Hi, I would still argue that double precision is important. My comment on SHAKE was based on CHARMM - in Gromacs it indeed doesn't matter for water.
Ran Justin A. Lemkul wrote: > > > Berk Hess wrote: >> Hi, >> >> Shake is not relevant for water and also a time step of 2 fs should >> be fine. >> The cut-off's are the problem. You have a buffer size of 0.1 nm, >> which is already smaller than 2 times the distance from the center of >> geometry >> of a water molecule to a hydrogen. You need some additional distance >> for water diffusion. I would use a buffer of 0.25 to 0.3 nm. >> You don't list you coulombtype setting. Use pme (or if you want >> perfect energy >> conservation: pme-switch), you can also use reaction-field-zero if >> you really don't >> want to use PME. >> Then use nstlist=-1, run a short simulation and check in at the end >> of your log file >> that the neighbor list lifetime is somewhere between 5 and 20 steps. >> >> We should have a wiki entry for such details. Maybe there is one, but >> I was too lazy >> to check or make one. > > There was a basic NVE page; I have updated it based on your notes above: > > http://www.gromacs.org/Documentation/Terminology/NVE > > -Justin > >> >> Berk >> >> > Date: Mon, 1 Mar 2010 09:16:36 +0100 >> > From: r.fried...@bioc.uzh.ch >> > To: gmx-users@gromacs.org >> > Subject: Re: [gmx-users] NVE of water >> > >> > Hi Andrea, >> > Did you use double precision? Also, I'd try a lower dt (say 1fs) and >> > shake tolerance (maybe 1E-8). >> > Good luck, >> > Ran >> > >> > Andrea wrote: >> > > Dear users, >> > > >> > > for test purposes in order to set up a bigger system, I try to >> run NVE >> > > simulations of SPC water, but the energy increases very >> rapidely. My >> > > guess is that the cutoffs I use are not good for water. I that the >> > > case ( I would be grateful for a good reference for suitable SPC >> water >> > > parameters) or do I miss something else? >> > > My parameter file for the NVE is: >> > > >> > > title = NVE >> > > cpp = /lib/cpp >> > > integrator = md >> > > dt = 0.002 ; ps ! = 2 fs >> > > nsteps = 50000 ; total 100 ps >> > > nstxout = 5000 >> > > nstvout = 5000 >> > > nstxtcout = 0 >> > > nstlog = 5000 >> > > nstenergy = 5000 >> > > nstlist = 10 >> > > ns_type = grid >> > > rlist = 1.1 >> > > >> > > unconstrained-start = yes >> > > constraints = all-bonds >> > > constraint_algorithm = shake >> > > shake_tol = 0.0001 >> > > ;VdW >> > > vdwtype = Switch >> > > rvdw = 1.0 ; rvdw+ (0.1:0.3)= rlist >> > > rvdw_switch = 0.9 >> > > gen_vel = no ; yes >> > > gen_temp = 300 >> > > gen_seed = -1 >> > > ;Temperature coupling >> > > tc_grps = system >> > > tcoupl = no ;nose-hoover >> > > tau_t = 0.1 >> > > ref_t = 300 >> > > ;Pressure coupling >> > > pcoupl = no >> > > optimize_fft = yes >> > > >> > > Any suggesions are really welcome. >> > > >> > > Thank you. >> > > >> > > Regards, >> > > Andrea Muntean >> > >> > >> > -- >> > ------------------------------------------------------ >> > Ran Friedman >> > Postdoctoral Fellow >> > Computational Structural Biology Group (A. Caflisch) >> > Department of Biochemistry >> > University of Zurich >> > Winterthurerstrasse 190 >> > CH-8057 Zurich, Switzerland >> > Tel. +41-44-6355559 >> > Email: r.fried...@bioc.uzh.ch >> > Skype: ran.friedman >> > ------------------------------------------------------ >> > >> > -- >> > gmx-users mailing list gmx-users@gromacs.org >> > http://lists.gromacs.org/mailman/listinfo/gmx-users >> > Please search the archive at http://www.gromacs.org/search before >> posting! >> > Please don't post (un)subscribe requests to the list. Use the >> > www interface or send it to gmx-users-requ...@gromacs.org. >> > Can't post? Read http://www.gromacs.org/mailing_lists/users.php >> >> ------------------------------------------------------------------------ >> New Windows 7: Find the right PC for you. Learn more. >> <http://windows.microsoft.com/shop> >> > -- gmx-users mailing list gmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users Please search the archive at http://www.gromacs.org/search before posting! Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. Can't post? Read http://www.gromacs.org/mailing_lists/users.php