Rust f64::asinh

Percentage Accurate: 30.3% → 99.9%
Time: 2.2s
Alternatives: 3
Speedup: 3.1×

Specification

?
\[\mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right), x\right) \]
(FPCore (x)
 :precision binary64
 (copysign (log (+ (fabs x) (sqrt (+ (* x x) 1.0)))) x))
double code(double x) {
	return copysign(log((fabs(x) + sqrt(((x * x) + 1.0)))), x);
}
public static double code(double x) {
	return Math.copySign(Math.log((Math.abs(x) + Math.sqrt(((x * x) + 1.0)))), x);
}
def code(x):
	return math.copysign(math.log((math.fabs(x) + math.sqrt(((x * x) + 1.0)))), x)
function code(x)
	return copysign(log(Float64(abs(x) + sqrt(Float64(Float64(x * x) + 1.0)))), x)
end
function tmp = code(x)
	tmp = sign(x) * abs(log((abs(x) + sqrt(((x * x) + 1.0)))));
end
code[x_] := N[With[{TMP1 = Abs[N[Log[N[(N[Abs[x], $MachinePrecision] + N[Sqrt[N[(N[(x * x), $MachinePrecision] + 1.0), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]], $MachinePrecision]], TMP2 = Sign[x]}, TMP1 * If[TMP2 == 0, 1, TMP2]], $MachinePrecision]
\mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right), x\right)

Local Percentage Accuracy vs ?

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

Accuracy vs Speed?

Herbie found 3 alternatives:

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

Initial Program: 30.3% accurate, 1.0× speedup?

\[\mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right), x\right) \]
(FPCore (x)
 :precision binary64
 (copysign (log (+ (fabs x) (sqrt (+ (* x x) 1.0)))) x))
double code(double x) {
	return copysign(log((fabs(x) + sqrt(((x * x) + 1.0)))), x);
}
public static double code(double x) {
	return Math.copySign(Math.log((Math.abs(x) + Math.sqrt(((x * x) + 1.0)))), x);
}
def code(x):
	return math.copysign(math.log((math.fabs(x) + math.sqrt(((x * x) + 1.0)))), x)
function code(x)
	return copysign(log(Float64(abs(x) + sqrt(Float64(Float64(x * x) + 1.0)))), x)
end
function tmp = code(x)
	tmp = sign(x) * abs(log((abs(x) + sqrt(((x * x) + 1.0)))));
end
code[x_] := N[With[{TMP1 = Abs[N[Log[N[(N[Abs[x], $MachinePrecision] + N[Sqrt[N[(N[(x * x), $MachinePrecision] + 1.0), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]], $MachinePrecision]], TMP2 = Sign[x]}, TMP1 * If[TMP2 == 0, 1, TMP2]], $MachinePrecision]
\mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right), x\right)

Alternative 1: 99.9% accurate, 1.6× speedup?

\[\mathsf{copysign}\left(\sinh^{-1} x, x\right) \]
(FPCore (x) :precision binary64 (copysign (asinh x) x))
double code(double x) {
	return copysign(asinh(x), x);
}
def code(x):
	return math.copysign(math.asinh(x), x)
function code(x)
	return copysign(asinh(x), x)
end
function tmp = code(x)
	tmp = sign(x) * abs(asinh(x));
end
code[x_] := N[With[{TMP1 = Abs[N[ArcSinh[x], $MachinePrecision]], TMP2 = Sign[x]}, TMP1 * If[TMP2 == 0, 1, TMP2]], $MachinePrecision]
\mathsf{copysign}\left(\sinh^{-1} x, x\right)
Derivation
  1. Initial program 30.3%

    \[\mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right), x\right) \]
  2. Step-by-step derivation
    1. lift-log.f64N/A

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right)}, x\right) \]
    2. lift-+.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \color{blue}{\left(\left|x\right| + \sqrt{x \cdot x + 1}\right)}, x\right) \]
    3. lift-sqrt.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \color{blue}{\sqrt{x \cdot x + 1}}\right), x\right) \]
    4. lift-+.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{x \cdot x + 1}}\right), x\right) \]
    5. lift-*.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{x \cdot x} + 1}\right), x\right) \]
    6. sqr-abs-revN/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{\left|x\right| \cdot \left|x\right|} + 1}\right), x\right) \]
    7. lift-fabs.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{\left|x\right|} \cdot \left|x\right| + 1}\right), x\right) \]
    8. lift-fabs.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\left|x\right| \cdot \color{blue}{\left|x\right|} + 1}\right), x\right) \]
    9. asinh-def-revN/A

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\sinh^{-1} \left(\left|x\right|\right)}, x\right) \]
    10. lower-asinh.f6499.9%

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\sinh^{-1} \left(\left|x\right|\right)}, x\right) \]
  3. Applied rewrites99.9%

    \[\leadsto \color{blue}{\mathsf{copysign}\left(\sinh^{-1} \left(\left|x\right|\right), x\right)} \]
  4. Step-by-step derivation
    1. lift-fabs.f64N/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\left|x\right|\right)}, x\right) \]
    2. rem-sqrt-square-revN/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{x \cdot x}\right)}, x\right) \]
    3. sqr-neg-revN/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{\color{blue}{\left(\mathsf{neg}\left(x\right)\right) \cdot \left(\mathsf{neg}\left(x\right)\right)}}\right), x\right) \]
    4. sqrt-prodN/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{\mathsf{neg}\left(x\right)} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right)}, x\right) \]
    5. lower-unsound-*.f64N/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{\mathsf{neg}\left(x\right)} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right)}, x\right) \]
    6. lower-unsound-sqrt.f64N/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\color{blue}{\sqrt{\mathsf{neg}\left(x\right)}} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right), x\right) \]
    7. lower-neg.f64N/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{\color{blue}{-x}} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right), x\right) \]
    8. lower-unsound-sqrt.f64N/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{-x} \cdot \color{blue}{\sqrt{\mathsf{neg}\left(x\right)}}\right), x\right) \]
    9. lower-neg.f6450.4%

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{-x} \cdot \sqrt{\color{blue}{-x}}\right), x\right) \]
  5. Applied rewrites50.4%

    \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{-x} \cdot \sqrt{-x}\right)}, x\right) \]
  6. Step-by-step derivation
    1. lift-asinh.f64N/A

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\sinh^{-1} \left(\sqrt{-x} \cdot \sqrt{-x}\right)}, x\right) \]
    2. asinh-defN/A

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\log \left(\sqrt{-x} \cdot \sqrt{-x} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right)}, x\right) \]
    3. lift-*.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\color{blue}{\sqrt{-x} \cdot \sqrt{-x}} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    4. fabs-sqrN/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\color{blue}{\left|\sqrt{-x} \cdot \sqrt{-x}\right|} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    5. lift-sqrt.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|\color{blue}{\sqrt{-x}} \cdot \sqrt{-x}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    6. lift-sqrt.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|\sqrt{-x} \cdot \color{blue}{\sqrt{-x}}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    7. rem-square-sqrtN/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|\color{blue}{-x}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    8. lift-neg.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|\color{blue}{\mathsf{neg}\left(x\right)}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    9. neg-fabsN/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\color{blue}{\left|x\right|} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
  7. Applied rewrites99.9%

    \[\leadsto \mathsf{copysign}\left(\color{blue}{-\sinh^{-1} x}, x\right) \]
  8. Step-by-step derivation
    1. lift-copysign.f64N/A

      \[\leadsto \color{blue}{\mathsf{copysign}\left(-\sinh^{-1} x, x\right)} \]
    2. lift-neg.f64N/A

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\mathsf{neg}\left(\sinh^{-1} x\right)}, x\right) \]
    3. copysign-other-negN/A

      \[\leadsto \color{blue}{\mathsf{copysign}\left(\sinh^{-1} x, x\right)} \]
    4. lower-copysign.f6499.9%

      \[\leadsto \color{blue}{\mathsf{copysign}\left(\sinh^{-1} x, x\right)} \]
  9. Applied rewrites99.9%

    \[\leadsto \color{blue}{\mathsf{copysign}\left(\sinh^{-1} x, x\right)} \]
  10. Add Preprocessing

Alternative 2: 65.1% accurate, 0.4× speedup?

\[\begin{array}{l} t_0 := \left|\left|x\right|\right|\\ \mathsf{copysign}\left(1, x\right) \cdot \begin{array}{l} \mathbf{if}\;\mathsf{copysign}\left(\log \left(t\_0 + \sqrt{\left|x\right| \cdot \left|x\right| + 1}\right), \left|x\right|\right) \leq 0.5:\\ \;\;\;\;\mathsf{copysign}\left(-1 \cdot \left|x\right|, \left|x\right|\right)\\ \mathbf{else}:\\ \;\;\;\;\mathsf{copysign}\left(\log t\_0, \left|x\right|\right)\\ \end{array} \end{array} \]
(FPCore (x)
 :precision binary64
 (let* ((t_0 (fabs (fabs x))))
   (*
    (copysign 1.0 x)
    (if (<=
         (copysign (log (+ t_0 (sqrt (+ (* (fabs x) (fabs x)) 1.0)))) (fabs x))
         0.5)
      (copysign (* -1.0 (fabs x)) (fabs x))
      (copysign (log t_0) (fabs x))))))
double code(double x) {
	double t_0 = fabs(fabs(x));
	double tmp;
	if (copysign(log((t_0 + sqrt(((fabs(x) * fabs(x)) + 1.0)))), fabs(x)) <= 0.5) {
		tmp = copysign((-1.0 * fabs(x)), fabs(x));
	} else {
		tmp = copysign(log(t_0), fabs(x));
	}
	return copysign(1.0, x) * tmp;
}
public static double code(double x) {
	double t_0 = Math.abs(Math.abs(x));
	double tmp;
	if (Math.copySign(Math.log((t_0 + Math.sqrt(((Math.abs(x) * Math.abs(x)) + 1.0)))), Math.abs(x)) <= 0.5) {
		tmp = Math.copySign((-1.0 * Math.abs(x)), Math.abs(x));
	} else {
		tmp = Math.copySign(Math.log(t_0), Math.abs(x));
	}
	return Math.copySign(1.0, x) * tmp;
}
def code(x):
	t_0 = math.fabs(math.fabs(x))
	tmp = 0
	if math.copysign(math.log((t_0 + math.sqrt(((math.fabs(x) * math.fabs(x)) + 1.0)))), math.fabs(x)) <= 0.5:
		tmp = math.copysign((-1.0 * math.fabs(x)), math.fabs(x))
	else:
		tmp = math.copysign(math.log(t_0), math.fabs(x))
	return math.copysign(1.0, x) * tmp
function code(x)
	t_0 = abs(abs(x))
	tmp = 0.0
	if (copysign(log(Float64(t_0 + sqrt(Float64(Float64(abs(x) * abs(x)) + 1.0)))), abs(x)) <= 0.5)
		tmp = copysign(Float64(-1.0 * abs(x)), abs(x));
	else
		tmp = copysign(log(t_0), abs(x));
	end
	return Float64(copysign(1.0, x) * tmp)
end
function tmp_2 = code(x)
	t_0 = abs(abs(x));
	tmp = 0.0;
	if ((sign(abs(x)) * abs(log((t_0 + sqrt(((abs(x) * abs(x)) + 1.0)))))) <= 0.5)
		tmp = sign(abs(x)) * abs((-1.0 * abs(x)));
	else
		tmp = sign(abs(x)) * abs(log(t_0));
	end
	tmp_2 = (sign(x) * abs(1.0)) * tmp;
end
code[x_] := Block[{t$95$0 = N[Abs[N[Abs[x], $MachinePrecision]], $MachinePrecision]}, N[(N[With[{TMP1 = Abs[1.0], TMP2 = Sign[x]}, TMP1 * If[TMP2 == 0, 1, TMP2]], $MachinePrecision] * If[LessEqual[N[With[{TMP1 = Abs[N[Log[N[(t$95$0 + N[Sqrt[N[(N[(N[Abs[x], $MachinePrecision] * N[Abs[x], $MachinePrecision]), $MachinePrecision] + 1.0), $MachinePrecision]], $MachinePrecision]), $MachinePrecision]], $MachinePrecision]], TMP2 = Sign[N[Abs[x], $MachinePrecision]]}, TMP1 * If[TMP2 == 0, 1, TMP2]], $MachinePrecision], 0.5], N[With[{TMP1 = Abs[N[(-1.0 * N[Abs[x], $MachinePrecision]), $MachinePrecision]], TMP2 = Sign[N[Abs[x], $MachinePrecision]]}, TMP1 * If[TMP2 == 0, 1, TMP2]], $MachinePrecision], N[With[{TMP1 = Abs[N[Log[t$95$0], $MachinePrecision]], TMP2 = Sign[N[Abs[x], $MachinePrecision]]}, TMP1 * If[TMP2 == 0, 1, TMP2]], $MachinePrecision]]), $MachinePrecision]]
\begin{array}{l}
t_0 := \left|\left|x\right|\right|\\
\mathsf{copysign}\left(1, x\right) \cdot \begin{array}{l}
\mathbf{if}\;\mathsf{copysign}\left(\log \left(t\_0 + \sqrt{\left|x\right| \cdot \left|x\right| + 1}\right), \left|x\right|\right) \leq 0.5:\\
\;\;\;\;\mathsf{copysign}\left(-1 \cdot \left|x\right|, \left|x\right|\right)\\

\mathbf{else}:\\
\;\;\;\;\mathsf{copysign}\left(\log t\_0, \left|x\right|\right)\\


\end{array}
\end{array}
Derivation
  1. Split input into 2 regimes
  2. if (copysign.f64 (log.f64 (+.f64 (fabs.f64 x) (sqrt.f64 (+.f64 (*.f64 x x) #s(literal 1 binary64))))) x) < 0.5

    1. Initial program 30.3%

      \[\mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right), x\right) \]
    2. Step-by-step derivation
      1. lift-log.f64N/A

        \[\leadsto \mathsf{copysign}\left(\color{blue}{\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right)}, x\right) \]
      2. lift-+.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \color{blue}{\left(\left|x\right| + \sqrt{x \cdot x + 1}\right)}, x\right) \]
      3. lift-sqrt.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \color{blue}{\sqrt{x \cdot x + 1}}\right), x\right) \]
      4. lift-+.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{x \cdot x + 1}}\right), x\right) \]
      5. lift-*.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{x \cdot x} + 1}\right), x\right) \]
      6. sqr-abs-revN/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{\left|x\right| \cdot \left|x\right|} + 1}\right), x\right) \]
      7. lift-fabs.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{\left|x\right|} \cdot \left|x\right| + 1}\right), x\right) \]
      8. lift-fabs.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\left|x\right| \cdot \color{blue}{\left|x\right|} + 1}\right), x\right) \]
      9. asinh-def-revN/A

        \[\leadsto \mathsf{copysign}\left(\color{blue}{\sinh^{-1} \left(\left|x\right|\right)}, x\right) \]
      10. lower-asinh.f6499.9%

        \[\leadsto \mathsf{copysign}\left(\color{blue}{\sinh^{-1} \left(\left|x\right|\right)}, x\right) \]
    3. Applied rewrites99.9%

      \[\leadsto \color{blue}{\mathsf{copysign}\left(\sinh^{-1} \left(\left|x\right|\right), x\right)} \]
    4. Step-by-step derivation
      1. lift-fabs.f64N/A

        \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\left|x\right|\right)}, x\right) \]
      2. rem-sqrt-square-revN/A

        \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{x \cdot x}\right)}, x\right) \]
      3. sqr-neg-revN/A

        \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{\color{blue}{\left(\mathsf{neg}\left(x\right)\right) \cdot \left(\mathsf{neg}\left(x\right)\right)}}\right), x\right) \]
      4. sqrt-prodN/A

        \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{\mathsf{neg}\left(x\right)} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right)}, x\right) \]
      5. lower-unsound-*.f64N/A

        \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{\mathsf{neg}\left(x\right)} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right)}, x\right) \]
      6. lower-unsound-sqrt.f64N/A

        \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\color{blue}{\sqrt{\mathsf{neg}\left(x\right)}} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right), x\right) \]
      7. lower-neg.f64N/A

        \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{\color{blue}{-x}} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right), x\right) \]
      8. lower-unsound-sqrt.f64N/A

        \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{-x} \cdot \color{blue}{\sqrt{\mathsf{neg}\left(x\right)}}\right), x\right) \]
      9. lower-neg.f6450.4%

        \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{-x} \cdot \sqrt{\color{blue}{-x}}\right), x\right) \]
    5. Applied rewrites50.4%

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{-x} \cdot \sqrt{-x}\right)}, x\right) \]
    6. Step-by-step derivation
      1. lift-asinh.f64N/A

        \[\leadsto \mathsf{copysign}\left(\color{blue}{\sinh^{-1} \left(\sqrt{-x} \cdot \sqrt{-x}\right)}, x\right) \]
      2. asinh-defN/A

        \[\leadsto \mathsf{copysign}\left(\color{blue}{\log \left(\sqrt{-x} \cdot \sqrt{-x} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right)}, x\right) \]
      3. lift-*.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\color{blue}{\sqrt{-x} \cdot \sqrt{-x}} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
      4. fabs-sqrN/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\color{blue}{\left|\sqrt{-x} \cdot \sqrt{-x}\right|} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
      5. lift-sqrt.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|\color{blue}{\sqrt{-x}} \cdot \sqrt{-x}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
      6. lift-sqrt.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|\sqrt{-x} \cdot \color{blue}{\sqrt{-x}}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
      7. rem-square-sqrtN/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|\color{blue}{-x}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
      8. lift-neg.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|\color{blue}{\mathsf{neg}\left(x\right)}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
      9. neg-fabsN/A

        \[\leadsto \mathsf{copysign}\left(\log \left(\color{blue}{\left|x\right|} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    7. Applied rewrites99.9%

      \[\leadsto \mathsf{copysign}\left(\color{blue}{-\sinh^{-1} x}, x\right) \]
    8. Taylor expanded in x around 0

      \[\leadsto \mathsf{copysign}\left(\color{blue}{-1 \cdot x}, x\right) \]
    9. Step-by-step derivation
      1. lower-*.f6452.2%

        \[\leadsto \mathsf{copysign}\left(-1 \cdot \color{blue}{x}, x\right) \]
    10. Applied rewrites52.2%

      \[\leadsto \mathsf{copysign}\left(\color{blue}{-1 \cdot x}, x\right) \]

    if 0.5 < (copysign.f64 (log.f64 (+.f64 (fabs.f64 x) (sqrt.f64 (+.f64 (*.f64 x x) #s(literal 1 binary64))))) x)

    1. Initial program 30.3%

      \[\mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right), x\right) \]
    2. Taylor expanded in x around -inf

      \[\leadsto \mathsf{copysign}\left(\log \color{blue}{\left(-1 \cdot \left(x \cdot \left(1 + -1 \cdot \frac{\left|x\right|}{x}\right)\right)\right)}, x\right) \]
    3. Step-by-step derivation
      1. lower-*.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(-1 \cdot \color{blue}{\left(x \cdot \left(1 + -1 \cdot \frac{\left|x\right|}{x}\right)\right)}\right), x\right) \]
      2. lower-*.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(-1 \cdot \left(x \cdot \color{blue}{\left(1 + -1 \cdot \frac{\left|x\right|}{x}\right)}\right)\right), x\right) \]
      3. lower-+.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(-1 \cdot \left(x \cdot \left(1 + \color{blue}{-1 \cdot \frac{\left|x\right|}{x}}\right)\right)\right), x\right) \]
      4. lower-*.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(-1 \cdot \left(x \cdot \left(1 + -1 \cdot \color{blue}{\frac{\left|x\right|}{x}}\right)\right)\right), x\right) \]
      5. lower-/.f64N/A

        \[\leadsto \mathsf{copysign}\left(\log \left(-1 \cdot \left(x \cdot \left(1 + -1 \cdot \frac{\left|x\right|}{\color{blue}{x}}\right)\right)\right), x\right) \]
      6. lower-fabs.f6427.4%

        \[\leadsto \mathsf{copysign}\left(\log \left(-1 \cdot \left(x \cdot \left(1 + -1 \cdot \frac{\left|x\right|}{x}\right)\right)\right), x\right) \]
    4. Applied rewrites27.4%

      \[\leadsto \mathsf{copysign}\left(\log \color{blue}{\left(-1 \cdot \left(x \cdot \left(1 + -1 \cdot \frac{\left|x\right|}{x}\right)\right)\right)}, x\right) \]
    5. Taylor expanded in x around 0

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right|\right), x\right) \]
    6. Step-by-step derivation
      1. lower-fabs.f6418.4%

        \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right|\right), x\right) \]
    7. Applied rewrites18.4%

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right|\right), x\right) \]
  3. Recombined 2 regimes into one program.
  4. Add Preprocessing

Alternative 3: 52.2% accurate, 3.1× speedup?

\[\mathsf{copysign}\left(-1 \cdot x, x\right) \]
(FPCore (x) :precision binary64 (copysign (* -1.0 x) x))
double code(double x) {
	return copysign((-1.0 * x), x);
}
public static double code(double x) {
	return Math.copySign((-1.0 * x), x);
}
def code(x):
	return math.copysign((-1.0 * x), x)
function code(x)
	return copysign(Float64(-1.0 * x), x)
end
function tmp = code(x)
	tmp = sign(x) * abs((-1.0 * x));
end
code[x_] := N[With[{TMP1 = Abs[N[(-1.0 * x), $MachinePrecision]], TMP2 = Sign[x]}, TMP1 * If[TMP2 == 0, 1, TMP2]], $MachinePrecision]
\mathsf{copysign}\left(-1 \cdot x, x\right)
Derivation
  1. Initial program 30.3%

    \[\mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right), x\right) \]
  2. Step-by-step derivation
    1. lift-log.f64N/A

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\log \left(\left|x\right| + \sqrt{x \cdot x + 1}\right)}, x\right) \]
    2. lift-+.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \color{blue}{\left(\left|x\right| + \sqrt{x \cdot x + 1}\right)}, x\right) \]
    3. lift-sqrt.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \color{blue}{\sqrt{x \cdot x + 1}}\right), x\right) \]
    4. lift-+.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{x \cdot x + 1}}\right), x\right) \]
    5. lift-*.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{x \cdot x} + 1}\right), x\right) \]
    6. sqr-abs-revN/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{\left|x\right| \cdot \left|x\right|} + 1}\right), x\right) \]
    7. lift-fabs.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\color{blue}{\left|x\right|} \cdot \left|x\right| + 1}\right), x\right) \]
    8. lift-fabs.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|x\right| + \sqrt{\left|x\right| \cdot \color{blue}{\left|x\right|} + 1}\right), x\right) \]
    9. asinh-def-revN/A

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\sinh^{-1} \left(\left|x\right|\right)}, x\right) \]
    10. lower-asinh.f6499.9%

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\sinh^{-1} \left(\left|x\right|\right)}, x\right) \]
  3. Applied rewrites99.9%

    \[\leadsto \color{blue}{\mathsf{copysign}\left(\sinh^{-1} \left(\left|x\right|\right), x\right)} \]
  4. Step-by-step derivation
    1. lift-fabs.f64N/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\left|x\right|\right)}, x\right) \]
    2. rem-sqrt-square-revN/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{x \cdot x}\right)}, x\right) \]
    3. sqr-neg-revN/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{\color{blue}{\left(\mathsf{neg}\left(x\right)\right) \cdot \left(\mathsf{neg}\left(x\right)\right)}}\right), x\right) \]
    4. sqrt-prodN/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{\mathsf{neg}\left(x\right)} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right)}, x\right) \]
    5. lower-unsound-*.f64N/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{\mathsf{neg}\left(x\right)} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right)}, x\right) \]
    6. lower-unsound-sqrt.f64N/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\color{blue}{\sqrt{\mathsf{neg}\left(x\right)}} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right), x\right) \]
    7. lower-neg.f64N/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{\color{blue}{-x}} \cdot \sqrt{\mathsf{neg}\left(x\right)}\right), x\right) \]
    8. lower-unsound-sqrt.f64N/A

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{-x} \cdot \color{blue}{\sqrt{\mathsf{neg}\left(x\right)}}\right), x\right) \]
    9. lower-neg.f6450.4%

      \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \left(\sqrt{-x} \cdot \sqrt{\color{blue}{-x}}\right), x\right) \]
  5. Applied rewrites50.4%

    \[\leadsto \mathsf{copysign}\left(\sinh^{-1} \color{blue}{\left(\sqrt{-x} \cdot \sqrt{-x}\right)}, x\right) \]
  6. Step-by-step derivation
    1. lift-asinh.f64N/A

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\sinh^{-1} \left(\sqrt{-x} \cdot \sqrt{-x}\right)}, x\right) \]
    2. asinh-defN/A

      \[\leadsto \mathsf{copysign}\left(\color{blue}{\log \left(\sqrt{-x} \cdot \sqrt{-x} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right)}, x\right) \]
    3. lift-*.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\color{blue}{\sqrt{-x} \cdot \sqrt{-x}} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    4. fabs-sqrN/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\color{blue}{\left|\sqrt{-x} \cdot \sqrt{-x}\right|} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    5. lift-sqrt.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|\color{blue}{\sqrt{-x}} \cdot \sqrt{-x}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    6. lift-sqrt.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|\sqrt{-x} \cdot \color{blue}{\sqrt{-x}}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    7. rem-square-sqrtN/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|\color{blue}{-x}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    8. lift-neg.f64N/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\left|\color{blue}{\mathsf{neg}\left(x\right)}\right| + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
    9. neg-fabsN/A

      \[\leadsto \mathsf{copysign}\left(\log \left(\color{blue}{\left|x\right|} + \sqrt{\left(\sqrt{-x} \cdot \sqrt{-x}\right) \cdot \left(\sqrt{-x} \cdot \sqrt{-x}\right) + 1}\right), x\right) \]
  7. Applied rewrites99.9%

    \[\leadsto \mathsf{copysign}\left(\color{blue}{-\sinh^{-1} x}, x\right) \]
  8. Taylor expanded in x around 0

    \[\leadsto \mathsf{copysign}\left(\color{blue}{-1 \cdot x}, x\right) \]
  9. Step-by-step derivation
    1. lower-*.f6452.2%

      \[\leadsto \mathsf{copysign}\left(-1 \cdot \color{blue}{x}, x\right) \]
  10. Applied rewrites52.2%

    \[\leadsto \mathsf{copysign}\left(\color{blue}{-1 \cdot x}, x\right) \]
  11. Add Preprocessing

Developer Target 1: 100.0% accurate, 0.4× speedup?

\[\begin{array}{l} t_0 := \frac{1}{\left|x\right|}\\ \mathsf{copysign}\left(\mathsf{log1p}\left(\left|x\right| + \frac{\left|x\right|}{\mathsf{hypot}\left(1, t\_0\right) + t\_0}\right), x\right) \end{array} \]
(FPCore (x)
 :precision binary64
 (let* ((t_0 (/ 1.0 (fabs x))))
   (copysign (log1p (+ (fabs x) (/ (fabs x) (+ (hypot 1.0 t_0) t_0)))) x)))
double code(double x) {
	double t_0 = 1.0 / fabs(x);
	return copysign(log1p((fabs(x) + (fabs(x) / (hypot(1.0, t_0) + t_0)))), x);
}
public static double code(double x) {
	double t_0 = 1.0 / Math.abs(x);
	return Math.copySign(Math.log1p((Math.abs(x) + (Math.abs(x) / (Math.hypot(1.0, t_0) + t_0)))), x);
}
def code(x):
	t_0 = 1.0 / math.fabs(x)
	return math.copysign(math.log1p((math.fabs(x) + (math.fabs(x) / (math.hypot(1.0, t_0) + t_0)))), x)
function code(x)
	t_0 = Float64(1.0 / abs(x))
	return copysign(log1p(Float64(abs(x) + Float64(abs(x) / Float64(hypot(1.0, t_0) + t_0)))), x)
end
code[x_] := Block[{t$95$0 = N[(1.0 / N[Abs[x], $MachinePrecision]), $MachinePrecision]}, N[With[{TMP1 = Abs[N[Log[1 + N[(N[Abs[x], $MachinePrecision] + N[(N[Abs[x], $MachinePrecision] / N[(N[Sqrt[1.0 ^ 2 + t$95$0 ^ 2], $MachinePrecision] + t$95$0), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]], $MachinePrecision]], TMP2 = Sign[x]}, TMP1 * If[TMP2 == 0, 1, TMP2]], $MachinePrecision]]
\begin{array}{l}
t_0 := \frac{1}{\left|x\right|}\\
\mathsf{copysign}\left(\mathsf{log1p}\left(\left|x\right| + \frac{\left|x\right|}{\mathsf{hypot}\left(1, t\_0\right) + t\_0}\right), x\right)
\end{array}

Reproduce

?
herbie shell --seed 2025192 
(FPCore (x)
  :name "Rust f64::asinh"
  :precision binary64

  :alt
  (! :herbie-platform c (let* ((ax (fabs x)) (ix (/ 1 ax))) (copysign (log1p (+ ax (/ ax (+ (hypot 1 ix) ix)))) x)))

  (copysign (log (+ (fabs x) (sqrt (+ (* x x) 1.0)))) x))