Average Error: 0.4 → 0.2
Time: 3.8s
Precision: binary64
\[\left(0 \leq u1 \land u1 \leq 1\right) \land \left(0 \leq u2 \land u2 \leq 1\right)\]
\[\left(\frac{1}{6} \cdot {\left(-2 \cdot \log u1\right)}^{0.5}\right) \cdot \cos \left(\left(2 \cdot \pi\right) \cdot u2\right) + 0.5 \]
↓
\[\sqrt{\mathsf{fma}\left(-0.05555555555555555, \log u1, 1\right) + -1} \cdot \cos \left(\left(2 \cdot \pi\right) \cdot u2\right) + 0.5 \]
(FPCore (u1 u2)
 :precision binary64
 (+
  (* (* (/ 1.0 6.0) (pow (* -2.0 (log u1)) 0.5)) (cos (* (* 2.0 PI) u2)))
  0.5))
↓
(FPCore (u1 u2)
 :precision binary64
 (+
  (*
   (sqrt (+ (fma -0.05555555555555555 (log u1) 1.0) -1.0))
   (cos (* (* 2.0 PI) u2)))
  0.5))
double code(double u1, double u2) {
	return (((1.0 / 6.0) * pow((-2.0 * log(u1)), 0.5)) * cos(((2.0 * ((double) M_PI)) * u2))) + 0.5;
}
↓
double code(double u1, double u2) {
	return (sqrt((fma(-0.05555555555555555, log(u1), 1.0) + -1.0)) * cos(((2.0 * ((double) M_PI)) * u2))) + 0.5;
}

function code(u1, u2)
	return Float64(Float64(Float64(Float64(1.0 / 6.0) * (Float64(-2.0 * log(u1)) ^ 0.5)) * cos(Float64(Float64(2.0 * pi) * u2))) + 0.5)
end
↓
function code(u1, u2)
	return Float64(Float64(sqrt(Float64(fma(-0.05555555555555555, log(u1), 1.0) + -1.0)) * cos(Float64(Float64(2.0 * pi) * u2))) + 0.5)
end

code[u1_, u2_] := N[(N[(N[(N[(1.0 / 6.0), $MachinePrecision] * N[Power[N[(-2.0 * N[Log[u1], $MachinePrecision]), $MachinePrecision], 0.5], $MachinePrecision]), $MachinePrecision] * N[Cos[N[(N[(2.0 * Pi), $MachinePrecision] * u2), $MachinePrecision]], $MachinePrecision]), $MachinePrecision] + 0.5), $MachinePrecision]
↓
code[u1_, u2_] := N[(N[(N[Sqrt[N[(N[(-0.05555555555555555 * N[Log[u1], $MachinePrecision] + 1.0), $MachinePrecision] + -1.0), $MachinePrecision]], $MachinePrecision] * N[Cos[N[(N[(2.0 * Pi), $MachinePrecision] * u2), $MachinePrecision]], $MachinePrecision]), $MachinePrecision] + 0.5), $MachinePrecision]

\left(\frac{1}{6} \cdot {\left(-2 \cdot \log u1\right)}^{0.5}\right) \cdot \cos \left(\left(2 \cdot \pi\right) \cdot u2\right) + 0.5
↓
\sqrt{\mathsf{fma}\left(-0.05555555555555555, \log u1, 1\right) + -1} \cdot \cos \left(\left(2 \cdot \pi\right) \cdot u2\right) + 0.5

Error

Derivation

  1. Initial program 0.4

    \[\left(\frac{1}{6} \cdot {\left(-2 \cdot \log u1\right)}^{0.5}\right) \cdot \cos \left(\left(2 \cdot \pi\right) \cdot u2\right) + 0.5 \]

  2. Applied egg-rr0.2

    \[\leadsto \color{blue}{\sqrt{\left(-2 \cdot \log u1\right) \cdot 0.027777777777777776}} \cdot \cos \left(\left(2 \cdot \pi\right) \cdot u2\right) + 0.5 \]

  3. Applied egg-rr0.4

    \[\leadsto \sqrt{\color{blue}{e^{\mathsf{log1p}\left(\log u1 \cdot -0.05555555555555555\right)} - 1}} \cdot \cos \left(\left(2 \cdot \pi\right) \cdot u2\right) + 0.5 \]

  4. Taylor expanded in u1 around 0 0.2

    \[\leadsto \sqrt{\color{blue}{\left(-0.05555555555555555 \cdot \log u1 + 1\right)} - 1} \cdot \cos \left(\left(2 \cdot \pi\right) \cdot u2\right) + 0.5 \]

  5. Simplified0.2

    \[\leadsto \sqrt{\color{blue}{\mathsf{fma}\left(-0.05555555555555555, \log u1, 1\right)} - 1} \cdot \cos \left(\left(2 \cdot \pi\right) \cdot u2\right) + 0.5 \]

  6. Final simplification0.2

    \[\leadsto \sqrt{\mathsf{fma}\left(-0.05555555555555555, \log u1, 1\right) + -1} \cdot \cos \left(\left(2 \cdot \pi\right) \cdot u2\right) + 0.5 \]

Reproduce

herbie shell --seed 2022192 
(FPCore (u1 u2)
  :name "normal distribution"
  :precision binary64
  :pre (and (and (<= 0.0 u1) (<= u1 1.0)) (and (<= 0.0 u2) (<= u2 1.0)))
  (+ (* (* (/ 1.0 6.0) (pow (* -2.0 (log u1)) 0.5)) (cos (* (* 2.0 PI) u2))) 0.5))