This blog and others have asked the format question before: In what format do you prefer to listen to and collect your music? It's a question that will be asked for a long time too, now that music is available in four formats.

I prefer digital music because it is cheap, easily backed up, easily transferred, and easily stored. More than half of my listening is done via iPod. When I must buy music in a physical format, I prefer compact discs because of the ease of creating mp3s from them.

I'd like to start collecting records, but the cost of buying records is the only hurdle. To illustrate how expensive records are, I went on to Amazon to check prices for random recordings in various formats.

Here's a sampling of what I found:

Enslaved, Vertabrae: CD, $14; mp3, $8; LP, $44
Death, Leprosy: CD, $9; mp3, $8; LP, $38
Christian Mistress, Possession: CD, $14; mp3, $8; LP, $19
Yob, Atma: CD, $13; mp3, $10; LP, $20
Revocation, Chaos of Forms: CD, $10; mp3, $9; LP, $27

Buying on vinyl assumes that a recording is available on vinyl at all. Vinyl albums are far more likely to be released in limited print runs, which further drives up the cost per copy. Per Recording Industry Association of America (RIAA) statistics, in 2009 just 4 million vinyl records were sold. Per those same RIAA statistics, 83.1 million mp3 albums were downloaded and 225.8 million CD albums were sold. From the recording industry's perspective, vinyl records are an affectation.

Fortunately, there's a technology which exists now that could drastically lower the manufacturing cost of vinyl records. The technology in question is called 3-D printing. A 3-D printer is similar to an inkjet printer. Instead of making what is essentially 2-D text on a piece of paper, a 3-D printer can create a replica of an object using digital specifications. The concept behind a 3-D printer is simple: make sure the machine's got the right precursor chemicals and is physically large enough for the end product, load the spec, and press Start. The machine will extrude the chemicals in layers until it achieves the desired result. There are other forms of 3-D printing, but I believe chemical extrusion is the method for printing a vinyl record.

3-D printing is a not a new concept, but it is slowly entering the public conscience. A cursory look around the internet shows that companies are using 3D printing for manufacturing prototypes, design work, and low volume but high quality manufacturing runs. As an example of how accurate and cost effective 3-D printing can be, Renault's Formula 1 team uses 3-D printers to manufacture components.

"Low volume but high quality" describes a metal label's vinyl record manufacturing needs. When a label pays for a printing run of hundreds or thousands of vinyl albums, it's taking a financial risk. I googled a vinyl record pressing service and checked the pricing of the first company that came up, which was Rainbo Records. Rainbo will print 1,000 vinyl albums for $2,500, with each subsequent pressing of 1,000 copies costing $1,700. Pressing 100 vinyl albums at a time will cost $879.

What happens if a label orders and sells 1,000 vinyl records but determines that there is additional demand? Ordering another 1,000 copies could result in hundreds of unsold copies occupying label and vendor shelves for years. For that matter, how does the label even know it will sell the original print run? What if the label could spend a few hundred dollars on a 3-D printer and chemicals, and then print 10 to 15 copies at a time? At that point, a label could essentially print to order: If three people order vinyl in a week, the label could quickly print three copies and mail them off. It's also conceivable that a traditional record manufacturer like Rainbo would buy the 3-D printers in order to reap the benefits of lowered production costs.

If the concept of a label printing its own vinyl albums a few at a time seems ridiculous, consider that you can already buy 3-D printers for home use. The MakerBot costs about $1,800, plus $90 for a spool of PVA wire (polyvinyl alcohol plastic) for manufacturing objects. If MakerBot or a similar system could print records, within just a few years, we might see the cost of vinyl records plummet. To be clear, MakerBot is the trilobite of home 3-D printing. Within a few years, the cost will drop and the technology will improve. It's even conceivable that you and I could print records at home.

I see three major technological and financial hurdles to the scenario I've described. The first is the cost of the actual chemical vinyl. I haven't been able to determine the cost per ton of vinyl, but if Rainbo can manufacture records as cheaply as $1.70 per copy, which includes costs besides the chemicals, vinyl itself cannot be that expensive.

The second issue is that somebody has to figure out how to translate a master into digital instructions that a 3-D printer could use. Right now, printing instructions are programmed by a human as dimensions and parameters, and I don't see how that would work with recreating a record. A traditional record master is a piece of metal with ridges on it; the ridges are pressed into vinyl to create the actual records. Creating a set of instructions that a 3-D printer could understand takes one of two avenues.

First, the label could pay for a single traditional master plate and then pay for a service to use lasers to scan the plate, essentially creating a digital relief of the master. The relief would include the grooves and the information in the grooves, just like an actual vinyl record. The relief itself would then be the digital specifications for recreating the vinyl record. Alternatively, during the process of creating digital masters of a recording for manufacturing CDs or creating mp3s, the recording could be translated into a set of instructions that the 3-D printer could use. Honestly, I have no idea how much those two scenarios would cost. The second method may not be technologically possible at all.

The third issue with using 3-D printers to make vinyl records is accuracy. For instance, MakerBot features accuracy of 2.5 microns on the Z-axis and 11 microns on the X and Y axis (a micron is 1 millionth of a meter). Since the grooves in a record are 2.5 to 3.5 millimeters (one thousandth of a meter) wide, MakerBot can make a record. The problem is getting MakerBot to reproduce the information etched in the grooves of the original record. The printer can't just make the grooves; it has to extrude chemicals in such a way that when the needle moves across the plastic, it reproduces the music. If the printer doesn't possess enough accuracy, the result would be a vinyl record that sounded like a low-quality mp3. Information would be lost, and so sound frequencies would disappear, instrument volumes in the mix could change, sound quality would degrade, and so on. I haven't been able to determine how fine the etchings in a vinyl record's grooves are. For comparison purposes, human hairs can be as thin as 1 micron across, so existing 3-D printers already possess remarkable accuracy.

Clearly there are technological and financial barriers in the way of using a 3-D printer to make a record. Even so, I can't help but be excited at the prospect of ordering a vinyl for $8, or buying a set of printing instructions for $5 and printing a record at home. Vinyl records are supposed to sound warmer than CDs or mp3s because there's no digital involvement in reproducing the sound. Would a record printed from digital instructions still sound warm and lifelike? Would printing your own records destroy the collecting experience? I have no idea, but I'd love to find out.

So what do you guys think? Will it ever be possible to use a 3-D printer to make records? Would you use one?

— Richard Street-Jammer

. . .

. . .

Here's a MakerBot printing out Darth Vader's head:

Thing-O-Matic at work on Darth Vader

. . .

Here's a professional grade 3-D metal printer that is used to make very intricate and detailed pieces of work:

How It's Made - 3-D Metal Printing

. . .

A final note: A quick YouTube search reveals at least one video purporting to depict a 3-D printer making a vinyl record. However, the "record" is visibly warped and "plays" a Rick Astley song. My point stands though: if audio information can be translated into a set of three-dimensional specifications that a printer can use, vinyl records could be printed.

More From Invisible Oranges