Tobias Hangleiter

Any updates on this? `v3` and `master` [also fail](https://github.com/qutech/filter_functions/runs/8086317862?check_suite_focus=true#step:6:64) while `v1` [works just fine](https://github.com/qutech/filter_functions/runs/8085120991?check_suite_focus=true#step:6:1).

The imaginary part of the result differs slightly from the numpy result (hence the `nulp=2500`) in the test. Strangely, this difference disappears if the exponential of the real part is...

I see, thanks. I think this is quite misleading since the whole point of a dedicated `expm1` is improved accuracy for small arguments due to special handling of the problematic...

> @thangleiter so I have not been able to replicate your problem. I made a new branch and pushed changes to it so I can test both functions simultaneously. Here's...

I can confirm this is still happening (`IPython 8.7.0`) on Python 3.9 and 3.10, but not on Python 3.11. The Spyder kernel is a little bit more forthcoming in the...

> @thangleiter since the other PR got merged, there are merge conflicts now - could you resovle those? > > > It might be worth discussing how to proceed (ping...

I merged the branch and kept both versions for discussion. The drivers @julienbarrier added would still need to be re-implemented, but the idea of this PR is that this should...

Thanks for approving this @einsmein. Before merging, it should still be discussed with @julienbarrier and @DCEM how to proceed with Kinesis drivers though, as there are now three different approaches;...

> I might not get everything correct in the following comment - it has been a while since I have been working on this. Same here, it's been a while....

> If I saw it correctly you chose to have a folder for each instrument and then a filename by implementation type? I chose to add a _DotNet at the...