By variable densities this video is referring to the density of the infill, not the density of the material extruded. At least that was my take on it.
Hasn’t the DIY 3D printing scene been doing this for quite some time now with the use of spars internal structures for solid objects?
Advertisement

Not exactly.
The example with plastic is the same principal as reprap where the infill pattern can be adjusted for a specific density. But the concrete example show that the material is self can be changed: on the outside is full concrete on the inside is concrete foam.
This is interesting as up to now we had multimaterial where each unit of volume must have the properties of one of the feed materials but here is an example where we can variate the properties of the material continuously in certain range.
One thing that would be interesting for reprap is to be able to change the elasticity of the material continuously, This would be great to embed springs into a design without a hard frontier between two materials that breaks easily.
Oh, I see. Thanks for clearing that up.
Altering the properties of plastic as its extruded would indeed provide for some interesting objects to be printed. I suppose this could be achieved by feeding two different plastics, with different mechanical properties, into the print head at different rates to control the properties of the product extruded. Or alternatively a more crude alternative could be to expose select areas to high intensity UV light during the printing process, making those areas more brittle while leaving others flexible.
They’re working on the feeding-multiple-plastics-into-a-mixing-chamber thing at MIT too. Just at initial stages. But the other thing is that it would be nice to be able to do some of this in analog form, i.e. you extrude a fairly coarse thing that has different skin/internal properties rather than having to make a zillion passes, each with a slightly different density or stiffness or whatever.