From @b2gills

  say i ** 3; # -1.83697019872103e-16-1i

  say i * i * i; # -0-1i

  my Complex $acc;
  $acc *= i for 1..3;
  say $acc; # -0-1i

Currently it is handled by

  multi sub infix​:<**>(Complex​:D \a, Num(Real) \b) returns Complex​:D {
  b == 0e0 ?? Complex.new(1e0, 0e0) !! (b * a.log).exp
  }

Which I think would be necessary for negative Ints ( I don't know enough about Complex numbers )

I came across this when creating a golf entry for https://codegolf.stackexchange.com/questions/102059/complex-binary-numbers/102523#102523

  *.base(2).comb(/1+/).map(i***.chars).sum

I had to go with the following instead

  *.base(2).comb(/1+/).map({[*] i xx.chars}).sum

From @zoffixznet

On Thu, 08 Dec 2016 07​:45​:18 -0800, brad wrote​:

say i ** 3; # -1.83697019872103e-16-1i

say i * i * i; # -0-1i

my Complex $acc;
$acc *= i for 1..3;
say $acc; # -0-1i

Currently it is handled by

multi sub infix​:<**>(Complex​:D \a, Num(Real) \b) returns Complex​:D {
b == 0e0 ?? Complex.new(1e0, 0e0) !! (b * a.log).exp
}

Which I think would be necessary for negative Ints ( I don't know
enough about Complex numbers )

I came across this when creating a golf entry for
https://codegolf.stackexchange.com/questions/102059/complex-binary-
numbers/102523#​102523

*.base(2).comb(/1+/).map(i***.chars).sum

I had to go with the following instead

*.base(2).comb(/1+/).map({[*] i xx.chars}).sum

As discussed on IRC https://irclog.perlgeek.de/perl6/2016-12-08#i_13705128 the current result is accurate, it's just not perfectly precise to give the -0-i answer.

Perl 5's Math​::Complex also gives the same noisy answer.

I've tried a couple of different algos that result in -0-i, but the fastest of them made the calculation 60x slower (others going as slow as 508x slower).

So I'm closing this as a won't-fix.

The RT System itself - Status changed from 'new' to 'open'

@zoffixznet - Status changed from 'open' to 'rejected'