Bad Pref

[sir spuddly buddly]

For the past 12 hours approx nearly every WU has come up "Bad pref". Is this a new thing, testing or what? Just interested to know. :cool:

[Edit - some 20 minutes later - now I'm starting to get "Good Pref" again. ]

[graeme]

Sometimes the system gets into an odd state. I'm not sure what the problem is, but I suspect it is trapped in a very shallow minimum. In this case, the curvatures at the minimum and saddles leading out can be particularly sensitive to the finite displacements used to calculate the Hessian matrix and then the prefactor. I think the clients are often finding the wrong number of negative modes, or a prefactor which is outside of a physical range specified. This never seems to halt the calculation, but it can take a while to collect enough good processes to get out of the state.

[sir spuddly buddly]

Thanks for the answer - had to be me in an "odd state"! :crazy:

[KWSN_Dagger]

Originally posted by sir spuddly buddly
Thanks for the answer - had to be me in an "odd state"! :crazy:

Of Course dear Sir. It's in our blood. Ni!:haddock: