Revisions mostly related to LAMMPS-GUI #54
Conversation
|
@simongravelle I just noticed that the files for tutorial 7 include a Linux binary. I don't think that that is such a good idea (out of principle and for the tutorial). We have to assume that a large number of tutorial users will be using Windows and they usually don't have access to a compiler (or the knowledge to compile software for Windows). How about I download and bundle the wham binary with LAMMPS-GUI? If you want to keep distributing a binary, then it should be a fully static portable binary. I have the tools to create that, since I also use them for portable x86_64 binaries (very helpful to verify if bugs have been fixed). |
Tutorial 7 primarily encourage the user to visit Grossfield Lab and follow the instruction from there. Initally the binary was given with the message 'you can also try your luck with the binary I compiled', but I understand that its not great. |
That would be awesome. Note that this code is not mine, but its documentation from the Grossfield lab says:
so with proper citation its probably fine. |
Not particularly :) |
It took a little while to make it work across platforms, but I am now building new packages that include the "wham" and "wham-2d" executables. Will upload them soon. I noticed that the wham executable uses a compile-time constant for k_B. In the context of LAMMPS it would be more consistent to have an optional command-line flag for the units and then use the same units name to value mapping of LAMMPS (with the same values as LAMMPS uses). What do you think? |
Yes, that would be clearly better, but do you have an idea of the amount of work required? I don't mind having a look at it in the next weeks. |
I don't expect a lot. There are only a few places where the define is used and that could be replaced by a global variable. There are several other global variables, so that won't be a big style change. |
Please have a look at this patch. This will add CMake support for building and installing, but also adds a "units" keyword followed by the choice of units (same as in a LAMMPS input). |
Thats really great, thank you for adding wham to the GUI in that way, it will make the tutorial so much cleaner to follow. |
FYI, I emailed my changes to upstream and they will be included in the next release. Apparently both, CMake support and setting units on the command line were planned features. |
lammps-tutorials.tex
Outdated
| step during energy minimization (a) and as a function of time during | ||
| molecular dynamics in the NVT ensemble (b). b)~Kinetic energy | ||
| ($k_\text{e}$) during energy minimization (c) and during molecular | ||
| dynamics (d).} |
There was a problem hiding this comment.
@simongravelle this labeling is rather confusing. How about dropping "a)" and "b)" and just refer to the individual panels as a, b, c, and d?
The use of p_e and k_e is a bit unusual. E_pot, E_kin, or U, K would be more conventional. Unfortunately, the LAMMPS manual for both compute pe and compute ke does not provide a reference. The potential for the individual styles are usually referred to as U, though.
Is there perhaps a consistent convention in any of the text books the tutorial recommends?
There was a problem hiding this comment.
The use of p_e and k_e is a bit unusual. E_pot, E_kin, or U, K would be more conventional. Unfortunately, the LAMMPS manual for both compute pe and compute ke does not provide a reference. The potential for the individual styles are usually referred to as U, though.
Good point, I created an issue, I will fix it #57
|
@simongravelle I flagging this as ready even though I am not finished, but the changes, though mostly cosmetic, are growing too fast and merging multiple smaller PRs reduces the risk of and effort to resolve potential merge conflicts. |
lammps-tutorials.tex
Outdated
| step during energy minimization (a) and as a function of time during | ||
| molecular dynamics in the NVT ensemble (b). b)~Kinetic energy | ||
| ($k_\text{e}$) during energy minimization (c) and during molecular | ||
| dynamics (d).} |
This pull request contains revisions to the paper text and the tutorial files to make using them with LAMMPS-GUI smoother. Also, some LAMMPS-GUI test is revised and simplified, where possible, to reduce redundancy.