... the effect is more pronounced with recordings on CD that are more compressed, which generally you want to listen to louder to give the sound more ’presence’ in the room. As you also pointed out, the more ’spacious’ recordings, within my system can definitely be played louder without the clipping effect emerging... no question about that.
I’m not sure that we’re using the term "compressed" in the same way. I use the term to refer to a recording that has been engineered such that there is a relatively small difference in volume between the loudest notes and the softest notes. The opposite of that would be a recording having wide "dynamic range," meaning a large difference in volume between the loudest notes and the softest notes.
The majority of pop and rock recordings are dynamically compressed to a considerable degree, and consequently have a relatively small difference in volume between the various notes. Less than 10 db in many cases, which corresponds to the loudest note requiring less than 10 times as much power to be reproduced as the softest note.
**Some** well engineered classical symphonic recordings, on the other hand, can have dynamic ranges of 30, 40, or even 50 db. Those numbers correspond respectively to requiring 1000 times, 10000 times and 100000 times as much power to reproduce the loudest notes as the softest notes, the loudest notes usually being very brief, and the softest notes usually being far softer than on most pop and rock recordings.
Therefore it is the latter recordings, having wide dynamic range, that will usually result in the user turning up the volume control higher than for the highly compressed recordings. And consequently much more power will usually be required to reproduce the (usually brief) high volume dynamic peaks on recordings having wide dynamic range than on recordings on which the volume of the notes doesn’t vary a great deal.
But if in fact you are finding that recordings having wide dynamic range do not cause the problem, while some recordings having narrow dynamic range do, it would lead me to suspect that the problem is not inadequate amplifier power. Rather it would suggest the possibility that either the amplifier section of the receiver is not able to handle the thermal (heating) effects of having to supply high power levels fairly constantly, or the speakers are not able to handle the thermal (heating) effects of receiving high power levels fairly constantly. And perhaps the recordings with the strong drum beats that you referred to are putting that over the edge.
But the situation I described in the previous paragraph would be unusual, compared to problems involving clipping of an underpowered amplifier on brief high volume dynamic peaks, or problems involving defective equipment. Which leads me to suspect, as I said at the beginning of this post, that we may be using the term "compressed" to mean different things.
Regards,
-- Al