Analog Revival Means Vinyl Will Outlive CD 800
An anonymous reader writes "In the age of the iPod, an unlikely revival is taking place — kids are turning to 7" vinyl to get their kicks. Sales of 7" singles are apparently through the roof. Bands like the White Stripes are releasing thousands of new singles on the format, and record purchases have risen by over a million units in the last year — back to 1998 levels. NME told CNET: "it's very possible that the CD might become obsolete in an age of download music but the vinyl record will survive,". The article explains how indie kids are drawn to vinyl because "the tactile joy of owning a physical object that represents your attachment to a band is infinitely more enjoyable than entering a credit card number into iTunes.""
WTF is an "indie kid?" (Score:2, Informative)
Probably kills brain cells, too. (Score:4, Informative)
It's the smell.
(sniffs record) Sweet, sweet acetate...
Re:Bell bottom jeans are back! (Score:5, Informative)
But vinyl won't die and with the latest download sites, independant labels have found a happy medium of producing less vinyl and offering their tracks online. Many labels are vinyl purists and haven't yet entered the digital realm. Some label owners whom I've talked with have had increased profits but most said it stays about the same margin-wise without as much overhead.
Re:Bah! Vinyl will never replace (Score:4, Informative)
i have a cherished record collection of older mid to late 80's hardcore and punk that is in the process of being ripped... and well, since the darn things wouldn't fit into the CD player... i went out to my local best buy... and 10 minutes later (after aggressively rejecting the extended warranty) had a sony turntable.
Over-romanticised rubbish. (Score:5, Informative)
- Modern day vinyl quality is *abysmal*. thin and cheap.
- Trying to fit a modern-day album onto vinyl drastically compresses the grooves. Albums aren't 35 minutes anymore, they're commonly 40-50 minutes.
- Vinyl can't replicate certain sounds. Try an out-of-phase bass signal across both channels, the needle would pop out of the groove.
- Think vinyl has a more "natural" sound? Then you're wilfully ignorant of the drastic equalisation mashing that is necessary to embed music on a record - the bottom end has to be all but removed, which the player then puts back in. Think any player gets it right? Or indeed the same as any other player?
There are many reasons to like vinyl, sound quality is not one of them.
Re:Vinyl was already immortal... (Score:5, Informative)
Re:Bah! Vinyl will never replace (Score:5, Informative)
Baloney (Score:5, Informative)
First of all, it's time to stop confusing a CD recording with a compressed, encoded recording. Compressing to too low a bit rate and/or with a poor algorithm will of course degrade sound quality. However, let's stick with the CD, since, like a vinyl album, it's also a physical object that one can own if one wants to.
A properly recorded CD can accurately reproduce the entire audible frequency range, from 20Hz to 20KHz with a completely flat response and with distortion that is far below detectability. No frequencies in this range are lost
On the other hand, the analog signal for a vinyl record goes through an intentional frequency and dynamic range distortion (i.e. intention dynamic range flattening to fit the capabilities of the medium, followed by an "undoing" of this process upon playback). The actual vinyl stampings are made from an original master, introducing further distortion. The stampings have an inherently higher noise floor compared with 16-bit/44.1KHz digital recordings and, in addition, are subject to artifacts from any dust or defects that might be present in the grooves. The grooves degrade further with each playing, too. Plus, there's the issue of wow and flutter from difficulty in controlling the rotation of the platter accurately.
Any preference for vinyl stems strictly from either comparing a poor CD recording to a great vinyl one, to preconceived notions that influence opinion, to nostalgia or to an actual preference for the types of distortion that vinyl produces. In the latter case, the vinyl sound can be completely simulated by intentionally applying the same distortions to CD output. As one poster mentioned, you could play back the signal from a vinyl album being played on a quality, high-end turntable and record it digitally onto a CD. The result would replicate all the effects that the vinyl lover formerly attributed to some superiority in the medium.
Here's an excerpt from the recording submission instructions of a commercial vinyl album-cutting facility that can be found online:
"As such, cutting a loud dynamic record presents many challenges not typical to the conventional recording and mixing process. Trutone's mastering engineers enjoy decades of experience specific to the analog format. This expertise facilitated by their use of our classic, vintage analog tube compressors, limiters and equalizers, afford our engineers the ability to provide all final EQ and level adjustments as your music is being transferred to the analog master. The result? A rich warm sound that transcends the digital phenomena, indicative of why vinyl remains the medium of choice for promoting and marketing music."
It's amazing that they make this last statement given that they practically tell you why and how they get this sound
David
Re:How is that any different... (Score:4, Informative)
Re:That begs the question... (Score:2, Informative)
Answer: BSW
Try this: http://www.bswusa.com/proditem.asp?item=TTUSB [bswusa.com]
This model has a direct USB output.
No, I don't work for BSW, but I have purchaced a LOT of audio stuff from them.
Re:Hard to find here (Score:2, Informative)
After all, when was the last time you saw a vinyl record player in Circuit City, Best Buy, SAMS Club, etc?
Re:Over-romanticised rubbish. (Score:3, Informative)
Digital is superior for storage. Analog is superior for interface. DJing is about interface.
Re:Vinyl has better audio quality (Score:1, Informative)
When an analog signal is converted to digital, all frequencies below (samplerate)/2 are encoded. For audio CD's, the sample rate is 44.1kHz, which means that all frequencies below 22.05kHz are represented. Nothing is "squared out". Now, the analog signal needs to be passed through a very steep lowpass filter at 22.05kHz to prevent aliasing, and that very steep filter causes distortion in the high end. However, vinyl poor response at very high frequencies, so you aren't really coming out ahead there. So your choices for fidelity are either poor high end phase response (audio CD) or a gentle rolloff above 10k (vinyl).
Secondly, there are very, very few albums that have been recorded in the last 10 years where the music is not digitized at some point in the production. I have been a mastering engineer for over 8 years, and have seen a sharp decline in analog delivery of mixes to my studio. I generally work on 4 albums a week, and have only seen two or three albums come in in the last several years that were tracked and mixed fully analog. Even on projects where the end result is vinyl, the master that gets sent to the cutting room is digital. The process of cutting the laquer for a disk master requires a delay line in the path. Old mastering tape players have two heads to create an analog delay (I am sitting next to such a machine, a Studer A-80 as I type this), however there are only one or two rooms in the world that are still set up for that. Even if a album is being cut to vinyl off an analog master, it is almost certianly being fed through a digital delay as part of the cutting process.
People like vinyl becuase records have a warm, fuzzy midrange boost that is flattering to many types of music. If what you want is fidelity, high sample rate audio such as DVD-Audio (24 bit, 96kHz) or SACD (DSD audio) are much, much more accurate recording mediums than vinyl will ever be.
-Matt Azevedo
M Works Mastering
www.m-works.com
Re:How is that any different... (Score:5, Informative)
ps: yes, i am blazed right now, thanks for asking
Minor correction: (Score:2, Informative)
This is not news to DJs (Score:2, Informative)
As for today's vinyl quality VS yesterdays, I'm the proud owner of 6 original pressing Beatles LPs and the first 3 Led Zepplin LPs, and none of them are pressed on vinyl that is as good of quality as some of my 12" singles of today's EDM music.
And yes, there are some very very low bass sounds that could make vinyl skip, but compare that to every sound ever put on CD, and RTFM on how sound waves are all naturally analog, and just what happens to sounds when they are digitally compressed. Read more about CD compression VS. Vinyl sound quality here - http://electronics.howstuffworks.com/question487.
Then, come back here, and we'll have an intelligent conversation.
Re:I'm living proof (Score:2, Informative)
No its not. In the major studios, digital is king these days. I've worked professionally in the field for about 20 years on and off.
Of course, a lot of people mix down to analogue to get that sound. It is an effect and nothing more than that. And its not terribly difficult to get in the digital world either -- just inconvienient.
Beyond that, there are many different aspects of 'recording in analogue'. Is it the tape sound? The nice bit of distortion you get as you've abused the tape? I can safely say that I love the sound of tape as it gets older. I had friends that would buy up crappy 2" masters that even the original artists decided it wasn't worth having around and using this stuff. One of the reasons mixing to tape is better is that you don't have idiots trying to make things 'perfect' and moving stuff around and trying to pitch shift the crap.
Beyond tape the other bit of 'analogue' folks generally refer to is analogue summing. There is some truth to the idea that analogue summing is special. It reacts a little differently than a simple digital addition. You know in an analogue summing, strong frequencies in one input might slightly distort the frequencies of other areas. Running correlational studies on these summing units, its not hard to reverse engineer how this stuff works. Gets a little crazy when trying to analyse the sums of more than a dozen inputs at the same time though (i.e., more computing power than I have access to) -- but the end result is something that can be equated out to DSP that plays on most modern computers with not problems.
Me? I like the sound of digital. I've worked in this medium for longer than I had to deal with analogue. The original CDs sucked because they used mastering that was intended for Tape or Phono without doing any deemphasis of the original filters. The RIAA had (has) specific filter curves for both mediums that one used to ensure proper playback. CDs had no such need for these. Beyond that, cheap bargin basement digital equipment that used non-matched parts 'because it was digital' ensured that the state of the art equipment actually sounded worse than their analogue equivelents.
This only took a few years to figure out, and soon you had folks using natural EQ that wasn't designed to either overload radio signals or jump the needles out of the track -- along with audiophile digital equipment (you'd be surprised to see how bad the original digital gear actually was) -- and once this was fixed, digital was proven to be and sound much better than its analogue equivelents. Unfortunately, the idea that analogue was better had taken hold and idiots decided to parry around memes such as dynamics and warmth.
These days, when I want analogue, I switch on the paper cones and I have what I need.
Re:Vinyl has better audio quality (Score:2, Informative)
I keep hearing this same ill-informed claptrap from people. You are simply wrong about many things. This "fact" is wrong. I refer you to the Nyquist-Shannon [wikipedia.org] theorem to refute your assertion and perhaps educate you at the same time. This also has been proven wrong (I'm sorry I can't link you to sources, you'd have to be a member of the AES). The reliability of auditory memory for the purposes of comparative listening tests is very short; a minute at the outside for even the most "trained" of audiophiles. Familiarity with the source material does not objectively affect a person's ability to differentiate* between sources. The "golden ear" is a myth.
* Notice I said "differentiate", not express a preference. All a good double-blind scientific test can do is tell whether a person can consistently tell the difference between two things (i.e., beyond the statistical probability that they are guessing). It does not try to determine which is "better". What you describe here in rather rough terms is the psycho-acoustic phenomenon called "frequency masking". This is but one of the tricks employed in lossy compression schemes like MP3 and Dolby AC3 encoding. This does not happen with the PCM encoding that is used for CDs.
And last, but certainly not least, you state:
I guarantee you that anyone can tell the difference between vinyl and CD. And in the majority of cases, they will prefer the CD recording.If you prefer vinyl, then that is your perogative. Some people enjoy Limburger cheese too, but they don't try to deny that it stinks or trot out pseduo-science to try to persuade people that stinky cheese is inherently better than non-stinky cheese.
Re:Sliders (Score:5, Informative)
- CDs have a much, much higher dynamic range than vinyl. Compare CD's 90db or so to vinyl's 45db on a good turntable.
- CDs lack an infinitely variable volume level. At 16 bits of resolution, there are 65,536 possible volume levels (including silence), in distinct steps. Normally one would never notice, but the limitations of digital DO have a profound effect while processing. This is one of many reasons a studio will work with 24, 32, 48, or even more bits of resolution, even if eventually it will be downsampled to 16 bit audio. All of the processing/mixing will normally be done at a higher resolution. Incidentally, this is why many bands still record using analog equipment, and some even do all of their mixes on analog. AAD or ADD is almost invariably going to be better than DDD if you listen to music with a lot of texture and dynamic range.
- CDs have a hard limit for frequency response, with an immediate cutoff at 22050hz, whereas vinyl's frequency response extends past 25000hz with a very gradual rolloff. This should be taken into account by the recording or mastering engineer with the top end attenuated on a gradual slope. This problem used to be evident with very early CD pressings where the CD would sound "harsh" or "overly bright" compared to cassette or vinyl pressings, until the recording or mastering engineer rolled off the highs with a gradual curve. Of course, if you blasted your eardrums with headphones at 120db, you won't hear the difference anyhow because you probably can't hear much beyond 12000hz, plus it wouldn't be evident with most pop anyhow, mainly with classical, jazz, and progressive rock.
Something to consider... (Score:5, Informative)
Something to consider: Vinyl can be read by archeologists; by looking at the groove under a microscope, they can infer that it's sound. CDs use a complex error correction algorithm that will take years to reverse engineer, and decoding an MP3 off of a hard drive will be even more difficult.
For more information, I've written an extensive study of the merits and drawbacks of vinyl: http://www.andrewrondeau.com/Writings/My%20Love-Ha te%20Relationship%20With%20Vinyl%20-%20Or%20-%20Wh y%20We%20Should%20Keep%20Making%20Vinyl.html [andrewrondeau.com]
From my article about its limits:
Re:Sliders (Score:4, Informative)
Re:Baloney (Score:2, Informative)
Digital signals are all averages. For example, the signals from 0.0001 to 0.0002 KHz will be averaged (using a variety of algorithms for signal processing, such as weighted averages, boxcars, etc.). One signal will emerge, and is assigned, lets say, a frequency of 0.00015 KHz. Sound waves from musical instruments are produced over a continuum of frequencies, i.e. there are an infinite number of frequencies between any two frequencies you pick. The digitized signals are discrete. The fineness of the divisions reflects the maximum attainable resolution for the digitized signal. Thus, digitizing sound results in the loss of an infinite amount of the original signal. Think integers versus set of real numbers.
The digititzation process has been optimized so that the losses are not detectable to the concious mind, for most people. This does not mean that they are undetectable by human ears, though. The sub-concious mind may discern differences, or it may not. The lack of clicks, pops, etc. may outweigh the loss. At some level, it obviously must become a subjective call.
One more word about signal processing - about signal to noise. Signals are 'cleaned up' when they are digitized to improve the signal to noise ratio. With music, I'm not sure I want all of the noise cleaned up, though. Sure, I don't want to hear the sound from the stylus dragging across vinyl or the tape sliding over the head. But I do want to hear harmonic dissonance and distortion, and maybe even amplifier hum when I'm listening to Hendrix.
I've lived through the conversion of sceintific instruments from analog to digital and would not go back. However, it is a mistake to think that a digital signal is just like the original, when in fact, there is a complete loss of fidelity.
MM
Re:Sliders (Score:2, Informative)
The main reason for using 24 bits at the tracking (recording) stage is to provide some headroom for unexpected signals. E.g. you can safely track at a much lower average level (-12dB say) say, than trying to stuff everything into the top 6dB of a 16-bit signal. 32 or 64-bits is used while processing, of course, to maintain resolution.
There's one reason that modern CDs sound crap: and that's excessive limiting (audio compression) caused by record companies' devotion to the 'loudness wars'. CDs can sound sublime, but for me vinyl is king. I just love the euphonic distortion of vinyl.
Re:Sliders (Score:2, Informative)
I agree that each format has it's ups and downs. How ever I would say that right now we are in the wax cylinder era of digital recording. Right now the high end analog recording equipment and reproduction gear is better than any of the high end digital equipment. How ever, that having been said, digital is ultimately the better format. It's advantages far out pace that of any analog recording device. The problem is we have not come to the point where digital recordings resolution, in both amplitude and time domains, is adeqaut enough to out paced the quality of it's previous analog counter parts. Once that happens, when we have much better digital gear for both the studio and consumer levels, digital will be the superior format.
CDs lack an infinitely variable volume level. At 16 bits of resolution, there are 65,536 possible volume levels (including silence), in distinct steps.
Actually, it's worse than that. There are only 32,768 "volume levels" (voltage amplitude). This is because the amplitude can have either a positive or negative phase, so the 16 bit range is divided into half to cover amplitude into either side of phase. You can only have up to a 32,767th voltage level in either direction of speaker movement. This poor amplitude resolution is one of the two major issues that need to be over come. In order to out pace high end analog tape deck, or especialy direct-to-disc (metal mother that is, NOT cd) recordings, the amplitude resolution needs to be at least 32bit if not 64bit. And we need more stable and accurate DAC and ADC chips. Many of the ones on the market are junk and have poor amplitude tracking (bad voltage jitter).
CDs have a hard limit for frequency response, with an immediate cutoff at 22050hz, whereas vinyl's frequency response extends past 25000hz with a very gradual rolloff.
Actually, there is no "cutoff" in the sampling process it self. As you mention this is done by the engineer. How ever it is possible to run a higher freq into a ADC, you will just start to get freq division. And in reality the usefull range of CDs is MUCH lower than the 22050hz virtual wall you describe. Most engineers start a hard slope cutoff from 18Khz up, some times even starting at 16Khz! These slopes end up with zero signal well before 22Khz, typicaly before 20Khz even. A GOOD vinyl recording can get out to 100Khz in upper range. Can you hear 100Khz tones? Not directly, no. How ever, you do notice the effect that these signals have on the audiable range, and if the presence is lacking you do notice the loss. This is part of why live music can sound so much more "open". A GOOD truntable, with a shabata style cartridge (say an old AT15), can easily reproduce this signal range. And the Kenwood L-07/L-09 system that I have can also reproduce this range. There are speakers that can reproduce this range (how ever my current cheaper home made ones cannot), for example Magneplanars from Magnapan. (I want these BAD!!! Some day...)
Any ways, this second comment of yours covers the second major problem with current digital formats, too low of a resolution in the time domain. We need higher sampling rates! Beyond the 18Khz or so ceiling, there are many other problems with the current method of digital sampling. For exmaple, by using a 44.1Khz sampling rate you end up getting a problem with phase shifting of freqs starting just above about the 1Khz area. There is nothing that can be done to get rid of this distortion, current methods of mathmatical ubfustraction simply create different more complex distrotions. The only solution is to crank the sampling rate so high that the problem of high freq phase shifting starts at around 100Khz. If I remember correctly, some friends and I came up with a 4.5Mhz requirement to get past this problem!
So, right now vynil pwn5! But some day digital will have awesome, better than vynil sound quality. Then and only then will the age of digital recording truely rock, and we can then retire our turntables and open reel decks.
Re:Baloney (Score:4, Informative)
Vinyl sounds "warmer" because vinyl DOES distort the sound. It is extremely imperfect, and those imperfections lend a certain quality to the reproduced sound that is common to vinyl but absent both from the original sound, and from the sound reproduced by CDs.
Digital sampling is capable of preserving 100% of the information, provided your sampling rate is double the maximum frequency you want to capture, and your analog sound source and speakers are of sufficiently high quality. (The latter problem is common to vinyl as well.) If you don't believe this, please don't make the mistake of assuming your anecdotes trump science. Take an introductory course in DSP and learn what it is doing.
Re:Sliders (Score:1, Informative)
No more amazing than what the heads in your harddrive are doing right now! We are talking about some of the finest phono cartridges out there, not all of them go out to 100Khz. Many only get out to 20Khz to 30Khz. But even cartrides under $100 (like the Grado Black) can get above 20Khz no problem. And there are many affordable cartirdges in the $150 that can hit 40Khz.
Now, I'll admit that this is just intuition on my part, and perhaps somebody will show some math to prove that I'm wrong (or a real-world measurement).
I am sure you can use Google to find the specs and tests done on many of the better phono cartridges. You will see that I am right about them getting out to 100Khz. If you look into the Shabata (type of cut for the diamond at the needle tip) design you will see what I am talking about. Audio-Technica made a lot of these. The original intent was to get the upper freq limit way out there so you could play back quadrophonic recordings. They would ebmed the two rear channels into higher freqs on the recording. Then you play that back through a box that filters these out from the two front chans, and extracts the singnals for the two rear chans. But a lot of people, including Audio-Technica eventualy, found that because the shabata style diamond cut made such consistent contact with the inner walls of the groove, they improved playback dramaticaly in two channel systems.
If you look into it you will find there are some amazing things going on during the playback of a record, that have been well documented from a physics stand point. You are right about there being some issues with playing back using a needle that vibrates at high freqs. One issue, if I remember correctly, is that the force at the point of contact between the needle and the vynil record grove creates an increase in temp over 400 degrees F at that spot! But of course this is a VERY small area, a very brief period of contact, and so it starts to cool down very quickly. How ever you can actually heat up and warp the grooves of a record if you play it too many times in one day!
Another issue with turntables is, really, their very design. Most employ an arm with a fixed pivot point. There is a problem with this in that the angle of the grove to the needle changes as you progress from the outter groove to the inner grooves. The shabata shaped diamond cut helps counter this much better than a standard eyliptical diamond cut needle, one of the reasons for it's better freq response. There are some "linear" turntables out there that attempt to solve this problem by having a arm that does perpendicular tracking by actualy moving from right-to-left as the record plays. But most of them are flawed by the counter weighting system they have to use in order to keep tracking preasure off the needle, it's almost imposible for them to properly judge how fast the arm should track as groove spacing was done dynamicaly on most modern (70's and later) vynil pressings. High volume signals require physicaly wider grooves than low volume ones, so in order to maximize recording time per surface they started using systems that would vary the spacing between grooves based on the volume of the track at that point in time. These system typically had to perform "look ahead" (second tape head that was ahead of the main tape heads for playback from master tapes) so it would know what the volume was going to be in the future as it was cutting. This way you could pack the groves as close together as possible with out accidently cutting through from one groove to the next due to a spike in volume, as you could adjust spacing far enough in adavnce.
Then there is the fact that it is almost impossible to g