The true test...

A recent edition of Engineering and Technology, the magazine of The Institution of Engineering and Technology, features a letter from Nick Arran, CEng. Basically Mr Arran was interested to read that loudspeaker manufacturer B&W had said that they trusted the ears of their master listener more than they trusted a spectrum analyser when they were designing equipment for the new Jaguar XF.

Mr Arran went on to question what he thinks is the Hi-Fi community’s ‘belief’ that the human ear and brain are a better measurement tool than any other available – including laboratory test equipment. He went on to say – with, I believe, more than just a hint of irony – how surprised he was that commonly accepted Hi-Fi practise hasn’t filtered through to engineering instrumentation – such as using Hi-Fi mains cables on testing equipment, specialist interconnect cables, supports and so on. His letter ends by questioning whether engineers, with their calibrated test equipment were missing something or whether the Hi-Fi consumers’ legs were ‘being pulled’.

This is a way of thinking that prevents progress and improvement. In its most extreme expression it says, “if we can’t measure it – it can’t exist”. The assumption being that we can measure everything of any importance. I believe, in fact, that we can measure very little of importance to the realistic reproduction of music.

At its root, conventional measurement is based on simple, steady state signals to quantify frequency response, harmonic distortion, signal to noise ratios etc. On the face of it, this approach produces quantifiable, repeatable, measurements of a range of important parameters. In reality all they do is tell you whether the device under test is working or faulty. They tell you nothing about how well the device handles music signals to deliver enjoyable and realistic music. In other words, they are not predictive and if they cannot describe exactly how something will sound, they are useless for that purpose.

I own and use lots of test equipment, but I don’t try and use it to tell me how good or bad something sounds. I use my ears for that. They are educated, trained and very discerning. They are consistent, predictable and I trust them.

Until we have a measurement system, based on dynamic, complex, real world, real time signals (like music!), we have no choice but to base our assessments on listening tests. To do that, we must trust our senses and know that what we hear is real world and not imaginary.