Why not go the other way?

Although the idea of a functional FF Foveon (fFFF) sensor is very appealing, hasn't one of the key plusses of the technology always been that the IQ (if the stars line up) can be so much better than from a comparable sized bayer sensor? (In fact I clearly remember the Merrill reviews that compared IQ to medium format.)

So, instead of chasing the fFFF - or trying to compete in a market saturated with very good FF  Bayer options - maybe Sigma could head in the opposite direction.

Specifically, why not turn conventional wisdom on its head and create a Foveon sensor at micro four thirds size.  Imagine something like an Em-5 (and all those excellent Olympus and Panasonic lenses) with a Foveon sensor (hell, even a shaved down Merrill sensor from 2012 would do me) bolted into it.

You'd think that the sort of problems that have affected Foveon cameras in the past (battery life, overheating, sluggish performance) could only be less in a modern body with a smaller sensor.

Sigma is not a huge player in MFT, but it wouldn't be creating this sort of body to sell lenses.  It would making an MFTF camera to sell it to a huge number of MFT users who would not be looking to replace their camera but to supplement it with something capable of doing something completely different.  For the cost of one or two new decent lenses it would be like doubling your kit