This subject comes up periodically on the forums and recently on this blog – the subject being how can modern music be used to test the sound of a sound system – and are we not really dependent on comparing what unamplified acoustical instruments sound like on our systems versus what they sound like in reality?

I.E. do we need a reference ‘perfect’ sound with which to determine just how much fidelity our high-fidelity reproduction systems have?

Definitely an aesthetically pleasing argument. But does it hold up? Is this the ONLY way to ascertain fidelity of a system?

Personally, I find it too limiting.

First, one can argue that there is no ‘perfect’ sound – that our recordings of, say, a piano are so very far from sounding like a real piano and that we cannot ask our systems to sound like a piano if the source material does not sound like a piano. A corollary to this is that 99% of people have no idea what a piano sounds like. And much fewer what the particular piano sounded like that is being recorded – in that particular hall – at the location of the particular microphone(s).

Second, that many people cannot tell the difference between a modest stereo system’s reproduction of a live band and the live band itself. I think Dunlavy used to hold these demos [using very modest albeit dynamic systems] in their factory. [I am very familiar what an acoustical guitar sounds like from 2 to, say, 10 feet away. I can be ‘fooled’ sometimes because my ear will focus on a limited number of cues: the note attack and note decay and note harmonics (of the guitar body and other strings, as well as buzzing against the pick and or frets, etc) ].

Third, music like Radiohead is recorded better and has a higher fidelity to the original music because so much of the original music is electronically generated – this is its natural medium. It doesn’t have to go from vibrations to sound to microphone to weak electrical signals. Radiohead is great for testing – and fairly unique – because it is both complex and recorded very very well. Classical music, when complex, often has issues because the instruments are so spread out they are hard to record with a single microphone – and multiple microphones introduce problems, etc. etc.

So a conclusion here, it seems , is that one has to pay freaking close attention to the sound of the reproduction even if it is purportedly of an acoustical instrument. Are the notes rendered well, how does the music make us feel, etc. And we also compare to other music we have heard on this system as well as this same music we have heard on other systems as well as other music by the same band as well as how other sources (LP, CD, Tape, SACD, etc.) of this same music has sounded in the past.

It is all about triangulating and interpolating and comparing and re-comparing sound to other sounds – and whether that sound is an acoustic instrument or a electronic keyboard or a voice or whatever – it hardly matters IMHO.

OK. My usual argument went like this: very, very few people give a darn whether their music is from an acoustical instrument or not – they just care about the MUSIC and THAT is what has to sound good and have high-fidelity. The question of whether a piano can be reproduced well, and whether reproducing a piano well means it can also reproduce Hendrix’s guitar well, is fine to debate but one could just as well shirt-circuit the whole mess and just listen to a recording of Hendrix’s guitar. I.E. one listens to music one likes and if that sounds correct, with respect to memories of live performances and memories of similar music on the system and similar systems, then one is happy happy.