Re: namd cvs compilation

From: BIN ZHANG (zhngbn_at_gmail.com)
Date: Mon Dec 01 2008 - 11:59:39 CST

Hi, Peter:
      Thanks for your reply.
       I have tried using 1-2 exclusions, though now I'm not getting
the error immediately, it still occurs at 85 steps.
:
       Warning: Bad global exclusion count, possible error!
       Warning: Increasing cutoff during minimization may avoid this.

       Another thing might deserve to mention is that, now I can start
to run MD, which was not possible before. But the MD will die with an
error:
ERROR: Atom 559 velocity is 588.344 1129.11 541.899 (limit is 800)
ERROR: Atoms moving too fast; simulation has become unstable.
ERROR: Exiting prematurely.

        I also attached my current configuration file in case you are
interested.

        BTW, I also added these 2 parameters to the rbcg-2007.par file:
       Nxg Nxx Nxg 2.988 92
       Nxg Nxg Nxg 2.988 92
       Is it safe to do this?

        Thanks a lot.
Bin


On Nov 29, 2008, at 4:52 AM, Peter Freddolino wrote:

> Hi Bin,
> you should be using 1-2 exclusions, not 1-4, for the RBCG model.
> This is likely the cause of your error, as it yields atoms which
> should have an exclusion but may be on the order of 20 A apart.
> Best,
> Peter
>
> BIN ZHANG wrote:
>> Hi,
>> Yes, I was getting that error.
>> About the "Bad global exclusion count", actually, I was
>> running a constant volume simulation. I also included my conf file
>> for the simulation:
>> Also, could you send me an example of conf file for CGMD?
>> Thanks a lot.
>> Bin
>>
>>
>> ===========================================
>> structure ../../common/cg_ionized.psf
>> coordinates ../../common/cg_ionized.pdb
>> temperature 0
>>
>> paraTypeCharmm on
>> parameters ../../common/rbcg-2007.par
>>
>> outputEnergies 100
>> outputTiming 1000
>> restartfreq 1000
>> xstFreq 1000
>> dcdFreq 1000
>> wrapAll on
>> wrapNearest on
>>
>> # 1 fs time step for computation of bonded interactions
>> timestep 15
>> # 2 fs time step for nonbonded electrostatic and short range vdW
>> nonBondedFreq 2
>> # 4 fs time step for long range interactions
>> fullElectFrequency 4
>>
>> # Switching function for decay of vdW forces (see Gumbart, Schulten
>> paper)
>> cutoff 12
>>
>> stepsPerCycle 20 ;# redo pairlists every twenty steps
>>
>> cellBasisVector1 110.00 00.00 00.00
>> cellBasisVector2 00.00 110.00 00.00
>> cellBasisVector3 00.00 00.00 100.00
>> cellOrigin 0.00 0.00 -5.00 ;# the *center* of
>> the cell
>>
>> # the z dimension is going to shrink so pad sufficiently
>> # the margin could be reduced once the cell is equilibrated
>> margin 5
>>
>> exclude scaled1-4
>> 1-4scaling 1.0
>>
>> cosangles on
>>
>> langevin on
>> langevinDamping 0.5
>> langevinTemp 300
>> langevinHydrogen no
>>
>> binaryoutput off
>> binaryrestart on
>> outputname cg_equil
>>
>> minimize 2000
>> ============================================
>>
>>
>>
>> On Nov 27, 2008, at 8:46 PM, Anirban Ghosh wrote:
>>
>>>
>>> Hello Bin,
>>>
>>> Are you getting the same error for Nxg Nxg Nxg also?
>>> If so, the I thin Anton or Peter can answer that one better,
>>> because some
>>> coefficient values are also associated with these angle parameters.
>>>
>>> For "Bad global exclusion count", you can try to run a constant
>>> volume
>>> simulation, that might help.
>>>
>>>
>>> Regards,
>>>
>>>
>>>> Hi, Anirban:
>>>> Can I do the same thing for Nxg Nxg Nxg?
>>>> Also, I kept getting the error: "FATAL ERROR: Bad global
>>>> exclusion count!" while running the CGMD using NAMD CVS version
>>>> with
>>>> "cosangles on".
>>>> Does anyone know why this is happening?
>>>> I also checked the NAMD troubleshooting page, (http://www.ks.uiuc.edu/Research/namd/wiki/index.cgi?NamdTroubleshooting
>>>> ), which said "These errors generally indicate that two atoms
>>>> that are
>>>> (ususally implicitly) excluded are on non-neighboring patches or
>>>> are
>>>> more than the cutoff distance apart"
>>>> Can anyone explain, at what situation, this will happen?
>>>>
>>>> Thanks a lot.
>>>> Bin
>>>>
>>>>
>>>> On Nov 27, 2008, at 12:27 AM, Anirban Ghosh wrote:
>>>>
>>>>>
>>>>> Hi,
>>>>>
>>>>> Add the following line to youparameter file, in the section
>>>>> "ANGLES":
>>>>>
>>>>> Nxg Nxx Nxg 2.988 92
>>>>>
>>>>> These parameters are pretty much universal for the protein
>>>>> backbone in
>>>>> the RBCG model (the angle Nxg Nxx Nxg corresponds to the
>>>>> connection
>>>>> glycine-arbitrary amino acid-glycine).
>>>>> Or you can use the older .par and .top files successfully.
>>>>>
>>>>> I was facing the same problem and Anton told me to do this. Hope
>>>>> it
>>>>> helps.
>>>>>
>>>>> Regards,
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>> Sorry about this, but I think that's the problem.
>>>>>> Now I can compile namd, but when I tried to run the CG model, one
>>>>>> peculiar
>>>>>> error occured:
>>>>>>
>>>>>> "Unable to find angle parameter for Nxg Nxx Nxg"
>>>>>>
>>>>>> Is this a reasonable error ?
>>>>>>
>>>>>> Thanks .
>>>>>> Bin
>>>>>>
>>>>>> On Wed, Nov 26, 2008 at 5:24 PM, BIN ZHANG <zhngbn_at_gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Hi, Axel:
>>>>>>>
>>>>>>> I just saw your response to the thread "NAMD crashed with
>>>>>>> TclForce
>>>>>>> turned on"
>>>>>>>
>>>>>>>> the namd cvs code defines USE_COMPAT_CONST in the headers that
>>>>>>>> include tcl.h. defining USE_NON_CONST is incompatible with that
>>>>>>>> (see tcl.h). that same should be true for tcl8.4 if you compile
>>>>>>>> the cvs. hence my suspicion that you compiled against an
>>>>>>>> inconsistent version of the tcl header. this can easily cause
>>>>>>>> segfaults regardless of the compiler.
>>>>>>>
>>>>>>> Does this mean when I try to compile cvs version of NAMD, the
>>>>>>> *.tcl
>>>>>>> file in the arch folder should no longer use "TCLFLAGS=-
>>>>>>> DNAMD_TCL
>>>>>>> -DUSE_NON_CONST", but instead should be :
>>>>>>> "TCLFLAGS=-DNAMD_TCL"?(-DUSE_NON_CONST no longer needed)
>>>>>>> Thanks a lot and Happy Thanksgiving.
>>>>>>>
>>>>>>> Bin
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Nov 26, 2008, at 4:55 PM, BIN ZHANG wrote:
>>>>>>>
>>>>>>> Hi, all:
>>>>>>>> I was trying to build NAMD CVS version(downloaded yesterday) on
>>>>>>>> a linux
>>>>>>>> cluster, with the arch file:
>>>>>>>>
>>>>>>>> ==========================================
>>>>>>>> NAMD_ARCH = Linux-amd64
>>>>>>>> CHARMARCH = mpi-linux-amd64-icc-mpich
>>>>>>>> CHARMOPTS = -thread context -memory os
>>>>>>>>
>>>>>>>> FLOATOPTS = -O3 -fno-rtti -ipo0 -xW -mtune=pentium4
>>>>>>>>
>>>>>>>> CXX = /opt/mpich/infiniband/intel/bin/mpiCC
>>>>>>>> CC = /opt/mpich/infiniband/intel/bin/mpicc
>>>>>>>> CXXOPTS = -static-intel $(FLOATOPTS)
>>>>>>>> CXXNOALIASOPTS = -fno-alias $(FLOATOPTS)
>>>>>>>> COPTS = -static-intel $(FLOATOPTS)
>>>>>>>> ==========================================
>>>>>>>>
>>>>>>>> But it failed with the error:
>>>>>>>>
>>>>>>>> /usr/include/tcl.h(256): catastrophic error: #error directive:
>>>>>>>> define at
>>>>>>>> most one of USE_NON_CONST and USE_COMPAT_CONST
>>>>>>>> # error define at most one of USE_NON_CONST and
>>>>>>>> USE_COMPAT_CONST
>>>>>>>> ^
>>>>>>>>
>>>>>>>> compilation aborted for src/mainfunc.C (code 4)
>>>>>>>> make: *** [obj/mainfunc.o] Error 4
>>>>>>>>
>>>>>>>> Does anyone know what is this error about? How to solve the
>>>>>>>> problem
>>>>>>>> then?
>>>>>>>>
>>>>>>>> Thanks a lot and Happy Thanksgiving.
>>>>>>>> Bin
>>>>>>>>
>>>>>>>> PS: the same configuration works for NAMD-2.6
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> -------------------------------------------------------------
>>>>>>>> The tree of liberty must be refreshed from time to time with
>>>>>>>> the
>>>>>>>> blood of
>>>>>>>> patriots and tyrants.
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> -------------------------------------------------------------
>>>>>>> The tree of liberty must be refreshed from time to time with the
>>>>>>> blood of
>>>>>>> patriots and tyrants.
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>> --
>>>>> Anirban Ghosh
>>>>> Grade Based Engineer
>>>>> Bioinformatics Team
>>>>> Scientific & Engineering Computing Group
>>>>> Centre for Development of Advanced Computing
>>>>> Pune, India
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> -------------------------------------------------------------
>>>> The tree of liberty must be refreshed from time to time with the
>>>> blood
>>>> of patriots and tyrants.
>>>>
>>>
>>> --
>>> Anirban Ghosh
>>> Grade Based Engineer
>>> Bioinformatics Team
>>> Scientific & Engineering Computing Group
>>> Centre for Development of Advanced Computing
>>> Pune, India
>>>
>>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> -------------------------------------------------------------
>> The tree of liberty must be refreshed from time to time with the
>> blood of patriots and tyrants.

-------------------------------------------------------------
The tree of liberty must be refreshed from time to time with the blood
of patriots and tyrants.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:21:32 CST