Real leaps and bounds will be made in 3d printing when it goes to using 5 axis machines… I’ve seen alot lately bashing the use of gcode and how it is out dated, and that something new needs to happen… But if you look at it as a machinist you are printing in 2d layer by layer giving you the limitations you are seeing. If you don’t believe me look up 5 axis milling you will see amazing things from it where under hangs aren’t an issue. Sorry for the rant and I hope I didn’t offend anyone, but Mechanical limitations in the FDM market is where you need to focus community efforts on
Advances do need to be made but you have to see both sides. Milling is subtractive and you start with a rigid structure and remove things which is why multiple axis can easily work. With 3D printing it is additive and you start with nothing so layer by layer tends to work in that situation. There has been attempts at multi axis printing but it is in its infancy and no where near as good as layering yet.
Even metal SLS currently has some benefits over milling so layer by layer is not going away and will only get better.
I stronggly disagree with you on this point.
The major benefit of FDM is that it is super easy to generate toolpaths one layer at a time, with zero chance for collisions.
Once you add additional axis of movement this no longer holds true, and the process becomes nearly as complicated as machining, but without the speed and material variety.
I’m not sure why you would need additional axes, since a machine with support can practically make any geometry already.
@Andrew_Diehl_Landru Which is why I think layer by layer printing is only going to get better and more refined. Like the evolution of FDM to SLS so that other materials can be used (e.g. printing titanium alloy). Multi-axis FDM printing will be useful for printing on already made objects like a bottle (yes terrible example) but as a replacement to current methods, unlikely.
Mechanical limitations? I’m seeing more chemical and thermodynamic limitations than anything else. Nobody has invented the perfect compound to print with yet.
@Francis_Lee I agree. If you want to change the way fdm is performed, you have to consider the medium as well, if not first.
Making 5-axis 3D printer hardware is easy. It’s toolpath generation that’s hard. 5-axis CNC CAM software is very, very complex and usually high-dollar.
5axis only works bests on subtractive manufacturing in terms of effectiveness, if applied to an fdm based machines, would only increase costs, and pherhaps defeats the purpose of low cost prototyping.
But I think the software needs more features, as in changing the toolpath in realtime, without ever touching the gcode (to change colors, adds new materials, resolution, etc) which might need more processing power on the machine side (also the developer of the software)
I think a bigger revolution will happen when conductive filament is perfected and a pick and place is added to a 3 printer. Imagine dropping an FPGA and sensors inside a 3D printed item.
@SirGeekALot Didn’t they did just that at autodesk?
Some people are trying multi-axis FDM https://www.kickstarter.com/projects/2003668803/5axismaker-first-ever-affordable-5axis-multi-fabri?token=d39a4343. The question is cost-benefit - do the benefits justify the cost?
@Aria_C_Bramanta not that I’m aware of. But they did make a huge multiple independent extruder setup.
@SirGeekALot Yep, Voxel8 :
@Aria_C_Bramanta I’ve seen that, but I didn’t realize AutoDesk was involved. Also, there is no pick and place integration, which was my main point. 
There’s one robotic arm style (forgot the name) even deposits solder paste, and place parts on printed traces