Yeah John I get all that and your tongue-in-cheek comment on "marketing embellishments" .
That's the stuff that gets me a bit riled up and is pretty common in marketing dental mills.
You bring up some good points and its cool that you seem to have good luck with the smaller shank tools and I'm sure you are not alone. I can only say that does not make up the lion's share of my experience (and that of my customers) who are less fortunate to the point I would not recommend them. Of course there are many factors that can mitigate success including non-machine related issues (like design topology, indications, CAM Software toolpath quality, leads/links, climb vs conventional cutting, depth of cut, step-over, poor parameters etc.) and machine related issues such as frame, fixturing and the biggies of: spindle, tool concentricity/run-out and production requirements that require high SFM for minimum cycle time.
A very important spec to consider for these types of tools tool run-out and as you rightly mentioned, tool deflection which is significantly more prominent in a 3mm shank tool and actually forces the tool to bend. I'm sure you know this, but the sake of those who do not, like a paperclip, if bent back and forth too much, it will break. If the tool bends more deeply into the cut, it can increase chipload to the point the tool generally breaks immediately or tool life suffers as a result of constant pounding, engaging, disengaging of the cutting flutes. The bending means the tool won’t follow the toolpath that was intended, which can lead to all manner of problems. Other problems include shorter tool life, decreased accuracy and poorer surface finish.
From a purely theoretical stand point rigidity increases as the 4th power of diameter and deceases as the 3rd power of length.
A 3mm shank tool is 3X less rigid than a 4mm shank tool. As an example if you had a 3mm shank 2mm ball end mill sticking out 30mm performing a roughing operation at 18000 rpm at 13000 m/min with a depth of cut of .5 and width of cut of 1.2mm, tool deflection would on the order of .12mm. Deflection of a 4mm would be almost 10X less.
Please understand that I am not trying to disparage any machine. My intent is to merely point out the importance of particular specifications and why they are important. My thought is that this information and that from any follow up discussions would better educate those looking to implement a CNC dental mill.
I believe this can enable new adopters to make more informed decisions and pay a little less attention to the "marketing embellishments" or at least be able to identify them.