[Pw_forum] FW: cell optimization results seems to be a little strange

[vega lew]

Dear Holger,

Thank your for responding. 

>   next to the suggestion of Lorenzo to use the bfgs algorithm, the point in 
> your input is the missing timestep for the cell dynamics with the 
> Wentzcovitch algorithm. (see quantum-wiki) It looks like you got a 
> non-evolving (or almost non-evolving) dynamics.

I have done the BFGS. And get the similar results. Both of Wentzcovitch and BFGS of QE
undervalue the cell parameters of anatase from my test. I'll show you my BFGS results soon.
Thank you for your point out that  timestep is missing. But I don't know how to choose a proper
value for it. I'll google it later but not now. I think BFGS maybe handle the cell parametter calculation.

> A second, more general remark on you input: Set up only the variables you 
> really need (as Axel pointet out in several mails before: every non-written 
> line is per definition a correct one).
> It doesn't make any sense to set numbers with more than 14 digits (the code 
> does not handle this) and in most cases 8 digits is more than enough, 

I see. Thank you

> Think about, why you set etot_conv_thr so tight (don't mix it up with 
> conv_thr, which seemed to be sloppy for this calculation ), while you set 
> forc_conv_thr to be really sloppy? (Btw, is there any reason to set the 
> cutoffs/thresholds to such precision?)

Many litterature doing so. The force threshold is usually set as 0.03eV and 0.05eV
for larger system. the cutoff in literature can be 380 or 400eV sometimes.
I'm only doing repeat work and learning how to do a perfect calculation.
Also the conv_thr and etot_conv_thr is taken from CASTEP calculation directly. Others
doing so of the same system is the only reason. I'm a beginner of PWscf and DFT
calculation. So I think doing repeat work of others and discuss with skilled people may be the 
best way for me. What I'm doing now is something more like a homework rather than
the research work. I'm learning to using PWscf. Thank you for your help.


Vega Lew
PH.D Candidate in Chemical Engineering

State Key Laboratory of Materials-oriented Chemical Engineering
College of Chemistry and Chemical Engineering
Nanjing University of Technology, 210009, Nanjing, Jiangsu, China



_________________________________________________________________
Connect to the next generation of MSN Messenger 
http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20080704/fc8dd4a9/attachment.html>