Dear Iannuzzi<br /><br /><div>Many thanks for your detailed explanation.</div><div>I have realized how to evaluate the accuracy of SGCPMD.</div><div><br /></div><div>One more question, in your paper "<span style="font-weight: bold;">Second generation Car-Parrinello MD: application to the </span><span style="font-weight: bold; font-style: italic;">h</span><span style="font-weight: bold;">-BN/Rh(111) nanomesh</span>", </div><div>you wrote that " Configurations have been extracted every 12 fs from a trajectory of 180 fs" in the note for Fig. 3.</div><div>So there should be 15 data corresponding to single point energy calculation of 15 configurations. </div><div>Why are there so many data in Fig. 3 ?</div><div><img alt="20230113110245.png" width="529px" height="193.447px" src="cid:8d969427-73c5-455e-bd35-ff53b46ed33e" /><br /></div><div><br /></div><div><br /></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">在2023年1月12日星期四 UTC+9 18:25:42<Marcella Iannuzzi> 写道：<br/></div><blockquote class="gmail_quote" style="margin: 0 0 0 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">Dear Moon, <div><br></div><div>no, I mean that one should take the coordinates generated by SGCP and recalculate energies and forces by requesting a tight convergence of the SCF,</div><div>which means, retrieving the BO electronic structure and forces for the exact same coordinates.</div><div>These BO energy and forces can be compared to the energy and forces computed along the SGCP, which in turn should have been saved during the SGCP run.</div><div>From the average and distribution of the deviations one can evaluate the accuracy of the SGCP run. </div><div><br></div><div>Regards</div><div>Marcella<br><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">On Thursday, January 12, 2023 at 2:42:47 AM UTC+1 <a href data-email-masked rel="nofollow">yue...@gmail.com</a> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Dear Marcella<div><br></div><div>As you said, the accuracy of the SGCP description should be determined respect to the BO value at the same coordinates.</div><div>So do you mean that the BOMD should be set with ENSEMBLE LANGEVIN and converge per SCF step ? </div><div><br></div><div>Moon<br><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">在2022年11月25日星期五 UTC+9 18:15:43<Marcella Iannuzzi> 写道：<br></div><blockquote class="gmail_quote" style="margin:0 0 0 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div>Dear Dmitrii</div><div><br></div>The parameters are determined by testing. <div>One has to find the conditions for which the noise term generates the correct average temperature, as measured by the equipartition theorem.</div><div>This leads to the correct sampling. </div><div>Gamma should be small, as long as the error in the forces is small. Under these conditions the dynamical properties are correct. </div><div>The first term of comparison to determine the accuracy of the SGCP description is obtained from the deviations in energy and forces with respect to the BO values at the same coordinates.<br></div><div>The force deviation has to have a vanishing average, and also that the distribution of errors should be Gaussian.<br></div><div><br></div><div>Regards</div><div>Marcella</div><div><br></div><div class="gmail_quote"><div dir="auto" class="gmail_attr">On Friday, November 25, 2022 at 8:09:07 AM UTC+1 <a rel="nofollow">dresear...@gmail.com</a> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Dear CP2k community,<div><br></div><div>I am trying to run the 2nd order CPMD. I have ionic liquid system in NVT ensemble. </div><div>Could you please advice how to select GAMMA and NOISY_GAMMA?</div><div><br></div><div>Also in this section below.</div><div><br></div><div>&MOTION<br> &MD<br> ENSEMBLE LANGEVIN<br> ...<br> &LANGEVIN<br> GAMMA 0.001<br> &END LANGEVIN<br> ...<br> &END MD<br></div><div><br></div><div><br></div><div>I have my BOMD script below.</div><div><br></div><div>Many thanks,</div><div>Dmitrii</div><div><br></div><div><br></div><div><br></div><div>&FORCE_EVAL<br> METHOD QS<br> &DFT<br> BASIS_SET_FILE_NAME BASIS_MOLOPT<br> POTENTIAL_FILE_NAME GTH_POTENTIALS<br> CHARGE 0<br> MULTIPLICITY 1<br> &MGRID<br> CUTOFF 400<br> NGRIDS 4<br> REL_CUTOFF 40<br> &END MGRID<br> &QS<br> METHOD GAPW<br> EPS_DEFAULT 1.0E-12<br> EXTRAPOLATION ASPC <br> EXTRAPOLATION_ORDER 3<br> &END<br> &SCF<br> SCF_GUESS ATOMIC<br> EPS_SCF 1.0E-6<br> MAX_SCF 15<br> &OT<br> MINIMIZER DIIS<br> PRECONDITIONER FULL_KINETIC<br> &END OT<br> &OUTER_SCF<br> EPS_SCF 1.0E-6<br> MAX_SCF 20<br> &END<br> &PRINT<br> &RESTART<br> &EACH<br> MD 0<br> &END EACH<br> &END<br> &END <br> &END SCF<br> <br> &XC<br> &XC_FUNCTIONAL BLYP<br> &END XC_FUNCTIONAL<br> &XC_GRID<br> XC_DERIV NN10_SMOOTH<br> XC_SMOOTH_RHO NN10<br> &END XC_GRID<br> &vdW_POTENTIAL<br> DISPERSION_FUNCTIONAL PAIR_POTENTIAL<br> &PAIR_POTENTIAL<br> PARAMETER_FILE_NAME dftd3.dat<br> TYPE DFTD3<br> REFERENCE_FUNCTIONAL BLYP<br> &END PAIR_POTENTIAL<br> &END vdW_POTENTIAL<br> &END XC<br> &POISSON<br> PERIODIC xyz<br> POISSON_SOLVER PERIODIC<br> &END POISSON<br> &END DFT<br> &SUBSYS<br> &CELL<br> ABC 19.3457 19.3457 19.3457<br> PERIODIC xyz<br> &END CELL<br> &TOPOLOGY<br> COORD_FILE_NAME <a href="http://npt.xyz" rel="nofollow" target="_blank" data-saferedirecturl="https://www.google.com/url?hl=zh-CN&q=http://npt.xyz&source=gmail&ust=1673661791411000&usg=AOvVaw1hVuH4LTdv5xI8-W-fyeld">npt.xyz</a><br> COORD_FILE_FORMAT XYZ<br> &END TOPOLOGY<br> &KIND H <br> BASIS_SET TZV2P-MOLOPT-GTH<br> POTENTIAL GTH-BLYP-q1 <br> &END KIND<br> &KIND Al<br> BASIS_SET DZVP-MOLOPT-SR-GTH<br> POTENTIAL GTH-BLYP-q3<br> &END KIND<br> &KIND F<br> BASIS_SET TZVP-MOLOPT-GTH <br> POTENTIAL GTH-BLYP-q7<br> &END KIND<br> &KIND O<br> BASIS_SET TZVP-MOLOPT-GTH<br> POTENTIAL GTH-BLYP-q6<br> &END KIND<br> &KIND C<br> BASIS_SET TZVP-MOLOPT-GTH<br> POTENTIAL GTH-BLYP-q4<br> &END KIND<br> &KIND S<br> BASIS_SET TZVP-MOLOPT-GTH<br> POTENTIAL GTH-BLYP-q6<br> &END KIND<br> &KIND N<br> BASIS_SET TZVP-MOLOPT-GTH<br> POTENTIAL GTH-BLYP-q5<br> &END KIND<br> &END SUBSYS<br>&END FORCE_EVAL<br><br>&GLOBAL<br> PROJECT IL_nvt<br> RUN_TYPE MD<br> PRINT_LEVEL LOW<br> FFTW_PLAN_TYPE EXHAUSTIVE<br>&END GLOBAL<br>&MOTION<br> &MD<br> ENSEMBLE NVT<br> STEPS 2000<br> TIMESTEP 0.5<br> &THERMOSTAT<br> TYPE NOSE<br> REGION MASSIVE<br> &NOSE<br> TIMECON 10.00<br> &END NOSE<br> &END THERMOSTAT<br> TEMPERATURE 303<br> &END MD<br> &PRINT<br> &TRAJECTORY<br> &EACH<br> MD 1<br> &END EACH<br> &END TRAJECTORY<br> &VELOCITIES OFF<br> &END VELOCITIES<br> &FORCES OFF<br> &END FORCES<br> &RESTART_HISTORY<br> &EACH<br> MD 500<br> &END EACH<br> &END RESTART_HISTORY<br> &RESTART<br> BACKUP_COPIES 3<br> &EACH<br> MD 1<br> &END EACH<br> &END RESTART<br> &END PRINT<br>&END<br>&EXT_RESTART<br> RESTART_FILE_NAME nvt-1.restart<br>&END<br></div></blockquote></div></blockquote></div></blockquote></div></blockquote></div> <p></p> -- <br /> You received this message because you are subscribed to the Google Groups "cp2k" group.<br /> To unsubscribe from this group and stop receiving emails from it, send an email to <a href="mailto:cp2k+unsubscribe@googlegroups.com">cp2k+unsubscribe@googlegroups.com</a>.<br /> To view this discussion on the web visit <a href="https://groups.google.com/d/msgid/cp2k/d5ce5e41-d42a-4332-be0c-d08ac84b6809n%40googlegroups.com?utm_medium=email&utm_source=footer">https://groups.google.com/d/msgid/cp2k/d5ce5e41-d42a-4332-be0c-d08ac84b6809n%40googlegroups.com</a>.<br />