Favourite Printer Features?

Just wanting to gather some info for the printer I'm designing. What are your guys' favourite features to have on a printer (besides the ability to print)? This includes both when using and assembling/moding it.

Examples include auto leveling, auto calibration (for steps/mm etc), minimal wiring, lots of printable (moddable!) parts in the printer's design, fewer printed parts in the printer. Anything goes. Anything you feel is over rated?

I'm trying to include as many of the things I see mentioned/can think of without compromising too much on the low cost.

Hmm, well an auto calibration that involved just clicking one button would be amazing.

Interchangeable tool heads - primarily print nozzle and laser engraver. Given how much mess a milling head would make I'd probably not bother with that, as you'd also need an industrial vacumn system set up as well. Something that never seems to be discussed with these all in one machines, but would be essential.

Other than that, a sheet of printbite is - to my mind - absolutely essential. Nobody needs glue or tape or scrapers any more - so why stiil use them ?

Speed is also good. My current favourite speed and layer combo is 150mm/s and 0.4mm layer height. Amazing how quick you can knock functional stuff out.

One thing I would love in a future printer are dual independant extruders, no idea how easy that would be to do. There are at least three companies doing this now, so you could probably 'borrow' some firmware and software :-)

Or for a delta a 2 into 1 print head. Can't really see any way you can have completely seperate extruders on a delta :-) But a 2 into 1 nozzle extruder setup would be really cool as i'm currently experimenting with 'Vanish' soluble support filament.

I haven't set up any wifi yet - the bb will have it - but remote uploading of files to print would be a good idea.
Remote print monitoring also a good idea - I've been looking at some cheap ip cameras for this but you could probably mount a raspberry pi and use the camera from that.

As far a sprintable parts goes - obviosuly a good idea for a self made machine. And just a sobviously the fewer parts the better.

I think I've got it down to about seven componets for my big delta with, various other bits for mounting stuff on it. But the purely functional parts are will definitely be less than 10. I mean you'll need 6 corner pieces, 12 magnet holders, three carriage assemblies etc. But only 8 or 9 seperate designs.

So from a hobbyist view point:
Cost,but not penny pinching so I am unhappy out of the shoot and upgrading right away
I love the delta design point but I would like to see a larger base but not necessarily a taller build.
Some level of enclosure for the build chamber (printable so it would be a 'free upgrade' )
Some real auto calibration with one-time ( well not-very-often ) manual bed adjustment.
32 bits processing power
I would like the 2-1 print head (Hey CA for dual independent extruders, did you see this https://www.thingiverse.com/thing:1087944 )
Fast SDcard write via either wifi or USB

Just my thoughts

Thanks for the opinions guys, I've thought about a lot of those as future upgrades/add one, like the dual extruding and stuff. As for the 32 bit processor the reason I'm opting for 8 bit is because of the massive price gap, the cheapest 32 bit board I've found is at least four times more expensive than a ramps/mega setup. That's a fairly big cost difference to me, but I will offer a 32 bit version for those willing to make that compromise.

I'm currently working on auto calibration and how to do it. I'm having trouble figuring out how exactly delta auto calibration works since it's the closest thing there is to the GUS simpson style design. I'm mostly having trouble figuring out what they use for a reference to calibrate against. Do they rely on steps/mm being accurate, knowing the precise bed size or what.

Edit:by auto calibration I mean more than just bed leveling, I mean steps/mm correction and stuff, so that assembly and set up is as hassle free as possible (I'm sure we've all had trouble with skewed or dimensionally inaccurate printers). Relying on 3d printed parts for the frame and motion system introduces a lot of places for error compared to linear rods, rails or extrusions, and manually accounting for all of it would result in most people signing the printer up for some flying lessons out the window.

check out the mks sbase board.
https://www.ebay.com/itm/Smoothiewar...4AAOSwVohZ3ZPj
$55

Actually the single best thing about asmoothieware compatible board is the support and install and adjusting manual. Hell for a delta it even tells you what and how to measure your arms - with pictures and annotations in english !
All calibrations and configuration is by a simple text file uploaded via the memory card.
And it'll run pretty much all configurations.

It's worth a few extra dollars.

Check out auotwhiz's build. He printed the frame and most other structural pieces. About as close as you can get to a printed printer.

Having seen the new monoprice mini delta I reckon it would be possible to print a mini delta frame in one or 2 pieces. Complete with channels for wheels.
Look at all those dirt cheap i3's with perspex frames. I'm sure pla is more rigid than perspex.

$55 dollars compares to $10 or less for mega/ramps, it's a fairly big price hike. Not saying they aren't worth it or don't have their advantages, I'd just like to stick to the spirit of the design/build and support the low cost option (I will include a 32 bit option though). I can't seem to find Autowhiz's build anywhere. If you're interested in a printed printer, check out the snappy v2 or the Dollo (not that Ono kickstarter resin printer thing with the phones), they are cartesian machines that focus on being printable. If memory serves only the electronics and steppers on the snappy v2 are not printed, it doesn't use linear rods, belts, pulleys, bearings or lead screws.

What I meant with using printed parts and inaccuracies was that in my build the linear actuators would be 3d printed. Basically, in most designs the 3d printed parts just act as brackets to hold the motion components like linear rods (the critical components) and so the precision of the printed parts themselves isn't critical to the machine functioning. In the GUS simpson other than some bearings for the pulleys, the linear motion system (actuator and 'guide rail') is entirely 3d printed an so the dimensions of the 3d printed parts are critical to the machine's operation.

That being said GUS is fairly robust when it comes to these inaccuracies, the only things that are critical are that the gears are round and the pulley holes and shoulder/effector (depending on which end of the arm you're looking at) all lie on the same line through the middle of the gear. That and that both halves of the same arm have the same dimensions (which shouldn't be a problem if they're printed on the same printer). You also need the shoulder offset and hub offset to be the same (for simplifying the math) but again, as long as they are printed on the same machine this shouldn't be an issue. Basically as long as your machine can print a square that is square (X and Y are the same length) you should be able to print a functioning GUS, just your holes/slots may be the wrong size.