• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Old Records

TWills

Senior member
#1 Why can't I make my own records at home?

#2 Do they really deliver "purer" sound than digital? My A+ teacher said they did...
 
#1 - You can. The equipment to do so is expensive and extremely specialized. Get out a magnifying glass, and examine the grooves of an LP some day, then hypothesize how you might create those in a flat sheet of vinyl. Kinda tough, eh?

#2 - No, they don't. Leaving aside the clicks and pops from dirt, debris, and manufacturing errors, and the wow and flutter caused by warps and other imperfections, that every LP I've ever listened to had, reproduction from an LP is subject to all kinds of mechanical limits. There is high-frequency roll-of due to the vinyl "giving" as the needle goes past. There is random noise from needle movement caused by imperfections in the surface of the vinyl or mechanical movement of the turntable. There is random noise introduced due to the extremely low-level signals from the cartridge that require huge amplifications. There are tracking errors caused by the inertia of the stylus any time you have large excursions in the sound levels. There is cross-talk between left and right channels due to the fact that the same stylus is tracking two different sound waveforms, one on the left side of the groove and one on the right. There are errors introduced in the tracking of the stylus as it sweeps from the outer grooves to the inner grooves.

And that's the truth.

/frank
 
Yes, they do, since they contain analog signals rather than digital.

CDs use a digital approximation - they take an exact analogue measurement and round it to the nearest 1/65000th.

You'd think that vinyl would give better precision because it doesn't quantize (round) in this way. But what is the smallest difference that could be reliably recorded on a record? It's the difference represented by the size of the molecules that make up the plastic. In fact, the molecules are sufficiently large, and the groove so shallow/narrow that this limit is considerably worse than the digital limit of CDs.
 
Don't know if it is still commercially available, but there was a turntable/pickup that used a laser to read the groove in LP records. While that doesn't eliminate the problem of dirt and dust in the grooves, it did get rid of the wear problem. Very expensive because of the tracking servos required.

http://www.elpj.com/about/

A big problem with LPs was that they tried to put too many minutes of music on each side and that made the grooves too close, therefore causing crosstalk between grooves. Half-speed mastering helped, but it was more expensive. You could probably pick up an old Vinyl cutter at antique auctions. If you set the parameters for maximum fidelity, it could be impressive.

Would I bother? HeII no. 16bit, 44khz sample rate is good enough for me.
 
#2 - Yes, they do, since they contain analog signals rather than digital.

Since when is analog better quality than digital? Why would the entire entertainment industry be moving to digital if analog was better?

Some people prefer the sound of an analog recording to that of a digital one, but I'm pretty sure nobody would say that analog is purer or more accurate.
 
Originally posted by: icarus4586
#2 - Yes, they do, since they contain analog signals rather than digital.

Since when is analog better quality than digital? Why would the entire entertainment industry be moving to digital if analog was better?

Some people prefer the sound of an analog recording to that of a digital one, but I'm pretty sure nobody would say that analog is purer or more accurate.

Well, theoretically, since the waveform was analog to begin with, recording it on an analog medium avoids an A->D conversion step (which is inherently lossy, and limited by the bitrate of the digital medium). You also avoid going through a D->A conversion on playback (since speakers are analog devices). So, if everything was done perfectly, and you had an analog recording setup with very high bandwidth and very low noise, you could have a 'better' recording than any digital representation.

However, analog mediums themselves are generally impossible to read and write in a lossless fashion (whereas digital information can be copied, read, and manipulated without loss of quality), and in the case of records, they wear out over time (so the sound quality degrades every time it is played back).

So, while analog media are theoretically better-suited for storing analog data, it's almost always easier and more reliable these days to digitally encode the analog data once (in a very high-quality way) and then store it in a digital format.
 
Modern LPs are actually better than what was used 20 years ago or so (before the CD), the fabrication technology is much better and the quality of the material etc is generally good. This is of course in part because most buyers of new LPs are into hifi and are very demanding when it comes to sound quality (LPs are usually more expensive then CDs).
Turntables are also expensive since all the new models are targeting the mid- and high-end market, an "entry-level" model costs at least $3-400 but "real" high-end models from e.g. Clearaudio costs $2-4000.

I think the company (some japanese high-end brand) have actually started selling the "laser-turntable" again.

However, while it is much cheaper to build a good system around a CD-player I would say the media is hardly ever the limiting factor when it comes to sound-quality, the CD and the LP have more or less the same "real-world" performance and you need to move to "superaudio" (SACD, DVD-A) in order to hear any real improvment.
But even then you need a well-engineered record (which rules out most modern productions) and a high-end system in order to be able to tell the difference.




 
FrankSchwab is completely wrong:

Analog format delivers an ideally one hundred percent "pure" signal.

Digital format delivers a discrete representation of a signal. The only way a digital signal could be as "pure" as an analog signal is if the digital signal had an infinite amount of bits.

If you research the keywords: "discrete representation" on google I bet you would learn a thing or two.
 
In theory yiou are right. However, in reality there is always a noise floor and only signals with an amplitude higher then the noise-floor contains usefull information.
Hence, even analog signals are "discretized" and do not contain an infinte amouint of information. In reality even 16 bits is usually more than enough to represent all the information in a signal, the last bit is lost in the noise.
In system with very high-precision 18-bits can sometimes be used.
 
The voice coil that makes most speakers work is very inductive. The digital artifacts that most people worry about have high frequency components and speaker inductance eliminates the problem. If you have electrostatic speakers (no voice coil), they have high capacitance values and therefore have trouble with low frequencies. The whole audio reproduction process is full of compromises. You just tune your system and select your components to fit your tastes.

Or, like me, you get old enough that your ears can't tell the difference regardless of how much money you spend. 🙁
 
Originally posted by: MAW1082
FrankSchwab is completely wrong:

Analog format delivers an ideally one hundred percent "pure" signal.

Digital format delivers a discrete representation of a signal. The only way a digital signal could be as "pure" as an analog signal is if the digital signal had an infinite amount of bits.

If you research the keywords: "discrete representation" on google I bet you would learn a thing or two.

not true, an analog signal can be perfectly reproduced as long as the original signal was sampled at atleast twice the Nyquist Rate.
 
Originally posted by: JonB
The voice coil that makes most speakers work is very inductive. The digital artifacts that most people worry about have high frequency components and speaker inductance eliminates the problem. If you have electrostatic speakers (no voice coil), they have high capacitance values and therefore have trouble with low frequencies. The whole audio reproduction process is full of compromises. You just tune your system and select your components to fit your tastes.

Or, like me, you get old enough that your ears can't tell the difference regardless of how much money you spend. 🙁

or years of too much volume 🙁

Analog just plain sounds better

 
Originally posted by: MAW1082
FrankSchwab is completely wrong:

Analog format delivers an ideally one hundred percent "pure" signal.

Digital format delivers a discrete representation of a signal. The only way a digital signal could be as "pure" as an analog signal is if the digital signal had an infinite amount of bits.

If you research the keywords: "discrete representation" on google I bet you would learn a thing or two.

Man, that made my morning :laugh:

If you research the terms "noise floor", "crosstalk", "click", "pop", "wow", "flutter", "signal to noise ratio", "electromagnetic interference", and others that you'll come across in your research, I bet you would learn a thing or two.

In a perfect world, analog is great. In the real world, trying to make analog great requires exponential effort and money, especially in the low-level signals that one recovers from tape, from a disk, or whatever your storage medium.

In the real world, making digital audio great for the most part requires a few more bits and 2-4 times the sampling rate of CD-Audio (already pretty damned good). A simpler and cheaper solution.

/frank
 
Originally posted by: Loki726
Originally posted by: MAW1082
FrankSchwab is completely wrong:

Analog format delivers an ideally one hundred percent "pure" signal.

Digital format delivers a discrete representation of a signal. The only way a digital signal could be as "pure" as an analog signal is if the digital signal had an infinite amount of bits.

If you research the keywords: "discrete representation" on google I bet you would learn a thing or two.

not true, an analog signal can be perfectly reproduced as long as the original signal was sampled at atleast twice the Nyquist Rate.

Not entirely true. While you may be capturing the correct frequency spectrum of the song, you are still discretizing the amplitude. Although for all practical purposes, you shouldn't be able to hear the added noise due to quantization error.... supposedly.
 
Originally posted by: icarus4586
#2 - Yes, they do, since they contain analog signals rather than digital.

Since when is analog better quality than digital? Why would the entire entertainment industry be moving to digital if analog was better?

Some people prefer the sound of an analog recording to that of a digital one, but I'm pretty sure nobody would say that analog is purer or more accurate.

CDROM media is more convenient than the one-feet large vinyl discs. Also, it resists better in times (or at least in time lines like up to 5 years). Did you ever thought of having a vinyl turntable in your car? Or a portable one in your pocket?
 
Originally posted by: Calin
Originally posted by: icarus4586
#2 - Yes, they do, since they contain analog signals rather than digital.

Since when is analog better quality than digital? Why would the entire entertainment industry be moving to digital if analog was better?

Some people prefer the sound of an analog recording to that of a digital one, but I'm pretty sure nobody would say that analog is purer or more accurate.

CDROM media is more convenient than the one-feet large vinyl discs. Also, it resists better in times (or at least in time lines like up to 5 years). Did you ever thought of having a vinyl turntable in your car? Or a portable one in your pocket?

Well, we have Hard drives that would fit in your pocket...
 
digital = faster to manipulate
analog = more accurate

given a function f(x),

with digital, x is a set of discrete values, 1, 2, 3 etc. of couse it could also be 1/100000, 2/100000 (where there are 100000 values between 1 and 2) where f(x) would also give you a set of discrete values, where there is a fixed number of points between any value you choose.
with analog, x is a set of continuous values, where there are an infinite number of values between 1 and 2 where f(x) gives you an infinite number of values between any 2 values you choose.

the caveat is this, our ears can only hear so much, and distinguish between so much. so digital is fine, as long as the bitrate is high enough. that is, there are enough points between any 2 values that our brains\ears cannot tell the difference. The problem is, each person's ears are unique.

When your bitrate is low, you run into tons of problems. take my mp3 player. I have "30" settings for volume. Sure that's enough, but there are times when i run into a situation where 10 is too low, and 11 is too high. the same can be said of my tv, where 5 is too low, but 6 is too high. these are just examples. encode an mp3 at 96k, and then one at 320k. tell me you cant tell the difference. 😉
 
Oh great, another vinyl vs. the world debate.

Yes, in theory analog media have perfect reproduction. In real life this does not happen for a number of reasons the most important of which is that the medium (in this case vinyl) has very real physical limitations. For one thing, you have a limited dynamic range because of the way the media is both recorded and read. Since there is a needle that has to follow the groove you can only make the groove "wiggle" so much before the needle can't follow properly and you get distortion. Now, a laser turntable would be better, but when they're recording they're assuming that most people have mechanical turntables not optical ones so they have to limit the dynamic range. Secondly your record is vulnerable to all kinds of environmental issues. If the room gets warm, cold, moist, etc. you are directly affecting the medium and by consequence the sound since it's an analog medium. Next, because of the way the record is both written and read, you have to live with relatively large tracks that mean a large record. It also makes it impractical for any portable uses as was mentionned abve.

Moving on to CDs then. These are definitely not perfect either, nonetheless they have the practical advantage that the data stored is much more robust than on vinyl. Yes the both the ADC and the DAC add some amount of noise to the signal, this is inevitable. What is more important however is that the quality of both the ADC (at the recording end) and the DAC in your CD can be implemented in different ways that have different compromises. Depending on the amount of effort that you put into the reconstruction of the signal the quantization noise caused by the conversion can be reduced. On top of this, you have to remember that there are analog components to your CD player or amplifier since sooner or later you have to convert the signal back to analog and do some power amplification before you send it out to your speakers. The way you handle each step and the quality of every component plays a factor in the final sound quality. Obviously, you can't expect a $30 cheapo CD player to sound as good as a $400 turntable.

With all this said, the CD format is very good but not without it's imperfections. In most cases, people prefer vinyl not because it's more accurate, but because the they *like* the distortions caused during the playback (eg the "warm sound" of vinyl). Nonetheless, if you look at a format like DVD-A, it's such ridiculous overkill compared to what the ear can perceive that these arguments are pretty much moot.

To put things in perspective, it is generally agreed that the human ear can perceive sounds up to 20 KHz, hence Sony and Philips originally used a 44.1 KHz sampling rate specification for the CD. DVD-Audio, on the other hand allows for sampling rates of up to 192 KHz which means that you can faithfully reproduce sounds of about approximately 96 KHz. Now, without going into the details of filter design, this makes it easier to have a flat passband for the relatively low frequency audible sound. Next, DVD-A allows for words to be up to 24 bits long; a 24-bit DAC would give you a noise floor of about -144dB. Meaning that the softest sound is about 16 million times softer than the louest sound. In practical terms, with a noise floor that low, the quantization noise isn't what's limiting your sound quality.

So, long story short the people who claim that analog is better than digital are only right in theory.
 
The sound warmth is a common reason for people to like vinyl. This is the same train of thought that goes in to a lot of high-end audio equipment where people don't strive for perfect audible reproduction, but rather for a desirable 'coloring' that the hardware does to the sound.
 
Back
Top