Waltern8tor wrote: ↑Tue Sep 19, 2023 8:08 am
I wasn't aware that such a thing as selectable filter characteristics for DAC even existed.
Yes, it does. Several D-A (and even some A-D) converter chip manufacturers offer different selectable filter characteristics, and some product manufacturers extend that selection capability to the end user.
If so why would you want to play around with filtering characteristics? Surely all of this takes part out of audible frequency spectrum (20Hz to 20kHz)
Yes, most of the options will impinge on the very top of the audio band.
The reasons for different filter characteristics is because they sound (slightly) different with different material (predominately material with a lot of energy at very high frequencies) and some listeners prefer different compromises to others.
In theory, the D-A reconstruction filter should have a brick wall response just below the Nyquist frequency (half the sample rate). So in a 44.1kHz system it should be flat up to 21 or 22kHz and offer upwards of 100dB attenuation at 22.05kHz (the Nyquist frequency) and above.
However, this is close to impossible and largely impractical to achieve in hardware.
Instead, the vast majority of standard converter filters are so-called 'half-band' filters. Although their response is flat up to 21 or 22kHz, they only manage 6dB of attenuation at the Nyquist frequency.
Manufacturers use this implementation because it's much easier to build in hardware and there's usually so little audio around the Nyquist frequency that the aliasing resulting from only 6dB of attenuation is negligible.... with most music. This kind of design is often called a 'fast' filter.
For those unhappy with this almost ubiquitous filter incarnation, one alternative filter response is to start the roll-off slightly earlier, to give a much better level of attenuation at the Nyquist frequency. This is often called a 'slow' filter.
Obviously, that approach will curtail the audible HF response around 20kHz very slightly, but remove aliasing artefacts with some material — and some people prefer that particular compromise.
Another issue addressed by different filter characteristics is the time domain response. The linear-phase filters traditionally employed preserve the phase relationships between harmonics across the audio bandwidth and allow incredibly steep filter cut-offs. These are good things. However, they also introduce pre-ringing artefacts which obviously can't happen in the natural causal world of acoustic sound. There is no pre-echo (pre-ringing) when someone hits a drum or plucks a guitar string!
Some people think they can hear this pre-ringing unnaturalness, so many converter chips now include one or more 'minimum-phase' filter options. These filter designs replicate analogue filters and have only (natural) post-ringing artefacts. Sadly, they also have much more gentle filter slopes, and consequently the compromises involve even greater HF roll-off, or more aliasing, or both.
In general, although a 44.1kHz digital system has a nominal 20kHz bandwidth, few people can hear sounds that high and so are willing to sacrifice a little of their inaudible HF extension to avoid audible aliasing or pre-ringing.
...and audiophiles love nothing more than tinkering with the settings and adjustments of their hi-fi equipment, convincing themselves they can hear improvements
