Christian, > i got two different results with step-13 if i switch debug mode on/ > off. My deal version is 6.2.1.
As others have already mentioned, this is something that is hard to avoid given that floating point arithmetic is not exact and in particular not commutative -- if a compiler reorders some additions or multiplications you may get different results on the order of round-off. I don't see this effect running here on my laptop. Can you be more specific as to what differences you see? It would also be interesting to know where exactly the two computations start to diverge, i.e. which operation/statement/function call it is that causes the problem? Best Wolfgang ------------------------------------------------------------------------- Wolfgang Bangerth email: [email protected] www: http://www.math.tamu.edu/~bangerth/ _______________________________________________ dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
