AMBER Archive (2005)

Subject: AMBER: Energy output problem!

From: Pradipta Bandyopadhyay (pradipta_at_iitg.ernet.in)
Date: Tue Oct 04 2005 - 07:08:04 CDT


Hi,

   I am doing NVE ensemble simulation using AMBER 7 in a linux machine.
While doing the simulation (I am using IREST=0, so generating
velocities again), I am seeing that the POTENTIAL
ENERGY of the NSTEP=0 of the new run is NOT the same as that of the LAST
STEP of the previous run.

My input is:
----------------------------------------------------------
&cntrl
    IREST = 0,
    NTX = 1,
    NTB = 1,
    NTT = 0,
    TEMPI = 300.0,
    TEMP0 = 300.0,
    NTP = 0,
    NSTLIM= 5, DT = .001,
    NTC = 1,
    NTF = 1,
    SCEE = 1.2,
    NTPR = 1,
  &end
---------------------------------------------------
the output of the last step of the 1st run

NSTEP = 5 TIME(PS) = 1249.005 TEMP(K) = 289.57 PRESS = 0.0
  Etot = -60031.7577 EKtot = 24678.0294 EPtot = -84709.7871
  BOND = 13635.3053 ANGLE = 1582.7727 DIHED = 1966.2843
  1-4 NB = 675.5404 1-4 EEL = 8394.8117 VDWAALS = 10183.8304
  EELEC = -121148.3319 EHBOND = 0.0000 RESTRAINT = 0.0000
  Ewald error estimate: 0.2660E-04
-------------------------------------------------------------------------
the output of the 1st step of the 2nd run
-----------------------------------------------------------------------
NSTEP = 0 TIME(PS) = 1249.005 TEMP(K) = 299.27 PRESS = 0.0
  Etot = -61663.9185 EKtot = 25504.5228 EPtot = -87168.4413
  BOND = 14340.6624 ANGLE = 1585.6018 DIHED = 1966.8589
  1-4 NB = 673.7908 1-4 EEL = 8392.2467 VDWAALS = 10191.5754
  EELEC = -124319.1772 EHBOND = 0.0000 RESTRAINT = 0.0000
  Ewald error estimate: 0.3724E-04
-------------------------------------------------------------------------

I thought that NSTEP=0 means no change of coordinates, so Potential energy
should remain the same. I turned off SHAKE also. Can someone point out my
mistake?

thanks.

    Pradipta
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu