Anne Heimes

sound good, @h-chmeruk can you also include @jf-som suggestion in your pr?

> [@h-chmeruk](https://github.com/h-chmeruk) are you also working on deprecating the old [pyfar.dsp.filter.fractional_octave_frequencies](https://pyfar.readthedocs.io/en/stable/modules/pyfar.dsp.filter.html#pyfar.dsp.filter.fractional_octave_frequencies)? I could take care of this, in case you're not. not yet, you can take care if you like.

We should also add ``__neg__(self)`` and ``__pos__(self)``, this will allow the use of ``b=-signal`` and ``c=+signal`` which is not supported atm

open discussion: - [ ] do we need separate classes for arithmetics and matrix multiplications? - [ ] should we do do simplie operations with numbers? ```python def __mul__(self, data):...

shift discussing into the correct issue: We could create a dedicated filter constructor for each filter method in ``pf.dsp.filter`` into the related Filter classe. For example: ``` filter = FilterSOS.from_butterworth(N,...

yes you are right, new try: We could create a dedicated filter constructor for each filter method in ``pf.dsp.filter`` into the related Filter classe. For example: ``` low_pass = FilterSOS.from_butterworth(N,...

beautiful!

small correction: merge it as soon as it has at least 2 approves.

sounds good, this would be also useful for #849, but can be added later. It could be implemented like this: ```python def find_nearest_frequency(self, value, method='nearest'): """Return the index that is...

good catch. I would suggest to first define what we want. ``frequency_range=(20, 200)`` seems quite easy, but how about do we want e.g. ``frequency_range=(20, 250) -> [31.5, 63, 125, 250]...