Re: 答复: [SIESTA-L] how to accerlarate MD calculations?

2010-09-09 Por tôpico Michael R. C. Williams
Dear Marcos, Do you have an example of an arch.make file using Intel compilers (any version) and -ipo? I haven't had much luck getting that to work, even with older versions of ifort. Specifically, if I add -ipo to FCFLAGS for ifort 10.0.x or 10.1.x, configure crashes with the following error:

Re: [SIESTA-L] Band Gap calculation in siesta

2010-09-09 Por tôpico deepak srinivasan
> > i tried changing the lattice vectors to a > 100 0 0 0100 0 0 0 100 and other higher values but still the alloc error persists however if run the simulation for low samples like 10 it runs fine and if i increase to about 700 it shows out of memory so is this too an out of memory e

[SIESTA-L] How to read charge density at a given position?

2010-09-09 Por tôpico leila
Hi, everyone! I managed to plot the charge density distribution, from rho2xsf and xcrysden. The generated file *.XSF seems to be only charge density at a rank of grids, but without the positions of these grids. How to read the charge value at a given (x, y, z). I will be really appreciate

答复: 答复: [SIESTA-L] how to accerlarate MD calculations?

2010-09-09 Por tôpico leila
Dear Marcos, Thank you very much! I will follow your advice on the compilation. PS: Yes, I am using the intel mkl, compiling with ifort. Thank you again! Best wishes! Yours sincerely, leila 发件人: Marcos Veríssimo Alves [mailto:marcos.verissimo.al...@gmail.com] 发送时间: 2010年9月8日 19:

Re: 答复: [SIESTA-L] how to accerlarate MD calculations?

2010-09-09 Por tôpico Michael R. C. Williams
Hi Marcos, Reading your suggestions for compiling SIESTA reminded me of a question I'd meant to ask at one point. In one of your messages about compiling SIESTA with ifort and MKL ( http://www.mail-archive.com/siesta-l@uam.es/msg01702.html ) you include an example arch.make file that links to a

Re: [SIESTA-L] COOP

2010-09-09 Por tôpico Nguyen Doan Sau
Hi everyone, I am a new user. I have some related questions on SIESTA executive program First, If you use COOP.write .true. the siesta program will omit the pdos block calculation. I saw it because I could not find the systemname.pdos after the convergence. Second, I think I got something wrong

[SIESTA-L] IR activity

2010-09-09 Por tôpico Michael R. C. Williams
Hi everyone, I'm interested in calculating the infrared transition intensity (and LO/TO) splitting of some phonons. The SIESTA manual says that there is a version of the Vibra utility that has these features (provided you calculate the Born effective charges along with force constants). There i

Re: [SIESTA-L] Band Gap calculation in siesta

2010-09-09 Por tôpico Henrik Löfås
Dear Deepak, For the input file you sent there is something wrong with the lattice vectors. LatticeConstant 2.46 Ang %block LatticeVectors 2.130 1.230 0.000 2.130 -1.230 0.000 0.000 0.000 0.000 %endblock LatticeVectors First the third one is zero, meaning you don't have any periodi

Re: [SIESTA-L] help to analyse the result of wavefunction...

2010-09-09 Por tôpico shizheng wen
Dear all, These days I am trying to use the Denchar program. It seems that the program can only deal with the orthogonal crystal vector. So if I want to plot the wavefuction of the non-orthogonal lattice vectors based system, how can I do this? Best regards, Shizheng Wen 2010/5/31 Jackie Wan >

Re: [SIESTA-L] COOP

2010-09-09 Por tôpico Nicolas Leconte
It is available in Siesta 3. Unless you have specific reasons, I'd advice you to upgrade your current version. Check my separate mail for the COOP utility. Nicolas 2010/9/9 Gregorio García Moreno > Sincerely, Thank for your answer. > However, I have been searching in my SIESTA/Util directory,

Re: [SIESTA-L] COOP

2010-09-09 Por tôpico Gregorio García Moreno
Sincerely, Thank for your answer. However, I have been searching in my SIESTA/Util directory, but I have not seen any program called COOP o similar. Can you send me the program so I compile it in my computer? Thanks Gregorio El 09/09/2010 12:44, Nicolas Leconte escribió: Hello Gregorio, A Mu

Re: [SIESTA-L] COOP

2010-09-09 Por tôpico Nicolas Leconte
Hello Gregorio, A Mulliken analysis is a charge integration method (not a very accurate one though, as it is basis-dependent). Don't put to much fate in its absolute values. Bader, Hirschfeld or Voronoi analyses are more accurate. Several topics on the mailing list already discuss this. A COHP/CO