- Posts: 272
about verlet pairlist scheme
- diaobk
-
Topic Author
- Visitor
8 years 9 months ago #4710
by diaobk
about verlet pairlist scheme was created by diaobk
Hi,
I am trying to run CGMD on GPU computer.
While GPU acceleration is only supported with Verlet.
In martini_v2.x_example.mdp, rvdw from 1.2 to 1.1 is suggested.
It seems change is made to compromise the lack of shift potential in verlet pairlist scheme.
But when I used vdwtype=Shift
a note came out
NOTE 1 : Replacing vdwtype=Shift by the equivalent combination of vdwtype=Cut-off and vdw_modifier=Force-switch [b]Did this mean the origin form of martini is now supported in Verlet scheme?[/b] While in the article: MARTINI Coarse-Grained Models of Polyethylene and Polypropylene We tested the use of the Verlet scheme with the MARTINI CG force field, using a reduced cutoff of 1.1 nm (instead of the usual 1.2 nm). we anyway suggest some caution when looking at dynamics within the Verlet scheme Caution from the change of potential form?[file md.mdp]:
Replacing vdwtype=Shift by the equivalent combination of vdwtype=Cut-off
and vdw_modifier=Force-switch
Did this mean the origin form of martini is now supported in Verlet scheme?
While in the article: MARTINI Coarse-Grained Models of Polyethylene and Polypropylene
We tested the use of the Verlet scheme with the
MARTINI CG force field, using a reduced cutoff of 1.1 nm
(instead of the usual 1.2 nm).
we anyway suggest
some caution when looking at dynamics within the Verlet
scheme
Caution from the change of potential form?
I am trying to run CGMD on GPU computer.
While GPU acceleration is only supported with Verlet.
In martini_v2.x_example.mdp, rvdw from 1.2 to 1.1 is suggested.
It seems change is made to compromise the lack of shift potential in verlet pairlist scheme.
But when I used vdwtype=Shift
a note came out
NOTE 1 : Replacing vdwtype=Shift by the equivalent combination of vdwtype=Cut-off and vdw_modifier=Force-switch [b]Did this mean the origin form of martini is now supported in Verlet scheme?[/b] While in the article: MARTINI Coarse-Grained Models of Polyethylene and Polypropylene We tested the use of the Verlet scheme with the MARTINI CG force field, using a reduced cutoff of 1.1 nm (instead of the usual 1.2 nm). we anyway suggest some caution when looking at dynamics within the Verlet scheme Caution from the change of potential form?[file md.mdp]:
Replacing vdwtype=Shift by the equivalent combination of vdwtype=Cut-off
and vdw_modifier=Force-switch
Did this mean the origin form of martini is now supported in Verlet scheme?
While in the article: MARTINI Coarse-Grained Models of Polyethylene and Polypropylene
We tested the use of the Verlet scheme with the
MARTINI CG force field, using a reduced cutoff of 1.1 nm
(instead of the usual 1.2 nm).
we anyway suggest
some caution when looking at dynamics within the Verlet
scheme
Caution from the change of potential form?
Please Log in or Create an account to join the conversation.
- djurre
-
- Offline
- Admin
Less
More
8 years 8 months ago #4805
by djurre
Replied by djurre on topic about verlet pairlist scheme
We have recently put online a few different sets of input parameters, which make it possible to simulate Martini system with the Verlet cut-off scheme (offering a performance gain as well). We have tested these parameters on several system (paper submitted) and see that properties are conserved. But caution is always good of course...
cgmartini.nl/cgmartini/index.php/force-f...ers/input-parameters
cgmartini.nl/cgmartini/index.php/force-f...ers/input-parameters
Please Log in or Create an account to join the conversation.
Time to create page: 0.088 seconds