Search found 133 matches

by magnus
15 Apr 2019, 08:52
Forum: Running Dalton
Topic: Only PCM CH2CL2 fails
Replies: 20
Views: 1668

Re: Only PCM CH2CL2 fails

You're welcome :-) Thank you for the useful link!
by magnus
14 Apr 2019, 20:07
Forum: Running Dalton
Topic: Only PCM CH2CL2 fails
Replies: 20
Views: 1668

Re: Only PCM CH2CL2 fails

As far as I know the default alpha factor will be applied unless explicitly changed through the input, so you should not do it in advance. Regarding your new calculation. First of all, since you are only adding spheres on heavy atom, i.e. not hydrogens, I assume you want to use the united-atoms appr...
by magnus
11 Apr 2019, 08:19
Forum: Running Dalton
Topic: Only PCM CH2CL2 fails
Replies: 20
Views: 1668

Re: Only PCM CH2CL2 fails

The problem in the first case is that there are some static array allocations that are not big enough for a molecule of this size. To fix this you have to edit the DALTON/include/pcmdef.h file and increase the MXSP value. It may also require that you increase MXVER . It is perhaps better to use the ...
by magnus
03 Apr 2019, 13:15
Forum: Running Dalton
Topic: CC3/aug-cc-pVTZ linear response segfaults
Replies: 18
Views: 429

Re: CC3/aug-cc-pVTZ linear response segfaults

I see. Those are really old test cases meant to be used only for performance benchmarks.
by magnus
03 Apr 2019, 08:53
Forum: Running Dalton
Topic: CC3/aug-cc-pVTZ linear response segfaults
Replies: 18
Views: 429

Re: CC3/aug-cc-pVTZ linear response segfaults

I'm not exactly sure which new CC method you are referring to but Hans Jørgen fixed the memory issue in the latest version (Dalton2018.2). You can find the manual for the latest version here: http://daltonprogram.org/manuals/dalton2018manual.pdf

FYI I'm not a prof :-)
by magnus
29 Mar 2019, 21:28
Forum: Running Dalton
Topic: double-hybrid PBE0DH
Replies: 3
Views: 172

Re: double-hybrid PBE0DH

The fix is also available as the latest commit on the release/2018 branch and will be part of an upcoming patch release of Dalton2018.
by magnus
28 Mar 2019, 11:06
Forum: Running Dalton
Topic: EEF affects TPA but not OPA
Replies: 2
Views: 120

Re: EEF affects TPA but not OPA

The input looks correct. When using EEF with OPA, you should look for XLFDIPLN, YLFDIPLN, and ZLFDIPLN. This will give the transition dipole moment including EEF. From those you can compute the transition or oscillator strength.
by magnus
27 Mar 2019, 11:03
Forum: Running Dalton
Topic: Only PCM CH2CL2 fails
Replies: 20
Views: 1668

Re: Only PCM CH2CL2 fails

That does indeed sound very strange. Is it the same error in all cases?
by magnus
26 Mar 2019, 09:55
Forum: Installing Dalton
Topic: Dalton 64-bits error
Replies: 6
Views: 238

Re: Dalton 64-bits error

The error you had in your first message was (most likely) because of wrong input so that does not necessarily mean that you need 64-bit integers. Note that the CC module in Dalton is not parallelized so MPI will not give you anything. You may however gain some speedup by using a threaded math librar...
by magnus
25 Mar 2019, 21:14
Forum: Installing Dalton
Topic: Although Dalton2018.1 has been successfully build by “setup --mpi", all the tests failed
Replies: 1
Views: 168

Re: Although Dalton2018.1 has been successfully build by “setup --mpi", all the tests failed

It seems to be a problem with your local setup rather than an issue in Dalton. At least the "Error: Permission denied" points in that direction.
by magnus
25 Mar 2019, 14:25
Forum: Installing Dalton
Topic: Dalton 64-bits error
Replies: 6
Views: 238

Re: Dalton 64-bits error

Can you provide the output from the setup part? Note that Dalton itself is only MPI parallel and NOT OpenMP threaded, so you have to be careful using parallel MKL together with MPI-enabled Dalton. Most often you get the best performance using just MPI, unless you are using non-parallel parts of Dalt...
by magnus
18 Mar 2019, 09:41
Forum: Installing Dalton
Topic: Dalton 64-bits error
Replies: 6
Views: 238

Re: Dalton 64-bits error

The reason for the error is most likely not related to 64-bit integers but rather that you are requesting to freeze 30 core orbitals which is difficult when you only have 8 occupied orbitals ;) In this case it seems more reasonable to just freeze the two core orbitals. Try to change the relevant par...
by magnus
13 Mar 2019, 09:59
Forum: Running Dalton
Topic: CCEQ_SOL VECTORS NOT CONVERGED
Replies: 2
Views: 140

Re: CCEQ_SOL VECTORS NOT CONVERGED

You made an error in your molecule input. You specify just two atomtypes but there are four so your calculation skips the last two.
by magnus
11 Mar 2019, 13:12
Forum: Running Dalton
Topic: CC3/aug-cc-pVTZ linear response segfaults
Replies: 18
Views: 429

Re: CC3/aug-cc-pVTZ linear response segfaults

Not sure if you need 14 GB though. I just chose a rather high amount.
by magnus
11 Mar 2019, 11:29
Forum: Running Dalton
Topic: CC3/aug-cc-pVTZ linear response segfaults
Replies: 18
Views: 429

Re: CC3/aug-cc-pVTZ linear response segfaults

Most likely Kenneth is correct. The calculation completes using 14 GB with Dalton2018.1. Output attached.
by magnus
11 Mar 2019, 09:40
Forum: Running Dalton
Topic: CC3/aug-cc-pVTZ linear response segfaults
Replies: 18
Views: 429

Re: CC3/aug-cc-pVTZ linear response segfaults

I'm trying to reproduce the error. At what point in the calculation did it fail? It would make it easier if you attached the output file (which btw also contains the contents of the dal and mol file). You could consider CCSDR(3) as an alternative which is not as accurate but very close according to ...
by magnus
21 Feb 2019, 10:16
Forum: Running Dalton
Topic: One-photon oscillator strengths with response
Replies: 3
Views: 228

Re: One-photon oscillator strengths with response

I'm not sure why you get the huge output file (this one looks fine?), but to get oscillator strengths and transition dipoles, add .DIPLEN under *LINEAR.
by magnus
19 Feb 2019, 19:26
Forum: Running Dalton
Topic: CUBE file from PE/TDDFT calculations
Replies: 2
Views: 154

Re: CUBE file from PE/TDDFT calculations

It is correct that the embedding potential depends on the state through the induced dipoles, however, the cube file will always contain the embedding potential where the induced dipoles have been converged with respect to the ground state.
by magnus
12 Feb 2019, 13:39
Forum: Installing Dalton
Topic: Build subdirectory cannot be created on Ubuntu18.04
Replies: 11
Views: 467

Re: Build subdirectory cannot be created on Ubuntu18.04

Great. Good luck with your calculations :-)
by magnus
12 Feb 2019, 13:18
Forum: Installing Dalton
Topic: Build subdirectory cannot be created on Ubuntu18.04
Replies: 11
Views: 467

Re: Build subdirectory cannot be created on Ubuntu18.04

I wouldn't worry about the benchmark tests since they are just timing out and not failing as such. You could look into the other two tests to see why they are failing. The outputs from the tests are in the tests directory in your build directory. Not sure I understand what is the issue with MPI. Eve...
by magnus
12 Feb 2019, 09:03
Forum: Installing Dalton
Topic: Build subdirectory cannot be created on Ubuntu18.04
Replies: 11
Views: 467

Re: Build subdirectory cannot be created on Ubuntu18.04

And the reason is that some files are changed. Now I don't know if those changes have been made intentionally or not, so I'm hesitant to make any recommendations about what you can do with those changes. The only recommendation I have is that you try from a fresh clone, checking out the latest relea...
by magnus
11 Feb 2019, 15:24
Forum: Installing Dalton
Topic: Build subdirectory cannot be created on Ubuntu18.04
Replies: 11
Views: 467

Re: Build subdirectory cannot be created on Ubuntu18.04

Ok, you are using the master branch which is for development so it can happen that some tests fail. For instance the dft_ac_multpole fails because of a known bug.
by magnus
11 Feb 2019, 14:40
Forum: Installing Dalton
Topic: Build subdirectory cannot be created on Ubuntu18.04
Replies: 11
Views: 467

Re: Build subdirectory cannot be created on Ubuntu18.04

The reason it fails is because you have local changes in those files, so if you were to change branch or checkout a tag those changes would be overwritten. You can check the current status using the command: git status
by magnus
11 Feb 2019, 10:15
Forum: Installing Dalton
Topic: Build subdirectory cannot be created on Ubuntu18.04
Replies: 11
Views: 467

Re: Build subdirectory cannot be created on Ubuntu18.04

Assuming you cloned the repository using the following command: git clone --recursive https://gitlab.com/dalton/dalton.git You can checkout the latest released version using the command: git checkout Dalton2018.0 and then update submodules by: git submodule update Then you can build the code as foll...
by magnus
09 Feb 2019, 18:50
Forum: Running Dalton
Topic: How to use SIRIFC
Replies: 6
Views: 325

Re: How to use SIRIFC

Ok, I misunderstood. In that case the restart will not help you either. I would assume that you would not need the **WAVE FUNCTIONS section but not sure if it makes a difference if it is not there.