Re: [gmx-users] pressure dependence on volume.

it so? Or manually setting the volume makes some magical differences? On Sun, Oct 19, 2014 at 10:24 AM, Johnny Lu wrote: > Thanks for answering my questions too. > > On Sun, Oct 19, 2014 at 10:18 AM, Johnny Lu > wrote: > >> The averaged volume of the 1bar pressure NPT simu

Re: [gmx-users] pressure dependence on volume.

Thanks for answering my questions too. On Sun, Oct 19, 2014 at 10:18 AM, Johnny Lu wrote: > The averaged volume of the 1bar pressure NPT simulation over 150ns was > very clearly 298.5 nm^3. > But now, when I set the volume of the NVT simulation to 298.7 nm^3 by > starting from a NP

Re: [gmx-users] pressure dependence on volume.

at 2:20 AM, Tsjerk Wassenaar wrote: > Hi Johnny, > > The density is M/V. The only parameter changing with pressure coupling is > V, so the density and volume are fully correlated. No surprise there. > > Cheers, > > Tsjerk > On Oct 19, 2014 2:27 AM, "Johnny Lu" wrot

Re: [gmx-users] pressure dependence on volume.

The frames with density closest to the average density also have volumes closest to the averaged volume. On Sat, Oct 18, 2014 at 8:18 PM, Johnny Lu wrote: > I searched the energy file made by g_energy, and found that NPT frames > with very similar densities have very similar volume. >

Re: [gmx-users] pressure dependence on volume.

I searched the energy file made by g_energy, and found that NPT frames with very similar densities have very similar volume. So, I guess using density is same as using volume in this case. On Sat, Oct 18, 2014 at 9:09 AM, Johnny Lu wrote: > Would manually setting the box size make vac

Re: [gmx-users] Conserved energy ("Conserved En.") in NVT simulation (Wade)

d ensemble. On Sat, Oct 18, 2014 at 7:02 PM, Johnny Lu wrote: > The bussi paper is at [ > marge.uochb.cas.cz/~roesel/teach/Clanky/paper5_Bussi_v_rescale.pdf > <http://marge.uochb.cas.cz/%7Eroesel/teach/Clanky/paper5_Bussi_v_rescale.pdf> > ]. > > I seem to vaguely remember som

Re: [gmx-users] Conserved energy ("Conserved En.") in NVT simulation (Wade)

ot sure now. On Sat, Oct 18, 2014 at 6:58 PM, Johnny Lu wrote: > Figure 2 of the Bussi paper has a diagram with constant tilde H for NVT > simulation with 2fs time step, and another one with rising tilde H for NVT > simulation with 40fs timestep, and say: "for tilde H in the NVT >

Re: [gmx-users] Conserved energy ("Conserved En.") in NVT simulation (Wade)

e old mail-list post at [ https://mailman-1.sys.kth.se/pipermail/gromacs.org_gmx-users/2013-August/083330.html ] On Sat, Oct 18, 2014 at 6:51 PM, Johnny Lu wrote: > Does this old maillist post say tilde H is conserved energy? [ > https://mailman-1.sys.kth.se/pipermail/gromacs.org_gmx-use

Re: [gmx-users] Conserved energy ("Conserved En.") in NVT simulation (Wade)

Does this old maillist post say tilde H is conserved energy? [ https://mailman-1.sys.kth.se/pipermail/gromacs.org_gmx-users/2013-August/083330.html ] On Sat, Oct 18, 2014 at 12:46 PM, Mark Abraham wrote: > On Sat, Oct 18, 2014 at 4:32 PM, Wade wrote: > > > Dear Mark, > > Thank your very muc

Re: [gmx-users] Use NVT to mimic NVE

I mean here seems to say conserved energy is tilde H [ https://mailman-1.sys.kth.se/pipermail/gromacs.org_gmx-users/2013-August/083330.html ] On Sat, Oct 18, 2014 at 6:44 PM, Johnny Lu wrote: > The conserved energy from g_energy, which I saw someone say is tilde H, > drop more crazily if

Re: [gmx-users] Use NVT to mimic NVE

The conserved energy from g_energy, which I saw someone say is tilde H, drop more crazily if I use 1s coupling constant. On Mon, Oct 13, 2014 at 5:52 PM, Mark Abraham wrote: > Hi, > > What is your target observable? What "got worse?" > > Mark > > On Mon, Oct

Re: [gmx-users] Conserved energy ("Conserved En.") in NVT simulation (Wade)

Does double precision cause a less drop/rise in tilde H over time, when compared to a mixed precision run under same condition ? On Sat, Oct 18, 2014 at 11:28 AM, Johnny Lu wrote: > Oh, integrating with rectangles under the curve causes error. I guess > larger time-steps can cause er

Re: [gmx-users] Conserved energy ("Conserved En.") in NVT simulation (Wade)

Oh, integrating with rectangles under the curve causes error. I guess larger time-steps can cause error in tilde H by this method of integration alone. On Sat, Oct 18, 2014 at 11:25 AM, Johnny Lu wrote: > For V-rescale thermostat, the bussi paper seems to say the tilde H is the > integ

Re: [gmx-users] Conserved energy ("Conserved En.") in NVT simulation (Wade)

For V-rescale thermostat, the bussi paper seems to say the tilde H is the integral of the auxiliary dynamics. Not sure if this is same as the conserved energy in gromacs, which is also the integral of the auxiliary dynamics. I guess numerical error causes the system to lose/gain total energy in a

[gmx-users] Check for Vacuum Bubbles & Density

Hi. The NPT simulation has 9K TIP3P water with 173 amino acid residues and uses Amber99SB-ILDN force field, and use Berendsen barostat and V-rescale thermostat. I plan to run NVT with the same pressure an temperature after this. The average density is 1015 kg/m^3 over 35ns. The density was near 1

Re: [gmx-users] pressure dependence on volume.

Would manually setting the box size make vacuum bubbles? On Sat, Oct 18, 2014 at 9:08 AM, Johnny Lu wrote: > my bad. that formula for density is wrong. But I till don't see why using > the same averaged density (instead of volume) for the NVT and NPT > simulations would give a di

Re: [gmx-users] pressure dependence on volume.

my bad. that formula for density is wrong. But I till don't see why using the same averaged density (instead of volume) for the NVT and NPT simulations would give a different result. On Sat, Oct 18, 2014 at 8:52 AM, Johnny Lu wrote: > Was the density calculated by # particles / volume

Re: [gmx-users] pressure dependence on volume.

Was the density calculated by # particles / volume ? If so, using volume (as I did), and using density would give the same wrong pressure, upon transitioning from NPT to NVT. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_Lis

Re: [gmx-users] Understanding comments in pdb2gmx output

Does the pdb file have alternative conformations for some side chains or atoms? Like a side chain may have 50% chance to be rotated left, and 50% chance to be rotated right. The crystallographer recognized these two possibilities, and give each possibility a 0.5 occupancy. It may look like LYSA

Re: [gmx-users] pressure dependence on volume.

olumes, and see if the NVT simulations continuing from those frames would give me the 1 bar pressure. On Fri, Oct 17, 2014 at 3:53 PM, Justin Lemkul wrote: > > > On 10/17/14 11:49 AM, Johnny Lu wrote: > >> At least for real water, a 0.2% error in volume can mean the press

Re: [gmx-users] pressure dependence on volume.

At least for real water, a 0.2% error in volume can mean the pressure is ~30 bar when the target pressure is 1 bar. http://docs.engineeringtoolbox.com/documents/309/water-density-temperature-pressure_2.png On Fri, Oct 17, 2014 at 10:49 AM, Johnny Lu wrote: > Hi. > > Mainly TIP3P wa

[gmx-users] pressure dependence on volume.

Hi. Mainly TIP3P water, 300K. The pressure is near 1 bar. A less than 1% error in volume would cause how much error in average pressure? >From three npt simulations of 72 ns with Berendsen barostat, I get the average volume is 298.5 nm^3. Continuing from a frame with volume very close to 298.5

Re: [gmx-users] Not achieving any speedup using Open MPI

i heard that mixing mpi and open mp together is bad idea (run more slow), unless the simulation has large number of atoms, and use many cpu. May be comparing gromacs running on a single computer, with and without openmp will show a difference in speed. On Fri, Oct 17, 2014 at 10:05 AM, Da-Wei Li

Re: [gmx-users] How to add iron parameter?

/msg24046.html On Wed, Oct 15, 2014 at 2:05 PM, Johnny Lu wrote: > Pure md doesn't deal with complex orbitals well, I guess. Hydrogen bonds > in pure md is not directional, as it should be. > > can molecular dynamics (md) simulate heavy metals, like iron, which have d > oribta

Re: [gmx-users] How to add iron parameter?

Pure md doesn't deal with complex orbitals well, I guess. Hydrogen bonds in pure md is not directional, as it should be. can molecular dynamics (md) simulate heavy metals, like iron, which have d oribtals? I heard that even calcium is a bit troublesome in md. On Wed, Oct 15, 2014 at 1:22 PM, Gua

Re: [gmx-users] Use NVT to mimic NVE

The simulation get worse in a new way. I see that thermostat scales velocity and may not fix numerical error in potential energy. On Mon, Oct 13, 2014 at 2:07 PM, Johnny Lu wrote: > I just want something that will patch up the energy lost due to numerical > error of NVE simulation

Re: [gmx-users] Use NVT to mimic NVE

e Bussi comment is merely to show that his thermostat correctly reduces > to Newton's law in the limit, not that it would be useful to run it in that > limit. > > On Mon, Oct 13, 2014 at 10:28 AM, Johnny Lu > wrote: > > > On page 014101-3, the Bussi paper (http://d

Re: [gmx-users] Use NVT to mimic NVE

On page 014101-3, the Bussi paper (http://dx.doi.org/10.1063/1.2408420) mentioned: "On the other hand, for coupling constant approaching infinity,the Hamiltonian dynamics is recovered." Does that means that for a large enough coupling constant, the velocities are nearly not rescaled, and the dynami

Re: [gmx-users] Regarding umbrella samplling

Hi. Does that mean the umbrella don't overlap enough at those z ? Can you make an overlay plot of the histograms of each umbrella? On Mon, Oct 13, 2014 at 1:54 AM, Sathish Kumar wrote: > Dear gromacs usres, > > I have run the umbrella sampling with the help of tutorial of justin. In > my system

Re: [gmx-users] Use NVT to mimic NVE

I am not quite in the business of folding/unfolding protein. On Sun, Oct 12, 2014 at 4:07 PM, Johnny Lu wrote: > Thank you for pointing me to that paper. I took a quick look of it > yesterday. > Would the results in the paper stay the same, if the side chain rotations > of

Re: [gmx-users] Use NVT to mimic NVE

es through some > sort of error accumulation is not known, nor which properties they might > affect. > > > On Sun, Oct 12, 2014 at 3:07 PM, Johnny Lu wrote: > > > oh, i still mix up things. the numerical error might act as dissipation. > > > > On Sun, Oct 12, 2014

Re: [gmx-users] Use NVT to mimic NVE

oh, i still mix up things. the numerical error might act as dissipation. On Sun, Oct 12, 2014 at 3:01 PM, Johnny Lu wrote: > Sorry that I misunderstood your post. > > Can compensating energy lost caused by numerical error with thermostat > cause error in the probability dis

Re: [gmx-users] Use NVT to mimic NVE

; Dr. Vitaly V. Chaban > > > > Виталий Витальевич ЧАБАН > > > > > > On Sun, Oct 12, 2014 at 7:20 PM, Johnny Lu > wrote: > > > Why non-equilibrium can cause lost of total energy in NVE simulation? > > > (baring the case that the non-equilibrium causes

Re: [gmx-users] Use NVT to mimic NVE

Why non-equilibrium can cause lost of total energy in NVE simulation? (baring the case that the non-equilibrium causes very large values that causes over flow / getting out of the stable range of algorithms of gromacs). A closed system should have constant energy, regardless of whether it is at eq

Re: [gmx-users] Use NVT to mimic NVE

Firstly, what are the possible errors/artifacts caused by masking "the energy leakage problem with the thermostat" in this case? Secondly, are there any other mdp setting that I can try to make the mixed precision gromacs NVE run have conserved energy over 100 ns ? (and... is it because of some e

Re: [gmx-users] Use NVT to mimic NVE

; > >> If energy transfer phenomena are in focus, NVE is an ensemble of > >> choice, no matter which relaxation constant is used for the T-coupling > >> algorithm. > >> > >> > >> Dr. Vitaly V. Chaban > >> > >> Виталий Витальевич ЧАБАН

Re: [gmx-users] Use NVT to mimic NVE

io Baptista > wrote: > > Then, run true NVE, for the reasons we already pointed. > > > > On Sat, 11 Oct 2014, Johnny Lu wrote: > > > >> For dynamics with correct rate and correct fluctuation. > >> > >> On Sat, Oct 11, 2014 at 8:09 AM, Dr. Vitaly

Re: [gmx-users] Use NVT to mimic NVE

For dynamics with correct rate and correct fluctuation. On Sat, Oct 11, 2014 at 8:09 AM, Dr. Vitaly Chaban wrote: > what is it needed for? > > > > Dr. Vitaly V. Chaban > > Виталий Витальевич ЧАБАН > > > On Fri, Oct 10, 2014 at 7:41 PM, Johnny Lu wrote: > &

Re: [gmx-users] intel compiler for AMD CPU?

gt; make the best efficiency... > > but I am not so sure for this under AMD CPU > > > On 10/10/2014 08:47 PM, Johnny Lu wrote: > >> The mailling list said that for intel xeon, gcc+fftw3 make faster gromacs >> than intel c/c++ compiler + intel math kernel librar

Re: [gmx-users] Use NVT to mimic NVE

seems to be a good idea, as the "canonical sampling through velocity rescaling" paper by G.Bussi said. arxiv.org/pdf/0803.4060 On Fri, Oct 10, 2014 at 1:41 PM, Johnny Lu wrote: > Hi. > > Is it a good idea to mimic NVE by a NVT simulation with a large > temperature coup

Re: [gmx-users] intel compiler for AMD CPU?

I heard that intel compiler intentionally makes a slower program, if it detects the cpu is not genuine intel. The mailling list said that for intel xeon, gcc+fftw3 make faster gromacs than intel c/c++ compiler + intel math kernel library. Not sure about amd. On Fri, Oct 10, 2014 at 2:04 PM, Albe

[gmx-users] Use NVT to mimic NVE

Hi. Is it a good idea to mimic NVE by a NVT simulation with a large temperature coupling time constant, to reduce the effect of the thermostat ? If I use V-rescale thermostat, what artifacts will the simulation get if I use a large coupling time (say, 500 ps) and single precision gromacs ? Thank

Re: [gmx-users] NVE Energy drop in Mixed Precision 4.6.7

It's relatively easy to feel good about solving the one problem that you > can solve. But only solve it if it's the one that matters ;-) > > Mark > > On Mon, Oct 6, 2014 at 9:03 PM, Johnny Lu wrote: > > > here is the energy from nve1.edr > > http://oi62.tiny

Re: [gmx-users] Pres-XX and Pres-YY difference

at 10:24 AM, Johnny Lu wrote: > I also ran another NPT simulation with Berendsen barostat, after those 300 > ns of NPT with Parrinello-Rahman barostat. So the two simulations in these > two emails have the same starting point. > This simulation uses 2 fs time step, while the one ab

Re: [gmx-users] Pres-XX and Pres-YY difference

in both simulations are very close to 1 bar, the target pressure, while the YY and ZZ part are not as close to 1 bar. On Fri, Oct 10, 2014 at 10:03 AM, Johnny Lu wrote: > Hi. The following seems a bit strange to me. It was extracted from the edr > file with g_energy of mixed precision g

[gmx-users] Pres-XX and Pres-YY difference

Hi. The following seems a bit strange to me. It was extracted from the edr file with g_energy of mixed precision gromacs 4.6.7. Statistics over 31921501 steps [ 0. through 31921.5000 ps ], 9 data sets All statistics are over 319216 points Energy Average Err.Est. R

Re: [gmx-users] NVE Energy drop in Mixed Precision 4.6.7

The run was on 12 Xeon CPU with a GPU. On Mon, Oct 6, 2014 at 3:03 PM, Johnny Lu wrote: > here is the energy from nve1.edr > http://oi62.tinypic.com/2pqw70l.jpg > > On Mon, Oct 6, 2014 at 3:01 PM, Johnny Lu wrote: > >> Hi. >> >> The system has 9000 water

Re: [gmx-users] NVE Energy drop in Mixed Precision 4.6.7

here is the energy from nve1.edr http://oi62.tinypic.com/2pqw70l.jpg On Mon, Oct 6, 2014 at 3:01 PM, Johnny Lu wrote: > Hi. > > The system has 9000 water and a ~160 aa amino acid and uses amber99SB-ildn. > > After extensive NPT (300ns, 2ns ts), and then NVT equilibrium (300 ns,

[gmx-users] NVE Energy drop in Mixed Precision 4.6.7

Hi. The system has 9000 water and a ~160 aa amino acid and uses amber99SB-ildn. After extensive NPT (300ns, 2ns ts), and then NVT equilibrium (300 ns, 2ns ts), the NVE simulation still has energy drop, despite of the very small time step. Is double precision gromacs the only solution? The mdp f

Re: [gmx-users] NPT Equilibrium?

dynamics is for sure not reached yet. > > Cheers, > > Tsjerk > On Oct 2, 2014 7:52 PM, "Johnny Lu" wrote: > > > Hi. > > > > The system is a ~ 165 amino acid protein with a 7 amino acid substrate, > in > > an about 9000 water rhombic dod

[gmx-users] NPT Equilibrium?

Hi. The system is a ~ 165 amino acid protein with a 7 amino acid substrate, in an about 9000 water rhombic dodecahedron water box. I used thermostat for the whole system. 1. minimization to force < 100. 2. NVT with position restrain on protein 0.3 ns 3. NPT with position restrain on protein 4. 3

Re: [gmx-users] performance 1 gpu

mance. > > You may want to try gcc 4.8 or 4.9 and FFTW 3.3.x, you will most > likely get better performance than with icc+MKL. > > > On Thu, Sep 25, 2014 at 12:50 PM, Johnny Lu > wrote: > > Hi. > > > > I wonder if gromacs 4.6.7 can run faster on xsede.or

Re: [gmx-users] Add missing residue

Sometimes the forcefield can not directly use the residue names of a pdb file. In Amber forcefield, a glycine at c-terminal should be 'CGLY' instead of just GLY. Similarly, there is 'NGLY'. And histidine of different protonation states have different names. Cysteine with and without disulfide bond

Re: [gmx-users] Using a 4.6.5 cpt file in 5.0 to extend a run?

Hi. Is it possible to compile gromacs 4.6 there? On Thu, Sep 25, 2014 at 9:59 AM, Nathan Scott wrote: > Mark > > Thank you very much for your help. The one part of your reply that confuses > me a bit is that I *did* actually pass the old cpt file to grompp to > generate a new tpr file before beg

[gmx-users] performance 1 gpu

Hi. I wonder if gromacs 4.6.7 can run faster on xsede.org because I see cpu waits for gpu in the log. There is 16 cpu (2.7 GHz), 1 phi co-processor, and 1 GPU. I compiled gromacs with gpu and without phi and with intel compiler and mkl. I didn't install for 5.0.1 because I worry this bug might

Re: [gmx-users] Commands to run simulations using multiple GPU's in version 5.0.1

what happened when you ran without gpu? I installed 5.0.1 on a single machine without gpu. It used threaded mpi and no real mpi and ran fine. On Wed, Sep 24, 2014 at 12:21 PM, Johnny Lu wrote: > well... I think i read it somewhere that threaded MPI is a dropped in > replacement for re

Re: [gmx-users] Commands to run simulations using multiple GPU's in version 5.0.1

above) of my post. > > Is there any other command similar to the one you posted above that I can > use with OpenMPI? Because it looks like threadMPI and OpenMPI are not > compatible to me. > > Thanks, > > > > > On Wed, Sep 24, 2014 at 10:50 AM, Johnny Lu >

Re: [gmx-users] Box Vectors in .gro file

how about this one? http://en.wikipedia.org/wiki/Primitive_cell On Wed, Sep 24, 2014 at 10:53 AM, Hardy, Adam wrote: > Dear all, > > I'm trying to set up a simulation which uses a parallelepiped box. The > trouble I'm having is I don't quite understand the notation used to > describe it in the .

Re: [gmx-users] Commands to run simulations using multiple GPU's in version 5.0.1

, Johnny Lu wrote: > Actually i am trying to find the answer to the same question now. > > manual 4.6.7/appendix D/mdrun > says > > -gpu_id > string > List of GPU device id-s to use, specifies the per-node PP rank to GPU > mapping > > On Tue, Sep 23, 2014 a

Re: [gmx-users] Commands to run simulations using multiple GPU's in version 5.0.1

As I have mentioned in the post, I have already tried those options but it > didn't work. Kindly please let me know if you have anymore suggestions. > > Thank you, > > On Tue, Sep 23, 2014 at 8:41 PM, Johnny Lu wrote: > > > Try -nt, -ntmpi, -ntomp, -np (one at a time)

Re: [gmx-users] 4.6.7 can't get regression test.

my bad, the link works. and wget it works, guess will try REGRESSIONTEST_PATH option instead. http://www.gromacs.org/Documentation/Installation_Instructions#Testing_GROMACS_for_correctness On Wed, Sep 24, 2014 at 8:52 AM, Johnny Lu wrote: > When I use cmake option "-DREGRESSIONTEST_

[gmx-users] 4.6.7 can't get regression test.

When I use cmake option "-DREGRESSIONTEST_DOWNLOAD", cmake gives: CMake Error at tests/CMakeLists.txt:69 (message): error: downloading 'http://gerrit.gromacs.org/download/regressiontests-4.6.7.tar.gz' failed how to fix that? -- Gromacs Users mailing list * Please search the archive at htt

Re: [gmx-users] Different energies of GPU and CPU-only runs with Gromacs 5.x

hi. can i get a link to the redmine issue? I decided to install 4.6.7 instead of 5.0.1 because of this. On Mon, Sep 22, 2014 at 5:52 AM, Mark Abraham wrote: > On Mon, Sep 22, 2014 at 11:02 AM, Machtens, Jan-Philipp < > j.macht...@fz-juelich.de> wrote: > > > Thanks for the reply! > > However, the

Re: [gmx-users] Commands to run simulations using multiple GPU's in version 5.0.1

Try -nt, -ntmpi, -ntomp, -np (one at a time) ? I forget about what I tried now But I just stop the mdrun, and then read the log file. Also can look for the mdrun page in the offical manual (pdf) and try this page: http://www.gromacs.org/Documentation/Gromacs_Utilities/mdrun?highlight=mdrun O

Re: [gmx-users] I: How to rolled up a polymer chain

Molecular dynamics software, like gromacs, use a set of rules to simulate the real world. This set of rules is a force-field. Force field is the simplified physical rules of the real world that are hopefully approximately right. Such simplified rules usually can not describe everything in the real

Re: [gmx-users] Compile and Run on Xsede

It works now. I didn't know there are "gpudev" and "gpu" queues. Have to use those to run cuda programs. There are 3 gpu per node... i loaded the correct intel version and then the one without cuda ran ok. On Tue, Sep 23, 2014 at 1:07 PM, Justin Lemkul wrote: > >

[gmx-users] Compile and Run on Xsede

Hi. On stampede of xsede.org , I compiled cmake, and then gromacs 4.6.7 with and without gpu, with the following cmake configuration: without gpu: module load mkl module load cuda/6.0 module load mvapich2 export MKLROOT=$TACC_MKL_DIR export MKL_TARGET_ARCH=

Re: [gmx-users] I: How to rolled up a polymer chain

Hi. I noticed mis-spellings in manual of another MD software that persisted throughout its versions. Googling for "polyethylene simulation" gave this article as first result: http://arxiv.org/pdf/1401.4295v1.pdf, which use lammps. I'm not sure about the availability, accuracy, and computation cost

Re: [gmx-users] Restart Arbitrary Frame & Thermostat.

I mean is the "last frame ... time", the time for the frame of the cpt file, or the frame right before the frame of the cpt file? On Fri, Sep 19, 2014 at 3:28 PM, Johnny Lu wrote: > I ran gmxcheck on an old cpt file: > > Command line: > gmxcheck -f nve1.cpt >

Re: [gmx-users] Restart Arbitrary Frame & Thermostat.

:23 PM, Johnny Lu wrote: > > > Sorry for not being clear about the question. > > > > 1. If I start with NVT with thermostat A, and then continue with NPT with > > thermostat A, and barostat B, will the internal variables of thermostat A > > stays the same, upon transi

Re: [gmx-users] Restart Arbitrary Frame & Thermostat.

simulation. Thanks. On Fri, Sep 19, 2014 at 1:04 PM, Mark Abraham wrote: > On Fri, Sep 19, 2014 at 6:39 PM, Johnny Lu wrote: > > > Hi. > > > > If I change ensemble (say, from NVT to NPT) but use the same > > thermostat/barostat, will the internal variables of the &

[gmx-users] Restart Arbitrary Frame & Thermostat.

Hi. If I change ensemble (say, from NVT to NPT) but use the same thermostat/barostat, will the internal variables of the thermostat/barostat stays the same (like how many step later to couple the system to thermostat)? If I want to continue from a particular saved frame of the simulation (saved x

Re: [gmx-users] Compatibility of Files Between Version, Precision, and Compilation

Is it possible to run equilibration with mdrun of mixed precision, then run production md with mdrun of double precision? On Tue, Sep 16, 2014 at 8:24 PM, Johnny Lu wrote: > Will gromacs 4.5 double precision (which has altivec support) run faster > on the power7 cluster than 4.6.7 or

Re: [gmx-users] Compatibility of Files Between Version, Precision, and Compilation

Tue, Sep 16, 2014 at 4:19 PM, Johnny Lu wrote: > > > that machine without sse2 is a power7 cluster. > > It has altivec 128 simd. I wonder if gromacs can use that? > > > > Not right now. Maybe in future GROMACS versions. > > > > Where can I find the offici

Re: [gmx-users] Compatibility of Files Between Version, Precision, and Compilation

ah. not any more. http://redmine.gromacs.org/issues/573 On Tue, Sep 16, 2014 at 10:19 AM, Johnny Lu wrote: > that machine without sse2 is a power7 cluster. > It has altivec 128 simd. I wonder if gromacs can use that? > Where can I find the official compilation switches for power7 (x

Re: [gmx-users] Compatibility of Files Between Version, Precision, and Compilation

that machine without sse2 is a power7 cluster. It has altivec 128 simd. I wonder if gromacs can use that? Where can I find the official compilation switches for power7 (xlc or gcc)? On Tue, Sep 16, 2014 at 10:03 AM, Mark Abraham wrote: > On Tue, Sep 16, 2014 at 3:06 PM, Johnny Lu wr

[gmx-users] Compatibility of Files Between Version, Precision, and Compilation

Hi. For cpt, gro, edr, and trr files, can I continue from one version (4.66 or 5.01) and precision (single or double), with gromacs of another version (4.6.6, or 5.0.1), precision, and on another machine? So far, it seems that tpr file compiled on a machine with sse2 and gromacs 4.6.6 double prec

[gmx-users] precision of mixed precision

Hi. Does gromacs do the following? 1. I can only write down an integer with 5 digits. 2. I want to record 43.41 accurately 3. So, I wrote 4341, and remember the actual number should be 100 times smaller. 4. By writing a number with integer precision, the double preciion number 43.41 was recorded

Re: [gmx-users] QM / MM assistance

... I just have a similar problem at this moment. It seems that CMakeFiles/CMakeError.log has some clues. Usually I save the output of cmake too. On Sun, Sep 14, 2014 at 9:03 AM, Johnny Lu wrote: > The log file may tell a bit more. > Sometimes it says a library is missing or many

Re: [gmx-users] QM / MM assistance

The log file may tell a bit more. Sometimes it says a library is missing or many other possible reasons. On Sun, Sep 14, 2014 at 6:27 AM, Eric Smoll wrote: > Hello Gromacs users, > > As far as I understand, the most up-to-date reference for compiling gromacs > for QM/MM calculations is shown her

Re: [gmx-users] Problem Compiling mdrun parallel version for GROMACS 4.5.6

By the way, have you tried gromacs 4.6.7? I think that only needs a c compiler. Or 5.0.1 ? On Fri, Sep 12, 2014 at 8:38 PM, Johnny Lu wrote: > Hi. > > google "undefined reference to `_intel_fast_memset'" > gives: > > https://software.intel.com/en-us/forums/topic

Re: [gmx-users] Problem Compiling mdrun parallel version for GROMACS 4.5.6

Hi. google "undefined reference to `_intel_fast_memset'" gives: https://software.intel.com/en-us/forums/topic/310493 are you using intel compiler with gnu linker? if you are using intel compiler, I think they have something like setcompilervar.sh (or something like that... i don't remember now)

Re: [gmx-users] g_energy_d average

I see. Thank you. For the npt equilibration, I was previously equilibrating at 300K, 1 bar, 2 ns of 1 fs timesteps, and the pressure statistics was 1.03 bar, temperature 299.9 K, and everything looked well. Then, I run a second equilibration at 300K, 1 bar, 0.5 ns of 0.5 fs timesteps, the pressur

[gmx-users] g_energy_d average

Hi. g_energy_d: Statistics over 901701 steps [ 0. through 450.8500 ps ], 1 data sets All statistics are over 90171 points Energy Average Err.Est. RMSD Tot-Drift --- Pressure

Re: [gmx-users] mpi/openmp setup to avoid #1578 on 5.0

Thanks for the quick mail. If I use 20 openmp threads and 1 MPI rank, will that bug happen in gromacs 5.0 (not 5.0.1)? On Fri, Sep 12, 2014 at 12:10 PM, Szilárd Páll wrote: > On Fri, Sep 12, 2014 at 5:36 PM, Johnny Lu wrote: > > Hi. > > I wonder if I run with following, will

[gmx-users] mpi/openmp setup to avoid #1578 on 5.0

Hi. I wonder if I run with following, will I get bug 1578 of gromacs 5.0? The number of OpenMP threads was set by environment variable OMP_NUM_THREADS to 1 Using 12 MPI processes Using 1 OpenMP thread per MPI process And, is it ok to use tpr of version 5.0.1 double precision on gromacs 5.0 double

Re: [gmx-users] talk page on gromacs wiki

7;ll get graffittied. > > Mark > > > On Fri, Sep 12, 2014 at 2:27 PM, Johnny Lu wrote: > > > Hi. Is it a good idea to open the talk page on gromacs wiki for everyone > to > > edit, just to record the rumors and common practice (Or, we can already > > edit

[gmx-users] talk page on gromacs wiki

Hi. Is it a good idea to open the talk page on gromacs wiki for everyone to edit, just to record the rumors and common practice (Or, we can already edit that wiki?) ? -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before p

Re: [gmx-users] Hyper-threading Gromacs 5.0.1

/day. I used to compile gromacs 4.6.5 single precision with intel compiler and mkl. This gromacs 5.0.1 double precision was compiled with gcc 4.4.7 because installing the intel compiler now needs root. On Thu, Sep 11, 2014 at 9:48 AM, Johnny Lu wrote: > this mail list thread talks about

Re: [gmx-users] Hyper-threading Gromacs 5.0.1

this mail list thread talks about it: https://www.mail-archive.com/gromacs.org_gmx-users@maillist.sys.kth.se/msg06331.html On Thu, Sep 11, 2014 at 9:45 AM, Johnny Lu wrote: > The gromacs wiki also says that mixing mpi and openmp is bad on small > computers. > > On Thu, Sep 11, 201

Re: [gmx-users] Hyper-threading Gromacs 5.0.1

The gromacs wiki also says that mixing mpi and openmp is bad on small computers. On Thu, Sep 11, 2014 at 9:44 AM, Johnny Lu wrote: > Ah. Thanks a lot. > As suggested by ( > https://www.ibm.com/developerworks/community/blogs/brian/entry/linux_show_the_number_of_cpu_cores_on_your_system1

Re: [gmx-users] Hyper-threading Gromacs 5.0.1

Ah. Thanks a lot. As suggested by ( https://www.ibm.com/developerworks/community/blogs/brian/entry/linux_show_the_number_of_cpu_cores_on_your_system17?lang=en), $ cat /proc/cpuinfo | grep "physical id" | sort | uniq | wc -l 2 $ cat /proc/cpuinfo | egrep "core id|physical id" | tr -d "\n" | sed s/p

Re: [gmx-users] Hyper-threading Gromacs 5.0.1

Does writing frequently (like position, velocity, energy, log every 10 timesteps) significantly slow down the simulation? On Wed, Sep 10, 2014 at 10:31 PM, Johnny Lu wrote: > tried that, and the result was: > > Reading file npt.tpr, VERSION 5.0.1 (double precision) > Changing nstlis

Re: [gmx-users] Hyper-threading Gromacs 5.0.1

://www.gromacs.org/Documentation/Errors --- so. i guess that is not a good idea. On Wed, Sep 10, 2014 at 9:55 PM, Johnny Lu wrote: > that 8-20% performance increase was for gromacs 4.6.5 > > On Wed, Sep 10, 2014 at 9:52 PM, Johnny Lu wrote

Re: [gmx-users] Hyper-threading Gromacs 5.0.1

that 8-20% performance increase was for gromacs 4.6.5 On Wed, Sep 10, 2014 at 9:52 PM, Johnny Lu wrote: > Is it a good idea to use 48 OpenMP thread, under 1 MPI thread on 24 Xeon > Processors? > > The mail list say such practice give about 8-20% performance increase > > Shoul

[gmx-users] Hyper-threading Gromacs 5.0.1

Is it a good idea to use 48 OpenMP thread, under 1 MPI thread on 24 Xeon Processors? The mail list say such practice give about 8-20% performance increase Should I try g_tune_pme when I searched for "imbalance" in the log file and found nothing (24 OMP thread under 1 MPI thread on 24 Xeon Process

Re: [gmx-users] Constraint and Temperature Drop in NVE

Somehow I thought about mice, maze, cheese, and food poisoning. What are the advantages of using multiple thermostat? On Wed, Sep 10, 2014 at 9:35 PM, Justin Lemkul wrote: > > > On 9/10/14 4:29 PM, Mark Abraham wrote: > >> Hi, >> >> On Wed, Sep 10, 2014 a

Re: [gmx-users] Constraint and Temperature Drop in NVE

get that now. It is because the method they talk about for defining the temperature involves polynomial fitting of position, and computers will take a long time to do that. On Wed, Sep 10, 2014 at 1:05 PM, Johnny Lu wrote: > Reading that paper again, I saw it says that deviation from

Re: [gmx-users] Constraint and Temperature Drop in NVE

Reading that paper again, I saw it says that deviation from the equipartition (but not the generalized-equipartition) signifies truncation error. On Wed, Sep 10, 2014 at 11:40 AM, Johnny Lu wrote: > sorry, i mean the lysozyme tutorial that i found there... > > On Wed, Sep 10, 2014 at

Re: [gmx-users] Constraint and Temperature Drop in NVE

entum (as in > GROMACS) is intrinsically flawed - see > > https://www.deshawresearch.com/publications/Equipartition%20and%20the%20Calculation%20of%20Temperature%20in%20Biomolecular%20Simulations.pdf > > Mark > > > On Wed, Sep 3, 2014 at 6:21 PM, Johnny Lu wrote: > > >

<    1   2   3   >