@Simd do you mean how long should it take to cast a uint64 to float64 and then multiply it by another float64? apparently some processors are optimised for float64 multiply and can do it more quickly than int64 multiply. technically simd can perform four float multiplies in a single cpu cycle under optimal conditions.
Slightly cursed language design idea: function calls looks almost the same as function definition headers except for one character
// Fibonacci
function f(n) =
n = 0 => 1
n = 1 => 1
n = n => function f(n)! + function f(n)!
// Factorial
function F(N) =
n = 0 => 1
n = n => function F(N-1)!
// Fibonacci
function f(n) =
n = 0 => 1
n = 1 => 1
n = n => function f(n - 1)! + function f(n - 2)!
// Factorial
function F(n) =
n = 0 => 1
n = n => function n * F(n-1)!
And as I've learned from NPSP, maintaining the bot takes up a lot of time, and you never know when you'll wake up and realize you've failed the community for the last six hours because your server got hit by lightning and the bot went down
Sort int64 with float64 values
Your challenge is to create the following total ordering of int64 and float64 values:
Apart from -0.0 which sorts before 0.0, two values of the same type sort according to which one is less than the other.
Where two values of different types are exactly equal, the ...