Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Update SAD guess files added after release of QUICK-21.03 #399

Open
agoetz opened this issue Feb 20, 2025 · 0 comments · May be fixed by #401
Open

Update SAD guess files added after release of QUICK-21.03 #399

agoetz opened this issue Feb 20, 2025 · 0 comments · May be fixed by #401
Assignees
@agoetz
Labels
bug Something isn't working
Milestone
QUICK-25.03

Comments

@agoetz
Copy link
Collaborator

agoetz commented Feb 20, 2025

The old SAD guess routines were replaced after release of QUICK-21.03. This was done because the old SAD guess is extremely inefficient. However, the new SAD guess produces wrong results. As a result the SAD guess data for all basis sets added after release of QUICK-21.03 is wrong. This affects following basis sets:

  • 6-311G2DF2PD
  • cc-pVTZ
  • aug-cc-pVDZ
  • aug-cc-pVTZ
  • def2-TZVP
  • def2-TZVPD
  • def2-TZVPP
  • def2-TZVPPD
  • aug-pc-1
  • pc-2
  • aug-pc-2

As an example, here is the SCF for CH4 with def2-TZVP (input taken from test ene_CH4_rhf_631gs and basis set changed):

                                         SCF ENERGY
| ------------------------------------------------------------------------------------------------------------------------
| NCYC       ENERGY         DELTA_E      SCF_TIME  DII_CYC   DII_TIME   O_TIME  DIAG_TIME    MAX_ERR    RMS_CHG    MAX_CHG
| ------------------------------------------------------------------------------------------------------------------------
|  1    -36.528678827      ------         0.145     1        0.00      0.14      0.00    0.2815E+01  0.6412E+00  0.1981E+02
|  2    -39.507331285  0.297865E+01       0.172     2        0.00      0.17      0.00    0.5856E+00  0.2397E-01  0.4285E+00
|  3    -40.143877810  0.636547E+00       0.172     3        0.00      0.17      0.00    0.2026E+00  0.6789E-02  0.2346E+00
|  4    -40.210150530  0.662727E-01       0.171     4        0.00      0.17      0.00    0.2991E-01  0.1677E-02  0.1729E-01
|  5    -40.211628563  0.147803E-02       0.170     5        0.00      0.17      0.00    0.5051E-02  0.5151E-03  0.1024E-01
|  6    -40.211694689  0.661258E-04       0.169     6        0.00      0.17      0.00    0.7806E-03  0.4973E-04  0.5372E-03
|  7    -40.211696408  0.171971E-05       0.167     7        0.00      0.17      0.00    0.1112E-03  0.9551E-05  0.2015E-03
|  8    -40.211696471  0.629149E-07       0.165     8        0.00      0.16      0.00    0.2513E-04  0.1696E-05  0.2118E-04
|  9    -40.211696473  0.224409E-08       0.158     9        0.00      0.16      0.00    0.2006E-05  0.2279E-06  0.2639E-05
| ------------------------------------------------------------------------------------------------------------------------
| REACH CONVERGENCE AFTER   9 CYCLES
| MAX ERROR = 0.200640E-05   RMS CHANGE = 0.227860E-06   MAX CHANGE = 0.263925E-05
| -----------------------------------------------
 ELECTRONIC ENERGY    =    -53.439102853
 CORE_CORE REPULSION  =     13.227406379
 TOTAL ENERGY         =    -40.211696473

Compare to this output where I replaced the SAD guess files for H and C obtained with QUICK-21.03 (using old SAD guess routines):

                                         SCF ENERGY
| ------------------------------------------------------------------------------------------------------------------------
| NCYC       ENERGY         DELTA_E      SCF_TIME  DII_CYC   DII_TIME   O_TIME  DIAG_TIME    MAX_ERR    RMS_CHG    MAX_CHG
| ------------------------------------------------------------------------------------------------------------------------
|  1    -40.261660961      ------         0.143     1        0.00      0.14      0.00    0.1981E+01  0.1915E-01  0.2214E+00
|  2    -40.187559072  -.741019E-01       0.174     2        0.00      0.17      0.00    0.7824E-01  0.3692E-02  0.8307E-01
|  3    -40.209080189  0.215211E-01       0.173     3        0.00      0.17      0.00    0.2579E-01  0.1089E-02  0.1101E-01
|  4    -40.211387258  0.230707E-02       0.173     4        0.00      0.17      0.00    0.9625E-02  0.4347E-03  0.4641E-02
|  5    -40.211680659  0.293401E-03       0.171     5        0.00      0.17      0.00    0.1884E-02  0.1318E-03  0.1766E-02
|  6    -40.211695947  0.152881E-04       0.169     6        0.00      0.17      0.00    0.3269E-03  0.2864E-04  0.3058E-03
|  7    -40.211696461  0.514152E-06       0.168     7        0.00      0.17      0.00    0.4432E-04  0.5245E-05  0.1261E-03
|  8    -40.211696474  0.122682E-07       0.163     8        0.00      0.16      0.00    0.4807E-05  0.6377E-06  0.7700E-05
| ------------------------------------------------------------------------------------------------------------------------
| REACH CONVERGENCE AFTER   8 CYCLES
| MAX ERROR = 0.480749E-05   RMS CHANGE = 0.637722E-06   MAX CHANGE = 0.769951E-05
| -----------------------------------------------
 ELECTRONIC ENERGY    =    -53.439102853
 CORE_CORE REPULSION  =     13.227406379
 TOTAL ENERGY         =    -40.211696474

Notice how the energy in the fist SCF iteration is much closer to the converged SCF energy and the SCF converges in less iterations.
@agoetz agoetz added the bug Something isn't working label Feb 20, 2025
@agoetz agoetz self-assigned this Feb 20, 2025
@agoetz agoetz added this to the QUICK-25.03 milestone Feb 20, 2025
This was referenced Feb 20, 2025
Open
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@agoetz agoetz

Labels
bug Something isn't working
Projects
None yet
Milestone
QUICK-25.03
Development

Successfully merging a pull request may close this issue.

Update sad guess files agoetz/QUICK
1 participant
@agoetz