New Magico...if you win the lottery

[f1eng]

Why strength to weight ratio maters for speakers? I have some bear canister for backpacking out of carbon fiber and my back is acutely aware of weight to strength ratio! Aluminum does not have a good damping to stiffness ratio and carbon fiber is only slightly better (at least in a not constraint layer construction)

The stiffness (not strength) to weight ratio is important if one wishes a speaker driver to work as a piston, ie no resonance, over its entire operating frequency range. Until relatively recently it was not considered possible in affordable drivers so the choice was largely to choose a material with enough internal damping to reduce the amplitude of the breakup modes without it being too heavy (hence inefficient) or dead sounding.
Almost every loudspeaker operates with damped resonances giving peaks and troughs in its response and distortion, it is just a question of how much, not whether.
It is possible nowadays to produce very low distortion drivers without resonances in their passband but they can be quite expensive.
It is relatively easy to produce a resonance free bass unit, but it will be expensive anyway because it is big and a good linear low distortion motor system is expensive.
It is relatively inexpensive to produce a mid range driver with a hard anodised aluminium cone which remains pistonic over its entire passband (hard anodised cones tend to be marketed as ceramic/aluminium sandwich or composite cones which, strictly, they are ). The expensive bit is to make sure the crossover is very steep to make sure nothing gets it ringing outside its passband.
Tweeters which are pistonic over their passband are technically feasible using high stiffness/weight metals and diamond but they are expensive and, because they lack much internal damping, always have a peaky resonance at (inaudibly) high frequencies which can be notched out or avoided with a low pass filter. I don't think many manufacturers using this sort of tweeter bother though just assuming the resonance will be inaudible.
It is entirely plausible to make a speaker with distortion less than the notional audibility limit of 0.1% distortion in the mid band these days.
That is not to say that people will necessarily prefer this low distortion, some very highly regarded and expensive speakers have 0.5% mid band distortion, probably due to a choice of the sort of non-pistonic cone material we are used to hearing the sound of!

I get it absolutely for speaker drivers except for the not optimal damping behavior of carbon fiber (which makes great string instruments) but the new thing for Magico is the switch of the enclosure material from aluminium to carbon fiber except there are AL stiffening structures. I made some speaker boxes myself using a lot of carbon fiber and Al but now I would modify that type of construction.

True the new product is the cabinet but a dead cabinet is important IME. In my experiment fine drive units in an excellent enclosure outperformed superb drivers in a modest enclosure.
It is amazing how good the Fink associates designed "Q" acoustics speakers are, for example.
The only thing about moulded woven carbon skins superior to aluminium for a cabinet (IMO) is ease of moulding to a complex shape (and the look if you like it). Despite the carbon fibres having pretty good properties, once it is woven and laminated into a composite, which is the easy way to use it, the properties are not as good as Ally.
Admittedly my experience in Noise and Vibration research predates my composites experience in Formula 1 but the theory is still there!