Records and CDs

The truth about records versus CDS lies in the way sound is replicated. With vinyl, what you are making is an analogous (a.k. a. analog) representation of the audio waveform. This has the distinct ability to make the record much more similar to the original recording of the music. The other major merit is how humans perceive sound. With vinyl, the ear never "tires" because it's a analogous wave form. In other worlds, the sound wave becomes an electrical wave becomes a sound wave again. This is how we percieve sound naturally. With a CD, you have what is known as "digital fatigue," in which your hearing doesn't perceive the sound as smoothly as analog, becuase digital literally breaks down sound waves into binary and replicates them through bits and filtering and whatnot. To make a long story short, digital isn't REAL. But it's more convenient...more time, smaller, portable, etc. In the end, it depends on what you're listening to and what you want. If you want to hear a good live acoustic band, listen to vinyl. If you want to hear glitch electronica, vinyl will add a psychosomatic "body" to the work, but since it had a digital conception, it doesn't particularly matter. And plus, vinyl makes you an instant bad ass, right?

This has the distinct ability to make the record much more similar to the original recording of the music."

I disagree. There are physical limitations on how well the signal on the analogue tape can be replicated in the vinyl pressing, and how faithfully this can be transcribed on playback. Ultimately, you typically end up with *less* resolution in your average vinyl LP (65- 70dB dynamic range, 50-15k bandwidth) than you do on a compact disc (93-96dB DR, 20-22k bandwidth).

"The other major merit is how humans perceive sound. With vinyl, the ear never "tires" because it's a analogous wave form. In other worlds, the sound wave becomes an electrical wave becomes a sound wave again. This is how we percieve sound naturally. With a CD, you have what is known as "digital fatigue," in which your hearing doesn't perceive the sound as smoothly as analog, becuase digital literally breaks down sound waves into binary and replicates them through bits and filtering and whatnot. To make a long story short, digital isn't REAL."

Highly debatable. Is hearing really analog? How do the nerve impulses get to the brain? What comes out of a CD player has (or shouldn't have) *any* remnants of its digital origins - it'll be an analogue waveform, just like that emerging from yr RIAA amp, except (chances are), it'll be a whole lot closer to what was originally on the master tape.

Now, none of this necessarily means that folks shouldn't find vinyl more 'real' - quite aside from any number of non-audio factors, there are some fairly well-established mechanisms by which the shortcomings of vinyl replay can actually 'enhance' stereo reproduction. HF phase anomalies leading to a greater sense of spaciousness, tonearm/cart resonances leading a 'warming' of the lower midrange, etc.

I think most of the gripes about digital audio were down to lousy application of the technology (or maybe 80s production methods, though that's another aesthetic debate entirely), rather than the inherent flaws in the technology itself. Besides, if 16/44.1 really isn't/wasn't enough (and I suspect it isn't on the recording side, but fine as a delivery medium), 24/96 is near-ubiquitous now.

I still buy a lot of vinyl - partly for the packaging - but it's a heck of a long time since I regretted buying a CD over its LP equivalent. The reverse is, unfortunately, not true however.

and if digital fatigue occurs, which any engineer with a decent set of ears will tell you that it does, why is it happening?

love, dr. balls

I have probably bought three times more vinyl than CDs this year (I tried doing a count of CDs last night, and it was around 80). I seem to buy albums on vinyl that I probably never would on CD (I bought a reissued gatefold copy of What's Going On last year for $18 Can. brand new, in a store that sells the CD version for $12 brand new, which I would never give the time of day). This is amplified with the crazy number of reissues on high quality pressings. I bought the Shuggie Otis reissue on 180gm vinyl, with the original cover art/tracklisting instead of the Luaka Bop version with the extra tracks, even though they were the same price. As well, the stores I shop at are generally run by people who know what I like, and they tend to be more vinyl heavy, so they dictate the format I buy considerably.

It's 44100 discrete samples per second (for Red Book digital), and, thanks to Shannon and Nyquist, that means *perfect* representation of any waveform below 22.1kHz. I'm sorry, but I do not buy the 'digital ear fatigue' business. These chains of samples are reconstructed into analogue waveforms - and you can basically take your pick, the one with the higher noisefloor and FR anomalies (analogue), or the one that's flat to beyond the range of adult human hearing and then drops like a stone (digital).

Analogue recording has bandwidth limits too - you seem to assume that because it's a continuous-voltage representation rather than a discrete-value representation, it somehow sidesteps these physical limits. It doesn't - it has a certain signal/noise ratio and a certain bandwidth.

There's a hell of a lot more to this subject than we've touched on here (Direct Stream Digital [1-bit, 2.8MHz] is *very* interesting), and, after an evening's Xmas drinking, I'm certainly not prepared to go any further right here.

Merry Christmas to one and all - analog(ue) and digital.

you are totally not factoring in the effect of the frequencies that are outside of the 20/20 range of our hearing. When you actually record sound, you are not just recording the frequencies we can hear, you are recording partials outside of our hearing spectrum that reach in and create overtones with the ones you can hear. 16/44 recording has neutered the recordings of the last 15 years. If you don't believe it, you should try working with mastering tools that are 24/96. You can hear the difference the extra recording range makes, it does in fact sound fuller and richer.

Redbook might very well be a perfect representation of a wave at 22.1kHz, but it is not a perfect representation of the acoustical phenomena that people try and capture in the studio. There is a great deal of sonic information that is lost in red book, and it is not something that should be ignored.

to put it simply, the things you cannot hear very much effect the things that you can.

mt

I am quite torn, I do a radio show in Detroit and I would love to be able to abandon transporting 60lbs. of vinyl to the station every week. Vinyl is great for cover art, and just the general joy of playing. Vinyl is not fun when you have to carry 60lbs. of it up and down three flights of stairs each week to play music for a couple hours.

Cd's are far less soulful, they just are not as much fun as records(a truly objective statement if ever there was one). CD's are far lighter, they do not degrade with repeated plays, the packaging is more durable, and they allow you to easily exchange music with friends in genres that you would never buy in. They are also a lot more profitable for labels, they cost half as much to make and distributors buy them for twice as much as records.

I am in the process of CDr'ing my vinyl collection to make it easier to do the show. As far as home play goes, I still will play the records before I play the cd's.

Does this supra-aural information actually have to be captured ON THE RECORDING MEDIUM to have this effect on the audio range? Or merely by the microphones?

If the latter, then this extra info which 'reaches in and creates overtones' within the audible range, must surely affect the 20-20k range, yes? In which case, a sampling system which captures that range will preserve the AUDIBLE effects of such overtones. "16/44 recording has neutered the recordings of the last 15 years. If you don't believe it, you should try working with mastering tools that are 24/96. You can hear the difference the extra recording range makes, it does in fact sound fuller and richer."

I do work (occasionally) with 24/96 - I actually prefer to work at 24/44.1, but that's more of an issue with non-integer sampling-rate conversion (which one has to do if the final product is CD) and hard disk space. I'm a strong advocate of 24-bit recording - *certainly* that makes a difference, giving more headroom for a start (allowing later gain-scaling without ever bringing quantisation error into the picture), and allowing all subsequent DSPing to be carried out at greater resolution. I remain unconvinced by the virtues of recording material with a FR up to 48kHz, other than it may allow a less severe filtering implementation on conversion.

OK, this has always puzzled me: why 96 and not 88.2?

88.2 (and 176.4 for that matter) is actually available on some pro recorders. The 44.1kHz sampling rate arose because digital storage has its origins in video technology, and by using 525-line video at 60Hz, or 625-line video at 50Hz, you can attain this sampling rate. DAT presumably came a bit later and had its sampling rates rather more sensibly pegged at a multiple of 2. 96 and 192 are now available.

131072 ya mean?

I didn't mean that DAT's sample-rate was a power of 2, more that it was based on multiples of 2 in kHz, 8 and 16kHz being traditional sample-rates for telephone transmission. DAT has a 'long play' mode which samples at 32k, and I guess 48k was as good as they could get back in the mid-80s when the DAT standard was set (rather than going the whole hog to 64k). 96k is simply 'twice as good'.

I don't think base-2 has any great meaning with regard to sample- rate, unlike bit-depth.

Conversion between DAT multiples and CD multiples is indeed a pain, so I don't personally bother. I think some people actually prefer to dump to 1/2" 15/30ips tape from DAT (with all the lovely compression artefacts that such a transfer may introduce) and then re-digitise at 44.1 for CD prep. If I ever get my hands on a reel-to-reel I might try it myself.

I'm afraid this is pretty much the standard response of the pro-vinyl contingent on the newsgroups - once every avenue of argument wrt greater resolution/more natural sound is exhausted, it comes down to "we know what music is SUPPOSED to sound like, and you guys don't".

Preference is sacrosanct - if you like how your LP-12 (a turntable which I've never heard give what I consider good, neutral reproduction - but I'm prepared to believe that I've simply heard poor installations) sounds, then that is your reference and I can have no argument with that. As good as my Gyrodec is, my Copland CDP outstrips it in pretty much every area - though I fear it's more of a shortcoming in the media than the hardware.

I've made CD-Rs of LPs which sound just like the LPs; I don't believe there's anything to LP reproduction which CD is 'missing', other than (possibly) euphonic artefacts associated with the shortcomings of the medium and the mechanical aspects of its playback. I have a great personal fondness for the vinyl format, but I don't consider a superior carrier for music.

Preference is all.

I just wondered what Gage-o and his pro-vinyl audiophile ilk thought about vinyl that has been through a digital recording or mastering stage. Presumably, with this digital 'bottleneck' (as they would see it), the resultant LP would be as bad as (and probably worse than) its CD counterpart. I'd like them to listen blind to my digitally remastered LP of 'Beggar's Banquet' and be able to identify this 'digital fatigue' of which they speak.

Unfortunately Nick, as you've just told them it's digitally remastered, the test is no longer blind. Doh!

Perhaps a better test might be switching between an LP in good nick, and a CD-R made from that same LP.

Part of the problem I have with the notion that continuous-voltage media are *automatically* better and more natural than any format that involves sampling, is that cassette tape comes into the picture. Is anyone going to argue that tapes are better than CDs? And if they're not, why aren't they (from a fidelity point-of-view) and in what sense can this be related to the vinyl-CD debate?

Yeah - it's a pretty basic automatic gain control circuit on some of the portables (Sony ones, I think), but, as you say, it's defeatable. My portable doesn't have such a thing and I wouldn't use it if it did. (It occurs to me here that you're probably talking about playback limiting rather than record limiting; again, I've no experience of this).

"MiniDiscs also seem to lose out in the details of the music. i've listened to a lot of house/techno/jungle DJ mixes on MD, and when the music starts getting complicated tape definitely has the edge over MiniDisc. MDs good for convenience, though, obviously"

Each to their own, I suppose. I can't think of a single area in which I prefer cassette to MD (aside from the fact that I have lots of music exclusively on tape which is better than the stuff I exclusively have on MD). Most of MD's data reduction seems to be done in the 14kHz+ area (i.e. stuff you could barely hear even if it wasn't masked by lower-freq material), so that might account for the alleged lack of spatial resolution associated with the format. I don't hear much evidence of congestion on complex passages though.