Open delchere opened 3 months ago
flokno
commented
3 months ago Hi @delchere , did you read the TDEP paper [1]? We try to explain the difference between MD-TDEP (TDEP) and statistical-sampling-TDEP (sTDEP) there. The tutorials only cover sTDEP.
It is difficult to say a priori why your values are wrong. Things to check: lattice expansion, xc treatment, force quality.
delchere
commented
3 months ago Thank you so much. I will read that paper then. Thanks a lot.
Best Regards
On Thu, Jul 4, 2024 at 11:47 AM Florian Knoop @.***> wrote:
Hi @delchere https://github.com/delchere , did you read the TDEP paper [1]? We try to explain the difference between MD-TDEP (TDEP) and statistical-sampling-TDEP (sTDEP) there. The tutorials only cover sTDEP.
It is difficult to say a priori why your values are wrong. Things to check: lattice expansion, xc treatment, force quality.
[1] https://joss.theoj.org/papers/10.21105/joss.06150
— Reply to this email directly, view it on GitHub https://github.com/tdep-developers/tdep/issues/75#issuecomment-2209264927, or unsubscribe https://github.com/notifications/unsubscribe-auth/BDSFO3K6KMPE7V3YWRS6VDLZKVVBTAVCNFSM6AAAAABKLWYJYCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMBZGI3DIOJSG4 . You are receiving this because you were mentioned.Message ID: @.***>
-- Delchere D0N-TSA Phd Student| ARISE-Project
Alumnus| AIMS-Cameroon
University of Lome | Togo
The Next Einstein Initiative | www.aims-cameroon.org http://www.aims-cameroon.org/
Mobile: +237 656160101 | +228 92069498
P.O. Box 1515, Klikamè , Lome, Togo
delchere
commented
3 months ago I thought my response was wrong because I only use 1000 md steps. Can this be a problem?
On Thu, Jul 4, 2024 at 1:05 PM Delchere Dontsa < @.***> wrote:
Thank you so much. I will read that paper then. Thanks a lot.
Best Regards
On Thu, Jul 4, 2024 at 11:47 AM Florian Knoop @.***> wrote:
Hi @delchere https://github.com/delchere , did you read the TDEP paper [1]? We try to explain the difference between MD-TDEP (TDEP) and statistical-sampling-TDEP (sTDEP) there. The tutorials only cover sTDEP.
It is difficult to say a priori why your values are wrong. Things to check: lattice expansion, xc treatment, force quality.
[1] https://joss.theoj.org/papers/10.21105/joss.06150
— Reply to this email directly, view it on GitHub https://github.com/tdep-developers/tdep/issues/75#issuecomment-2209264927, or unsubscribe https://github.com/notifications/unsubscribe-auth/BDSFO3K6KMPE7V3YWRS6VDLZKVVBTAVCNFSM6AAAAABKLWYJYCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMBZGI3DIOJSG4 . You are receiving this because you were mentioned.Message ID: @.***>
-- Delchere D0N-TSA Phd Student| ARISE-Project
Alumnus| AIMS-Cameroon
University of Lome | Togo
The Next Einstein Initiative | www.aims-cameroon.org http://www.aims-cameroon.org/
Mobile: +237 656160101 | +228 92069498
P.O. Box 1515, Klikamè , Lome, Togo
-- Delchere D0N-TSA Phd Student| ARISE-Project
Alumnus| AIMS-Cameroon
University of Lome | Togo
The Next Einstein Initiative | www.aims-cameroon.org http://www.aims-cameroon.org/
Mobile: +237 656160101 | +228 92069498
P.O. Box 1515, Klikamè , Lome, Togo
I dont actually have any error but i just want to understand something. When i was following the minimal examples, i have to run MD simulation( i was running 10000 steps) before a parse the vasprun.xml to get the input files for tdep calculation. But when i went to the GitHub page of tdep i feel like i dont even need to run the MD steps but simply do the scf loop until my the phonon dispersion is not changing anymore. My questions are if i dont need MD simulation, how can i get the [infile.lotosplitting]? Which method is the best one? *Am i run to thing that it is 2 methods? if yes please can you tell me exactly how to proceed(like the step by step) ? i am interested of thermal conductivity. I would like to mention that i have compute the thermal conductivity of silicon but get 95w/mk. these are the steps i follow: create infile.ucposcar, infile.ssposcar, compute a 1000 steps of MD simulation and parse the vasprun.xml to get others inputs, afterward i have computed the thermal conductivity at 300k using the qpoints convergence . Why is my answer not correct? please clarify me