Accelerated MD

From: Felipe Merino (felipe.merino_at_mpi-muenster.mpg.de)
Date: Wed Nov 07 2012 - 10:08:15 CST

Dear all,

I was recently trying to run some dual boost aMD simulations and i
noticed a strange behavior. The calculated nonbonded interactions are
very different in the Energy and aMD output

for instance:

ACCELERATED MD: STEP 77048000 dV 0 dVAVG 0 BOND 1185.03 ANGLE 2836.47
DIHED 4198.44 IMPRP 0 ELECT -56107.6 VDW -1327.51 POTENTIAL -49215.2

ETITLE: TS BOND ANGLE DIHED
IMPRP ELECT VDW BOUNDARY MISC
KINETIC TOTAL TEMP POTENTIAL TOTAL3
TEMPAVG PRESSURE GPRESSURE VOLUME PRESSAVG GPRESSAVG

ENERGY: 77048000 1185.0331 2836.4667 4198.4370
0.0000 -392493.9129 45860.2205 0.0000 0.0000
60024.3445 -278389.4111 297.7610 -338413.7556
-278294.6736 297.7610 51.2260 51.2979 1009821.9108
51.2260 51.2979

So all the bonded interactions correspond to each other but neither the
non bonded nor the potential. Initially i thought that namd was ignoring
the waters but in that case the neither angle nor the bond term should
agree (or is it the settle?). In any case i am a little bit confused,
because as far as i remember from the manual it seems that the boost is
applied to the entire system.

Am i missing something in the implementation??

Regards

Felipe

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:22:43 CST