Actually, de M3 has an agitator. Check the reviews and reports here.
And it is availiable with a hopper. A really smart design too.

The M3 maybe isn't such a good starting point tho'; It's very expensive. The cost isn't easy to bring down because the burr carrier is a solid piece of brass. Which is not cheap to manufacture. And certainly not to tight tolerances, which it is. And it has to be massive, to keep the burrs in place while grinding.
There are a few things that can be done I guess.
-Give the burr carrier two supports
-Two driving axle bearings.
-Flat burr set. The belt drive doesn't have to be that large, and the burr carrier can be smaller too.
-Cover the belt drive.
-Aluminum burr carrier. Cheaper than brass.

All this said and done, I still think there are significant improvements possible with standard grinders. Teflon coating on a straighter grind path would alleviate a lot of the current problems i think. Maybe an auger to prevent popcorning and a Versalab-type doser.

Stuggi wrote:I've never seen one in person, but from what I can gather it's belt driven for some obscure reason (which means that the belt, at some point, has to be replaced, if it isn't made out of foreverlastium. It is also uncovered, which is a potential danger, belts can cut of fingers if the motor driving it is powerful enough) and I can't see why they didn't make a hopper for it, it must be a PITA to make sure the beans stay the small cavity above the burrs while grinding.
they might be just hot air.

There are a lot of answers to those questions on the Versalab site. <http://www.versalab.com/server/coffee/grindernew.html> Shows your single dosing idea in practice (although, it's by volume, not weight) and allows for two apps, to use for two coffees, one dose at a time. The belt is so that it can easily accommodate a grind-through design, instead of the problems that burrs mounted on top of a motor with a side discharge can initiate. Cruise the site, follow the process, and a lot of the stuff that you're looking for is already there. If I could find someone interested in paying decent money for both of my Majors, I'd use the cash to buy me one o' these.

I'm very much a newbie, but from my understanding:

In theory, would it make sense to have a grinder that distributes grounds in repeated, impossibly thin, layers?

The idea here would be to create a "snowfall" action where grounds are dropped in consistent, thin layers from a high location, perhaps through a long vertical tube. It would be similar to the idea that if you want to salt food more evenly, you sprinkle it from high above, instead of close to the food, forming localized clumps. I sort of see this idea in the Marzocco Swift.

Rather than re-inventing the wheel...

It's worthwhile to check out what's been done in the past with coffee (from the Alinox through the Swift to the Serch-modified Anfim to the prototypes of the new LM grinder) -- but more importantly to look at what's been done in other industries, in particular in pharmaceuticals.

More than that, however, you probably should start with defining goals and objectives - 'cause right now it looks like you're focusing purely on "a better version" style approaches. For example... if a critical goal is to create very consistent grind size (and I'm NOT saying that would be a good thing) you would probably look at either two-stage grinding (like the Alinox) or vertical burr sets. Both of these have trade-offs, however, that might compromise your other goals (easy maintenance, easy adjustment across wide ranges, little to no trapped grounds for example).