Re: Pike 8.0 beta
Marc dirix <[email protected]>
| Newsgroups | gmane.comp.lang.pike.user,gmane.comp.lang.pike.devel |
|---|---|
| Message-ID | <CAERi-x38PMe49YOrqWgQeQAw612YP_Ct3R5DWRHoyz9jUhbA_g@mail.gmail.com> |
FWIW, in Matlab "rand" always generates a pseudo-random-sequence using the
same random numbers. After a Matlab restart the output of rand is exactly
the same. I often rely on it to return the same numbers when the a script
is executed twice. (Using rng('default') the sequence can also be restarted
from the beginning during script execution).
As such I would also like for pike to have a simillar pseudo random number
generator, where random_seed() (or any other name) should always return the
same sequence at either 1. fresh start of pike 2. given a specific seed.
/M
Op ma 4 jan. 2016 om 18:28 schreef Lahtinen Kimmo (Valtori) <
[email protected]>:
>
> I think I can understand the need for your request, but I am wondering is
> the random function correct for that kind of purpose. I think the main
> function is to generate random sequence and more random is better. I think
> the random seed was invented to solve the problem of not enough randomness?
>
> Kimmo
>
> I'd say it would be prudent to switch to a predictable
> pseudo-random-sequence as soon as someone has called random_seed() with a
> non-zero parameter. If random_seed() is not being called, then it does not
> matter, and faster and more random is better.
>
>
>