You should test convergence of your results by using larger grids. This has been discussed on the forum many times. If your system (e.g. energy) has converged with a 1x1x1 sampling, then you should get the same results with a 1x1x4. You will need to experiment, until you get the level of accuracy with respect to computation time you require.
Chris On 13 April 2011 00:34, deepak srinivasan <[email protected]> wrote: > > Hi > Thanks for pointing out my mistake. I forgot to add Lattice Vectors to my > input file. > I have another doubt > I am trying to simulate a molecule(AGNR), and somewhere i read that k-grid > of 1 x 1 x 1 will suffice for molecules. > However i am trying to use 1 x 1 x 4 grid. Will it change the result by > much as it tends to take a lot of time when compared to 1 x 1 x 1 grid. > Which one would be the bset to use > > Also i am using a shifted molecule (it has no zero coordinates) and forcing > unshift into it by putting lambda points as 0.0 in k grid Monkhorst Pack. > Will it cause problems > > Sorry if my questions sound dumb > > > > -- > regards > deepak > 3rd Year Under Graduate > Electronics Design and Manufacturing > IIIT (D&M),Kancheepuram > IIT Madras Campus,Chennai >
