Critic's Corner: To Jim A RE: "Ringing" and time... by Archimago Discuss a review. Provide constructive feedback. Talk to the industry.

For Sale Ads FAQ / News / Events

Posted by Archimago (A ) on May 3, 2018 at 07:58:52

In Reply to: I'm confused posted by Jim Austin on May 2, 2018 at 16:55:01:

"Just go find any of the hundreds or thousands of illustrations of (e.g.) fast linear-impulse-response filters and slow-roll-off lir filters and look at their response in the time domain. I think it's perfectly clear that the slow roll-off filter "rings" less--has a shorter duration in the time domain, i.e., is a better "impulse"--than the fast roll-off filters that closely resemble the "sinc" filter you've described."

That's not correct, Jim.

Looking at impulse responses and extrapolating that the ringing in time correlates to what we actually hear with real music is inaccurate . Remember, an impulse is an "illegal" signal that is not properly bandwidth limited to below Nyquist.

As I wrote and demonstrated previously (linked below), with properly bandwidth limited music, the impulse duration does NOT translate into any actual ringing when processed through the digital filter. Try this yourself.

Instead, a steep linear phase filter will provide better transient time and no phase distortion compared to weak minimum phase filtering like with MQA. The only time you will see ringing is with typically highly compressed/clipped audio with square waves or simply poor bandwidth limited signals which are not what we typically regard as "audiophile quality" music demanding high resolution gear.

IMO, the only value in showing those impulse durations is simply companies like MQA making claims with no basis in reality. Yes, a slower impulse response "looks good" in not showing stuff like pre-ringing but in reality is smoke and mirrors. They are suboptimal - ultimately resulting in slower impulse rise time, introduce phase distortions, and of course are "leaky" allowing imaging/"up aliasing" artifacts to seep beyond Nyquist. [MQA seems to be simply using this to show that their playback has content above 22.05kHz - for example with MQA-CD's even though these are "false frequencies".]

"I understand the temptation for technical people to believe what they learned from textbooks is the "gold standard." Been there, done that. Often, though, that judgment is based on certain assumptions that may or may not be true, certain values that may not be universal."

So what makes Bob Stuart / MQA's "judgment" "based on certain assumptions" true? Technically they are purposely breaking the time and frequency accuracy of PCM playback by relaxing parameters that other audio designers hold dear (for example, look at what Chord is doing with their WTA filter). To force all MQA playback to use such filters based on simply the advertising benefits of the "look" of the impulse response and ignoring all the clear deficits is patently ridiculous!

• Audiophile Myth #260: The Detestable Digital Filter Ringing   (Open in New Window)

This post is made possible by the generous support of people like you and our sponsors:

---

[ Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2024, Audio Asylum, all rights reserved ]

---

[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]

---