List Commands Library for 50g

[DavidM]

(03-18-2018 06:05 PM)pier4r Wrote:  

Quote:I keep feeling like there must be a way to test the efficacy of the shuffle compared to the "<num> RAND * IP 1 +" alternative without porting everything to a different platform and investing huge amounts of time and effort to fine-tune the tests.

Here I cannot follow. Why should you test the things on another platform? For the efficacy, you need to find (or develop) a reasonable metric for the context.

...because the established test suites seem to require that the PRNGs exist as functions on the same platform that the suite itself is on. I've seen multiple comments about how they specifically exclude testing datasets produced by a PRNG as they aren't sufficient to determine effectiveness. So they apparently require the PRNG to exist as a function within the suite's environment so that both the input and output can be controlled and tested.

All of this amounts to waaaay more effort (and headache) than was ever intended for a simple command to help shuffle a list. If there's truly no other way to reasonably validate it, then I'll simply remove it from the library. I have no desire to misrepresent the effectiveness of the command by leaving it in place with uncertain results, nor do I intend to become an expert in PRNGs simply to include a command in a library. It's beginning to look like that expertise is required to create a reasonable shuffling routine, though.