Re: [SIESTA-L] bad pseudos...

Dear Pablo Why you don't get these from http://www.icmab.es/siesta/Pseudos-Bases? Of course you should login for download. In addition, the name of the pseudopotential file must be the same as your element and with capital letter. It means that for Carbon your file must be named as C.psf not c.ps

[SIESTA-L] what's the Qsol et.c

Dear All: There are four quantities (qc0, qcn0，Qsol,Qtot) about electric charge in transiesta calculation(shown as following). >From SIESTA: Efermi= -4.805 qc0 : 0.000 qcn0 : 456.635 nou,NGL,NGR:3731 832 832 TranSiesta: Qsol,Qtot: 483.480 957.0001.000 transies

Re: [SIESTA-L] bad pseudos...

Thank you very much Pablo! As for the Ni issue, I used this functional previously (not SIESTA but a different code), so I'm more or less confident that it should work for systems I want to study. But thanks for pointing out, this is very useful info. 2010/5/21 Pablo A. Denis > sorry Vassili bu

Re: [SIESTA-L] DRSLL

Dear Pablo, Do you also have an O.psf for DRSLL. Many Thanks John Dr. John S. Tse Professor Canada Research Chair (Tier I) Department of Physics and Engineering Physics University of Saskatchewan 116 Science Place Saskatoon, SK S7N 5E2 Canada (Tel) +1 306 966 6410 (Fax) +1 306 966 6400 Quot

[SIESTA-L] DRSLL

Dear colleagues, Would anyone mind sharing any details of their experience with the new functional type? How does it compare with PBE in terms of calculation speed, what the possible caveats are, etc. Also, I'd be extremely grateful if anyone could share their pseudos for C and H (and maybe some o

Re: [SIESTA-L] SieSTMulator

A similar tool is developed at the Alain Rochefort group in Montreal http://nanostructures.phys.polymtl.ca/Home.html I don't think they'll give you the code but they could also agree to do simulations as coauthors.

Re: [SIESTA-L] SieSTMulator

Guys, don't mix plotting and calculating. There was one more package for STM simulation based on SIESTA results http://www.icmm.csic.es/jcerda/ But it is not free anymore as well. Probably you could get it by contacting Jorge directly and including him as a coauthor.

RE: [SIESTA-L] siesta: WARNING: Qtot, Tr[D*S] =

To add to Marcos's suggestions, try to link your MPI compilation with the sequential version of mkl. MPI + threads can be tricky and they can certainly interfere with each other. Although I do not foresee this as the problem, it is something to try. ITHREAD=-L/opt/intel/cmkl/10.1.1.019/lib/em

Re: [SIESTA-L] SieSTMulator

Also WSXM can produce constant current STM images from LDOS files of Siesta (this is not mentioned in siesta manual). Ruslan Zhachuk

Re: [SIESTA-L] SieSTMulator

The PLSTM is simple, it just integrates LDOS between Fermi and eU. The package http://arxiv.org/pdf/cond-mat/0604540 can model tip shape, material. It also uses much more reasonable distances between tip and sample. I think the package is not public. Regards Ruslan Zhachuk 2010/5/21 Ian Shuttlew

Re: [SIESTA-L] SieSTMulator

Henrik I dont know if there is now another STM package for SIESTA, but last year I successfully used the PLSTM.F package (bundled in the 'Util' folder) to simulate STM images, obtaining: Deduction of a three-phase model for the (sqrt(3)xsqrt(3))R30°-Cu2Si/Cu(111) surface alloy I. G. Shuttleworth,

[SIESTA-L] SieSTMulator

Hi, I saw this question had been up sometime before, but I am wondering if the STM simulator described here http://arxiv.org/pdf/cond-mat/0604540 is still in development or is available somewhere? It seems to give very good results, and it would be nice if one could use it. Regards Henrik

Re: [SIESTA-L] How to see the ANI file?

Also, molden works pretty well and is quite lightweight. Funnily enough, coded in Fortran :) On Fri, May 21, 2010 at 1:12 PM, Mehmet Topsakal < topsa...@unam.bilkent.edu.tr> wrote: > use JMOL (http://jmol.sourceforge.net ) > > > > > 2010/5/21 Gregorio García Moreno > > Can anyone tell me how to

Re: [SIESTA-L] How to see the ANI file?

use JMOL (http://jmol.sourceforge.net ) 2010/5/21 Gregorio García Moreno > Can anyone tell me how to see the ANI file as a molecular graph? > > Thanks >

Re: [SIESTA-L] How to see the ANI file?

ANI is identical to xyz. Any molecular viewer that reads xyz files (xmakemol is the simplest one, gdis is what works best with SIESTA) can show those. HTH, Herbert Gregorio García Moreno wrote: Can anyone tell me how to see the ANI file as a molecular graph? Thanks -- Herbert Fruchtl Se

[SIESTA-L] How to see the ANI file?

Can anyone tell me how to see the ANI file as a molecular graph? Thanks <>

Re: [SIESTA-L] siesta: WARNING: Qtot, Tr[D*S] =

If you are really plagued that much about the error, you can try a few things. One of them is what Daniel Backlund has just suggested: lower the optimization level. Sometimes faster is not better, especially when it comes to ifort. Second, perhaps this could be a problem with precision? By defaul

Re: [SIESTA-L] Results of parrellel-complied siesta3.0b is wrong,different from results of serial version

Kemal, -O2 and -O1 are less aggressive optimizations than -O3. With -O3, the compiler will do changes to the code at compile-time which can lead to many sorts of errors - from sometimes unreliable numerical results due to changing the order of instructions, to run-time errors sometimes. This is do

Re: [SIESTA-L] siesta: WARNING: Qtot, Tr[D*S] =

Hey, I keep getting the same error as well. IFort and F90, won't matter. But with F90, things are even worse. The electron density never converges. For both cases, the serial and parallel versions differ. Is anyone aware of a reply by the development on the matter? This is not a new topic as f

Re: [SIESTA-L] Results of parrellel-complied siesta3.0b is wrong,different from results of serial version

Hello, Alexander, could you please explain why the -O2 or -O1 options? What issue does it solve? Cheers, Kemal