in) :: d_m
    real(8), intent (in) :: m
    real(8) :: tmp
    if (d_m <= 2.85d+256) then
        tmp = (((c0 / d) * (d_m / w)) * (c0 * (d_m / (d * w)))) / h
    else
        tmp = (c0 * 0.0d0) / (w * 2.0d0)
    end if
    code = tmp
end function

Java

d_m = Math.abs(d);
public static double code(double c0, double w, double h, double D, double d_m, double M) {
	double tmp;
	if (d_m <= 2.85e+256) {
		tmp = (((c0 / D) * (d_m / w)) * (c0 * (d_m / (D * w)))) / h;
	} else {
		tmp = (c0 * 0.0) / (w * 2.0);
	}
	return tmp;
}

Python

d_m = math.fabs(d)
def code(c0, w, h, D, d_m, M):
	tmp = 0
	if d_m <= 2.85e+256:
		tmp = (((c0 / D) * (d_m / w)) * (c0 * (d_m / (D * w)))) / h
	else:
		tmp = (c0 * 0.0) / (w * 2.0)
	return tmp

Julia

d_m = abs(d)
function code(c0, w, h, D, d_m, M)
	tmp = 0.0
	if (d_m <= 2.85e+256)
		tmp = Float64(Float64(Float64(Float64(c0 / D) * Float64(d_m / w)) * Float64(c0 * Float64(d_m / Float64(D * w)))) / h);
	else
		tmp = Float64(Float64(c0 * 0.0) / Float64(w * 2.0));
	end
	return tmp
end

MATLAB

d_m = abs(d);
function tmp_2 = code(c0, w, h, D, d_m, M)
	tmp = 0.0;
	if (d_m <= 2.85e+256)
		tmp = (((c0 / D) * (d_m / w)) * (c0 * (d_m / (D * w)))) / h;
	else
		tmp = (c0 * 0.0) / (w * 2.0);
	end
	tmp_2 = tmp;
end

Wolfram

d_m = N[Abs[d], $MachinePrecision]
code[c0_, w_, h_, D_, d$95$m_, M_] := If[LessEqual[d$95$m, 2.85e+256], N[(N[(N[(N[(c0 / D), $MachinePrecision] * N[(d$95$m / w), $MachinePrecision]), $MachinePrecision] * N[(c0 * N[(d$95$m / N[(D * w), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]), $MachinePrecision] / h), $MachinePrecision], N[(N[(c0 * 0.0), $MachinePrecision] / N[(w * 2.0), $MachinePrecision]), $MachinePrecision]]

Derivation

1. Split input into 2 regimes

2. if d < 2.8499999999999999e256

   1. Initial program 24.2%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in h around 0

      \[\leadsto \color{blue}{\frac{\frac{-1}{4} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot {h}^{2}\right)}{{d}^{2}} + \frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h}} \]
   4. Step-by-step derivation

      1. lower-/.f64 N/A

         \[\leadsto \frac{\frac{-1}{4} \cdot \frac{{D}^{2} \cdot \left({M}^{2} \cdot {h}^{2}\right)}{{d}^{2}} + \frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{\color{blue}{h}} \]

   5. Applied rewrites 15.6%

      \[\leadsto \color{blue}{\frac{\mathsf{fma}\left(\frac{c0 \cdot c0}{D \cdot D}, \frac{d \cdot d}{w \cdot w}, \frac{\left(\left(M \cdot M\right) \cdot \left(h \cdot h\right)\right) \cdot \left(D \cdot D\right)}{d \cdot d} \cdot -0.25\right)}{h}} \]

   6. Taylor expanded in c0 around inf

      \[\leadsto \frac{\frac{{c0}^{2} \cdot {d}^{2}}{{D}^{2} \cdot {w}^{2}}}{h} \]
   7. Step-by-step derivation

      1. frac-times N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{{d}^{2}}{{w}^{2}}}{h} \]

      2. pow2 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{{w}^{2}}}{h} \]

      3. pow2 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{\frac{{c0}^{2}}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      5. pow2 N/A

         \[\leadsto \frac{\frac{c0 \cdot c0}{{D}^{2}} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      6. pow2 N/A

         \[\leadsto \frac{\frac{c0 \cdot c0}{D \cdot D} \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      7. times-frac N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      8. lower-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      9. lower-/.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      10. lower-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \frac{d \cdot d}{w \cdot w}}{h} \]

      11. times-frac N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      12. lower-*.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      13. lower-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      14. lower-/.f64 41.7

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

   8. Applied rewrites 41.7%

      \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      2. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      3. lift-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{c0}{D}\right) \cdot \left(\frac{d}{w} \cdot \frac{d}{w}\right)}{h} \]

      4. unswap-sqr N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

      5. lower-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

      6. lower-*.f64 N/A

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

      7. lower-*.f64 49.0

         \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

   10. Applied rewrites 49.0%

       \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]
   11. Step-by-step derivation

       1. lift-*.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

       2. lift-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

       3. lift-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(\frac{c0}{D} \cdot \frac{d}{w}\right)}{h} \]

       4. frac-times N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \frac{c0 \cdot d}{D \cdot w}}{h} \]

       5. associate-/l* N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(c0 \cdot \frac{d}{D \cdot w}\right)}{h} \]

       6. lower-*.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(c0 \cdot \frac{d}{D \cdot w}\right)}{h} \]

       7. lower-/.f64 N/A

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(c0 \cdot \frac{d}{D \cdot w}\right)}{h} \]

       8. lower-*.f64 47.8

          \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(c0 \cdot \frac{d}{D \cdot w}\right)}{h} \]

   12. Applied rewrites 47.8%

       \[\leadsto \frac{\left(\frac{c0}{D} \cdot \frac{d}{w}\right) \cdot \left(c0 \cdot \frac{d}{D \cdot w}\right)}{h} \]

   if 2.8499999999999999e256 < d

   1. Initial program 20.5%

      \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
   2. Add Preprocessing

   3. Taylor expanded in c0 around -inf

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-1 \cdot \left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right)} \]
   4. Step-by-step derivation

      1. mul-1-neg N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\mathsf{neg}\left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right) \]

      2. lower-neg.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right) \]

      3. *-commutative N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

      4. lower-*.f64 N/A

         \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

   5. Applied rewrites 0.0%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-\left(0 \cdot \frac{d \cdot d}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot c0\right)} \]

   6. Taylor expanded in c0 around 0

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   7. Step-by-step derivation

   8. Applied rewrites 36.2%

      \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
   9. Step-by-step derivation

      1. lift-*.f64 N/A

         \[\leadsto \color{blue}{\frac{c0}{2 \cdot w} \cdot 0} \]

   10. Applied rewrites 41.3%

       \[\leadsto \color{blue}{\frac{c0 \cdot 0}{w \cdot 2}} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 9: 34.8% accurate, 7.1× speedup

Math

\[\begin{array}{l} d_m = \left|d\right| \\ \frac{c0 \cdot 0}{w \cdot 2} \end{array} \]

FPCore

d_m = (fabs.f64 d)
(FPCore (c0 w h D d_m M) :precision binary64 (/ (* c0 0.0) (* w 2.0)))

C

d_m = fabs(d);
double code(double c0, double w, double h, double D, double d_m, double M) {
	return (c0 * 0.0) / (w * 2.0);
}

Fortran

d_m =     private
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(c0, w, h, d, d_m, m)
use fmin_fmax_functions
    real(8), intent (in) :: c0
    real(8), intent (in) :: w
    real(8), intent (in) :: h
    real(8), intent (in) :: d
    real(8), intent (in) :: d_m
    real(8), intent (in) :: m
    code = (c0 * 0.0d0) / (w * 2.0d0)
end function

Java

d_m = Math.abs(d);
public static double code(double c0, double w, double h, double D, double d_m, double M) {
	return (c0 * 0.0) / (w * 2.0);
}

Python

d_m = math.fabs(d)
def code(c0, w, h, D, d_m, M):
	return (c0 * 0.0) / (w * 2.0)

Julia

d_m = abs(d)
function code(c0, w, h, D, d_m, M)
	return Float64(Float64(c0 * 0.0) / Float64(w * 2.0))
end

MATLAB

d_m = abs(d);
function tmp = code(c0, w, h, D, d_m, M)
	tmp = (c0 * 0.0) / (w * 2.0);
end

Wolfram

d_m = N[Abs[d], $MachinePrecision]
code[c0_, w_, h_, D_, d$95$m_, M_] := N[(N[(c0 * 0.0), $MachinePrecision] / N[(w * 2.0), $MachinePrecision]), $MachinePrecision]

Derivation

1. Initial program 23.7%

   \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
2. Add Preprocessing

3. Taylor expanded in c0 around -inf

   \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-1 \cdot \left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right)} \]
4. Step-by-step derivation

   1. mul-1-neg N/A

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\mathsf{neg}\left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right) \]

   2. lower-neg.f64 N/A

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right) \]

   3. *-commutative N/A

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

   4. lower-*.f64 N/A

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

5. Applied rewrites 5.6%

   \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-\left(0 \cdot \frac{d \cdot d}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot c0\right)} \]

6. Taylor expanded in c0 around 0

   \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
7. Step-by-step derivation

8. Applied rewrites 30.1%

   \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
9. Step-by-step derivation

   1. lift-*.f64 N/A

      \[\leadsto \color{blue}{\frac{c0}{2 \cdot w} \cdot 0} \]

10. Applied rewrites 34.8%

    \[\leadsto \color{blue}{\frac{c0 \cdot 0}{w \cdot 2}} \]
11. Add Preprocessing

Alternative 10: 30.1% accurate, 7.8× speedup

Math

\[\begin{array}{l} d_m = \left|d\right| \\ \frac{c0}{w + w} \cdot 0 \end{array} \]

FPCore

d_m = (fabs.f64 d)
(FPCore (c0 w h D d_m M) :precision binary64 (* (/ c0 (+ w w)) 0.0))

C

d_m = fabs(d);
double code(double c0, double w, double h, double D, double d_m, double M) {
	return (c0 / (w + w)) * 0.0;
}

Fortran

d_m =     private
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(c0, w, h, d, d_m, m)
use fmin_fmax_functions
    real(8), intent (in) :: c0
    real(8), intent (in) :: w
    real(8), intent (in) :: h
    real(8), intent (in) :: d
    real(8), intent (in) :: d_m
    real(8), intent (in) :: m
    code = (c0 / (w + w)) * 0.0d0
end function

Java

d_m = Math.abs(d);
public static double code(double c0, double w, double h, double D, double d_m, double M) {
	return (c0 / (w + w)) * 0.0;
}

Python

d_m = math.fabs(d)
def code(c0, w, h, D, d_m, M):
	return (c0 / (w + w)) * 0.0

Julia

d_m = abs(d)
function code(c0, w, h, D, d_m, M)
	return Float64(Float64(c0 / Float64(w + w)) * 0.0)
end

MATLAB

d_m = abs(d);
function tmp = code(c0, w, h, D, d_m, M)
	tmp = (c0 / (w + w)) * 0.0;
end

Wolfram

d_m = N[Abs[d], $MachinePrecision]
code[c0_, w_, h_, D_, d$95$m_, M_] := N[(N[(c0 / N[(w + w), $MachinePrecision]), $MachinePrecision] * 0.0), $MachinePrecision]

Derivation

1. Initial program 23.7%

   \[\frac{c0}{2 \cdot w} \cdot \left(\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} + \sqrt{\frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} \cdot \frac{c0 \cdot \left(d \cdot d\right)}{\left(w \cdot h\right) \cdot \left(D \cdot D\right)} - M \cdot M}\right) \]
2. Add Preprocessing

3. Taylor expanded in c0 around -inf

   \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-1 \cdot \left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right)} \]
4. Step-by-step derivation

   1. mul-1-neg N/A

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(\mathsf{neg}\left(c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right)\right) \]

   2. lower-neg.f64 N/A

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-c0 \cdot \left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right)\right) \]

   3. *-commutative N/A

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

   4. lower-*.f64 N/A

      \[\leadsto \frac{c0}{2 \cdot w} \cdot \left(-\left(-1 \cdot \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)} + \frac{{d}^{2}}{{D}^{2} \cdot \left(h \cdot w\right)}\right) \cdot c0\right) \]

5. Applied rewrites 5.6%

   \[\leadsto \frac{c0}{2 \cdot w} \cdot \color{blue}{\left(-\left(0 \cdot \frac{d \cdot d}{\left(\left(h \cdot w\right) \cdot D\right) \cdot D}\right) \cdot c0\right)} \]

6. Taylor expanded in c0 around 0

   \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
7. Step-by-step derivation

8. Applied rewrites 30.1%

   \[\leadsto \frac{c0}{2 \cdot w} \cdot 0 \]
9. Step-by-step derivation

   1. lift-*.f64 N/A

      \[\leadsto \frac{c0}{\color{blue}{2 \cdot w}} \cdot 0 \]

   2. count-2-rev N/A

      \[\leadsto \frac{c0}{\color{blue}{w + w}} \cdot 0 \]

   3. lower-+.f64 30.1

      \[\leadsto \frac{c0}{\color{blue}{w + w}} \cdot 0 \]

10. Applied rewrites 30.1%

    \[\leadsto \frac{c0}{\color{blue}{w + w}} \cdot 0 \]
11. Add Preprocessing

Reproduce

herbie shell --seed 2025092 
(FPCore (c0 w h D d M)
  :name "Henrywood and Agarwal, Equation (13)"
  :precision binary64
  (* (/ c0 (* 2.0 w)) (+ (/ (* c0 (* d d)) (* (* w h) (* D D))) (sqrt (- (* (/ (* c0 (* d d)) (* (* w h) (* D D))) (/ (* c0 (* d d)) (* (* w h) (* D D)))) (* M M))))))