Nick

@CromwellEnage : [This](https://github.com/boostorg/math/blob/develop/include/boost/math/tools/univariate_statistics.hpp) also implements the correct algorithm.

@jzmaddock : I did test it! @mborland , @jzmaddock : Quick question: Given a floating point type `T`, how to I form the associated complex type in the multiprecision case?...

> The static_assert on is_floating_point would have to go as well... Meh, who cares. Got rid of it. > Note that in order to use multiprecision types with Eigen we...

> We really shouldn't need that, and it might get rid of the error? Got rid of it; now I'm here: ``` In file included from test/polynomial_roots_test.cpp:12: /opt/homebrew/opt/llvm/bin/../include/c++/v1/complex:730:62: fatal error:...

Ok, upon further investigation is appears that Eigen has the following lines of code everywhere: ```cpp typedef std::complex ComplexScalar; ``` I think we could probably convince them to do: ```cpp...

> Well, I was going to say we should file a bug report again Eigen... and tried but gitlab tied me up in circles and I gave up :( Done:...

A quick `git grep` indicates that we only have standard Laguerre and associated Laguerre. I do like the idea of adding them; though I might wish the user to form...

@jzmaddock : I've had this code for polynomial rootfinding lying around: ```cpp #if __has_include() template [[nodiscard]] std::vector roots() const { if (c_.size() == 1) { return std::vector(); } // There...

> Since you got this to work previously for 128 bit float do you see anything obvious I am missing? IIRC, this was something I looked up in the [quadmath...

@mborland : Does it need a `fast_float_distance` header? Could we not just improve the performance of the current implementation? @jzmaddock : This is looking like it's about ready to go;...