Because it wasn't converging well, if I remember correctly. Sometimes even in insulators there are band crossings during the self-consistency. A broadening smooothens the "jumps" that make self-consistency slow to converge.
Paolo On Wed, Jun 8, 2016 at 2:03 AM, Jaret Qi <[email protected]> wrote: > Thank you Paolo for replying. Why did you use smearing for a gap structure, > TiS3 is non-metal. > > > On Saturday, June 4, 2016 3:59 AM, Paolo Giannozzi <[email protected]> > wrote: > > > I have run your input with > - a smearing and a less dense k-point grid > - reduced cutoffs (30 and 300) > on a small PC with no problem, apart from the fact that the final cell > is very different in form (not in volume) from the initial one > > Paolo > > On Fri, Jun 3, 2016 at 8:20 AM, Jaret Qi <[email protected]> wrote: >> Dear QE users, >> I am relaxing monoclinic TiS3 but I got an error after 5 steps says: >> problems computing cholesky >> I included the input below, please any suggestions: >> &control >> calculation = 'vc-relax' >> restart_mode = 'from_scratch' >> pseudo_dir = './' >> outdir = './' >> >> / >> &system >> ibrav= 12, >> celldm(1) = 9.41273, >> celldm(2) = 0.681, >> celldm(3) = 1.7888, >> celldm(4) = 0, >> vdw_corr='dft-D' >> nat=8 >> ntyp=2 >> ecutwfc=60 >> ecutrho=600 >> occupations = 'fixed' >> degauss=0.00 >> / >> &electrons >> conv_thr = 1e-8, >> mixing_beta=0.1, >> electron_maxstep=200 >> / >> &IONS >> / >> &CELL >> / >> >> ATOMIC_SPECIES >> Ti 47.867 Ti.pw91-sp-van_ak.UPF >> S 32.065 S.pw91-van_ak.UPF >> >> ATOMIC_POSITIONS {crystal} >> S 0.250000 0.104252 0.199565 >> S 0.750000 0.895748 0.800435 >> S 0.750000 0.758849 0.453327 >> S 0.250000 0.241151 0.546673 >> S 0.750000 0.489814 0.800644 >> S 0.250000 0.510186 0.199356 >> Ti 0.250000 0.722972 0.641355 >> Ti 0.750000 0.277028 0.358645 >> K_POINTS {automatic} >> 7 7 7 0 0 0 >> >> >> ------------------- >> Jaret Q. >> ASU > >> >> >> >> >> >> _______________________________________________ >> Pw_forum mailing list >> [email protected] >> http://pwscf.org/mailman/listinfo/pw_forum > > > > -- > Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche, > Univ. Udine, via delle Scienze 208, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 > > > -- Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche, Univ. Udine, via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222 _______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
