The
audible differences between a metal dome with a resonance at 30kHz and one with a resonance at 40kHz will be marginal, if any. Well and truly in the last 1-2% of diminishing returns territory.
I'm not denying that some diaphragm materials are technically better than others, nor that certain design attributes are objectively an improvement over others, nor that these things are of technical interest and importance (of course they are).
However, these differences we're talking about here (and even the differences between any well-engineered soft dome and a well-engineered hard dome) are just not highly relevant to perceived sound quality in a complete loudspeaker system.
@Zvu re: your comparison between the T25D and T25S, I honestly don't know why Bliesma released that S version - it's a disaster, and not representative of what a soft dome is capable of. To go back to the SS 970000, which is actually a good example of a soft dome, we have a slight ripple in the frequency response around the break-up, which is easily EQ'd out (as it appears in the same place both on and off-axis).
Now, at risk of seeming a hypocrite here, I'll admit that I don't tend to use soft dome tweeters because of the issues already discussed, and because metal domes that perform better are no longer necessarily more expensive. So yeh, I'll make choices in my designs based on what is objectively better. But I don't believe that these choices are significant at the end of the day in terms of sound quality. If the SS 97000, for example, were not more expensive than a few similarly performing, cheaper tweeters that happen to be metal/ceramic, I would happily use it.
I think we are getting into objectivist audio nervosa when we start to split hairs over these things. At the end of the day, if the tweeter can be made to have a flat frequency response, a controlled off-axis response, and no excessive distortion in its passband, it can be made to perform as part of an excellent loudspeaker.
Those are my 2c