To anybody using GlobalStiffnessTimeStepper : The default safety coefficient on dt is 0.25, which means that the actual dt in the simulation will be 0.25*(true critical dt). 0.25 is very low. The same value in PFC for instance is 0.8 by default. We are investigating this with Janek atm, and in my last tests, I could run simulation with safety coefficient = 1 without any trouble. So try modifying this and if you don't see any problem go for less safety. The same simulation wich need 100 iterations with safety=0.25 will need only 25 iterations with safety=1. This is not a negligeable difference...
I think I will change this default safety coefficient soon, to prevent people from using such a conservative value. Bruno -- _______________ Chareyre Bruno Maitre de conference Institut National Polytechnique de Grenoble Laboratoire 3S (Soils Solids Structures) - bureau E145 BP 53 - 38041, Grenoble cedex 9 - France Tél : 33 4 56 52 86 21 Fax : 33 4 76 82 70 43 ________________ _______________________________________________ Yade-users mailing list [email protected] https://lists.berlios.de/mailman/listinfo/yade-users
