Since I have a budget priority (e.g. spend money on Project A, then B, etc) and the 8-in-1-out Coaxial Hotend --- the variant I'm most excited for -- is looking to be £200+, it gets hard to justify over spending on #SecSavr Suspense [gd0105] instead.

At the same time, it's not like the 4-in-1-out has gone anywhere. It still exists and should have enough inputs for "most-colour" (almost full-colour, but subpar blacks) or multi-material and/or abrasives with glue, but not both at the same time. It should also be possible to enlarge the M6 into an M7 (or reduce the M7 to an M6 on the coaxial-to-nozzle heatblock-break) to test out such a configuration, adding some M2 threads to hold the nozzle block in place. I'd likely need to use PTFE tape to seal the heatbreaks, and from the thermal simulation, it looks like I could also use PTFE to seal the heatblock-break. This is the cheapest solution in both money and time, and it's probably best to find if there are any more hurdles with a 4-in-1-out coaxial hotend before I double the amount of inputs. 

Another option is to turn the unibody hotend into a printed heatblock, aimed at using as little material as possible (whilst not drastically increasing convectionable surface area). I'd imagine that this would be less than £100.

The last option is to design some jigs and manually lathe out the 8-in-1-out hotend. The main concern is milling out flat faces to be able to drill into the stock without the drill bit moving. Costs for the tools and materials should be rather low, but I'd have to hope that the machines I need are available at the university. If I go down this route, I'd likely make multiple as I'd imagine it'll be trivial, since everything would already be set up.

The aforementioned 2 options still have somewhat unsolved problems, such as mounting to the gantry and the design of the 8 extruders.

[13 Jan] I've found out about xyzdims' parametric printed mixing hotend, which has some nice insights on what characteristics a unibody hotend could have.