On Thu, 23 Jun 2005, Denny Esterline wrote:

>> A simple way I would probably use is to use a compressor, like (g)zip,
>> to try and compress the output. The more random the output is, the
>> less likely (g)zip will be able to compress it. Apart from that, you
>> would want to make sure that a histogram of all the output values
>> looks like a reasonably flat line (i.e. all values are equally
>> represented).
>>
>> Greetings,
>> Maarten Hofman.
>
> Hmm.. I like that, quick and easy. I was thinking in binary -there
> should be about the same number of zeros and ones, and the length of
> strings of the same bit should fit a bell shaped curve. But definitly
> more research is in order.

Look into chi-squared analysis. Look for a program called ent.exe (by 
John Walker) on the Internet.

Peter
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist