From: David Hardy (dhardy_at_ks.uiuc.edu)
Date: Fri May 21 2021 - 00:43:32 CDT
[Resending my reply to Sara from earlier today so that it is received by namd-l — had unintentionally sent from my non-list-subscribed Illinois.edu address.]
I expect that you are talking about using the GPU-resident mode of NAMD 3, with “CUDASOAintegrate on”? The constraint routines and dummy atom site needed for the TIP4P water model are not yet supported on the GPU. We do have plans to support as many features as we can for GPU-resident NAMD, including TIP4P water, but it will take time for us to do so.
I do appreciate you pointing out this issue to us -- one would hope that NAMD would immediately terminate with an error message rather than compute invalid results.
-- David J. Hardy, Ph.D. Beckman Institute University of Illinois at Urbana-Champaign 405 N. Mathews Ave., Urbana, IL 61801 dhardy_at_ks.uiuc.edu, http://www.ks.uiuc.edu/~dhardy/ > On May 19, 2021, at 11:24 PM, Sara Omar <sara.ibrahim.omar_at_gmail.com> wrote: > > Greetings, > > When I try to minimize my system composed of a protein complex solvated in OPC water (Amber parameters - and I set watermodel to tip4 and rigidatoms in the config file), I get 'Nan' values in the output when running on NAMD3 CUDA version. > I do not get this issue when running the system on NAMD2.14 and the system gets minimized successfully. > > Any suggestions? > > Kindest regards, > Sara >
This archive was generated by hypermail 2.1.6 : Fri Dec 31 2021 - 23:17:11 CST