@KamilaSzewczyk I agree APL is really fun to golf in, I don't really get the appeal of doing something like osabie where it's more a 'pick the right builtins for the job'
it's so easy to make something peak slow in APL, just use the stencil operator in a convoluted way so the interpreter doesn't recognize the special combination
I wondered if you can parallelise MT to work on multiple seeds on once in APL but it seems far too iterative to be doable in some sort of array oriented way
I mean all I would see in here is that once you compare just omega with floor omega and the while later you compare 2 = +/ thing with floor u which is being replaced by omega÷omega
or am I too exhausted after physics and do I see angle speed in here
@Konrad'Unrooted'Klawikowski OK, lets make a thought experiment. It isn't so, but let's imagine that $was a primitive monadic function that checked if the argument is integer. How might you use that in { 2 = +/ u = ⌊ u ← ⍵÷⍳⍵ } to simplify the code?
@KamilaSzewczyk Towards the end of the bounty week.
@ngn on modern hardware more people care about ease of use than speed unless what you're doing absolutely needs speed but that's probably a very rare scenario
@rak1507 performance doesn't have to be at the cost of ease of use. i find the opposite is true - simpler code/interpreters/editors/tools.. are faster.
@Konrad'Unrooted'Klawikowski Think of the flow of data from the right. ⍵×⍳⍵ returns a result which we call u and can be be fed to $ which returns a result which is fed to +/ etc.
@Konrad'Unrooted'Klawikowski Hold on, never mind an implementation. If you are (in your head) to determine if a number is an integer, how many numbers do you need from me?
problems.tryapl.org @Konrad'Unrooted'Klawikowski if you ever want some fairly simple problems to try have a look at these, there are sample solutions available as well if you get really stuck (or people can help here)
