Does It have to be loud?

I am with the crowd that thinks the ability to sound good at lower volume is a high priority.  Yes, the ability to play at high volume is a criterion of performance, but, the ability to sound dynamic, full and exciting at low volume is more important to me.  I know one can show that classical music has an enormous dynamic range such that the peaks can be well over 100 db.  But, recordings don't come close to delivering the full dynamic range; if you set the peaks at 100 db, the quiet passages are unrealistically loud and inappropriate.  I don't ever play such music at "realistic" peak volume.  For amplified music, I don't ever want to attempt to duplicate a live performance; frankly, if my system only sounded as good as perfectly reproducing such performances, I would dump it.

The very best systems at sounding good at low volume levels also tend to ones that are highly efficient and easy to drive.  That allows the use of low-powered amps.  That is a really big deal to me because ALL of the amps I really like happen to be quite low in output (the most powerful is a custom-built OTL amp at 30 watts).  I haven't heard most of the high-powered candidates out there, but, I've heard a few and they have the same problem--they sound dull and lifeless and un-engaging, particularly if you don't crank up the volume.  

@larryi   +1 on your post. 

Question regarding:

That allows the use of low-powered amps. That is a really big deal to me because ALL of the amps I really like happen to be quite low in output

Are you primarily referring to tubed amplification? If so, does the same apply to Solid State amplification? Thank you.

David,

The amps I have liked the most are tube amps.  I own a custom-built push-pull 45 amp, the Audio Note (uk) Kageki (parallel 2a3 SET) and a custom-built push-pull 349 amp.  The Kageki, rated at 6.5 watts/channel, is the most powerful of the trio.  But, I had in my system a solid state First Watt J2 for about two weeks (borrowed from a friend).  The J2 sounded very good--it was vivid and engaging and only a touch hard on the initial attack of notes. I could easily live with the J2.  That same friend built for himself a low-powered SIT amp based on schematics made available by First Watt.  I think that this is also an excellent low-powered solid state amp. 

In the past, the common complaint raised against solid state is that the sound tended to be harsh and "grainy."  I don't think that is the case these days.  Most solid state amps are smooth sounding and are not at all unpleasantly harsh or shrill.  To me, the complaint about harsh and shrill can more often be leveled against high-powered tube gear than solid state gear.  But, when even decent sounding solid state amps are played at the lower levels I prefer, they sound a bit lifeless, and I tend to lose interest.  A friend of mine, who also loves low-powered tube gear, and who has heard much more stuff than I have (he goes to the Munich show), says that the Bridge Audio Laboratories (BAlabo) high-powered amps sound good.  I have not heard these amps, and in any case, I could never afford to buy amps that cost six figures.

@gdhal

The mathematics you use are correct. However speakers do not behave linearly after a certain point. Worse speakers distort terribly. That point is surprisingly low. Soundstage do NOT test speakers above 95 dB SPL as a level above that will damage most speakers!! (Soundstage conclusions and not my opinion).

Drivers have a limited excursion where they are fairly linear (Xmax). Driver voice coils get extremely hot and that causes significant compression (non linearity). There are a huge amount of challenges for high fidelity (no added distortion) at rock concert levels.

My speakers play at 115 dB SPL linearly even at a continuous level with similar low distortion to playing at modest 85 dB levels. This is extremely rare in a speaker and requires large woofers, large drive motors on the drivers (huuuge magnets), short voice coils in a large gap (to preserve linearity),large diameter voice coils (for better cooling), extremely tight tolerances (better cooling) and lots of clean power - in short a huge amount of engineering is required to achieve this over a speaker that is designed to only play up to 95 dB SPL before starting to distort heavily (the majority of designs). To achieve this performance requires specific engineering that you won’t find in 99.9% of home audio.

Since our ears and brain interprete distortion as loudness most audiophiles think they have speakers that play extremely loud however a dB meter will confirm to them that what they think is loud is actually just huge amounts of distortion giving the appearance of loudness.

 audiophiles think they have speakers that play extremely loud however a dB meter will confirm to them that what they think is loud is actually just huge amounts of distortion giving the appearance of loudness

You make a great argument for horn design many sound amazing at low levels but have extremely low distortion as SPL increases