\[{\left(\tan \left(\log_* (1 + a)\right)\right)}^{\left(\log \left({a}^{\left(\sinh a\right)}\right)\right)}\]
Test:
(pow (tan (log1p a)) (log (pow a (sinh a))))
Bits:
128 bits
Bits error versus a
Bits error versus b
Bits error versus c
Time: 16.4 s
Input Error: 0.2
Output Error: 0.1
Log: ⚲
Profile: 🕒
\({\left({\left({\left(\sqrt[3]{\sqrt[3]{\tan \left(\log_* (1 + a)\right)}}\right)}^3\right)}^3\right)}^{\left(\log a \cdot \sinh a\right)}\)
  1. Started with
    \[{\left(\tan \left(\log_* (1 + a)\right)\right)}^{\left(\log \left({a}^{\left(\sinh a\right)}\right)\right)}\]
    0.2
  2. Applied taylor to get
    \[{\left(\tan \left(\log_* (1 + a)\right)\right)}^{\left(\log \left({a}^{\left(\sinh a\right)}\right)\right)} \leadsto {\left(\tan \left(\log_* (1 + a)\right)\right)}^{\left(\log a \cdot \sinh a\right)}\]
    0.0
  3. Taylor expanded around 0 to get
    \[{\left(\tan \left(\log_* (1 + a)\right)\right)}^{\color{red}{\left(\log a \cdot \sinh a\right)}} \leadsto {\left(\tan \left(\log_* (1 + a)\right)\right)}^{\color{blue}{\left(\log a \cdot \sinh a\right)}}\]
    0.0
  4. Using strategy rm
    0.0
  5. Applied add-cube-cbrt to get
    \[{\color{red}{\left(\tan \left(\log_* (1 + a)\right)\right)}}^{\left(\log a \cdot \sinh a\right)} \leadsto {\color{blue}{\left({\left(\sqrt[3]{\tan \left(\log_* (1 + a)\right)}\right)}^3\right)}}^{\left(\log a \cdot \sinh a\right)}\]
    0.0
  6. Using strategy rm
    0.0
  7. Applied add-cube-cbrt to get
    \[{\left({\color{red}{\left(\sqrt[3]{\tan \left(\log_* (1 + a)\right)}\right)}}^3\right)}^{\left(\log a \cdot \sinh a\right)} \leadsto {\left({\color{blue}{\left({\left(\sqrt[3]{\sqrt[3]{\tan \left(\log_* (1 + a)\right)}}\right)}^3\right)}}^3\right)}^{\left(\log a \cdot \sinh a\right)}\]
    0.1

Original test:


(lambda ((a default) (b default) (c default))
  #:name "(pow (tan (log1p a)) (log (pow a (sinh a))))"
  (pow (tan (log1p a)) (log (pow a (sinh a)))))