Has anyone experimented with low-slop linkages in order to create micromills? I've never specifically looked for it, but neither have I come across it in my reasonably extensive travels. I figure if it's been done, someone on this forum will know about it! Given that high speed micro rotary tools in both electrical and pneumatic formats exist, it seems like the obvious method of low-cost experimentation with >3-axis CNC milling would be to create really small-scale machines. A 2" square platform could easily be held in place by a couple of NEMA 23s, even for alu milling. It would be perfect for tabletop gaming models, high-accuracy scale miniatures, jewellery, etc. in ways that 3D printing just can't compete with. Obviously 0.1mm of backlash or say, 0.01mm of rotational error with standard proportional control over the course of a 10mm object would be relatively large percentage errors, but using a mechanical advantage of say, 10:1 would drop them by probably nearly an order of magnitude (assuming imperfect linkages) whilst allowing weaker drive motors to be used because the forces are both much smaller to begin with and far less able to transmit back to the motors through a 10:1 lever and a screw drive. The 3D models would probably be sent to the machine scaled according to the machine's "normal" size, and then machined at 10% scale. I envisage a tiny platform in the middle of a large cube. The regular axes are linked to the much smaller versions of the axes holding the actual tool via the levers. The platform itself rotates in both the A and C (I think? about z and y) axes like a DMG DMU-50. With the tiny 1/16" collet tools available to miniature carvers and a little mechanical ingenuity, could micromachining be attainable for DIYers? And even if so, is there a practical point to doing so other than being able to create small complex objects?