alt.lawyers

bealoid wrote:
> Cynic wrote in
> news:scqcg31a9bpoip4jhtk17qb1hgobgjc6l9@ :
>
> [snip]
>
>>> , that LSBs are not as random as people think
>> That is true of *compressed* data, but I can assure you that the low 4
>> bits of a noisy WAV file is about as random as you'll get.
>
> I now find that there are a gajillion different types of WAV in common use
> and many of them use compression.
>
> WAV 'data chunks' are 'left justified' and padded to multiples of 8 bits
> with 0s.
>
> The Data chunk bytes of a 16 bit PCM WAV would have a minimum value of
> 0800h and a maximum value of 7FFFh.
>
> I dunno, that looks pretty not-random to me. Especially if you're cramming
> other stuff in there that's outside those limits.

It's quite easy to take up an interest in signal processing, and amazing
how much noisy data can be created by just a few hours of processor
time. I've no few gigabytes myself of raw (recorded noisy) and processed
sound data kicking around (the latter quite unreproducible because the
code has since changed and I can't remember the exact parameters used
anyway.....). Easy enough to 'run off' a few more gig's if needed; all
available with genuine random bits, sir :-)

Just because you can't see how something can be done, or someone else
has done it provably wrongly, doesn't mean it can't be done. But yes,
you do have to be careful!!

--
Mike Scott (unet )
Harlow Essex England