I noticed a peculiar thing the other day.  At least, I thought it was peculiar at the time.  Let me tell you all about it.

One of the tricks you can easily use to fine-tune the positioning of your loudspeakers is to move your head instead.  Moving your head from side-to-side, from back-to-front, and up-and-down, you can listen to how dependent your system’s sonic balance and imaging are with regard to listening position.  These things are not so much governed by your system as by your system’s interaction with your listening room.  This is why identical systems can sound radically different when installed in different listening rooms.

Changes in sound balance as you move your head are usually caused by the inescapable fact that sound generated by a loudspeaker driver is not uniformly distributed into the room.  Most of it propagates straight ahead out of the loudspeaker, but as you move away from the straight-ahead position the output starts to fall off.  Complicating this behaviour is the fact that as the frequency rises, so this off-centre drop-off gets progressively worse.  We refer to this phenomenon using the term ‘dispersion’, and it is a natural consequence of the fact that the loudspeaker driver is not infinitely small.  One consequence of this dispersion is that the listener’s perceived frequency balance will depend to some extent on where in the room he is located.

The perception of a good ‘holographic’ spatial image is a much more complex matter, and, if we are honest about it, is not fully understood.  The spatial image is a construct that our brains create for us, rather than a specific property of the system, and so is very much a matter that dwells within the realm of psychoacoustics.  Having said that, there are a number of things that we do know to have a positive impact on a system’s ability to generate a holographic spatial image.  Chief among those is timing coherence.  It seems that the more extreme the measures taken to improve timing coherence, the better the imaging we end up with.  The real problem arises because we cannot actually measure this ‘timing coherence’ at all.  Frankly, we can only wave our arms in attempting to define what it actually is.

The best way to rationalize timing coherence is to think of a loudspeaker.  Modern speaker design theory takes great pains to minimize cabinet resonance.  These days even the most budget-friendly designs from the better manufacturers have non-resonant cabinets that respond with a dead thud when you rap them with your knuckles, a property that was evident only on the best of high-end designs as little as 20 years ago.  A resonant cabinet will store energy and release it as sound waves a faction of a second later.  This, after all, is what you hear when rapping a cabinet with your knuckles produces a distinctive sound.  By contrast, rapping the cabinets of my B&W 802 Diamonds produces nothing more than sore knuckles.

Understanding these concepts in loudspeakers is quite simple, but extending them to electronic components is less so.  Even so, some concepts are well understood.  Removing capacitors from the signal path is one such example.  Mechanically isolating the chassis, less so.  But if you get the chance to listen to Nordost’s Sort Füts and Sort Kones it can be very instructive.

Anyway, all this is to say that if your audio components are well designed they can generate that holographic sense of image that many of us crave from our systems.  But you still need to set the system up correctly in the listening room in order to make it happen.  This is because the sound that reaches the listening position is a composite of direct sound and a combination of different reflected sounds.  If you ever get to hear a high-end loudspeaker inside an anechoic chamber – which I recognize very few of you ever will – you would be amazed as to how awful it sounds.  It will sound so dry you’ll need to take a bottle of water in there with you.  When you come out, you’ll feel like you have cotton wool in your ears.  So it is important to recognize the dominant effect of the room interaction on how your system actually sounds.

It also explains how the concept of a ‘sweet spot’ actually arises.  There is usually only one place in your listening room where the combination of direct and reflected sound comes together to generate the optimum image.  When you set up your listening room, your challenge is to make it such that this optimum spot coincides with where you place your listening chair.  Usually, when things are close to ideal, the optimum spot will move with the loudspeakers, so if it is two feet in front of your listening chair, you can correct the situation by moving the speakers two feet forward.  Or you could just move your chair.

I have one last observation to make here, and it is quite an important one.  Think about your listening chair.  If it has a high back, then reflections off the back of the chair will tend to dominate the sound field, and you may find that regardless of where you position it, you just don’t get a good image.  In general, you should always strive to use a listening chair with a low back.  In my own listening room, therefore, I have a rather stylish Italian white leather sectional sofa with a low back that comes below my shoulder line.

When a new component comes along which makes a significant change to your system, such as my new DirectStream DAC, its contribution may be such as to require a reassessment of where that optimal listening position is located.  It is quite an easy process – or at least it should be.  Sitting in your favourite listening position, you move your head from side-to-side, then back-and-forth, and finally up-and-down, until you locate the new optimal position.  You then adjust your speaker position, and/or move your listening chair, to correct for the offset.

It should be easy, but in my case it has proven not to be so.  You see, regardless of the adjustments I make, the optimum position is always about 10-12 inches higher than where I am sitting.  I have come to realize that the culprit is my much-loved sofa.  Even though its back doesn’t even come up to my shoulders, it appears that it still manages to contribute significant reflections up from its seat cushions.  Also, as I sit on it with my palms lightly touching the seat cushions, I can plainly pick up vibrations from the leather surface.  These are not at all evident if I instead place my hands on fabric cushions.  Right now I have co-opted a pair of seat cushions from another of my sofas to raise my listening position by about 10 inches.  It will do for some listening tests, but of course I now have no back support whatsoever. I have a bad back, so that is not the basis of a long-term solution.

So is my problem down to reflections from the leather surfaces, or re-radiation from the vibrating surface?  I am working on the notion of the former, because reflections tend to disrupt imaging, whereas vibrations tend to disrupt tonal neutrality, and in any case are surely too heavily damped.  For reasons of practicality (and in the interests of sustaining a 36-year marriage which is worth more than my stereo) the sofa needs to stay.  I am contemplating a solution to damp the source of these refections by judicious placement of an absorptive panel on the ceiling above the sofa.  Last year I placed one on the ceiling above and between my speakers to great effect, so I am thinking along the lines of something similar.

Meanwhile, I plan to experiment with covering the sofa’s leather surfaces with some absorptive material just to see what that does.  Such are the joys of the high end.  Your system and your room are like two top drivers on the same Formula One team.  Getting them to cooperate can be a challenge.