Last year, shares of 3D Systems, a maker of industrial and consumer 3D printers, shot up 270 percent. These gains turned 3D Systems’ stock into one of the best performers of 2012 and broadened investor interest in 3D printing technology. The question for 2013 will be whether that interest can hold as investors, enthused by what they’ve been reading about 3D printing technology, encounter some of the realities of the marketplace for these machines.
(Folks just coming up to speed on this topic can check out a profile I wrote on 3D Systems and 3D printing early last year. Those who read it and invested can thank me later.)
3D Systems makes the vast majority of its money selling large printers to companies that want to crank out quick prototypes of parts. Aeronautics and auto companies have been longtime users of this technology. Today so, too, are consumer electronics companies and even orthodontists making custom braces. For all of 2011, 3D Systems reported revenue of $230.4 million. One chunk ($137.3 million) came from selling the actual machines, while the second chunk ($93.1 million) came from selling what amount to proprietary plastics and powders that go into the machines, much as Hewlett-Packard sells toner and ink to its printer customers.
Here’s the rub: 3D Systems spent just $14.3 million on research and development in 2011. That’s a paltry 6 percent of revenue. Its main rival Stratasys posted 2011 revenue of $155.9 million and spent $14.4 million on R&D.
Step back and think about these totals for a minute, and you might come away disheartened. 3D printing rightly gets billed as one of the most exciting areas of technology, and it’s simply not receiving the level of investment that you would expect.
A major problem with 3D printing ignored by most is that there’s no Moore’s Law-type mechanics underlying the technology. The physics behind fusing a powder with a laser, or melting plastic and squeezing it through a nozzle, pose severe limitations on the speed at which you can print an object. And once you’ve applied a layer of plastic, the printer must take time to move and adjust a platform supporting the object.
You can add a nozzle or two to speed up the work, but this is a far cry from doubling the amount of transistors on a chip every 18 months, or doubling the storage on a hard drive. There may come a spectacular advance to change these equations, but no one seems to have found any yet. And it seems unlikely that the amount spent on R&D today by the major makers of 3D printers—3D Systems, Stratasys, and Germany’s EOS—will be enough to turn up an unexpected, exponential price-per-performance booster.
Evidence of just how antiquated the technology is can be seen at 3D Systems’ headquarters in South Carolina. There the company has dozens of machines lined up to receive orders from customers via its 3D-printing service arm. Technicians stand in front of computers and try to figure out which products can be printed together most efficiently. It’s basically just as fast to print lots of things as it is to print one thing, so you want to line up products that can fit geometrically alongside each other and maximize the amount of printing done per run.
Much of this process is done manually, with technicians twisting objects around on a screen to see how they fit together. When the actual print is complete, the technicians carry the parts back to a different room to manually clean them off and manually pack them.
To the extent that there are big price/performance advances taking place today, they are happening at such companies as Shapeways. Based in New York, the company operates about 10 3D printers at factories in Europe and the U.S. People go to its website, pick objects they want, and Shapeways prints them and mails them. The company has developed its own algorithms to automate how products are arranged in the 3D printers and has more automated shipping systems. “The manual process that used to take one technician two to three hours now takes place in the background,” says Peter Weijmarshausen, chief executive officer at Shapeways.
Right now, Weijmarshausen says makers of 3D printers charge about $100 per kilo for their proprietary plastics, while essentially identical materials can be found in other industries for about $10 per kilo. Why? It’s just that the off-the-shelf materials are not certified to run in the 3D-printer makers’ machines. Shapeways has managed to command better prices through the volume of goods it is printing—about 125,000 objects per month. “The goal is to get more machines, have more distribution, and be more efficient,” he says.
The fundamental gains that come through running a business at scale have helped cut the cost of a 3D-printed custom iPhone case from $250 to $20 in the last four years, Weijmarshausen says.
The consumer end of the 3D printing market may be the great hope for technology advancement in the years to come. Dozens of startups are building machines that, while rudimentary and less capable than their industrial counterparts, do let people create an awful lot of things for a low cost. Companies such as MakerBot and Afinia and some machines on KickStarter have drawn outside investment into the industry.
(How awesome, mind you, would it be to see Hewlett-Packard apply decades of printing expertise and its vast supply chain to this area? Nothing would repair HP’s reputation as an innovator so much as a super-easy-to-use, cool-looking home 3D printer. Or perhaps, Apple could throw $100 million or so at a 3D printing gambit. I digress. Sigh.)
The big hope, as I see it, is that venture capitalists that have long ignored 3D printing may finally jump in and help spark a big step forward with the technology. Until that happens, it’s hard to imagine the 3D printing market exploding the way shareholders seem to hope will happen.