DAC filters


Doesn't it make sense that if filters are barely perceived as currently implemented, then shouldn't there be a amplification adjustment  to each filter to accentuate or decrease it's effect.

And for that matter, why can't a filter setting  be created and saved as one of the settings for the user.  Maybe also the ability to rename mind numbing tags like apodizing linear minimum phase slow etc.
recluse

Showing 2 responses by nekoaudio

The different analog filter choices on DAC chips should not have an impact on the volume, just on the signal. So it doesn't make sense to change the signal amplitude unless you're specifically trying to make the user prefer one filter over the other in a way that has nothing to do with the filter itself. In which case you might as well just recommend a specific filter, or pick a specific filter and not allow the user to change it.

Choice of filter could be saved with different settings but that requires including additional software and hardware with the product as well as developing a management UI, all of which will increase the cost. Versus simple hardware push buttons / toggles which is significantly less expensive to implement. But yes, there are certainly products that do that (not necessarily very expensive products, but it kind of depends on the company/market/product).

As for the name, I personally think it is fine for the filter names to be approximate descriptions of the filter. It also avoids having to design and implement text-input, which is usually very inconvenient unless you are using a mobile device or computer.
@recluse "amplification" of the "effect" of the analog filter choice is no longer the analog filter choice, but rather intentionally introducing distortion.

Which designers (and sound engineers) do, but not by trying to mimic the effect of a filter but rather by just designing the product to have distortion that is audible and pleasing (in the opinion of the designer or target customer).

As for the audibility of different filters, that's a different discussion, but there are definitely measurable impacts on the frequency response which fall within the amount of audible detection when using a filter at 44.1kHz or 48kHz sample rates, and in my experience also measurable and audible impacts on the phase response. If you know what to listen for.

Just because many people don't hear differences doesn't mean those differences aren't audible. It also doesn't mean those people couldn't hear those differences without additional education or training. You may be unaware of it, but there are both professional and consumer markets that makes experimentation into these things and implementing different features worth doing.

For consumers who don't care, they can pick whichever less expensive product is easy to use where the designer has simply followed a reference implementation and uses a traditional brick-wall linear filter. Possibly making choices based on differences in distortion. And this question doesn't even come up.