Josh Vermaas

Martin and I have been talking, and we think the generalization for triclinic boxes is something like this. Written out this way, it looks like it ought to be performant,...

All of them I think would be ideal, to make colvars properly GPU-resident the whole time the simulation is running (except for file I/O). Otherwise you still incur a memory...

@orbeckst and @hmacdope , I've got a draft that I've tested to work on the trajectories I had on hand with a orthogonal box. I'm pretty sure it would work...

Naw, I figured it out. @hmacdope , its good to go now, assuming I wrote the test correctly. See #4011

Yes, I did. Thanks @orbeckst. I've gotten the citations reformatted.

Have you looked at the [docs](https://vmd.robinbetz.com/molecule.html)? What you are doing in VMD depends on behavior described in the [user guide](https://www.ks.uiuc.edu/Research/vmd/current/ug/node182.html), where `molecule.load` in VMD 1.9.3 (or I think 1.9.4) will...

I think that is normal. The best test is of course to write a short python script. Something like: `import vmd` in an interactive or batch python session should now...

Ok, serves me right for not testing things. Shoulda been `help(vmd)`, but that isn't the problem. I *think* the .so wasn't compiled as a dynamically loaded executable, since it was...

Sure. You can already do a clone off of my fork, it just isn't in the mainline yet. `git clone --branch ppc64le [email protected]:jvermaas/vmd-python.git`

Weird. I assume the web version shows up fine? https://github.com/jvermaas/vmd-python I know I have my ssh keys setup, but it looks like above you use the https version instead, so...