Like my father is wont to say: "The marvel is not that the bear dances well, but that the bear dances at all." And as someone who's been interested in home 3D printing since the first rep-rap, I can say that the bear is starting to do a workable foxtrot and may breakdance before the decade's out.
Yes, maching is better for most parts. Machining however isn't turnkey. You can buy an assembled 3d printer for
$499, fully assembled dual-head* model with a build plate large enough for an ar-15 receiver?
under 3k. Both allow you to plug in, fire up and print parts.
outside of attention grabbing news stories, 3D printings use for guns would be more for prototyping and accessory-making than for part manufacturing. Got an idea for a new locking system and want to demonstrate it? 3d print the part and show it off. Want to test a grip shape or sight geometry? same thing. Want a replacement part for a gun that had stopped being manufactured before the great war? print a hundred parts for a few dollars to test the fit and functionality and then send the specs off to a machinist to get the real deal .
assuming a certain gunbroker auction doesn't take all of my funds, I'll be getting a desktop 3D printer soon. first thing I'm going to design and print? a mag loading tool for my makarov. There's one on the market but it's something that's both a good learning exercise and a functional object that other shooters may find useful.
Long term gun plans with this tool include hard to find parts such as the magazine for a .22lr conversion for the Mak and a replacement front sight for my ar-7 designed to hold a fiber optic light pipe. There's not enough market demand to make manufacture of either profitable and once I've finished designing and testing the parts I can release the design for free an anyone who cares to make their own.
Less design-oriented and more mundane use would be snap caps (my kitten tends to steal mine when they're ejected). You can buy them for most calibers but printing would be cheaper and easier, not to mention more available for less off the shelf ammo and faster than waiting for shipping.
One more quick note:
Investment casting from 3d printed parts is quite possible. one of the driving forces behind 3d printing is the thousands hobbyists thinking “let's see if this is possible”, if it is it's another tool to use, if it isn't the next question becomes “how can we make this possible” and the end result may well be “look at what I can do!”
*main non-decorative use for a dual-head 3d printer is to print soluble supports. Print the main part using standard plastic, print the supports using a special plastic that dissolves in water and you can print more complex shapes. the recent AR-15 receiver defense distributed made was printed using this method, albeit with a much more expensive hunk of hardware.