Re: Query regarding colvar distanceZ

From: Peter Mawanga (peter.mawanga.lagos_at_gmail.com)
Date: Mon Dec 26 2016 - 11:39:48 CST

I tried with 10, 15, 20 as lower boundary; but it still is same. Anyway I
had *expandboundaries* on to fix that. I presume I am missing something
small but significant.

On Tue, Dec 27, 2016 at 3:49 AM, Vermaas, Joshua <Joshua.Vermaas_at_nrel.gov>
wrote:

> Hmm… What happens if you ask for run 0? That will do an energy evaluation,
> and should, if I remember correctly, also write the colvars.traj just for
> that frame. Again, the only thing I can think of is that the distance along
> the x axis is so small that distanceZ is hitting the lower boundary
> potential and causing a mess. Others might have better ideas.
>
> Josh Vermaas
>
> Director’s Postdoctoral Fellow
> National Renewable Energy Laboratory
> joshua.vermaas_at_nrel.gov<mailto:joshua.vermaas_at_nrel.gov>
>
>
>
>
> On Dec 26, 2016, at 8:17 AM, Peter Mawanga <peter.mawanga.lagos_at_gmail.com
> <mailto:peter.mawanga.lagos_at_gmail.com>> wrote:
>
> colvars.traj output:
>
> # step polymerbeadseparation
> 0 3.00000000000000e+01
> 10000 2.91237590280440e+01
> 20000 3.00346690073403e+01
> 30000 2.82243021688464e+01
>
> I couldn't figure out why is this occurring, given both have the same
> input files and are running on NAMD 2.12.
>
> Thanks
> [https://ssl.gstatic.com/ui/v1/icons/mail/images/cleardot.gif]Peter
>
> On Tue, Dec 27, 2016 at 2:07 AM, Peter Mawanga <
> peter.mawanga.lagos_at_gmail.com<mailto:peter.mawanga.lagos_at_gmail.com>>
> wrote:
> Hello Joshua
>
> The load balancing doesn't start for distanceZ case, hence I am not
> getting any colvars.traj output either. This is the error:
>
> PRESSURE: 0 -2045.01 -9.4077 22.7297 -9.4077 -29.1219 -2.91884 22.7297
> -2.91884 1.2482
> GPRESSURE: 0 -2045.01 -9.4077 22.7297 -9.4077 -29.1219 -2.91884 22.7297
> -2.91884 1.2482
> ETITLE: TS BOND ANGLE DIHED
> IMPRP ELECT VDW BOUNDARY MISC
> KINETIC TOTAL TEMP POTENTIAL TOTAL3
> TEMPAVG PRESSURE GPRESSURE VOLUME
> PRESSAVG GPRESSAVG
>
> ENERGY: 0 7.0528 11.4618 0.0000
> 0.0000 -75.4197 -155137.7401 0.0000 0.0000
> 23072.9575 -132121.6877 309.2266 -155194.6452
> -131515.7116 309.2266 -690.9601 -690.9601
> 3468098.5652 -690.9601 -690.9601
>
> OPENING EXTENDED SYSTEM TRAJECTORY FILE
> ERROR: Atom 2684 velocity is -312318 -220584 242621 (limit is 1200, atom
> 11 of 43 on patch 345 pe 8)
> ERROR: Atoms moving too fast; simulation has become unstable (1 atoms on
> patch 345 pe 8).
> ERROR: Atom 51 velocity is 311893 220572 -242609 (limit is 1200, atom 7 of
> 43 on patch 355 pe 8)
> ERROR: Atoms moving too fast; simulation has become unstable (1 atoms on
> patch 355 pe 8).
> ERROR: Exiting prematurely; see error messages above.
>
> Whereas for the case of distance colvar:
>
> PRESSURE: 0 -12.0398 -9.41329 22.7792 -9.41329 -29.1219 -2.91884 22.7792
> -2.91884 1.2482
> GPRESSURE: 0 -12.0398 -9.41329 22.7792 -9.41329 -29.1219 -2.91884 22.7792
> -2.91884 1.2482
> ETITLE: TS BOND ANGLE DIHED
> IMPRP ELECT VDW BOUNDARY MISC
> KINETIC TOTAL TEMP POTENTIAL TOTAL3
> TEMPAVG PRESSURE GPRESSURE VOLUME
> PRESSAVG GPRESSAVG
>
> ENERGY: 0 7.0528 11.4618 0.0000
> 0.0000 -75.4197 -155137.7401 0.0000 0.0000
> 23072.9575 -132121.6877 309.2266 -155194.6452
> -132107.7606 309.2266 -13.3045 -13.3045
> 3468098.5652 -13.3045 -13.3045
>
> OPENING EXTENDED SYSTEM TRAJECTORY FILE
> LDB: ============= START OF LOAD BALANCING ============== 3.92203
> LDB: ============== END OF LOAD BALANCING =============== 3.92287
> Info: useSync: 0 useProxySync: 0
> LDB: =============== DONE WITH MIGRATION ================ 3.96309
> LDB: ============= START OF LOAD BALANCING ============== 5.12916
> LDB: Largest compute 547 load 0.001950 is 0.2% of average load 0.822949
> LDB: Average compute 0.000365 is 0.0% of average load 0.822949
> LDB: Partitioning computes with target load 0.082295
> LDB: Increased migratable compute count from 11400 to 11400
> LDB: Largest unpartitionable compute is 0.001106
> LDB: ============== END OF LOAD BALANCING =============== 5.1428
> .
> .
> .
>
> colvars.traj output:
>
> # step polymerbeadseparation
> 0 3.00000000000000e+01
> 10000 2.91237590280440e+01
> .
> .
> .
>
>
> I couldn't figure out why is this occurring, given both have the same
> input files and are running on NAMD 2.12.
>
> Thanks
> Peter
>
> On Tue, Dec 27, 2016 at 1:12 AM, Vermaas, Joshua <Joshua.Vermaas_at_nrel.gov<
> mailto:Joshua.Vermaas_at_nrel.gov>> wrote:
> Hi Peter,
>
> What are the colvars values at time zero if you do it either way? It
> should be reported in the (very short) colvars.traj file. It sounds like
> what are reasonable boundaries and centers for the distance are not
> reasonable for distanceZ, and so the simulation gets a gigantic force and
> leads to atoms moving too fast.
>
> Josh Vermaas
>
> Director’s Postdoctoral Fellow
> National Renewable Energy Laboratory
> joshua.vermaas_at_nrel.gov<mailto:joshua.vermaas_at_nrel.gov><mailto:
> joshua.vermaas_at_nrel.gov<mailto:joshua.vermaas_at_nrel.gov>>
>
>
>
>
> On Dec 26, 2016, at 5:57 AM, Peter Mawanga <peter.mawanga.lagos_at_gmail.com
> <mailto:peter.mawanga.lagos_at_gmail.com><mailto:peter.
> mawanga.lagos_at_gmail.com<mailto:peter.mawanga.lagos_at_gmail.com>>> wrote:
>
> Hello everyone
>
> I want to apply colvar distanceZ for my NAMD ABF simulation instead of
> distance. But each time the simulation keeps crashing (atoms moving too
> fast) for distanceZ but works perfectly fine with distance colvar. Please
> check the input file below:
>
>
> colvarsTrajFrequency 10000
> colvarsRestartFrequency 10000
>
> colvar {
> name polymerbeadseparation
> width 0.1
> lowerboundary 5.0
> upperboundary 30.0
> lowerwallconstant 100.0
> upperwallconstant 100.0
> expandBoundaries on
>
> //distance
>
> distance {
> group1 {
> atomnumbers {17}
> }
> group2 {
> atomnumbers {51}
> }
> oneSiteTotalForce yes
> }
>
> //distanceZ
>
> distanceZ {
> main {
> atomnumbers {17}
> }
> ref {
> atomnumbers {51}
> }
> axis (1, 0, 0)
> oneSiteTotalForce yes
> }
>
> abf {
> colvars polymerbeadseparation
> fullSamples 1000
> hideJacobian yes
> historyFreq 10000
> }
>
> Please suggest some solution.
>
> Cheers
> Peter
>
>
>
>
>

-- 
Cheers
Peter

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:20:55 CST