Richard Jernigan -> RE: Do the Classics Suppress Contemporary, Creative Works of Music? Literature? Art? (Sep. 11 2015 18:24:10)
|
quote:
ORIGINAL: Ricardo There are two main reasons IMO, and it is not really an illusion about taste or preference for old school fidelity. The first reason is sampling rate...a CD or any digital sound will only be squared off sound wave type thing. The higher the sampling the smaller the little squared edges and the more "round" it might seem. So the potential is there to fool the ear into hearing a smooth curved soundwave (that you get with vinyl), but there are limits depending the source. Old CD's always admit this on a small lable on the disc. The second reason is far less acceptable, and the main problem IMO. The overuse of COMPRESSION....this is used to boost volume mainly and destroys the natural dynamics of the music...the reason old vinyl sounds more "live". With the un needed extra compression comes a noticeable increase in "hiss" in the high frequency such that necessary noise reduction becomes the final death blow to the original intended audio that was captured. The worst is the old analog recordings "remastered" or "restored" by some modern engineer that puts it in the computer program and compresses the crap out of it and dumps noise reduction liberally until it sounds like some weird robotic music. This overall horrifying process in the recording chain is usually called "mastering"... there couldn't be a more inappropriate name for the event of utterly destroying a musician's work by an un caring hearing impaired "proffessional". I do not dispute that some vinyl audio setups sound better than some digital setups. My first extended experience with digital was with a portable Sony player and Sony headphones. It sounded somewhat worse than my very good LP setup at the time, but I started using digital as soon as CDs came out in 1983 because I traveled a lot for work, and it sounded better than most cassettes on my high end Walkman. But this is not due to inherent limitations in digital technology. More about this later. In 1989 I bought a pair of Stax Signature Lambda electrostatic headphones. They are still the cleanest transducer I have ever heard. They don't kick you in the chest on the bass like the top of the line Linn speakers, but the headphones are just a little cleaner in the mids and highs. I started reading Stereophile magazine at about the same time. I thought they were nuts. They went on at some length about how different CD players sounded different. But I had accumulated a fair number of CDs, and I went around to the shops in Santa Barbara with my headphones and listened to CD players. Guess what? They really did sound different. I bought a midrange Sony player, about $350 in 1988. The $1,000 Sony sounded a little better, but I wasn't ready to put that much money into digital, given my experience up until then. Having learned a bit more about the technical details of digital audio, on my travels I started carrying a half dozen CDs that sounded good on the Sony, along with the Stax headphones. In the early 1990s top of the line Meridian and Sony players sounded better than mine, but still not enough for me to fork over the cash. Then at a little hi-fi shop in Honolulu I plugged the Staxes into a Linn CD player. I listened for about an hour, then got out my checkbook. In Houston I listened to the same CD player through Linn amps and top of the line Linn speakers. I got out my checkbook again. When the speakers arrived at Kwajalein the pallet with the two speakers, packed in their tri-wall boxes, weighed 440 pounds (200 Kg). I had to get a little help the get them up to my third floor apartment. Technical details: Square edge soundwaves: Nope. The square edged output of the digital-to-analog converter is fed to a reconstruction filter in the analog domain. This smoothes out the jaggies. You might think this is just an attempt to make the best of a bad situation but, "In the field of digital signal processing, the sampling theorem is a fundamental bridge between continuous-time signals (often called "analog signals") and discrete-time signals (often called "digital signals"). It establishes a sufficient condition for a sample rate that permits a discrete sequence of samples to capture ALL the information from a continuous-time signal of finite bandwidth." https://en.wikipedia.org/wiki/Nyquist–Shannon_sampling_theorem In theory there is no limit on how close you can come to PERFECTLY reconstructing an analog signal from digital samples, provided the analog signal is band limited in frequency. Human hearing is band limited in frequency. Nobody can hear above 22 KHz. Technology and ingenuity dictate how close you can get. Modern technology and design let you get closer with digital technology than with vinyl LPs. With a good digital system and good recordings, you can hear it. Technology limitations and some unfortunate design decisions led to digital sound in 1983 that was generally inferior to the best LP systems. I put up with it for portability. The best that could be done with the CD in 1983 was to store 78 minutes of 44 KHz sample rate stereo (total 88 KHz), 16-bit audio. Turns out this is plenty, but the design of the reproduction systems suffered from two serious design flaws. The first was the design of the anti-aliasing filter that came before the analog-to-digital converter. These had flat frequency response up to around 18-19 KHz, and then dropped off very steeply. You have to cut off below half the sampling frequency. The problem was that these "brick wall" filters inevitably introduced severe phase shifts in the highest frequencies. You could hear the ill effects. The second problem took a little longer to figure out. The samples have to be clocked into the digital-to-analog converter at precisely the sample rate, 44.1 KHz. The first few generations of CD players tried to recover the clock from the analog signals themselves via a phase locked loop. The result was microscopic variations in the the clock rate. Worse yet, the clock jitter was correlated with the loudest signals. When people looked into this they found that the resulting distortion was highly audible. One reason the Linn player sounds so good is that it doesn't try to recover the clock from the CD, whose spin rate varies slightly. The player itself generates the clock, and regulates the disc's spin rate from it. With late 1980's technology this was expensive and bulky--the player comes in two boxes. But it sounded great. The anti-aliasing filter problem was solved by oversampling. But I won't bore you with the details. You can look it up. I listened to a lot of really good LP systems from the 1960s through the 1990s. I never heard one that sounded as good as the best digital systems. I also heard quite a few $20-50K systems, both analog and digital, that sounded really, really terrible. My theory was that those systems were designed, built and sold by, and bought by people who didn't know what live music actually sounded like. I went to one of the three expensive hi-fi shops in Santa Barbara and listened to a pair of 6-foot tall electrostatic speakers that cost about $15,000 a pair. I put on a classical guitar CD that sounded pretty good on my Stax headphones. On the speakers it sounded dreadfully screechy. It was the middle of the afternoon on a weekday. I was the only customer. The store owner was an engineer who had worked for Delco in Santa Barbara. I asked him if he wanted to hear what a live classical guitar sounded like. He said, "Sure." I went home, got my Contreras "doble tapa", came back to the store, sat between the speakers and played. I asked the owner if it sounded different. Then I played the CD through the same CD player and the Staxes. Being an engineer he understood when I pointed out the flaw in the speaker design. The big panel was moved as a plane, generating a sharply focussed beam of high frequency energy. Good electrostatic speaker design moves the membrane like part of a pulsating sphere, leading to a more natural sound dispersion pattern. I agree with Ricardo that a lot of recorded music sounds bad because it is compressed. A lot of commercially recorded classical guitar music sounds like some completely different instrument. But the available dynamic range from a good CD player nowadays is much, much wider than is remotely possible with vinyl technology, so compression on CDs is an esthetic decision, not a technical limitation. All the same, a good vinyl system is likely to sound better than a bad digital one. And the evaluation of recorded sound is, in the end, subjective. RNJ Now that I've got that off my chest, it's time to go swim a few laps....
|
|
|
|