@alexberger wrote:
In my experience, each pair of speakers that isn’t driven easily by 30 Watt amplifier - sounds too compressed to me.
That outcome is also very much dependent on the specific amplifier used. I’ve heard a very good 30W pure class amplifier drive a pair of notoriously heavy load S.P. Technology Revelation’s rather effortlessly, whereas coupling them to a pair of 200W NuForce monoblocks felt like those amps struggled by comparison - even though, on paper, they’d have a +6dB headroom advantage.
Which is also to say that the sensitivity rating is only a partial signifier; a complex passive cross-over can bring many a wattage proficient (but apparently PSU weak) amp to its knees, with the same amp performing closer to its full potential and rated specs presented to an easier load. A more load benign amp requires a sturdy, powerful PSU, and those things come at a cost.
On principle though we fully agree. A difficult to drive pair of speakers, likely due to the "load severe-ness" as caused by its XO, presents a significant problem to me that only highlights why I’d rather have it configured actively sans passive XO, but that’s another discussion. High eff. passively configured, heavy load speakers would likely prove less of a hassle to the amp given that it has more headroom to deal with a complex XO.
And it easy to explain. When you put dozens watts on speakers voice coil - it overheats that leads to very sever and clearly listened compression.
In practicality I’m inclined to believe thermal modulation (as term also used by poster @audiokinesis) is the more general problem, as this comes into effect with peak heat build-ups in voice coils (and XO?) with short term transients, that dulls the perceived transients somewhat. It seems this phenomenon is less a studied field, but it could explain with why transient snap sounds more convincing with high eff. speakers.
Not to say thermal compression isn’t an issue with low eff. speakers. Dynamic prowess and their fuller swings certainly takes a hit here.