I am truly at a loss as to why we are still arguing about this, but we somehow still are! (See “Retailers Giving Vinyl Records Another Spin.”) Here is the quick answer: digital-audio techniques and media can capture and reproduce sonic events far more faithfully than any analogue technique and medium. Note that the focus of the above statement is fidelity not preference, a distinction that most “LPs versus CDs” debates wrongly blur.
Old news
More than twenty years have gone by since the first time CD sales surpassed those of LPs, and several highly qualified acousticians and engineers have since weighed in on the LPs-versus-CDs topic, outlining in numerous books, scholarly journal articles, and presentations the mathematical, acoustical, signal-processing, and perceptual issues involved. (See the relevant, well-written article on Wikipedia for a partial bibliography. See, also, an intelligent talk on the topic by Princeton University’s Paul Lansky.) Some of these experts have also sent relevant letters to the popular press or have published blogs and other online resources. (See this post by analog-integrated-circuit designer Mike Demler.)
Still, self-proclaimed “experts” and die-hard lovers of popular myths, who seem to approach knowledge almost exclusively through what C. S. Peirce, in the 19th century, called “the method of tenacity” (holding on to one’s already established beliefs at all cost), insist on keeping the analogue-versus-digital-sound debate alive. See, for example, this thread on Audiokarma.org’s discussion forum, which includes a typical example of the types of arguments used by LP advocates: “I know vinyl is better because… it just is” (emphasis in the original!).
I will not waste any time here repeating in detail the arguments for the superior audio fidelity of CDs versus LPs. Interested readers can find more information through a relevant Google search, assuming they know how to weed through the returned results and evaluate Internet resources. (e.g. Does the author identify him/herself? What are his/her credentials? Are the arguments supported by references to credible, peer-reviewed sources? Are the sources of information properly cited? etc.)
I will simply outline the inherent and unsurpassable limitations of analog media, such as vinyl LPs, and highlight an important distinction between fidelity and preference that seems to be overlooked in digital-versus-analog debates.
Limitations of LPs
The mechanical nature of sound-signal capture and reproduction in LPs and the associated issues of inertia, momentum, and interference impose frequency and dynamic response limits (i.e. limits in both the range and fineness by which a signal’s frequency and amplitude content can be captured and reproduced without interfering with adjacent signals) that constitute an unavoidable fidelity bottleneck within the medium. CDs completely bypass these issues thanks to optical methods of sound-signal capture and reproduction, assuming appropriate digitization (sampling-rate and bit-depth choice), storage (CD-surface and surface-coating choice), and handling (CD-surface protection during use to minimize the need for digital error correction).
Fidelity versus preference
Advocates of LPs and other analog sound media often cite the analog sound’s greater “warmth,” “smoothness,” and “fullness” as the main reasons for choosing analog over digital. Interestingly, these subjective sound-quality characteristics are related to acoustic side effects imposed on live, sonic events by the analog media themselves. Preference for such sonic qualities may be based on familiarity and habit (having grown up listening to music exclusively through analogue media, showing a conditioned preference towards the “familiar”) or may constitute a conscious aesthetic choice (intentionally altering a sonic event, through the sound-quality distortions introduced by analog media, to achieve a given aesthetic result). Regardless of the reasons behind some listeners’ preference for the sound-quality distortions introduced by analog media, the fact is that the sound quality carried by such media is exactly that: distorted. Preference for analog over digital and the other way around occupies an inherently subjective, gray area, and discussions on it can and will continue. However, when it comes to sound fidelity (how accurately an acoustic vibration is represented by a sound signal) it’s just black and white, with digital coming out the clear and undisputable winner.
Hi!
Thanks for the link to my blog. I just happened to find your site during a Google search today.
Regards,
Mike
I fail to follow your connection that the technology limitations inherent to vinyl LP playback (you actually generalize that to all analog media, egads) somehow leads to your conclusion that analog is inferior to digital. The only myth that needs to be discussed is the one originally promised by CD – that of “perfect sound forever”. Some 25 years later, commercial aspects aside, the predominant feeling by those in the industry who live and breathe music reproduction every day is that CD is far from perfect, and really isn’t forever. But don’t take it from me – listen to the recording engineers and attentive producers and artists who actually care about the quality of the product they produce. Or better yet, listen to the music itself. It’s all right there – layed out for you to hear. I refuted it at first too. My engineering training told me Nyquist had to be right. My common sense told me that the high end audio press was in the business of propagating these lies to keep selling copy. But ultimately, my own experiences as a musician, and as a passionate listener convinced me that something, indeed, had gone astray when the world of music and digital information theory collided. And the emotional bond that happens when you are hearing music that most closely approximates the “real thing in real space”, is unmistakably more common in the world of LP playback than digital playback. As for why, I can only attribute it to things we as yet do not understand about the nature of the information being processed. But people far smarter than I are actively working that problem, and I for one, am willing to wait for them to figure it out.
Thanks for your comment Barry.
CDs are indeed far from perfect. And as I indicate in my original post, my statement on their superiority over LPs assumes appropriate digitization (sampling-rate and bit-depth choice), storage (CD-surface and surface-coating choice), and handling (CD-surface protection during use to minimize the need for digital error correction).
Digital sound processing and storage, as implemented on CDs, does introduce its own artifacts/distortions relative to the original signal it attempts to represent. The two most obvious problems are with the assumptions underlying the Nyquist theorem and the dynamic range that can be represented by 16 bit audio. Engineers and acousticians, some of whom I interact regularly with at the meetings of the Acoustical Society of America and the Audio Engineering Society, are indeed working on these issues as we speak, and I too am doing work that contributes to these collaborative efforts. [My relevant research is with regards to spectral analysis of sound signals and mathematical representation of aspects of timbre – see http://www.acousticslab.org if interested.]
However, these issues aside, the stricking lack of fidelity in LP storage and reproduction makes CDs definately the better (even if not the ultimately best) choice.
Although I focused on LPs, the drawbacks I describe have direct analogues in all other analog media (cassett tapes, reel-to-reel tapes) and result in analogously compromised fidelity.
As I mentioned in the original post, fidelity is not equivalent to preference. I do not doubt that you and many others may find the LP listening experience more satisfying. I too feel the same way for certain musical pieces. But when it comes to strictly comparing ability/potential to more faithfully capture, store, and reproduce a given complex signal, analog media are so poor performers that the undoubtedly imperfect and artifact-prone CDs still come out the winners.
Pantelis