> And your conclusion is? It looks to me like all those perls are pretty well behaved, including my old ones

Yes it seems so. My conclusion was in Re^2: is rand random enough to simulate dice rolls?: as haukex and other said is unwise to break a test for something related to a core function which might have problems in preistoric installations. A big warning, as above, is more than nothing in case of test failure.

L*