Hello,
What you see is probably a bug.
I have rerun the test case in $AMBERHOME/test/sebomd/AM1/water1 and
indeed, if you turn 'method = 0' to 'method = 1', sander hangs forever.
That said, please be careful about two things:
1) The divide and conquer method is designed to be used on "large"
system. Here your system is very small (just 1 water molecule in gas phase)
2) as said in the Amber documentation, 'method = 2' (and 'method = 3')
are to be preferred over 'method = 1'. In my experience, they scale
better and converge also better. If you use 'method = 2', sander does
not hang anymore :-)
Sincerely,
Gerald.
On 04/06/2018 01:47 PM, Mingyuan wrote:
> Dear Amber members:
>
>
>
> I have meet several questions about SEBOMD with Divide and Conqure SCF
> method, i want to test the efficiency of this method, so i just simply
> prepare a water in gas phase, the mdout file is stopped at this place:
>
>
>
> 4. RESULTS
>
> ----------------------------------------------------------------------------
> ----
>
>
>
> | # of SOLUTE degrees of freedom (RNDFP): 9.
>
> | # of SOLVENT degrees of freedom (RNDFS): 0.
>
> | NDFMIN = 9. NUM_NOSHAKE = 0 CORRECTED RNDFP = 9.
>
> | TOTAL # of degrees of freedom (RNDF) = 9.
>
> ---------------------------------------------------
>
>
>
> eedmeth=4: Setting switch to one everywhere
>
>
>
> ---------------------------------------------------
>
> | Local SIZE OF NONBOND LIST = 0
>
> | TOTAL SIZE OF NONBOND LIST = 0
>
>
>
> The sander program still work on the background and there are not any other
> output files. I am puzzled with this strange question. My calculations could
> finished using SEBOMD with option 'method 0' quickly but failed with option
> ¡®method 1¡¯. And my md.in file is shown as follows:
>
>
>
> heat ras-raf
>
> &cntrl
>
> imin=0,irest=0,ntx=1,
>
> nstlim=1,dt=0.000001,
>
> ntc=1,ntf=1,
>
> cut=999, ntb=0,
>
> ntpr=1, ntwx=1,
>
> ntt=3, gamma_ln=2.0,
>
> tempi=300.0, temp0=300.0,
>
> ifqnt=1
>
> /
>
> &qmmm
>
> qm_theory='SEBOMD'
>
> /
>
> &sebomd
>
> hamiltonian='AM1'
>
> charge=0
>
> method=1
>
> dbuff1=6.0
>
> dbuff2=0.0
>
> ncore=1
>
> /
>
>
>
> Are there any mistakes in my md.in file?
>
> Could you help me? It's really my honor to receive any letter from
> you£¬thank you ¡£
>
> Best wishes for you!
>
>
> Mingyuan
>
>
> 2017.12.18
>
>
>
>
>
>
>
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
____________________________________________________________________________
Prof. Gerald MONARD
Directeur du m¨¦socentre EXPLOR
Universit¨¦ de Lorraine
Boulevard des Aiguillettes B.P. 70239
F-54506 Vandoeuvre-les-Nancy, FRANCE
e-mail : Gerald.Monard.univ-lorraine.fr
phone : +33 (0)372.745.279
mobile : +33 (0)678.006.443
web : http://www.monard.info
____________________________________________________________________________
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Apr 06 2018 - 08:30:04 PDT
