Something occurred to me as I skimmed the arguments below.I think it would be more technically accurate to call Meridian's proposed ideal system impulse response a blurring filter. It has a slow roll-off response and its shape is designed to be as close as possible to a Gaussian. In image processing, Gaussian filtering is also called Gaussian blur. It is used to soften and smooth images.
In their 2014 AES paper, Stuart and Craven say that their ideal system response is equivalent to the dispersion of sound passing through 10m of air. They speak of it like it's a good thing. But how about zero meters? Why should we accept a system that adds any dispersion at all? Conventional high sample rate PCM systems (I would say 88.2k and above, but it depends on the source of music) can already pass the full musical spectrum without adding any dispersion. Compared to that, Stuart's system adds temporal dispersion and high frequency roll-off, smoothing over transients somewhat. I know it's "only" equivalent to 10m of air, which seems like it should be pretty benign, but why adopt a system whose design target is less transparent than what's already available?
Second, the idea of selecting a reconstruction filter on playback that's matched to the anti-aliasing filter used in recording was a fine idea when it was implemented for HDCD because of the limitations of the Redbook format, where both filters had to be right at the top of the audible band. It is completely unnecessary at higher sample rates where the filters' stop bands begin at a frequency higher than the musical content.
Edits: 11/02/16 11/03/16
This post is made possible by the generous support of people like you and our sponsors:
Topic - On the blurriness of MQA - Dave_K 12:38:32 11/02/16 (13)
- Guys, pleeeze... - 13th Duke of Wymbourne 18:46:13 11/02/16 (3)
- I'll stick to the technical points then - Dave_K 08:57:34 11/03/16 (2)
- RE: I'll stick to the technical points then - 13th Duke of Wymbourne 15:07:29 11/03/16 (1)
- RE: I'll stick to the technical points then - Dave_K 12:30:14 11/08/16 (0)
- RE: On the blurriness of MQA - mkuller 12:46:44 11/02/16 (8)
- MQA compression is lossy - Dave_K 14:59:27 11/02/16 (1)
- RE: MQA compression is lossy - mkuller 16:41:25 11/02/16 (0)
- RE: On the blurriness of MQA - BubbaMike 13:08:13 11/02/16 (5)
- The point of these points is lost on me. - PAR 01:16:04 11/03/16 (4)
- The point is to discuss the merits of an emerging product that is starting to penetrate the market - Dave_K 11:07:23 11/03/16 (1)
- RE: The point is to discuss the merits of an emerging product that is starting to penetrate the market - fmak 12:06:10 11/03/16 (0)
- What is the point - fmak 07:31:32 11/03/16 (1)
- Hear, hear! - Doug Schneider 07:34:23 11/03/16 (0)