You bring up some good points and I'm familiar with the Stratasys products (though it's been a few years). You can only slightly "buy your way out of the speed

You bring up some good points and I'm familiar with the Stratasys products (though it's been a few years). You can only slightly "buy your way out of the speed problem", which is as much dependent on the material as it is the capabilities of the machine. An 18 hour print for a small item is not unusual at all, for me, but I'm on a pretty simple printer. A stratasys-style machine wouldn't make sense for this application. I think "high-end desktop quality" FDM would suffice for accessories/GameStop (with, perhaps, an option to get items printed in aluminum/other exotic materials shipped -- partner with someone), so the aim would be to find a reliable, enclosed (with appropriate filtration[0]), FDM (or similar price/quality), pick up 3-5 of them per store -- different sizes -- and spend a mint building software for your customers to use to interface with the printers[1] and automate the hell out of the ugly parts of the printing process (print-bed surface prep/part removal, maintenance and cleaning). And make absolutely sure the model has every convenience feature (like bed leveling, temperature controlled build chamber) I'd probably limit material types to one (production) and one (support, when necessary). Optimize all of the hardware for the printers to the specific characteristics of that material and choose a material that prints easily/quickly -- PLA is a reasonable choice, prints pretty quickly but is brittle/fails over time and has a low melting point, ABS has several toxicity concerns and is finicky as hell to print with, but it's cheap and when it prints "right", it's durable. It sounds crazy to do something like this with desktop 3D printers, but it makes the most sense. $1000 will get you one hell of a box with the ability to print in multiple colors and with more than one hot-end. It's probably, as you mention, not quite there yet -- I know from personal experience, messing around with the slicer software can take some time. There'd be a lot of software to be written and probably some additional hardware to design to fit the purposes of the stores printing needs. But, I guess you'd give the staff something to do while waiting for customers to not come in and buy games. /s :) While they're at it, throw in a large CNC/laser cutter, re-tool to sell dev boards/robotics equipment, replace "Game" with "Hacker" and open up several hundred hacker-spaces around the country. Perhaps that would have worked better for Radio Shack. But I'm only half-joking -- every time I walk into small-time game/hobby/computer shops I find that the (often young) staff there are makers and would fit right in with a little training. [0] This, too, depends on the material. And you might need to call in the lawyers/etc to figure out how things are going to have to be designed to avoid a future class-action from your employees/customers due to exposure to fumes. ABS is known to be problematic; it's also cheap and has many other desirable properties. [1] The good news is that there is a ton of open-source pieces to start with.