Point charges are not being used in TC jobs

[ChristopherAMyers]

Problem Description

It doesn't see like the TCPB-Client is property sending the point charge information over to TeraChem correctly. I can't tell if this is a TCPB or a TC issue, but when the pointcharges option is given, the job will run and say that the point charge file has been found. However, then actual SCF energy is unchanged from a job that does not have any point charge files specified.

Demonstration

I put together a small example of two water molecules so demonstrate what I'm talking about. In the server directory, there are three (renamed) jobs. One run with TCPB that specifies the point charge file, one that does not, and another that calls TeraChem (the same version being used with TCPB) using the same files generated by the TCPB jobs.

The repo has some more details, but in short, the energies generated by these three jobs are as follows:

job_with_charges/tc.out:  FINAL ENERGY: -76.3434353532 a.u.
job_no_charges/tc.out:    FINAL ENERGY: -76.3434353532 a.u.
job_TC_file_mode/tc.out:  FINAL ENERGY: -76.4700644036 a.u.

TeraChem Information

TeraChem version: TeraChem v1.9-2022.03-dev
This is run form a Docker image:
* Compiled by coltonbh Sat Mar 26 12:11:31 PDT 2022 *
* Supported architectures: sm_37 sm_52 sm_60 sm_61 sm_70 sm_75 sm_80 sm_86 sm_87*
* Build cuda_11.4.r11.4/compiler.30188945_0
NVRM version: NVIDIA UNIX x86_64 Kernel Module 470.57.02 Tue Jul 13 16:14:05 UTC 2021
GCC version: gcc version 4.8.5 20150623 (Red Hat 4.8.5-44) (GCC)

[coltonbh]

@henryw7 can you see if the server is correctly handing these values? Sounds like TeraChem is receiving them, since the stdout claims the point charge file is found, but the calculation is not incorporating them somehow.

[tmartine68]

You don't seem to have included the charges.xyz file in the repository?

…

On Mon, Feb 26, 2024 at 11:29 AM Christopher A. Myers < ***@***.***> wrote: Problem Description It doesn't see like the TCPB-Client is property sending the point charge information over to TeraChem correctly. I can't tell if this is a TCPB or a TC issue, but when the pointcharges option is given, the job will run and say that the point charge file has been found. However, then actual SCF energy is unchanged from a job that does not have any point charge files specified. Demonstration I put together a small example <https://github.com/ChristopherAMyers/tcpb_pt_charges_bug> of two water molecules so demonstrate what I'm talking about. In the server directory, there are three (renamed) jobs. One run with TCPB that specifies the point charge file, one that does not, and another that calls TeraChem (the same version being used with TCPB) using the same files generated by the TCPB jobs. The repo has some more details, but in short, the energies generated by these three jobs are as follows: job_with_charges/tc.out: FINAL ENERGY: -76.3434353532 a.u. job_no_charges/tc.out: FINAL ENERGY: -76.3434353532 a.u. job_TC_file_mode/tc.out: FINAL ENERGY: -76.4700644036 a.u. TeraChem Information TeraChem version: TeraChem v1.9-2022.03-dev This is run form a Docker image: * Compiled by coltonbh Sat Mar 26 12:11:31 PDT 2022 * * Supported architectures: sm_37 sm_52 sm_60 sm_61 sm_70 sm_75 sm_80 sm_86 sm_87* * Build cuda_11.4.r11.4/compiler.30188945_0 NVRM version: NVIDIA UNIX x86_64 Kernel Module 470.57.02 Tue Jul 13 16:14:05 UTC 2021 GCC version: gcc version 4.8.5 20150623 (Red Hat 4.8.5-44) (GCC) — Reply to this email directly, view it on GitHub <#41>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABHZXRKX2UP232YYS355YKTYVTPBTAVCNFSM6AAAAABD2ZDQ3KVHI2DSMVQWIX3LMV43ASLTON2WKOZSGE2TIOJVGQZTGNQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- ======================================= Todd Martinez Ehrsam and Franklin Professor of Chemistry Department of Chemistry Stanford University Stanford, CA 94305-5080 ***@***.*** 650-736-8860 ======================================

[ChristopherAMyers]

woops, sorry about that. Should be fixed now. Must have removed it before I initialized the repo while I was still playing around with file names