Guillaume Fraux
Guillaume Fraux
> I do not see how 3 is more readable than 2. Using the 'trick' `use types::Unchecked as U;` and then `self.rho[U(i, j, k)]`.
I we can get to a point where not every access is annotated with `unsafe` while retaining the performance, I think I can live with the `unsafe Array::at(usize, usize) ->...
Ok, I agree with you =)
Just removing the unsafe version did not worked out, so we might want to go the `at`, `at_mut` way =)
This sounds nice! Especially for this part: > It was mentioned in this paper that it is sufficient to approximate the total force by only computing short range contributions. Maybe...
> In smart MC you have to evaluate the forces twice - in the new as well as the old state - to compute both forwards and backwards probabilities. OK,...
No, my code example had no sense. This should not be part of the output, rather of the Compute trait. This could be decomposed in a two-steps process: 1. Get...
Tagging `L-Easy` as this can be an easy first contribution to the code. Pick a function missing example, check for usage in the code/unit tests and write the example.
> SPC/TIP4P Water These can only be done using Monte-Carlo simulations for now, because there is no code for rigid molecules in the MD part.
This is as simple as defining a new `struct` and implementing the `Potential` trait on it. See [`src/core/src/energy/functions.rs`](https://github.com/lumol-org/lumol/blob/master/src/core/src/energy/functions.rs) for examples.