Active crossover feature

[mirekti]

1 Why would they bother as the development as the development budget for software would be huge and the grief if it didn't work perfectly first time even bigger.

Given the existence of Phantom, it tells us Devialet already has software developed to the certain level. I don't believe they would need to invest too much for this to work for masses. If you check current solutions, the UI is quite simple.
Something similar to this:

2 The grief about not using their people to sort out expert range/Phantom problems first.

This is a growing company, and this shouldn't be an issue.

3 How many would they sell to recoup the cost?

Not sure as we don't know what the cost is.

All in all, Devialet is an electronics company which tries to set its own footprint which they already did.
I don't see active crossover as something only DIY people would use. Devialet only needs to enter the exiting market which will grow.
Different brands could start to produce speakers without passive crossover. Those brands could invest more in driver development rather than passive crossover. They can do some measurements in house and provide you with data like: HF pass, LF pass, delay etc. and only thing one has to do is to enter these parameters in the DSP or Devialet can make profiles as it does now, irrelevant. There are many ways to do it.
More and more companies go in that direction. Look at the new core business statement of this company: http://www.holmacoustics.com/news.php

Some will say this is complicated. I'd say, Windows were complicated to start with. Nowdays, this can be as trivial as a Spark setup.
We'll see. I am not saying this has to be done by Devialet, but it would be nice. And I am not trying to convince anyone to agree with me. The whole idea of DSP crossovers is already happening. Devialet started to expand by introducing Phantom, but I want more