PItchfork Article on High Resolution Audio

[Loop 7]

Well written and likely understandable by non-audio types. I passed it on to a few friends.

Link: What Exactly Is Hi-Res Audio, and Should I Care About It? (Pitchfork)

[Casey Leedom]

Definitely for beginners ... and probably useful at that ... but way too easy on the MQA scam ...

Casey

[pknaz]

I have yet to see any double blind tests that show hi res files are "better" or even distinguishable from CDDA

Edit: with a high enough sample size as to be statistically valid.

[Casey Leedom]

Well, it's definitely the case that the "highest bit" is the quality of the mastering. I've heard insanely good CD (44.1kHz/16bit) recordings and sub-standard higher resolution recordings. To truly judge this, as with MQA, you'd need to compare the exact same master at several sampling rates.

It's certainly "conceivable" that a higher-than-CD resolution recording could sound better. Shannon tells us that the maximum reproducible frequency with CDs is 22kHz, but no one uses "Brick Wall Filters", so in reality we're already many dB down by that point. So if you've got extremely good hearing, there's at least prima facia justification for somewhat higher frequency sampling. The question in my mind is, is that any justification for going beyond 96kHz/24bit? (To pick a point which should trivially cover human hearing rage.)

Of course, in human eyesight there's a similar issue with the minimum arc resolution which the human eye is capable of resolving. But there was an interesting article a year ago which showed macroscopic patterns evolving out of microscopic (sub-visual) details. I've always wondered if a similar effect could occur in aural space.

Casey

[KeithL]

While I agree entirely with the basic message in the article, and some of the information was useful, I also thought that some was rather misleading.

For example, the biggest problem with MP3 files is NOT cutting off high frequencies at 18.5 kHz.
The biggest problem is the mayhem that gets done to the rest of the audio spectrum.
(Taken at face value, the article might make you think that, if you can't hear past 18k, then MP3 files will sound just fine.)

[Casey Leedom]

By the way, Home Theater High Fidelity recently published an article on An Audiophile's Guide to Quantization Error, Dithering, and Noise Shaping in Digital Audio which you might find interesting in this domain. To some extent, one might be tempted to say that the higher the sampling rate, the better in order to avoid Quantization Error Distortion Spikes and to improve the Noise Floor ...

Casey