Re: [SIESTA-L] [***Posible SPAM***]Sin Asunto/Subject is blank
N.B. the siesta list moderator: Please do not post further correspondence in this thread. Fast has its own mailing list referenced below. Thank you. Hello Soumaia! As mentioned previously, no help can in general be provided about installation. Systems are much too variable. Check the documentation provided in the package and consult you local system manager. BTW, as pointed out in my earlier reply, do NOT use the siesta list for queries about fast. DO use the FAST list. You can subscribe at http://lists.gforge.inria.fr/mailman/listinfo/nossi-tddft-users Yours Ross Brown On 10/01/17 19:42, Soumaia Djaadi wrote: Hello Users I' m trying to install FAST TDDFT code but I have some problams, can you help me ? Djaadi Soumaia. . Generators The following generators are available on this platform: Ninja = Generates build.ninja files (experimental). Unix Makefiles = Generates standard UNIX makefiles. CodeBlocks - Unix Makefiles = Generates CodeBlocks project files. Eclipse CDT4 - Unix Makefiles = Generates Eclipse CDT 4.0 project files. KDevelop3 = Generates KDevelop 3 project files. KDevelop3 - Unix Makefiles = Generates KDevelop 3 project files. [ilaf@localhost build]$ -DSIESTA_XC_DIR=/ABSOLUTEPATH/home/siesta-3.2-pl-5/Obj/SiestaXC ../ bash: -DSIESTA_XC_DIR=/ABSOLUTEPATH/home/siesta-3.2-pl-5/Obj/SiestaXC: No such file or directory [ilaf@localhost build]$ -DSIESTA_XC_DIR=/ABSOLUTEPATH//Obj/SiestaXC ../ bash: -DSIESTA_XC_DIR=/ABSOLUTEPATH//Obj/SiestaXC: No such file or directory [ilaf@localhost build]$ ..
Re: [SIESTA-L] Fast TDDFT
On 05/01/17 16:11, Soumaia Djaadi wrote: Hi all I'm trying calculate the optical properties of clusters by siesta.3 code but this latter is based in DFT method. After the search, l found Fast code, this version is designed to work with data produced by the siesta dft code. But I do not know anything for him. Where possible, it is found ?and how they are installed? and how the process works with siesta3 code ?Can any one help me. With Best rogards Soumaia Djaadi. Hello Soumaia! Your query is timely, since the authors of FAST were just in process of announcing the code! FAST is a TDDFT code developed a few years ago. It provides transition energies, polarizations and oscillator strengths. It is based on products of finite atomic orbitals, which explains why it was designed to work with siesta. It is indeed fast, scaling as N^3 where N is the size of the atomic basis set. Of course, these days the most accurate spectral calculations use hybrid functionals, not yet available in siesta. The main caveat is therefore that FAST is restricted to local density approximation, since the opinion of its authors was that the generalized gradient approximation was not notably better than LDA, at least on the systems they were interested in. This said, LDA may be sufficient for your needs. Indeed, depending on the kind of insight one is seeking, pursuit of the nth decimal place can be a self-defeating exercise. Fast was funded by the French ANR, as part of the Nossi project. It is thus distinct from siesta, with its own "CecilA license", issued by the French Agence pour la protection des programmes. The terms of the license are generous, similar to the GPL license, and aimed at allowing interested parties to use the code freely and carry it forward with new developments. Fast may or may not fit your requirements. But first be aware that all the authors of the code are busy on other projects. Note in particular that they cannot offer help with installation and that, while sympathetic to help requests, they can offer support only on the basis of ''when we can spare a moment'' , /via/ the following list ( N.B. _distinct_ from the siesta list): http://lists.gforge.inria.fr/mailman/listinfo/nossi-tddft-users The most recent version of the code (2016) was matched to siesta-3.2-pl-5, the Intel 15.X compilers and Cmake 3. It is freely available at https://gforge.inria.fr/frs/?group_id=1179 Yours, Ross Brown
[SIESTA-L] Post-doc position SW France: QM/MM simulation of emission switching in fluorescent probes for neurotransmitters
Post-doctoral position in molecular simulation, south-west France QM/MM simulation of emission switching in fluorescent probes for neurotransmitters Understanding commutation in molecular systems is important in many fields, ranging from alternative computer memory technologies to molecular biology, where fluorescent probes are used widely in microscopy to elucidate the presence, motion and interactions of biomolecules. Desirable features of such probes include sensitivity to the substrate and specific recognition in presence of competing species in physiological conditions. The position offered is to further comprehension of these features using DFT, TDDFT and QM/MM simulations. The simulations will be applied to recently developed cyclotriveratrylenes for recognition of the neurotransmitter acetylcholine. These molecules behave as fluorescent molecular baskets with emission properties switched by the docking of the substrate. The postdoc position involves both development of new code and use of established codes to determine the docking of the substrate and its impact on the spectrum of the probe. A QM/MM scheme coupling the parallel MD and DFT codes dl_poly and siesta has been developed in the ANR funded HPC project Nossi. The coupling scheme is additive, with electrostatic embedding. The candidate will be responsible for bookkeeping of classical force field interactions along QM/MM frontiers through covalent bonds. The frontier region consists of design atoms with lone pairs and suitably adjusted pseudopotentials (already available) to replace bonds cut by the frontier. A new linear response TDDFT algorithm is also coupled into the QM/MM simulation. The position also involves DFT and TDDFT computations on the probe molecules, using recently developed van der Waals functionals, with standardcodes. These computations will serve as references for the QM/MM simulations. The position features regular interaction with experimenters and first-hand access to data. Candidates should be: competent in molecular dynamics, ab initio and DFT methods, including experience in computer programming. Familiarity with biomolecules would be an advantage. Location: IPREM, Université de Pau et des pays de l'Adour, with regular trips to Bordeaux (ISM, Université Bordeaux 1) for discussion with experimenters. Duration: 1 year, starting before the end of 2011. Pay: 2168€/month net Enquiries candidatures to: Ross Brown, ross.brown @ univpau.fr; Isabelle Baraille, isabelle.baraille @ univpau.fr; Isabelle Gosse, gosse @ enscbp.fr
