The moral objections to state-sponsored gambling were long ago swept away by courts and the apparently insatiable appetites of humans to wager on everything from horses and dogs to the whims of

random-number generators and bouncing pingpong balls.

Large-scale studies, which rely on high-speed

random-number generators and quicken the pace of volunteers' responses, seem to interfere with any psychokinetic effects, contends a group of psychokinesis researchers led by Dean Radin of the Institute of Noetic Sciences in Petaluma, Calif.

Using graph partitioning algorithms as an example, we demonstrate the often underestimated influence of these random-number generators on the result of the heuristic algorithms.

graph partitioning, heuristic algorithms, pseudo random-number generator

Monte Carlo applications are widely perceived as embarrassingly parallel.(1) The truth of this notion depends, to a large extent, on the quality of the parallel

random-number generators used.

These

random-number generators are the algorithms of choice for applications such as encrypting credit card numbers in Internet transactions, for which unpredictability is of paramount importance.

However, defects in underlying uniform

random-number generators may result in poor quality of the nonuniform random variate.

Faulty randomness Researchers provided new mathematical insight into why certain

random-number generators give wrong results in some computational experiments and simulations (www.sciencenews.org/ 20030927/mathtrek.asp).

We tried the other LCGs mentioned above, as well as several other recommendable

random-number generators (such as those proposed by L'Ecuyer [1999a], for example), and their displays were similar to that of Figure 4, i.e., they behaved in agreement with [H.sub.0].

However, the most widely used multiplicative, congruential

random-number generators with modulus [2.sup.31] - 1 have a cycle length of about 2.1 x [10.sup.9].

Ferrenberg, a computational physicist at the University of Georgia in Athens, and his co-workers have discovered that even "high-quality"

random-number generators, which pass a battery of randomness tests, can yield incorrect results under certain circumstances.

"We've got a new situation in which conventional

random-number generators just won't do," says George Marsaglia, a computer scientist and statistician at Florida State University in Tallahassee.