100 microns resolution?

it's theorectically good enough.

I think most of the demand for printing won't be prototypes to cast - laser sintering will still be better (if more expensive)

Where print on demand might work as a business model is unmouldable shapes - like complex industrial terrain, wrecked giant machines and that sort of thing.

_________________
My shifting projects

It very much depends on what you want to do.
A building might work well - a vehicle depends on - a mini will not work.

Those macines work with a sausage of plastic.
And while you may make it as thin as 0,1mm you still can see where they touch.
Even if you can position the printhead to 0,1mm or even finer, you can't print details of that resolution.
Starting and stopping the sausage flowing out of the printhead is not to accurate. Also you can expect the sausage to start with a kind of blob, however small.
You eill get the best results with continous streams of plastic, which means when you have something sticking out horizontally, you must getthe saussage out and in again without ovlapping. The resolution will be at least halved by that.
Corners must have some kind of a radius, because the plastic will build up in a tight corner.
The old software had difficulties with small objects, an other no go.
It's an interesting toy, but I'd stay away from it for making Epic stuff. For 2k plus material you get a some models made at shapeways.