Octave 3.8, jcobi/4, as called

Percentage Accurate: 27.5% → 99.9%

Time: 6.5s

Alternatives: 8

Speedup: 71.0×

Specification

\[i > 0\]

Math

\[\begin{array}{l} \\ \begin{array}{l} t_0 := \left(2 \cdot i\right) \cdot \left(2 \cdot i\right)\\ \frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{t\_0}}{t\_0 - 1} \end{array} \end{array} \]

FPCore

(FPCore (i)
 :precision binary64
 (let* ((t_0 (* (* 2.0 i) (* 2.0 i))))
   (/ (/ (* (* i i) (* i i)) t_0) (- t_0 1.0))))

C

double code(double i) {
	double t_0 = (2.0 * i) * (2.0 * i);
	return (((i * i) * (i * i)) / t_0) / (t_0 - 1.0);
}

Fortran

real(8) function code(i)
    real(8), intent (in) :: i
    real(8) :: t_0
    t_0 = (2.0d0 * i) * (2.0d0 * i)
    code = (((i * i) * (i * i)) / t_0) / (t_0 - 1.0d0)
end function

Java

public static double code(double i) {
	double t_0 = (2.0 * i) * (2.0 * i);
	return (((i * i) * (i * i)) / t_0) / (t_0 - 1.0);
}

Python

def code(i):
	t_0 = (2.0 * i) * (2.0 * i)
	return (((i * i) * (i * i)) / t_0) / (t_0 - 1.0)

Julia

function code(i)
	t_0 = Float64(Float64(2.0 * i) * Float64(2.0 * i))
	return Float64(Float64(Float64(Float64(i * i) * Float64(i * i)) / t_0) / Float64(t_0 - 1.0))
end

MATLAB

function tmp = code(i)
	t_0 = (2.0 * i) * (2.0 * i);
	tmp = (((i * i) * (i * i)) / t_0) / (t_0 - 1.0);
end

Wolfram

code[i_] := Block[{t$95$0 = N[(N[(2.0 * i), $MachinePrecision] * N[(2.0 * i), $MachinePrecision]), $MachinePrecision]}, N[(N[(N[(N[(i * i), $MachinePrecision] * N[(i * i), $MachinePrecision]), $MachinePrecision] / t$95$0), $MachinePrecision] / N[(t$95$0 - 1.0), $MachinePrecision]), $MachinePrecision]]

Initial Program: 27.5% accurate, 1.0× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} t_0 := \left(2 \cdot i\right) \cdot \left(2 \cdot i\right)\\ \frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{t\_0}}{t\_0 - 1} \end{array} \end{array} \]

FPCore

(FPCore (i)
 :precision binary64
 (let* ((t_0 (* (* 2.0 i) (* 2.0 i))))
   (/ (/ (* (* i i) (* i i)) t_0) (- t_0 1.0))))

C

double code(double i) {
	double t_0 = (2.0 * i) * (2.0 * i);
	return (((i * i) * (i * i)) / t_0) / (t_0 - 1.0);
}

Fortran

real(8) function code(i)
    real(8), intent (in) :: i
    real(8) :: t_0
    t_0 = (2.0d0 * i) * (2.0d0 * i)
    code = (((i * i) * (i * i)) / t_0) / (t_0 - 1.0d0)
end function

Java

public static double code(double i) {
	double t_0 = (2.0 * i) * (2.0 * i);
	return (((i * i) * (i * i)) / t_0) / (t_0 - 1.0);
}

Python

def code(i):
	t_0 = (2.0 * i) * (2.0 * i)
	return (((i * i) * (i * i)) / t_0) / (t_0 - 1.0)

Julia

function code(i)
	t_0 = Float64(Float64(2.0 * i) * Float64(2.0 * i))
	return Float64(Float64(Float64(Float64(i * i) * Float64(i * i)) / t_0) / Float64(t_0 - 1.0))
end

MATLAB

function tmp = code(i)
	t_0 = (2.0 * i) * (2.0 * i);
	tmp = (((i * i) * (i * i)) / t_0) / (t_0 - 1.0);
end

Wolfram

code[i_] := Block[{t$95$0 = N[(N[(2.0 * i), $MachinePrecision] * N[(2.0 * i), $MachinePrecision]), $MachinePrecision]}, N[(N[(N[(N[(i * i), $MachinePrecision] * N[(i * i), $MachinePrecision]), $MachinePrecision] / t$95$0), $MachinePrecision] / N[(t$95$0 - 1.0), $MachinePrecision]), $MachinePrecision]]

Alternative 1: 99.9% accurate, 1.8× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;i \leq 100000:\\ \;\;\;\;\frac{\left(i \cdot i\right) \cdot 0.25}{\mathsf{fma}\left(4 \cdot i, i, -1\right)}\\ \mathbf{else}:\\ \;\;\;\;0.0625\\ \end{array} \end{array} \]

FPCore

(FPCore (i)
 :precision binary64
 (if (<= i 100000.0) (/ (* (* i i) 0.25) (fma (* 4.0 i) i -1.0)) 0.0625))

C

double code(double i) {
	double tmp;
	if (i <= 100000.0) {
		tmp = ((i * i) * 0.25) / fma((4.0 * i), i, -1.0);
	} else {
		tmp = 0.0625;
	}
	return tmp;
}

Julia

function code(i)
	tmp = 0.0
	if (i <= 100000.0)
		tmp = Float64(Float64(Float64(i * i) * 0.25) / fma(Float64(4.0 * i), i, -1.0));
	else
		tmp = 0.0625;
	end
	return tmp
end

Wolfram

code[i_] := If[LessEqual[i, 100000.0], N[(N[(N[(i * i), $MachinePrecision] * 0.25), $MachinePrecision] / N[(N[(4.0 * i), $MachinePrecision] * i + -1.0), $MachinePrecision]), $MachinePrecision], 0.0625]

Derivation

1. Split input into 2 regimes

2. if i < 1e5

   1. Initial program 32.6%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing
   3. Step-by-step derivation

      1. lift-/.f64 N/A

         \[\leadsto \color{blue}{\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1}} \]

      2. lift-/.f64 N/A

         \[\leadsto \frac{\color{blue}{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]

      3. lift-*.f64 N/A

         \[\leadsto \frac{\frac{\color{blue}{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]

      4. associate-/l* N/A

         \[\leadsto \frac{\color{blue}{\left(i \cdot i\right) \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]

      5. lift--.f64 N/A

         \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\color{blue}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1}} \]

      6. lift-*.f64 N/A

         \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\color{blue}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)} - 1} \]

      7. difference-of-sqr-1 N/A

         \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\color{blue}{\left(2 \cdot i + 1\right) \cdot \left(2 \cdot i - 1\right)}} \]

      8. times-frac N/A

         \[\leadsto \color{blue}{\frac{i \cdot i}{2 \cdot i + 1} \cdot \frac{\frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{2 \cdot i - 1}} \]

      9. associate-*r/ N/A

         \[\leadsto \color{blue}{\frac{\frac{i \cdot i}{2 \cdot i + 1} \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{2 \cdot i - 1}} \]

      10. lower-/.f64 N/A

          \[\leadsto \color{blue}{\frac{\frac{i \cdot i}{2 \cdot i + 1} \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{2 \cdot i - 1}} \]

   4. Applied rewrites 99.9%

      \[\leadsto \color{blue}{\frac{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right) \cdot 0.25}{2 \cdot i - 1}} \]
   5. Step-by-step derivation

      1. lift-/.f64 N/A

         \[\leadsto \color{blue}{\frac{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right) \cdot \frac{1}{4}}{2 \cdot i - 1}} \]

      2. lift-*.f64 N/A

         \[\leadsto \frac{\color{blue}{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right) \cdot \frac{1}{4}}}{2 \cdot i - 1} \]

      3. associate-/l* N/A

         \[\leadsto \color{blue}{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right) \cdot \frac{\frac{1}{4}}{2 \cdot i - 1}} \]

      4. lift-*.f64 N/A

         \[\leadsto \color{blue}{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right)} \cdot \frac{\frac{1}{4}}{2 \cdot i - 1} \]

      5. lift-/.f64 N/A

         \[\leadsto \left(i \cdot \color{blue}{\frac{i}{\mathsf{fma}\left(2, i, 1\right)}}\right) \cdot \frac{\frac{1}{4}}{2 \cdot i - 1} \]

      6. associate-*r/ N/A

         \[\leadsto \color{blue}{\frac{i \cdot i}{\mathsf{fma}\left(2, i, 1\right)}} \cdot \frac{\frac{1}{4}}{2 \cdot i - 1} \]

      7. lift-*.f64 N/A

         \[\leadsto \frac{\color{blue}{i \cdot i}}{\mathsf{fma}\left(2, i, 1\right)} \cdot \frac{\frac{1}{4}}{2 \cdot i - 1} \]

      8. frac-times N/A

         \[\leadsto \color{blue}{\frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\mathsf{fma}\left(2, i, 1\right) \cdot \left(2 \cdot i - 1\right)}} \]

      9. lift-fma.f64 N/A

         \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\color{blue}{\left(2 \cdot i + 1\right)} \cdot \left(2 \cdot i - 1\right)} \]

      10. lift-*.f64 N/A

          \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\left(\color{blue}{2 \cdot i} + 1\right) \cdot \left(2 \cdot i - 1\right)} \]

      11. lift--.f64 N/A

          \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\left(2 \cdot i + 1\right) \cdot \color{blue}{\left(2 \cdot i - 1\right)}} \]

      12. difference-of-sqr-1 N/A

          \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\color{blue}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1}} \]

      13. lift-*.f64 N/A

          \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\color{blue}{\left(2 \cdot i\right)} \cdot \left(2 \cdot i\right) - 1} \]

      14. lift-*.f64 N/A

          \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\left(2 \cdot i\right) \cdot \color{blue}{\left(2 \cdot i\right)} - 1} \]

      15. lower-/.f64 N/A

          \[\leadsto \color{blue}{\frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1}} \]

      16. lower-*.f64 N/A

          \[\leadsto \frac{\color{blue}{\left(i \cdot i\right) \cdot \frac{1}{4}}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]

      17. lift-*.f64 N/A

          \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\color{blue}{\left(2 \cdot i\right)} \cdot \left(2 \cdot i\right) - 1} \]

      18. lift-*.f64 N/A

          \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\left(2 \cdot i\right) \cdot \color{blue}{\left(2 \cdot i\right)} - 1} \]

      19. difference-of-sqr-1 N/A

          \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{1}{4}}{\color{blue}{\left(2 \cdot i + 1\right) \cdot \left(2 \cdot i - 1\right)}} \]

   6. Applied rewrites 100.0%

      \[\leadsto \color{blue}{\frac{\left(i \cdot i\right) \cdot 0.25}{\mathsf{fma}\left(4 \cdot i, i, -1\right)}} \]

   if 1e5 < i

   1. Initial program 19.5%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   3. Taylor expanded in i around inf

      \[\leadsto \color{blue}{\frac{1}{16}} \]
   4. Step-by-step derivation

   5. Applied rewrites 100.0%

      \[\leadsto \color{blue}{0.0625} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 2: 99.9% accurate, 1.6× speedup

Math

\[\begin{array}{l} \\ \frac{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right) \cdot 0.25}{\left(i + i\right) - 1} \end{array} \]

FPCore

(FPCore (i)
 :precision binary64
 (/ (* (* i (/ i (fma 2.0 i 1.0))) 0.25) (- (+ i i) 1.0)))

C

double code(double i) {
	return ((i * (i / fma(2.0, i, 1.0))) * 0.25) / ((i + i) - 1.0);
}

Julia

function code(i)
	return Float64(Float64(Float64(i * Float64(i / fma(2.0, i, 1.0))) * 0.25) / Float64(Float64(i + i) - 1.0))
end

Wolfram

code[i_] := N[(N[(N[(i * N[(i / N[(2.0 * i + 1.0), $MachinePrecision]), $MachinePrecision]), $MachinePrecision] * 0.25), $MachinePrecision] / N[(N[(i + i), $MachinePrecision] - 1.0), $MachinePrecision]), $MachinePrecision]

Derivation

1. Initial program 26.3%

   \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
2. Add Preprocessing
3. Step-by-step derivation

   1. lift-/.f64 N/A

      \[\leadsto \color{blue}{\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1}} \]

   2. lift-/.f64 N/A

      \[\leadsto \frac{\color{blue}{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]

   3. lift-*.f64 N/A

      \[\leadsto \frac{\frac{\color{blue}{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]

   4. associate-/l* N/A

      \[\leadsto \frac{\color{blue}{\left(i \cdot i\right) \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]

   5. lift--.f64 N/A

      \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\color{blue}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1}} \]

   6. lift-*.f64 N/A

      \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\color{blue}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)} - 1} \]

   7. difference-of-sqr-1 N/A

      \[\leadsto \frac{\left(i \cdot i\right) \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\color{blue}{\left(2 \cdot i + 1\right) \cdot \left(2 \cdot i - 1\right)}} \]

   8. times-frac N/A

      \[\leadsto \color{blue}{\frac{i \cdot i}{2 \cdot i + 1} \cdot \frac{\frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{2 \cdot i - 1}} \]

   9. associate-*r/ N/A

      \[\leadsto \color{blue}{\frac{\frac{i \cdot i}{2 \cdot i + 1} \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{2 \cdot i - 1}} \]

   10. lower-/.f64 N/A

       \[\leadsto \color{blue}{\frac{\frac{i \cdot i}{2 \cdot i + 1} \cdot \frac{i \cdot i}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{2 \cdot i - 1}} \]

4. Applied rewrites 100.0%

   \[\leadsto \color{blue}{\frac{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right) \cdot 0.25}{2 \cdot i - 1}} \]
5. Step-by-step derivation

   1. lift-*.f64 N/A

      \[\leadsto \frac{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right) \cdot \frac{1}{4}}{\color{blue}{2 \cdot i} - 1} \]

   2. count-2-rev N/A

      \[\leadsto \frac{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right) \cdot \frac{1}{4}}{\color{blue}{\left(i + i\right)} - 1} \]

   3. lower-+.f64 100.0

      \[\leadsto \frac{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right) \cdot 0.25}{\color{blue}{\left(i + i\right)} - 1} \]

6. Applied rewrites 100.0%

   \[\leadsto \frac{\left(i \cdot \frac{i}{\mathsf{fma}\left(2, i, 1\right)}\right) \cdot 0.25}{\color{blue}{\left(i + i\right)} - 1} \]
7. Add Preprocessing

Alternative 3: 99.9% accurate, 1.8× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;i \leq 100000:\\ \;\;\;\;\frac{i}{\mathsf{fma}\left(4, i \cdot i, -1\right)} \cdot \left(0.25 \cdot i\right)\\ \mathbf{else}:\\ \;\;\;\;0.0625\\ \end{array} \end{array} \]

FPCore

(FPCore (i)
 :precision binary64
 (if (<= i 100000.0) (* (/ i (fma 4.0 (* i i) -1.0)) (* 0.25 i)) 0.0625))

C

double code(double i) {
	double tmp;
	if (i <= 100000.0) {
		tmp = (i / fma(4.0, (i * i), -1.0)) * (0.25 * i);
	} else {
		tmp = 0.0625;
	}
	return tmp;
}

Julia

function code(i)
	tmp = 0.0
	if (i <= 100000.0)
		tmp = Float64(Float64(i / fma(4.0, Float64(i * i), -1.0)) * Float64(0.25 * i));
	else
		tmp = 0.0625;
	end
	return tmp
end

Wolfram

code[i_] := If[LessEqual[i, 100000.0], N[(N[(i / N[(4.0 * N[(i * i), $MachinePrecision] + -1.0), $MachinePrecision]), $MachinePrecision] * N[(0.25 * i), $MachinePrecision]), $MachinePrecision], 0.0625]

Derivation

1. Split input into 2 regimes

2. if i < 1e5

   1. Initial program 32.6%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   4. Applied rewrites 100.0%

      \[\leadsto \color{blue}{\frac{i}{\mathsf{fma}\left(4, i \cdot i, -1\right)} \cdot \left(0.25 \cdot i\right)} \]

   if 1e5 < i

   1. Initial program 19.5%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   3. Taylor expanded in i around inf

      \[\leadsto \color{blue}{\frac{1}{16}} \]
   4. Step-by-step derivation

   5. Applied rewrites 100.0%

      \[\leadsto \color{blue}{0.0625} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 4: 99.5% accurate, 2.6× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;i \leq 0.5:\\ \;\;\;\;\left(i \cdot i + 0.25\right) \cdot \left(\left(-i\right) \cdot i\right)\\ \mathbf{else}:\\ \;\;\;\;\frac{0.015625}{i \cdot i} + 0.0625\\ \end{array} \end{array} \]

FPCore

(FPCore (i)
 :precision binary64
 (if (<= i 0.5)
   (* (+ (* i i) 0.25) (* (- i) i))
   (+ (/ 0.015625 (* i i)) 0.0625)))

C

double code(double i) {
	double tmp;
	if (i <= 0.5) {
		tmp = ((i * i) + 0.25) * (-i * i);
	} else {
		tmp = (0.015625 / (i * i)) + 0.0625;
	}
	return tmp;
}

Fortran

real(8) function code(i)
    real(8), intent (in) :: i
    real(8) :: tmp
    if (i <= 0.5d0) then
        tmp = ((i * i) + 0.25d0) * (-i * i)
    else
        tmp = (0.015625d0 / (i * i)) + 0.0625d0
    end if
    code = tmp
end function

Java

public static double code(double i) {
	double tmp;
	if (i <= 0.5) {
		tmp = ((i * i) + 0.25) * (-i * i);
	} else {
		tmp = (0.015625 / (i * i)) + 0.0625;
	}
	return tmp;
}

Python

def code(i):
	tmp = 0
	if i <= 0.5:
		tmp = ((i * i) + 0.25) * (-i * i)
	else:
		tmp = (0.015625 / (i * i)) + 0.0625
	return tmp

Julia

function code(i)
	tmp = 0.0
	if (i <= 0.5)
		tmp = Float64(Float64(Float64(i * i) + 0.25) * Float64(Float64(-i) * i));
	else
		tmp = Float64(Float64(0.015625 / Float64(i * i)) + 0.0625);
	end
	return tmp
end

MATLAB

function tmp_2 = code(i)
	tmp = 0.0;
	if (i <= 0.5)
		tmp = ((i * i) + 0.25) * (-i * i);
	else
		tmp = (0.015625 / (i * i)) + 0.0625;
	end
	tmp_2 = tmp;
end

Wolfram

code[i_] := If[LessEqual[i, 0.5], N[(N[(N[(i * i), $MachinePrecision] + 0.25), $MachinePrecision] * N[((-i) * i), $MachinePrecision]), $MachinePrecision], N[(N[(0.015625 / N[(i * i), $MachinePrecision]), $MachinePrecision] + 0.0625), $MachinePrecision]]

Derivation

1. Split input into 2 regimes

2. if i < 0.5

   1. Initial program 31.6%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   3. Taylor expanded in i around 0

      \[\leadsto \color{blue}{{i}^{2} \cdot \left(-1 \cdot {i}^{2} - \frac{1}{4}\right)} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\left(-1 \cdot {i}^{2} - \frac{1}{4}\right) \cdot {i}^{2}} \]

      2. unpow2 N/A

         \[\leadsto \left(-1 \cdot {i}^{2} - \frac{1}{4}\right) \cdot \color{blue}{\left(i \cdot i\right)} \]

      3. associate-*r* N/A

         \[\leadsto \color{blue}{\left(\left(-1 \cdot {i}^{2} - \frac{1}{4}\right) \cdot i\right) \cdot i} \]

      4. lower-*.f64 N/A

         \[\leadsto \color{blue}{\left(\left(-1 \cdot {i}^{2} - \frac{1}{4}\right) \cdot i\right) \cdot i} \]

      5. lower-*.f64 N/A

         \[\leadsto \color{blue}{\left(\left(-1 \cdot {i}^{2} - \frac{1}{4}\right) \cdot i\right)} \cdot i \]

      6. lower--.f64 N/A

         \[\leadsto \left(\color{blue}{\left(-1 \cdot {i}^{2} - \frac{1}{4}\right)} \cdot i\right) \cdot i \]

      7. lower-*.f64 N/A

         \[\leadsto \left(\left(\color{blue}{-1 \cdot {i}^{2}} - \frac{1}{4}\right) \cdot i\right) \cdot i \]

      8. unpow2 N/A

         \[\leadsto \left(\left(-1 \cdot \color{blue}{\left(i \cdot i\right)} - \frac{1}{4}\right) \cdot i\right) \cdot i \]

      9. lower-*.f64 98.9

         \[\leadsto \left(\left(-1 \cdot \color{blue}{\left(i \cdot i\right)} - 0.25\right) \cdot i\right) \cdot i \]

   5. Applied rewrites 98.9%

      \[\leadsto \color{blue}{\left(\left(-1 \cdot \left(i \cdot i\right) - 0.25\right) \cdot i\right) \cdot i} \]
   6. Step-by-step derivation

   7. Applied rewrites 98.9%

      \[\leadsto \left(-1 \cdot \left(i \cdot i\right) - 0.25\right) \cdot \color{blue}{\left(i \cdot i\right)} \]

   if 0.5 < i

   1. Initial program 20.8%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   3. Taylor expanded in i around inf

      \[\leadsto \color{blue}{\frac{1}{16} + \frac{1}{64} \cdot \frac{1}{{i}^{2}}} \]
   4. Step-by-step derivation

      1. +-commutative N/A

         \[\leadsto \color{blue}{\frac{1}{64} \cdot \frac{1}{{i}^{2}} + \frac{1}{16}} \]

      2. lower-+.f64 N/A

         \[\leadsto \color{blue}{\frac{1}{64} \cdot \frac{1}{{i}^{2}} + \frac{1}{16}} \]

      3. associate-*r/ N/A

         \[\leadsto \color{blue}{\frac{\frac{1}{64} \cdot 1}{{i}^{2}}} + \frac{1}{16} \]

      4. metadata-eval N/A

         \[\leadsto \frac{\color{blue}{\frac{1}{64}}}{{i}^{2}} + \frac{1}{16} \]

      5. lower-/.f64 N/A

         \[\leadsto \color{blue}{\frac{\frac{1}{64}}{{i}^{2}}} + \frac{1}{16} \]

      6. unpow2 N/A

         \[\leadsto \frac{\frac{1}{64}}{\color{blue}{i \cdot i}} + \frac{1}{16} \]

      7. lower-*.f64 99.7

         \[\leadsto \frac{0.015625}{\color{blue}{i \cdot i}} + 0.0625 \]

   5. Applied rewrites 99.7%

      \[\leadsto \color{blue}{\frac{0.015625}{i \cdot i} + 0.0625} \]
3. Recombined 2 regimes into one program.

4. Final simplification 99.3%

   \[\leadsto \begin{array}{l} \mathbf{if}\;i \leq 0.5:\\ \;\;\;\;\left(i \cdot i + 0.25\right) \cdot \left(\left(-i\right) \cdot i\right)\\ \mathbf{else}:\\ \;\;\;\;\frac{0.015625}{i \cdot i} + 0.0625\\ \end{array} \]
5. Add Preprocessing

Alternative 5: 99.5% accurate, 2.6× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;i \leq 0.5:\\ \;\;\;\;\left(\left(i \cdot i + 0.25\right) \cdot \left(-i\right)\right) \cdot i\\ \mathbf{else}:\\ \;\;\;\;\frac{0.015625}{i \cdot i} + 0.0625\\ \end{array} \end{array} \]

FPCore

(FPCore (i)
 :precision binary64
 (if (<= i 0.5)
   (* (* (+ (* i i) 0.25) (- i)) i)
   (+ (/ 0.015625 (* i i)) 0.0625)))

C

double code(double i) {
	double tmp;
	if (i <= 0.5) {
		tmp = (((i * i) + 0.25) * -i) * i;
	} else {
		tmp = (0.015625 / (i * i)) + 0.0625;
	}
	return tmp;
}

Fortran

real(8) function code(i)
    real(8), intent (in) :: i
    real(8) :: tmp
    if (i <= 0.5d0) then
        tmp = (((i * i) + 0.25d0) * -i) * i
    else
        tmp = (0.015625d0 / (i * i)) + 0.0625d0
    end if
    code = tmp
end function

Java

public static double code(double i) {
	double tmp;
	if (i <= 0.5) {
		tmp = (((i * i) + 0.25) * -i) * i;
	} else {
		tmp = (0.015625 / (i * i)) + 0.0625;
	}
	return tmp;
}

Python

def code(i):
	tmp = 0
	if i <= 0.5:
		tmp = (((i * i) + 0.25) * -i) * i
	else:
		tmp = (0.015625 / (i * i)) + 0.0625
	return tmp

Julia

function code(i)
	tmp = 0.0
	if (i <= 0.5)
		tmp = Float64(Float64(Float64(Float64(i * i) + 0.25) * Float64(-i)) * i);
	else
		tmp = Float64(Float64(0.015625 / Float64(i * i)) + 0.0625);
	end
	return tmp
end

MATLAB

function tmp_2 = code(i)
	tmp = 0.0;
	if (i <= 0.5)
		tmp = (((i * i) + 0.25) * -i) * i;
	else
		tmp = (0.015625 / (i * i)) + 0.0625;
	end
	tmp_2 = tmp;
end

Wolfram

code[i_] := If[LessEqual[i, 0.5], N[(N[(N[(N[(i * i), $MachinePrecision] + 0.25), $MachinePrecision] * (-i)), $MachinePrecision] * i), $MachinePrecision], N[(N[(0.015625 / N[(i * i), $MachinePrecision]), $MachinePrecision] + 0.0625), $MachinePrecision]]

Derivation

1. Split input into 2 regimes

2. if i < 0.5

   1. Initial program 31.6%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   3. Taylor expanded in i around 0

      \[\leadsto \color{blue}{{i}^{2} \cdot \left(-1 \cdot {i}^{2} - \frac{1}{4}\right)} \]
   4. Step-by-step derivation

      1. *-commutative N/A

         \[\leadsto \color{blue}{\left(-1 \cdot {i}^{2} - \frac{1}{4}\right) \cdot {i}^{2}} \]

      2. unpow2 N/A

         \[\leadsto \left(-1 \cdot {i}^{2} - \frac{1}{4}\right) \cdot \color{blue}{\left(i \cdot i\right)} \]

      3. associate-*r* N/A

         \[\leadsto \color{blue}{\left(\left(-1 \cdot {i}^{2} - \frac{1}{4}\right) \cdot i\right) \cdot i} \]

      4. lower-*.f64 N/A

         \[\leadsto \color{blue}{\left(\left(-1 \cdot {i}^{2} - \frac{1}{4}\right) \cdot i\right) \cdot i} \]

      5. lower-*.f64 N/A

         \[\leadsto \color{blue}{\left(\left(-1 \cdot {i}^{2} - \frac{1}{4}\right) \cdot i\right)} \cdot i \]

      6. lower--.f64 N/A

         \[\leadsto \left(\color{blue}{\left(-1 \cdot {i}^{2} - \frac{1}{4}\right)} \cdot i\right) \cdot i \]

      7. lower-*.f64 N/A

         \[\leadsto \left(\left(\color{blue}{-1 \cdot {i}^{2}} - \frac{1}{4}\right) \cdot i\right) \cdot i \]

      8. unpow2 N/A

         \[\leadsto \left(\left(-1 \cdot \color{blue}{\left(i \cdot i\right)} - \frac{1}{4}\right) \cdot i\right) \cdot i \]

      9. lower-*.f64 98.9

         \[\leadsto \left(\left(-1 \cdot \color{blue}{\left(i \cdot i\right)} - 0.25\right) \cdot i\right) \cdot i \]

   5. Applied rewrites 98.9%

      \[\leadsto \color{blue}{\left(\left(-1 \cdot \left(i \cdot i\right) - 0.25\right) \cdot i\right) \cdot i} \]

   if 0.5 < i

   1. Initial program 20.8%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   3. Taylor expanded in i around inf

      \[\leadsto \color{blue}{\frac{1}{16} + \frac{1}{64} \cdot \frac{1}{{i}^{2}}} \]
   4. Step-by-step derivation

      1. +-commutative N/A

         \[\leadsto \color{blue}{\frac{1}{64} \cdot \frac{1}{{i}^{2}} + \frac{1}{16}} \]

      2. lower-+.f64 N/A

         \[\leadsto \color{blue}{\frac{1}{64} \cdot \frac{1}{{i}^{2}} + \frac{1}{16}} \]

      3. associate-*r/ N/A

         \[\leadsto \color{blue}{\frac{\frac{1}{64} \cdot 1}{{i}^{2}}} + \frac{1}{16} \]

      4. metadata-eval N/A

         \[\leadsto \frac{\color{blue}{\frac{1}{64}}}{{i}^{2}} + \frac{1}{16} \]

      5. lower-/.f64 N/A

         \[\leadsto \color{blue}{\frac{\frac{1}{64}}{{i}^{2}}} + \frac{1}{16} \]

      6. unpow2 N/A

         \[\leadsto \frac{\frac{1}{64}}{\color{blue}{i \cdot i}} + \frac{1}{16} \]

      7. lower-*.f64 99.7

         \[\leadsto \frac{0.015625}{\color{blue}{i \cdot i}} + 0.0625 \]

   5. Applied rewrites 99.7%

      \[\leadsto \color{blue}{\frac{0.015625}{i \cdot i} + 0.0625} \]
3. Recombined 2 regimes into one program.

4. Final simplification 99.3%

   \[\leadsto \begin{array}{l} \mathbf{if}\;i \leq 0.5:\\ \;\;\;\;\left(\left(i \cdot i + 0.25\right) \cdot \left(-i\right)\right) \cdot i\\ \mathbf{else}:\\ \;\;\;\;\frac{0.015625}{i \cdot i} + 0.0625\\ \end{array} \]
5. Add Preprocessing

Alternative 6: 99.2% accurate, 2.7× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;i \leq 0.5:\\ \;\;\;\;\left(-0.25 \cdot i\right) \cdot i\\ \mathbf{else}:\\ \;\;\;\;\frac{0.015625}{i \cdot i} + 0.0625\\ \end{array} \end{array} \]

FPCore

(FPCore (i)
 :precision binary64
 (if (<= i 0.5) (* (* -0.25 i) i) (+ (/ 0.015625 (* i i)) 0.0625)))

C

double code(double i) {
	double tmp;
	if (i <= 0.5) {
		tmp = (-0.25 * i) * i;
	} else {
		tmp = (0.015625 / (i * i)) + 0.0625;
	}
	return tmp;
}

Fortran

real(8) function code(i)
    real(8), intent (in) :: i
    real(8) :: tmp
    if (i <= 0.5d0) then
        tmp = ((-0.25d0) * i) * i
    else
        tmp = (0.015625d0 / (i * i)) + 0.0625d0
    end if
    code = tmp
end function

Java

public static double code(double i) {
	double tmp;
	if (i <= 0.5) {
		tmp = (-0.25 * i) * i;
	} else {
		tmp = (0.015625 / (i * i)) + 0.0625;
	}
	return tmp;
}

Python

def code(i):
	tmp = 0
	if i <= 0.5:
		tmp = (-0.25 * i) * i
	else:
		tmp = (0.015625 / (i * i)) + 0.0625
	return tmp

Julia

function code(i)
	tmp = 0.0
	if (i <= 0.5)
		tmp = Float64(Float64(-0.25 * i) * i);
	else
		tmp = Float64(Float64(0.015625 / Float64(i * i)) + 0.0625);
	end
	return tmp
end

MATLAB

function tmp_2 = code(i)
	tmp = 0.0;
	if (i <= 0.5)
		tmp = (-0.25 * i) * i;
	else
		tmp = (0.015625 / (i * i)) + 0.0625;
	end
	tmp_2 = tmp;
end

Wolfram

code[i_] := If[LessEqual[i, 0.5], N[(N[(-0.25 * i), $MachinePrecision] * i), $MachinePrecision], N[(N[(0.015625 / N[(i * i), $MachinePrecision]), $MachinePrecision] + 0.0625), $MachinePrecision]]

Derivation

1. Split input into 2 regimes

2. if i < 0.5

   1. Initial program 31.6%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   3. Taylor expanded in i around 0

      \[\leadsto \color{blue}{\frac{-1}{4} \cdot {i}^{2}} \]
   4. Step-by-step derivation

      1. unpow2 N/A

         \[\leadsto \frac{-1}{4} \cdot \color{blue}{\left(i \cdot i\right)} \]

      2. associate-*r* N/A

         \[\leadsto \color{blue}{\left(\frac{-1}{4} \cdot i\right) \cdot i} \]

      3. metadata-eval N/A

         \[\leadsto \left(\color{blue}{\left(\mathsf{neg}\left(\frac{1}{4}\right)\right)} \cdot i\right) \cdot i \]

      4. distribute-lft-neg-in N/A

         \[\leadsto \color{blue}{\left(\mathsf{neg}\left(\frac{1}{4} \cdot i\right)\right)} \cdot i \]

      5. *-commutative N/A

         \[\leadsto \left(\mathsf{neg}\left(\color{blue}{i \cdot \frac{1}{4}}\right)\right) \cdot i \]

      6. distribute-lft-neg-out N/A

         \[\leadsto \color{blue}{\left(\left(\mathsf{neg}\left(i\right)\right) \cdot \frac{1}{4}\right)} \cdot i \]

      7. lower-*.f64 N/A

         \[\leadsto \color{blue}{\left(\left(\mathsf{neg}\left(i\right)\right) \cdot \frac{1}{4}\right) \cdot i} \]

      8. distribute-lft-neg-out N/A

         \[\leadsto \color{blue}{\left(\mathsf{neg}\left(i \cdot \frac{1}{4}\right)\right)} \cdot i \]

      9. *-commutative N/A

         \[\leadsto \left(\mathsf{neg}\left(\color{blue}{\frac{1}{4} \cdot i}\right)\right) \cdot i \]

      10. distribute-lft-neg-in N/A

          \[\leadsto \color{blue}{\left(\left(\mathsf{neg}\left(\frac{1}{4}\right)\right) \cdot i\right)} \cdot i \]

      11. metadata-eval N/A

          \[\leadsto \left(\color{blue}{\frac{-1}{4}} \cdot i\right) \cdot i \]

      12. lower-*.f64 98.5

          \[\leadsto \color{blue}{\left(-0.25 \cdot i\right)} \cdot i \]

   5. Applied rewrites 98.5%

      \[\leadsto \color{blue}{\left(-0.25 \cdot i\right) \cdot i} \]

   if 0.5 < i

   1. Initial program 20.8%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   3. Taylor expanded in i around inf

      \[\leadsto \color{blue}{\frac{1}{16} + \frac{1}{64} \cdot \frac{1}{{i}^{2}}} \]
   4. Step-by-step derivation

      1. +-commutative N/A

         \[\leadsto \color{blue}{\frac{1}{64} \cdot \frac{1}{{i}^{2}} + \frac{1}{16}} \]

      2. lower-+.f64 N/A

         \[\leadsto \color{blue}{\frac{1}{64} \cdot \frac{1}{{i}^{2}} + \frac{1}{16}} \]

      3. associate-*r/ N/A

         \[\leadsto \color{blue}{\frac{\frac{1}{64} \cdot 1}{{i}^{2}}} + \frac{1}{16} \]

      4. metadata-eval N/A

         \[\leadsto \frac{\color{blue}{\frac{1}{64}}}{{i}^{2}} + \frac{1}{16} \]

      5. lower-/.f64 N/A

         \[\leadsto \color{blue}{\frac{\frac{1}{64}}{{i}^{2}}} + \frac{1}{16} \]

      6. unpow2 N/A

         \[\leadsto \frac{\frac{1}{64}}{\color{blue}{i \cdot i}} + \frac{1}{16} \]

      7. lower-*.f64 99.7

         \[\leadsto \frac{0.015625}{\color{blue}{i \cdot i}} + 0.0625 \]

   5. Applied rewrites 99.7%

      \[\leadsto \color{blue}{\frac{0.015625}{i \cdot i} + 0.0625} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 7: 99.0% accurate, 4.2× speedup

Math

\[\begin{array}{l} \\ \begin{array}{l} \mathbf{if}\;i \leq 0.5:\\ \;\;\;\;\left(-0.25 \cdot i\right) \cdot i\\ \mathbf{else}:\\ \;\;\;\;0.0625\\ \end{array} \end{array} \]

FPCore

(FPCore (i) :precision binary64 (if (<= i 0.5) (* (* -0.25 i) i) 0.0625))

C

double code(double i) {
	double tmp;
	if (i <= 0.5) {
		tmp = (-0.25 * i) * i;
	} else {
		tmp = 0.0625;
	}
	return tmp;
}

Fortran

real(8) function code(i)
    real(8), intent (in) :: i
    real(8) :: tmp
    if (i <= 0.5d0) then
        tmp = ((-0.25d0) * i) * i
    else
        tmp = 0.0625d0
    end if
    code = tmp
end function

Java

public static double code(double i) {
	double tmp;
	if (i <= 0.5) {
		tmp = (-0.25 * i) * i;
	} else {
		tmp = 0.0625;
	}
	return tmp;
}

Python

def code(i):
	tmp = 0
	if i <= 0.5:
		tmp = (-0.25 * i) * i
	else:
		tmp = 0.0625
	return tmp

Julia

function code(i)
	tmp = 0.0
	if (i <= 0.5)
		tmp = Float64(Float64(-0.25 * i) * i);
	else
		tmp = 0.0625;
	end
	return tmp
end

MATLAB

function tmp_2 = code(i)
	tmp = 0.0;
	if (i <= 0.5)
		tmp = (-0.25 * i) * i;
	else
		tmp = 0.0625;
	end
	tmp_2 = tmp;
end

Wolfram

code[i_] := If[LessEqual[i, 0.5], N[(N[(-0.25 * i), $MachinePrecision] * i), $MachinePrecision], 0.0625]

Derivation

1. Split input into 2 regimes

2. if i < 0.5

   1. Initial program 31.6%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   3. Taylor expanded in i around 0

      \[\leadsto \color{blue}{\frac{-1}{4} \cdot {i}^{2}} \]
   4. Step-by-step derivation

      1. unpow2 N/A

         \[\leadsto \frac{-1}{4} \cdot \color{blue}{\left(i \cdot i\right)} \]

      2. associate-*r* N/A

         \[\leadsto \color{blue}{\left(\frac{-1}{4} \cdot i\right) \cdot i} \]

      3. metadata-eval N/A

         \[\leadsto \left(\color{blue}{\left(\mathsf{neg}\left(\frac{1}{4}\right)\right)} \cdot i\right) \cdot i \]

      4. distribute-lft-neg-in N/A

         \[\leadsto \color{blue}{\left(\mathsf{neg}\left(\frac{1}{4} \cdot i\right)\right)} \cdot i \]

      5. *-commutative N/A

         \[\leadsto \left(\mathsf{neg}\left(\color{blue}{i \cdot \frac{1}{4}}\right)\right) \cdot i \]

      6. distribute-lft-neg-out N/A

         \[\leadsto \color{blue}{\left(\left(\mathsf{neg}\left(i\right)\right) \cdot \frac{1}{4}\right)} \cdot i \]

      7. lower-*.f64 N/A

         \[\leadsto \color{blue}{\left(\left(\mathsf{neg}\left(i\right)\right) \cdot \frac{1}{4}\right) \cdot i} \]

      8. distribute-lft-neg-out N/A

         \[\leadsto \color{blue}{\left(\mathsf{neg}\left(i \cdot \frac{1}{4}\right)\right)} \cdot i \]

      9. *-commutative N/A

         \[\leadsto \left(\mathsf{neg}\left(\color{blue}{\frac{1}{4} \cdot i}\right)\right) \cdot i \]

      10. distribute-lft-neg-in N/A

          \[\leadsto \color{blue}{\left(\left(\mathsf{neg}\left(\frac{1}{4}\right)\right) \cdot i\right)} \cdot i \]

      11. metadata-eval N/A

          \[\leadsto \left(\color{blue}{\frac{-1}{4}} \cdot i\right) \cdot i \]

      12. lower-*.f64 98.5

          \[\leadsto \color{blue}{\left(-0.25 \cdot i\right)} \cdot i \]

   5. Applied rewrites 98.5%

      \[\leadsto \color{blue}{\left(-0.25 \cdot i\right) \cdot i} \]

   if 0.5 < i

   1. Initial program 20.8%

      \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
   2. Add Preprocessing

   3. Taylor expanded in i around inf

      \[\leadsto \color{blue}{\frac{1}{16}} \]
   4. Step-by-step derivation

   5. Applied rewrites 99.3%

      \[\leadsto \color{blue}{0.0625} \]
3. Recombined 2 regimes into one program.
4. Add Preprocessing

Alternative 8: 51.4% accurate, 71.0× speedup

Math

\[\begin{array}{l} \\ 0.0625 \end{array} \]

FPCore

(FPCore (i) :precision binary64 0.0625)

C

double code(double i) {
	return 0.0625;
}

Fortran

real(8) function code(i)
    real(8), intent (in) :: i
    code = 0.0625d0
end function

Java

public static double code(double i) {
	return 0.0625;
}

Python

def code(i):
	return 0.0625

Julia

function code(i)
	return 0.0625
end

MATLAB

function tmp = code(i)
	tmp = 0.0625;
end

Wolfram

code[i_] := 0.0625

Derivation

1. Initial program 26.3%

   \[\frac{\frac{\left(i \cdot i\right) \cdot \left(i \cdot i\right)}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right)}}{\left(2 \cdot i\right) \cdot \left(2 \cdot i\right) - 1} \]
2. Add Preprocessing

3. Taylor expanded in i around inf

   \[\leadsto \color{blue}{\frac{1}{16}} \]
4. Step-by-step derivation

5. Applied rewrites 50.2%

   \[\leadsto \color{blue}{0.0625} \]
6. Add Preprocessing

Reproduce

herbie shell --seed 2024319 
(FPCore (i)
  :name "Octave 3.8, jcobi/4, as called"
  :precision binary64
  :pre (> i 0.0)
  (/ (/ (* (* i i) (* i i)) (* (* 2.0 i) (* 2.0 i))) (- (* (* 2.0 i) (* 2.0 i)) 1.0)))