Slashdot Mirror
Vinyl Gets Its Groove Back
theodp writes "Time reports that vinyl records are suddenly cool again. Vinyl has a warmer, more nuanced sound than CDs or MP3s; records feature large album covers with imaginative graphics, pullout photos, and liner notes. 'Bad sound on an iPod has had an impact on a lot of people going back to vinyl,' says 15-year-old David MacRunnel, who owns more than 1,000 records."
What's even more amusing is that almost all vinyls pressed today are mastered off the final digital master.
Most music is recorded digitally and then mastered digitally. The vinyl records pressed use a digital master. Now the digital master used is almost certainly of higher quality than version pressed onto a CD, but still - records are still an analog copy (of the original analog master) of a digital master.
Of course you could, theoretically, spin records at a faster speed and then pitch them down in software if you want -- but if you are going to transfer a record to digital, it is usually a better plan to record them at a slower speed and then pitch them up in software, as you'll have more samples available for each second of audio. Software like Audacity even includes processing presets for doing pitch manipulation among standard record speeds -- this is why the 33/45 turntable that Thinkgeek offers, for example, is marketed as being capable of transferring 78rpm records, even though it is not capable of playing them in real time.
Not true. Many people like to make this claim, but analog records have physical limitations as to the frequency content they can record. There is a noise level which limits the accuracy of recordings done on records, just as there is an associated noise power from continuous->discrete conversion in the A/D process. You can create digital recording which retain more of the originally produced sound than an analog record possibly can with increased sampling rates and low noise electronics.
I don't know if they actually sound better, but I personally just love the physical action of putting on a record.
They can sound better if you have a good turntable with a good cartridge, a good preamp and amp, and good speakers that are capable of resolving the differences between digital and analog audio. The problem is, you're talking about $20,000 worth of high-end audio equipment there.
And that's not taking into account wear and tear. Vinyl degrades with each use; there is no getting around it. You're putting two physical parts in contact and moving them against each other; over time, your records will sound worse and there is nothing you can do about it.
People who make blanket statements about vinyl sounding better just haven't taken real-world considerations into account. In the real world and under most conditions, a 128kbps mp3 played on an iPod is probably going to sound better than a well-worn vinyl record of the same recording.
That "warm" sound is distortion. Some listeners may like it, but from a quality/reproduction standpoint it is most definitely a bad thing. If record companies really are selling vinyl again, they're probably just trying to make a quick buck on nostalgic idiots who are actually dumb enough to buy vinyl records in 2008. Even the record companies realize that online distribution is the next big thing.
Most of the problems with CDs and digital audio can be blamed on poor compression and the loudness war. I'm really sick of hearing the same old rants from vinyl fanboys... why is this even worthy of Slashdot's front page?
Oh, and technically speaking, vinyl has a finite bitrate. Once you get down to the molecular level, that is... so I'll have no more of this "vinyl has infinite quality" nonsense.
One of these days, I'm going to cut you into little pieces.
Yup. 24-bit precision gives you almost 17 million values. Assuming a total groove width of 2 mil (50 microns), the maximum excursion is physically bounded at about half that or you'll end up with the cutter over in the next groove... maybe a little more, but not much. So 50 microns of width divided by 17 million gives ups about 3 × 10^-12 meters, or about 0.03 angstroms....
Now, to put that in perspective... The estimates I've seen for the diameter of a hydrogen atom are about 1 * 10^-10 meters, give or take. That would make the resolution of a 24-bit digital signal equivalent to an analog cutter whose resolution is just about a 30th the width of a hydrogen atom... well beyond what the laws of physics allow.
A typical particle of PVC, as best I could ascertain from a quick web search, would be 100,000 times as large. This puts vinyl at about 10-11 bits of resolution, practically speaking. Don't get me wrong, I think vinyl sounds better than CDs in many cases, but that's because of awful digital mastering practices---overcompressing the signal, audio engineers who can't hear above 12kHz doing the mix, overhyped highs and lows to compensate for craptastic sound systems, etc. It's not because vinyl is inherently better; it's because audio production from the vinyl area was inherently better. Don't get me started on the Disneyana AutoTune-until-your-ears-bleed style of recording we're getting out of the industry today. When it comes to an audio delivery format, there's a certain degree of "garbage in, garbage out" at work.....
Check out my sci-fi/humor trilogy at PatriotsBooks.
For me vinyl was always cool, but regardless of the arguments abount sound quality there's one feature that vinyl posesses for DJ's that's frequently overlooked - the user interface - the way you can control the music by dragging the record on the turntable, the way you can seek to the right point in the record just by dropping the needle in the right place - the way you can see the beats, the builds and the breakdowns on the media just by looking at the way the light reflects from the surface. That's why I still buy it, for performance purposes.
Now, there are many attempts to replicate the interface, either with the giant jog wheels on the CDJ's or vinyl control discs sending control signals to computers (Serato/MsPinky/Final Scratch) but while these bring advantages to the equation - mnamely being able to carry a larger selection in your record bag or laptop's disc - they still fall short of the pure vinyl experience in subtle ways.
Now I can listen to practically any track ever recorded, on demand and for free at sites like imeem.com when I love music I want the physical artifact and a vinyl version always gets more love from me.
Oh and vinyl is robust, I have 10 year old CD's that are turning brown and won't play, but I have 50 year old vinyl that still works just fine.
Vinyl degrades with each use; there is no getting around it.
:
Actually, you CAN get around it if you're willing to shell out $10k+
http://www.elpj.com/
That's a popular misconception. The human ear doesn't just stop hearing at any particular frequency, per se. It tapers off above a certain point. Beyond 15-20 kHz (I assume that's what you meant), depending on the individual, it starts falling off. By 22 or 23 kHz, you need a pretty massive volume to hear it, but most people who haven't blown their ears can still hear it.
Check out my sci-fi/humor trilogy at PatriotsBooks.
They can sound better if you have a good turntable with a good cartridge, a good preamp and amp, and good speakers that are capable of resolving the differences between digital and analog audio.
The ones I laugh at are the ones who get a USB turntable because they don't like digital sound and want the analog experiance.
They get better sound simply because most vinyl isn't in the loudness war to kill the dynamic range. A CD with about 96 DB of dynamic range should sould better than the about 65 DB dynamic range of a turntable. Unfortunately the advantage of the CD format is often engineered out to sound louder.
The irony is a USB analog turntable outputs a digital signal on the USB cable. Often the sample rate is the same as a CD. Even more often they are sold to the clueless without even listing the sample rate or bits. Quick, can you tell me if this is an 8 bit, 16 bit, 24 bit, sample size at 16K, 44.1, 48, 96, 128 Ksamples/sec?
http://www.thinkgeek.com/electronics/mp3/90a0/
They advertise it on a geek website without posting the important specs.. Guys, what's the wow & flutter and rumble levels?
For me, I'm sticking to my 1980's moving coil linear track turntable with a good reciever plugged into a quality mixer (to set levels) which is then fed into a pro USB a/d converter. I capture at 96KHZ 24bit and downconvert to CD quality to burn CD's. It works for me.
Here is another USB turntable with no specs listed.
http://www.amazon.com/Ion-iTTUSB-Turntable-USB-Record/dp/B000BUEMOO
and another;
http://www.amazon.com/Numark-TTUSB-Turntable-with-USB/dp/B000G3FNVM
Here is one that is reviewed and the A/D stats are known..
The sound quality was as good as can be expected from old, scratchy records. The built-in audio card records 16-bit at 44.1khz
http://reviews.cnet.com/turntables/stanton-t-90-usb/4505-7860_7-32417457.html
Wow, no better than CD quality...
Some of these turntables get poor marks for their conversion to digital quality.
"The TTUSB10 as a Turntable
After my disappointing experience with the TTUSB10 USB turntable's recorded sound quality, I plugged it into the phono input in my stereo, hoping for some sweeter sounds. This time around, the TTUSB10 did not let me down: smooth, rich audio came through the speakers and my test headphones without a trace of the harsh digital noise that plagued my test recordings. It would be a bit of a waste of money just to buy it as a standard turntable, but if nothing else, the TTUSB10 makes for an excellent unit for playing your vinyl music collection on your stereo system."
http://www.everythingusb.com/ion_ttusb10_usb_turntable_13231.html
The truth shall set you free!
Down-sampling is a *form* of compression, and it is one of the forms CDs employ (another main form is to reduce the resolution which is completely distinct from downsampling). In fact, it's a form a lossy compression. Which is exactly what I stated.
Yet another form of compression employed on CDs is dynamic range compression, which results in significantly reduced quality (far worse than the amount of downsampling and reduced resolution employed on audio CDs, interestingly enough), but is not inherent to the format and wasn't really brought up in my post. I only bring it up now to demonstrate at least *three* ways audio CDs are compressed.
Lets see... where to start.
1/ 16bit digital audio has about twice the dynamic range (numerically) of vinyl records. In fact 16bit digital audio has more dynamic range than the best professional analogue tape recorders - even when those tape recorders use good noise reduction techniques.
2/ 16bit digital audio has more than twice the channel separation (numerically) of vinyl records. In fact it has complete channel separation.
3/ 16bit digital audio does not require dynamic compression in order to fully capture and playback the entire dynamic range of a large orchestra. Vinyl requires dynamic compression for almost everything that is to be reproduced with vinyl, including capturing all frequencies below 1kHz.
4/ The simple process of tracking the "stylus" through the grove of the record damages the vinyl, deforms the grove and introduces distortion. It is simply not possible for a stylus to faithfully capture from the vinyl what was pressed onto the vinyl - even with the most expensive equipment. Most styli are not even capable of not jumping out of the grove in louder parts of the music.
5/ ALL music recorded in professional studios today is recorded digitally using either 16bit or 24bit recorders. The bit-rate is what determines the depth of the sound and the total dynamic range available. All vinyl does is introduce limitations and distortions.
AND they still say that vinyl has a superior sound. Well yes - it does - when you compare it with MP3s! But that really is not saying much.
That is the real problem. See also http://en.wikipedia.org/wiki/Loudness_wars.
If a CD is released without dynamic compression, it will sound fine.
Several years ago, the german HiFi magazine Stereoplay made an experiment to determine if the digitizing as such makes an audible difference. They took a high quality analog recording and played it two different ways:
1) Directly from turntable to amplifier and from there to loudspeaker, no digital equipment involved.
2) Somewhere in between, the signal went into an A/D converter and from there into a D/A converter. The other components were the same as in 1).
In a blind test (cannot remember if it was double blind) the test audience could not determine a difference. The equipment was quite high-quality BTW, they definitely used one of the $20.000 or more rigs that are often quoted as being necessary for hearing the differences.
Also, Vinyl is not immune against someone compressing the digital master before the recording is transferred to vinyl. Expect such stupidity to happen shortly
C - the footgun of programming languages
Because low frequency sounds have much more "energy" than high frequency sounds, the sound on an LP is equalized before encoding onto the record. This equalization is done according to a standard curve so all playback equipment handles it roughly the same, and the equalization boosts the high frequency sounds by 20db while REDUCING low frequency sounds by 20db, with a crossover point at roughly 1khz. The exact constants are 314uS and 3140uS, or about 100hz and 10khz, above and below which the equalization is "shelved," or flat.
If this equalization were not present, it would be almost impossible for the LP record to exist, as the grooves on a record would have to be so far apart. It would also be very, very hard to get playback equipment to reliably track such a record.
Now, records are not just "cut" in a dumb fashion. Since the 70s at least, mastering equipment has been smart enough to move the cutter head across the record at variable pitch. In this way, passages that had a lot of bass content (and thus produced wide excursion of the stylus) could be recorded at a wider pitch than "average" tracks. In fact, it is this equipment which allowed those "extra long play" records of the late 70s to come into existence. Radio Shack sold a few of these featuring such artists as Arthur Feidler and the Boston Pops, and Earth, Wind and Fire, and these albums could play a half hour or more on each side. This was done by careful equalization and record level settings combined with variable pitch cutting of the master disk.
So far as excursion goes, no, it aint limited at all to anything like 2 mils. If you can find an old copy of Telarc's recording of Stravinsky's Firebird Suite and look closely at the record, you will see places where the groove pitch is about fifty times that! This was considered one of the benchmark tests of the day as many cartridges and tonearms could not play it without skipping. In fact, if you simply read some old equipment reviews of the 70s and 80s you will often find this recording to be one of the standard reviewers tests.
But what you completely missed is electrical noise. See, a standard phono cartridge has an impedance of 600 ohms. A 600 Ohm source impedance, at room temperature, has a fairly well defined noise floor. That is, barring any other source of noise, the simple thermal noise of the transducer itself can never go below a certain level. Given a "0db" standard for most phono cartridges of roughly 4.5mV, the noise floor can never me more than 76db below zero. This was, in fact, the source of some amount of fraudulent advertising during the "numbers race" of the 70s and 80s, when many manufacturers would claim phono s/n rations of upward of 100db. While one can most certainly make a preamp that can prodice this low noise output with a SHORTED input, connecting an actual transducer to the input throws that right to the wind. As a result the FTC mandated phono S/N be specified with a standard input impedance of 600 Ohms.
None of which _really_ means anything. Zero db on a phonograph is not a hard limit (as shown by the Telarc recording) and that noise floor does not mean no information can exist below -76db. But likewise, Digital recordings are not so "hard limited" either. Noise shaping allows much greater than 96db s/n floor across the midrange where it is most needed at the expense of higher frequency noise floor where it is less likely to be audible.
Basically, the difference between these two - outside the distortions implicitly mandated by the RIAA EQ curve and the electronics needed to accommodate it - comes down to mastering. Which adds new meaning to the phrase "fool me once, shame on you, fool me twice..." When, in a few years, these kids buying vinyl have grown into twenty somethings with plenty of disposable income and are once again lured into replacing their "old vinyl collection" with new digitally mastered SACD recordings that are cut from the _analog_ masters (that sound good) rather than the CD masters where the signal was digitally comp
That's what the "audiophiles" claim, but that's not the way CDs are recorded. That mythical "number of bits" figure is mostly a marketing argument, digital recordings today are done in a way that eliminates quantization noise in the audible band
Digital recording technology isn't just a fad, if it were a new one would have replaced it by now. Digital is actually better than analog in *all* aspects, if done right. If done wrong, well, does a bad analog recording sound good to you?
The weakest link in sound recording and reproduction is almost always the conversion between electrical signals to sound and vice-versa. When people "compare" digital sound to analog they are often comparing listening to an ipod with earbuds with listening to a $100k analog system. Well, try to listen to the ipod in a pair of these $5350.00 speakers and tell me again again about those "warmer, more nuanced" sounds.
Unless you can be sure that the vinyl and CD masters were identical, your tests are invalid.
Due to the limitations of vinyl, and the current trend in CD mastering, they are unlikely to be the same masters.
A better comparison is to try comparing the direct output of the turntable to the same through 16bit 44.1KHz ad/da conversion.
In double blind tests, no one has ever been able to reliably tell the difference.
Of course vinyl can sound better than CDs. "Sounds better" merely means I prefer it. Saying that there's no way vinyl sounds better than CDs is like saying that there's no way strawberry ice cream tastes better than chocolate.
What you can say is that there's no way vinyl sounds more accurate than CDs. While you could make a contrary argument (say, that the distortions in LP playback cancel out distortions in the recording process), this is a defensible statement.
But the LP vs CD argument is different than other audiophile foolishness, like $7000 cables. People who sell expensive cables claim audible differences where no one has ever proven that there is any difference, so it's not reasonable to claim a preference. No one claims that there's no audible difference between LPs and CDs, so people should be expected to prefer one over the other. And it's not reasonable to claim that everyone else should share your preference. Some folks prefer "pretty" to "realistic".
Down sampling is down sampling, not lossy compression.
If it was the same as lossy compression, then that would imply would data on the CD would be uncompressed on playback to provide some resemblance to the original high sample rate master.
This does not happen on CD, as the missing information from the original master is irretrievably lost. There is no decompression on playback, and so no extra information is generated.
If you take a picture and remove half the pixels, you have not compressed it, you have removed half the pixels. This is equivalent to downsampling. There is no way of getting those missing pixels back.
If you use a compression scheme that allows assign more data to those pixels are more important to the way humans perceive images, you have used lossy compression. You can increase the perceived quality of the image after decompression.
Lossy compression also implies a trade off between human perception and available bandwidth. As perception does not factor in linear PCM audio, you cannot say CD uses lossy compression.
No, vinyl records are not better, they are just different. I grew up on the crux of the changeover from records to CDs. I've heard it all. I work in audio professionally. There is an entire spectrum of audio quality regardless of the format. If there is any issue at all it's that people are getting used to and are willing to forgive crappy compression on audio. Properly recorded digital still has warmth, depth, and has a far more "nuanced" sound because it isn't buried under a freaking mechanical noise floor. This resurgence is just another trend that will fade as quick as it started.
Now, when mp3 players came out, the producers of popular music started applying even more of this dynamic range compression to bring up the softer parts of the mix. Part of that is because music is more often heard in noisy environments rather than a quiet listening room. If you ever tried to listen to Wagner on the subway, you'll know what I mean. You'll only hear the loudest parts.
Yes, the data compression used can create artifacts of its own. But when you're talking about the way music is "mixed for mp3 players", you're talking about dynamic range compression.
You are welcome on my lawn.
My favorite thing is what tends to happen on certain audiophile boards. Some new, audiophile LP pressing of some overly venerated classic rock title is announced. MSRP: something like $30. Cue spooage, despite the fact that old copies of said release can be found for $.50 everywhere. Anyway, so the date approaches. The first person gets his record. Posts an almost glowing review, replete with WARMTH and TUBEY GOOSH and whatnot.
What's the almost?
There's always a caveat. "Pressing was a little warped; going to ask Classic to replace it." "There're a few nasty ticks on side 2." "There was, believe it or not, a skip in the third song."
And it just makes me laugh and laugh. Dude, you know what the problem is? You just bought...new vinyl! You're going to return it because it exhibits a lower grade of the problems inherent to the medium? Your search for perfection is going to yield anything but.
Bah. What's terrible is that it's apparently far easier to reissue things on vinyl than it is on CD (for independent labels, at least), so a lot of titles that haven't made it to CD get reissued on "audiophile vinyl" for $30, and then everyone acts as if that's solved the iniquity of its otherwise-unavailability.