The ABINIT Users Mailing List ( CLOSED )

["Eric J. Walter" <ewalter@particle.physics.wm.edu>]

Dear Matthieu, 


Thanks for the reply.

These results are posted on http://www.cpd.wm.edu/ejw_abinit/

In t1/, I have posted the results with the d/dk step at gamma with the patched
and unpatched versions.

In t2/, I have posted the same without the d/dk step.  

On the alpha platform, the unpatched anaddb code without the d/dk step at 
gamma
crashes.  But, on our itanium cluster it does not.  It simply runs to 
the end without imposing the ASR (ASR is set to 2).

I have updated the weblink to include the itanium results (labeled with 
"_intel").
Here is a diff of the anaddb out files with and without the patch as well
as no d/dk step:

------------------------------------------------------------------------
<5 piezo1->t2>diff -ub anaddb.out_orig_intel anaddb.out_patched_intel 
--- anaddb.out_orig_intel       Thu Jun 23 11:15:58 2005
+++ anaddb.out_patched_intel    Thu Jun 23 11:15:58 2005
@@ -34,7 +34,7 @@
 
  read the DDB information and perform some checks
 
--begin at tcpu      0.013  and twall      0.013 sec
+-begin at tcpu      0.014  and twall      0.051 sec
 
  Real(R)+Recip(G) space primitive vectors, cartesian coordinates 
(Bohr,Bohr^-1):
  R(1)= -2.8629351  2.8629351  2.8629351  G(1)=  0.0000000  0.1746459  
0.1746459
@@ -43,12 +43,14 @@
  Unit cell volume ucvol=  9.3863013E+01 bohr^3
  Angles (23,13,12)=  1.09471221E+02  1.09471221E+02  1.09471221E+02 degrees
  Now the whole DDB is in central memory
+COMMENT:  rfelfd set to 0 since due to metal occupations
+COMMENT:  this change made by EJW, NOT PART OF THE STANDARD ABINIT RELEASE!!!
 
 
================================================================================
 
  Dielectric Tensor and Effective Charges
 
--begin at tcpu      0.021  and twall      0.021 sec
+-begin at tcpu      0.021  and twall      0.081 sec
 
  Metallic case : effective charges are set to 0
 
@@ -56,19 +58,19 @@
 
  Treat the first list of vectors
 
--begin at tcpu      0.021  and twall      0.021 sec
+-begin at tcpu      0.022  and twall      0.082 sec
  
   Phonon wavevector (reduced coordinates) :  0.00000  0.00000  0.00000
  Phonon energies in Hartree :
-   3.974000E-04  3.974000E-04  3.974000E-04
+   0.000000E+00  0.000000E+00  0.000000E+00
  Phonon energies in meV     :
--  1.081381E+01  1.081381E+01  1.081381E+01
+-  0.000000E+00  0.000000E+00  0.000000E+00
  Phonon frequencies in cm-1    :
--  8.721923E+01  8.721923E+01  8.721923E+01
+-  0.000000E+00  0.000000E+00  0.000000E+00
  Phonon frequencies in Thz     :
--  2.614767E+00  2.614767E+00  2.614767E+00
+-  0.000000E+00  0.000000E+00  0.000000E+00
  Phonon energies in Kelvin  :
--  1.254889E+02  1.254889E+02  1.254889E+02
+-  0.000000E+00  0.000000E+00  0.000000E+00
 
  Analysis of degeneracies and characters (maximum tolerance=1.00E-06 a.u.)
  Symmetry characters of vibration mode #   1
@@ -79,18 +81,18 @@
  
   Phonon wavevector (reduced coordinates) :  0.05000  0.05000  0.00000
  Phonon energies in Hartree :
-   7.044141E-04  8.084519E-04  3.449701E-03
+   5.816119E-04  7.040367E-04  3.426734E-03
  Phonon energies in meV     :
--  1.916808E+01  2.199909E+01  9.387113E+01
+-  1.582647E+01  1.915781E+01  9.324618E+01
  Phonon frequencies in cm-1    :
--  1.546010E+02  1.774347E+02  7.571218E+02
+-  1.276491E+02  1.545182E+02  7.520812E+02
  Phonon frequencies in Thz     :
--  4.634822E+00  5.319358E+00  2.269794E+01
+-  3.826822E+00  4.632339E+00  2.254683E+01
  Phonon energies in Kelvin  :
--  2.224361E+02  2.552886E+02  1.089328E+03
+-  1.836583E+02  2.223169E+02  1.082076E+03
 
 
================================================================================
 
-+Total cpu time      0.023  and wall time      0.024 sec
++Total cpu time      0.023  and wall time      0.108 sec
 
  anaddb : the run completed succesfully.

------------------------------------------------------------------------
------------------------------------------------------------------------

Sorry I missed the 4.5.3 deadline. 

Let me know if there are other questions,

Thanks,

Eric

On Thu, Jun 23, 2005 at 04:56:59PM +0200, verstraete@pcpm.ucl.ac.be wrote:
> 
> Excellent point, and (probably) a good patch too. I think it is possible 
> to impose the ASR, though. I've always done it and the acoustic 
> frequencies are definitely often nonzero if you don't. Could you show us 
> the frequencies at Gamma and 0.05 0.05 0.0 without your patch and without 
> the d/dk step at gamma (still with asr 2)? Or does that crash? From your 
> mail I deduce anaddb finishes cleanly but did not impose the ASR - right?
> 
> just a week too late for 4.5.3
> 
> Matthieu
> 
> -- 
> ================================================================
> Matthieu Verstraete             mailto:verstraete@pcpm.ucl.ac.be
> PCPM, Boltzmann, pl. Croix du Sud, 1        tel: +32 10 47 33 59
> B-1348 Louvain-la-Neuve  Belgium            fax: +32 10 47 34 52