in reply to New software TRNG

For any random number generator it is important to know its statistical properties. Can you please elaborate a bit for your generator? For example, does it generate uniformly distributed numbers over the given range? What is the expected value (or average over many trials)? Are consecutively drawn numbers independent/uncorrelated?

Replies are listed 'Best First'.
Re^2: New software TRNG
by tim.qfs (Novice) on Jan 29, 2014 at 16:13 UTC
    It works by counting the number of for-loop cycles per eight microseconds several times and dividing the counts into two groups. A bit (0 or 1) is then determined by which group wins. A draw means "try again". This process is repeated for each bit.

    I have tested it and it does produce results which are fairly uniform and because of the way it operates, the results are theoretically independent of each other.
[reply]

      Is the exclusive use in the function of variables that are non-local-ized globals an essential source of the entropy it supplies? If so, please be aware that entropy may "leak" from the function and make the operation of any program employing it very random indeed. If not, please consider using lexical (my) variables.

[reply]

      I wanted to give it a try but as I run all my scripts under strict and warnings it did not compile.

[reply]
[d/l]
[select]
        I have just seen your reply. To be honest, there are some large gaps in my knowledge of Perl and I did not know the importance of strict and warnings. I am now working on my script to address this and I will repost it when I have done so. I hope that you will give it ago again once I have done this.

        Thank you very much.
[reply]

In Section Seekers of Perl Wisdom