Re: Phix Random Numbers

new topic     » goto parent     » topic index » view thread      » older message » newer message
RobertS said...

The Phix documentation says about rnd(), "You should note that the implementation is trivial and you should not assume that the results will be perfectly distributed or cryptographically secure. Several other random number generators can be found in PCAN (or the Euphoria archive)."

Probably my fault, but I didn't manage to find them. It's not hugely important to me, but if someone could help me to improve on the trivially implemented random numbers with without too much effort, I'd be grateful.

(Not meant for cryptographical purposes, rather for Monte Carlo experiments. If the existing implementation is good enough for that anyway, I'd be glad to hear that.)

Thanks,
Robert

Sorry, not been payin' attention. Yeah, the Phix ones are perfectly fine for Monte Carlo experiments, that's just a standard disclaimer against doin' stuff with them that could lead to financial loss.

Not that you should need them, or have personally tested any of them, I found these:
http://phix.x10.mx/pmwiki/pmwiki.php?n=Main.AssemblerVersionOfMersenneTwisterRng
http://phix.x10.mx/pmwiki/pmwiki.php?n=Main.DcScalableRandomNumberGenerator
http://phix.x10.mx/pmwiki/pmwiki.php?n=Main.MersenneTwister
http://phix.x10.mx/pmwiki/pmwiki.php?n=Main.RandomNumbers
http://phix.x10.mx/pmwiki/pmwiki.php?n=Main.Randomv100

I've updated that remark in the docs to ..."trivial and while fine for most amateur/experimental/personal use, you should "..., thanks.

new topic     » goto parent     » topic index » view thread      » older message » newer message

Search



Quick Links

User menu

Not signed in.

Misc Menu