What's the status of this... is it likely to make it into release 6.0 or should we postpone it? It seems that it ought to be in this release as it is a correction to the physics, i.e. a bugfix, rather than a release.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
There are two separate branches relating to this topic (apartest and apartest2). I'm not sure how much testing there has been relating to these branches or what the status is. If I remember correctly there were two separate issues in the collisions that are addressed by both these branches:
This is related to detecting when the only kinetic species are electrons.
The second is to do with calculating the ion bulk velocity when electrostatic (or something similar).
I'm pretty sure the first issue is addressed in both but I don't recall if the second issue was fully resolved in apartest2. The apartest2 branch was originally created to look at trying to resolve the first issue by tackling the underlying cause, i.e. always having to specify at least one species with type='ion'. The changes committed should allow(/require) one to choose type='electron' when running with adiabatic ions. This makes the logic relating to adiabatic species more transparent (such as selecting the correct adiabatic_response type automatically) but there are two outstanding issues:
This would (potentially) break backwards compatibility.
More testing is required to check there are no unexpected issues.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Having discussed this with Greg, I think that these changes should be included in 6.0, but with switches to turn them on, thus not breaking backwards compatibility. Since these are really fixes rather than features, I think we can merge them into the release candidate directly, so that we don't have to hold up the fork any longer. If it turns out that there isn't time to merge these changes into the release candidate for 6.0, they will just have to go into the next release. In this case, we could have 6.0 print warnings in the appropriate settings.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
What's the status of this... is it likely to make it into release 6.0 or should we postpone it? It seems that it ought to be in this release as it is a correction to the physics, i.e. a bugfix, rather than a release.
There are two separate branches relating to this topic (
apartestandapartest2). I'm not sure how much testing there has been relating to these branches or what the status is. If I remember correctly there were two separate issues in the collisions that are addressed by both these branches:I'm pretty sure the first issue is addressed in both but I don't recall if the second issue was fully resolved in
apartest2. Theapartest2branch was originally created to look at trying to resolve the first issue by tackling the underlying cause, i.e. always having to specify at least one species withtype='ion'. The changes committed should allow(/require) one to choosetype='electron'when running with adiabatic ions. This makes the logic relating to adiabatic species more transparent (such as selecting the correctadiabatic_responsetype automatically) but there are two outstanding issues:Having discussed this with Greg, I think that these changes should be included in 6.0, but with switches to turn them on, thus not breaking backwards compatibility. Since these are really fixes rather than features, I think we can merge them into the release candidate directly, so that we don't have to hold up the fork any longer. If it turns out that there isn't time to merge these changes into the release candidate for 6.0, they will just have to go into the next release. In this case, we could have 6.0 print warnings in the appropriate settings.