Can there possibly be a 'true' random number generator?

Question

Someone once said "Anyone who uses arithmetic methods to produce random numbers is in a state of sin."
Most random number generators use a measure of a randomly varying physical property (such as the internal 'tick time' of a computer) to 'seed' an arithmetic random number generator.
The argument against this is that even physical phenomena can lose their randomness, leading to unreliable results.
What are your comments on this random number generator:
Draw a table with 10 columns and 1 row. The first 'cell' is 'cell(1)', the second 'cell(2)' etc...
Take the ticker time, say 2114457852323654475
Take the last digit (5) and place the number '0' in cell(5).
Take the ticker time again, say 21153214578965428654
Take the last digit (4) and place the number '1' in cell(4) or the cell next to it if it's been filled....etc. The10 digit number containing digits 0-9 in random order is then used as a seed in an arithmetic random number generator.
How 'random' is the result?

Answer 1

In theory there could likely be a true random number generator. However, due to chaos, even a seemingly insignificant occurence can influence the outcome of the random number generator. Add to that Huygens uncertainty principle in which any measurement of an object will affect its state, and you've doubly messed up the true randomness of the generator.

Answer 2

Depends on how you intend to use your pseudo-generator. I remember calling millions of simulations in Qbasic and the ticker time thinggie gave recurring results if I called it too many times.



http://www.mymathforum.com

Answer 3

The only true random number generator is inside your skull.