Gearspace.com - View Single Post - MQA discussion at Denver RMAF
View Single Post
Old 31st August 2017 | Show parent
Lives for gear
FabienTDR's Avatar
Verified Member
Originally Posted by lucey ➡️
I think he said they turned off the DeBlur, that's a gimmick.
Worth pointing out that the whole idea of "Blur" bases on an intuitive misunderstanding of how filters affect a signal. The concept of "Blur" implies that a filter negatively affects what it passes in the form of a "time-smear". This idea is brutally wrong.

Filters fulfill their task by spreading the stopband (what they remove) over time. Not what they preserve! (the passband).

In case of a steep linear phase LP filter at say, 1kHz, this smear is very audible. That's because the stopband content itself was very audible from the beginning on. Say, 3kHz content is now spread over time, and this is easy to hear.

But, in case of MQA or DAs in general, we are messing with Nyquist filters purely removing content generously beyond the bandwidth limit of our perception. 20kHz and above. This ultrasonic content, no matter how strong it is and how strong it gets spread over time, can never ever become audible during filtering! It would require a linear filter to change pitch of the signal, which is completely impossible.

Filter blur is reality, but only with filters operating within the audible range. Nyquist filters typically do not! AD blur is nonsense.

The whole habit of nyquist filter impulse response plot interpretation is charlatanry. Not more, not less. It is a central aspect of MQA's argumentation. :D