Hi,
I am doing the same run on two different parallel machines on 4 CPU's each one build on xeon processor and another on 2 dual Opteron machine. All the values in the Free energy of the ion-electron system below, are the same except
V(xc)+E(xc) = XCENC
Free energy of the ion-electron system (eV)
---------------------------------------------------
alpha Z PSCENC = 2219.11116875
Ewald energy TEWEN = -24693.41994369
-1/2 Hartree DENC = -6333.65344275
-V(xc)+E(xc) XCENC = 806.77569300
PAW double counting = 18439.13923259 -17825.34637919
entropy T*S EENTRO = 0.00000000
eigenvalues EBANDS = -609.51666535
atomic energy EATOM = 29088.32474472
---------------------------------------------------
free energy TOTEN = 1091.41440808 eV
What could be the reason for this difference and what does the term V(xc), E(xc) and XCENC mean?
V(xc)+E(xc) = XCENC not same
Moderators: Global Moderator, Moderator
-
TAT
- Newbie
- Posts: 39
- Joined: Thu Aug 02, 2007 7:28 pm
- License Nr.: 691 5-1
V(xc)+E(xc) = XCENC not same
Last edited by TAT on Wed Sep 19, 2007 4:18 am, edited 1 time in total.
-
admin
- Administrator
- Posts: 2921
- Joined: Tue Aug 03, 2004 8:18 am
- License Nr.: 458
V(xc)+E(xc) = XCENC not same
maybe the reason is the default optimization of xcgrad.o and xcspin.o in the makefiles, on some machines (compilers) the optimization level is too high and has to be decreased to -O0.
Please re-compile vasp on both machines with -O0 set explicitely for the compilation of these 2 files, and check if the results are the same then. (if they are, please increase the optimization step by step, up to the highest level which gives the same results).
Please re-compile vasp on both machines with -O0 set explicitely for the compilation of these 2 files, and check if the results are the same then. (if they are, please increase the optimization step by step, up to the highest level which gives the same results).
Last edited by admin on Tue Sep 25, 2007 10:59 am, edited 1 time in total.