Re: [gmx-users] GMX 2018 and CUDA 9.1

That is not what I meant. PME on GPU was related to my interest in the latest GMX version. The interest in the newest CUDA version is coming from another group member, so I am just making sure things are compatible. Thanks! Alex On 2/6/2018 12:17 AM, Mark Abraham wrote: Hi, Yes, but the

Re: [gmx-users] regarding "-missing" option of pdb2gmx

Hi, A more considerate approach to asking for others' time is to describe your actions and output more completely when first asking. Mark On Tue, Feb 6, 2018, 05:22 Мижээ Батсайхан wrote: > Dear Mark, > > Thank you very much for your answer. In the modelled structure,

Re: [gmx-users] GMX 2018 and CUDA 9.1

Hi, Yes, but the GPU support for PME is not related to CUDA version. Mark On Tue, Feb 6, 2018, 03:00 Alex wrote: > Hi all, > > We're about to update our software, which includes CUDA libs. Before we > do, has the latest Gromacs version been tested with CUDA 9.1? Because I

Re: [gmx-users] regarding "-missing" option of pdb2gmx

Dear Mark, Thank you very much for your answer. In the modelled structure, there is no missing atoms. Actually, I used -ignh option for pdb2gmx, and it give this error. This error releated to the hydrogen atoms in the structure. It give me following warning: WARNING: atom HE3 is missing in

Re: [gmx-users] Using xy periodic boundary and 3dc Ewald summation

I believe that the 3DC method still applies summations over Z. From the abstract of the paper: "The proposed method adds a correction term to the standard Ewald summation formula." What the 3DC method does is add a correction term so that unrealistic electrostatic interactions between periodic

[gmx-users] GMX 2018 and CUDA 9.1

Hi all, We're about to update our software, which includes CUDA libs. Before we do, has the latest Gromacs version been tested with CUDA 9.1? Because I really want that PME offloaded to GPU! :) Thank you, Alex -- Gromacs Users mailing list * Please search the archive at

Re: [gmx-users] Setting restrain strength to be changed by time

Hello Mark Thank you for answering. I will try doing multiple runs. Kazushi 2018-02-05 20:44 GMT+09:00 Mark Abraham : > Hi, > > There's no way to do that other than a shell script that modifies the > temperature along with the force constant for multiple separate

[gmx-users] Jobs and Evens in CHARMM-GUI

Dear GROMACS users, Please check and post new job openings and events in CHARMM-GUI (http://www.charmm-gui.org/jobs ), and subscribe for any future new opening and event, as well as CHARMM-GUI updates. Best, Jumin CHARMM-GUI manager

[gmx-users] Using xy periodic boundary and 3dc Ewald summation

Dear GROMACS user, Currently I am simulating a Metal Organic Frameworks membrane, in which I have applied vacuum space at z direction (3 times of the current non-vacuum system). Following the instruction on GROMACS, I have set the pbc = xy and ewald-geometry = 3dc. So here is my first

Re: [gmx-users] Windows10+Visual studio 2017+CUDA8.0+Gromacs-2016.4

Thanks Mark，       I tried to download your modified nbnxn_cuda.cu file from https://gerrit.gromacs.org/c/7538/ , but also have same error (2>E:/User-software/gromacs-2016.1/gromacs-2016.1/src\gromacs/mdlib/nbnxn_cuda/nbnxn_cuda_kernel_fermi.cuh(406): error :

Re: [gmx-users] Windows10+Visual studio 2017+CUDA8.0+Gromacs-2016.4

Thanks Mark，       I tried to download your modified nbnxn_cuda.cu file from https://gerrit.gromacs.org/c/7538/ , but also have same error (2>E:/User-software/gromacs-2016.1/gromacs-2016.1/src\gromacs/mdlib/nbnxn_cuda/nbnxn_cuda_kernel_fermi.cuh(406): error : identifier

Re: [gmx-users] Windows10+Visual studio 2017+CUDA8.0+Gromacs-2016.4

Thanks Mark，       I tried to download your modified nbnxn_cuda.cu file from https://gerrit.gromacs.org/c/7538/ , but also have same error (2>E:/User-software/gromacs-2016.1/gromacs-2016.1/src\gromacs/mdlib/nbnxn_cuda/nbnxn_cuda_kernel_fermi.cuh(406): error : identifier

[gmx-users] gmx hbond

Hi I want to calculate *number of hydrogen bond* by *distance from center of bilayer*. i use gmx hbonds -dist but it doesn't give me the proper result.is it true? which options should i implement to get the best result? best regads -negar -- Gromacs Users mailing list * Please search the

Re: [gmx-users] regarding "-missing" option of pdb2gmx

Hi, You're modelling a physical system, so you have to know what you intend to model. What atoms are missing? Is that consistent? Why would you even consider running with a topology that is inconsistent with your intent? Mark On Mon, Feb 5, 2018, 07:58 Мижээ Батсайхан

Re: [gmx-users] Setting restrain strength to be changed by time

Hi, There's no way to do that other than a shell script that modifies the temperature along with the force constant for multiple separate runs. Mark On Mon, Feb 5, 2018, 11:36 Kazushi Hiranuka wrote: > Dear gromacs users > > I'm wondering if there is a way to set restrain

[gmx-users] Fwd: Umbrella sampling

Sent from my iPhone Begin forwarded message: > From: Rose > Date: February 5, 2018 at 1:16:29 GMT+3:30 > To: Gromacs > Subject: Umbrella sampling > > Thank you so much for your detailed and useful answer,like always > >

[gmx-users] Setting restrain strength to be changed by time

Dear gromacs users I'm wondering if there is a way to set restrain strength to be changed(gradually decreased) as time goes. I want to do that when performing simulated annealing where restrains are fully applied to specific atoms during temperature increase but desirable to be gradually

Re: [gmx-users] Windows10+Visual studio 2017+CUDA8.0+Gromacs-2016.4 package

Hi, I have had a quick look at the code, and don't see how your context could lead to this issue. But I have a theory that a necessary include file got changed as we updated code from the 5.1 branch to the 2016 branch. If I'm right, then if you get the code from the download link at

[gmx-users] Regarding DPCA

> Dear gromacs users > > I done simulations of peptide of 20 residues length for 300ns. As I would > like to explore the conformational flexibility I have chosen to do dPCA. I > have gone through the tutorial which is present in the gromacs cite. > Firstly, I have created the index file which

[gmx-users] Windows10+Visual studio 2017+CUDA8.0+Gromacs-2016.4 package

Hello,every one; Now, I am trying to build a windows version of gromacs with CUDA8.0. In deed, according to some information from GROMACS official website and other experience, and I have already built some version of gromacs 5.1.x and its GPU version or 2016.x with CPU. However, I meet some