AMBER Archive (2005)

Subject: AMBER: All the tests passed except prmtop and prepin...

From: mmv_at_whu.edu.cn
Date: Thu Mar 31 2005 - 03:59:26 CST

Hi all,
I installed amber8 on a k7 computer running debian based system.I run the configuration script by "-athlon ifort" option.
And MKL was installed.
When I "make test" ,something possible FAILURE message appeared.
I checked all the .dif files. Some error messages could be due to round off differences.While the others were dramatic different. For example:
vi antechamber/ash/prmtop.dif
17,19c17,19
< -1.1E+1 -1.0E+1 6.1 9.4E-1 1.8
< 1.1E+1 -1.1E+1 -1.0E+1 6.1 1.4
< 1.3 7.4E-1 4.9E-1 -2.8 2.0 

---
>  -1.1E+1 -1.0E+1  6.1  9.0E-1  1.9
>   1.1E+1 -1.1E+1 -1.0E+1  6.1  1.3
>   1.3  7.0E-1  5.0E-1 -2.7  1.9
I noticed all the failure tests associated with prmtop or prepin program except a really strange failure message:
vi $AMBERHOME/test/pheTI/out.p1.dif
190c190
<  Ewald error estimate:   0.5094E-16
---
>  Ewald error estimate:   0.0000

vi out.p1.save
......
      DV/DL, AVERAGES OVER      10 STEPS

 NSTEP =       10   TIME(PS) =       0.010  TEMP(K) =     0.00  PRESS =     0.0
 Etot   =         0.0000  EKtot   =         0.0000  EPtot      =         2.9891
 BOND   =         0.0000  ANGLE   =         0.0000  DIHED      =         0.0000
 1-4 NB =        -4.0247  1-4 EEL =         0.0000  VDWAALS    =         7.0139
 EELEC  =         0.0000  EHBOND  =         0.0000  RESTRAINT  =         0.0000
 Ewald error estimate:   0.5094E-16
......

vi out.p1
......
 ------------------------------------------------------------------------------

      DV/DL, AVERAGES OVER      10 STEPS

 NSTEP =       10   TIME(PS) =       0.010  TEMP(K) =     0.00  PRESS =     0.0
 Etot   =         0.0000  EKtot   =         0.0000  EPtot      =         2.9891
 BOND   =         0.0000  ANGLE   =         0.0000  DIHED      =         0.0000
 1-4 NB =        -4.0247  1-4 EEL =         0.0000  VDWAALS    =         7.0139
 EELEC  =         0.0000  EHBOND  =         0.0000  RESTRAINT  =         0.0000
 Ewald error estimate:   0.0000E+00
.......

I have no idea about  the abnormal value 0.0000E+00.

I have installed amber8 for 3 times,first on a p4 machine running redhat9,then on the same machine running debian-based system,and this time on a k7 computer running the same debian system.

At the first time,I met very very slightly differences when making test. At the second time,all the tests passed.And this time,here comes a lot of problem.I also noticed that opteron system didn`t work well in the benchmark test,while I had thought AMD`s processors were more powerful compared with corresponding INTEL`s. Is all of these could  be due to something called compatibility ?

Attachment is the output of "make test".
Thank you very much for help.

Best

Shuli.

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