search

grimme-lab / xtb

Semiempirical Extended Tight-Binding Program Package
https://xtb-docs.readthedocs.io/
GNU Lesser General Public License v3.0
554 stars 138 forks source link

Exact fixing not working for gfnff geometry optimization #809

Open pwging13 opened 1 year ago

pwging13 commented 1 year ago

using exact fixing with $fix with a gfnff optimization fails with the xtb 6.6.0 binaries.

 * calculating model hessian...
Using GFN-FF Lindh-Hessian
 * fragmented diagonalization... 10 fragments
 Shifting diagonal of input Hessian by   2.006057175050955E-002
 Lowest  eigenvalues of input Hessian
   -0.000061   -0.000028   -0.000000   -0.000000   -0.000000   -0.000000
   -0.000000   -0.000000   -0.000000   -0.000000   -0.000000   -0.000000
   -0.000000   -0.000000   -0.000000   -0.000000   -0.000000   -0.000000
 Highest eigenvalues
    2.884734    2.956798    2.972934    3.108546    3.121327    3.239710

 k=         144   nvar=        3087
 ANC generation failed
================
 final structure:
TariniHardikar commented 1 month ago

Hello! I was wondering if you ever found a fix for this issue? I'm running into the same problem with xtb 6.7.0, and I'm trying to do fixing as follows: $fix elements=[N,Zn] freeze=[1, 4, ..., etc]

pwging13 commented 1 month ago

Hello! I was wondering if you ever found a fix for this issue? I'm running into the same problem with xtb 6.7.0, and I'm trying to do fixing as follows: $fix elements=[N,Zn] freeze=[1, 4, ..., etc]

Unfortunately no, I have the same issue still.

TariniHardikar commented 1 month ago

Hey @pwging13, the solution mentioned here: https://github.com/grimme-lab/xtb/issues/800 worked for me.