[onerng talk] RNG designs

[Paul Campbell]

On Sun, 02 Feb 2014 14:21:05 ianG wrote:
> On 2/02/14 12:07 PM, Paul Campbell wrote:
> > On Sun, 02 Feb 2014 21:49:51 Jim Cheetham wrote:
> >> But even if the RF stream delivers less than 1 bit of entropy per bit,
> >> that doesn't really hurt us. It is still better than zero :-)
> >> 
> >> If the diode circuit delivers a better result, is it really 'perfect'?
> > 
> > arguably yes  - the avalanches in the diode are triggered by quantum
> > effects .... it's the sampling of the resulting data that introduces the
> > slight 1/0 difference
> 
> Huh.  Sampling effects the distribution.  Nice!


no not in that way - the signal coming out of the diode is not exactly 0s and 
1s - mostly it is but part of the time  it's transitioning, so some times when 
you sample it's half way in between a 1 and a 0 and the results are sort of 
random - it's actually much more complex than this (try searching for 
"metastability" ....) - in reality we get a 1 or 0

I've tweaked the spot where we sample so the result is pretty close to 50:50 
but it will never be perfect, one could spend a lot of money making something 
that was much better but it would probably require hand tweaking on them 
production like (that costs money)

	Paul