Good sound in reproduced music comes from an intelligent and well executed initial design. Often this crucial first step is given a cursory effort with the idea that all can be fixed at a later stage by throwing technological electronic wonder circuits as the final link in the chain to fix up all that is ailing. Time and time again, this essentially negative feedback loop has been shown not to work. It is the rare case where not doing it right to start with results in a better outcome.
One of the reasons that the simplest audio systems can often command the highest prices is because all the engineering and build expense have gone into the front end of the process. Quality parts are not cheap, but, since they are hidden from the users view, many manufacturers will skimp on them, preferring to put the money into the faceplate of the product and relying on electronic wizardry to 'fix up' what is flawed to start with. For instance, power supplies are a critical part of all audio equipment and one of the most critical when it comes to amps and preamps. Good power supply design requires top-notch transformers and smart use of their outputs, but users will rarely see this part. The hidden power supply can easily be compromised and the use of following circuits will give a competent facade to an otherwise weak design.
Negative feedback is often used in the output stages of amplifiers to help smooth out the output to better match the input. Once again, this is a cheap way to buy better looking specificiations. By its very nature negative feedback will always be chasing its own tail. The output signal is compared to the input and adjustments are essentially made on the fly to tweak it into shape. An example from the age of turntables illustrates this point very clearly. In an effort to produce turntables with better paper specs, manufacturers would put servo feedback controls in the tables to monitor the speed of the platter. Basically, if the platter were turning a little fast, the servo would slow it down until it was too slow, then the servo would speed it up. The principle is predicated on there being a state of error at all times. The result was turntables with average wow and flutter (speed stability) figures of remarkable precision. It was certainly true that these tables could average 33.3333333333 with errors measured in the 1000ths of a percent. But, more important to the actual sound of the turntable was that, while the average was spot on, the instantaneous speed measurement was almost never on. Music is not a function of getting the average right over time; getting it right at each instant is far more important. Would you want to listen to an instrument that was, on average, in perfect tune, or one that is just in tune? This constant point to point speed variation is what kept these tables from having solid stereo imaging and accurate spatial placement. A classic example of letting a paper specification overstate its importance, resulting in an inferior actual product.
A modern day example is the use of massive digital processing to allow a speaker/amplifier combination to "listen" to the actual playback environment and provide an inverse processing curve to adjust out all the actual listening space anomalies. This sounds like a great idea and some positive experiences with the concept seem to be cropping up. The problem with this system (as with all systems of this nature) is that, rather than producing a high fidelity playback device to start with, the effort will be spent on a post playback fix. All this processing is the very antithesis of a straight wire with gain, the holy grail of audio reproduction. This is the failure of surround sound as a high fidelity medium. Note I said as a high fidelity medium, not an effects medium, at which it does a great job. Two different purposes with different needs; fidelity is not the focus of the latter and certainly not its strong suit. Just as with food, starting with good ingredients nets a better tasting product than crap covered with flavoring.
As music becomes increasingly over produced and over processed there is less and less fidelity on which to base equipment choices and design. The same things that ruin playback gear ruin recorded music. On-the-fly pitch correction for singers and auto-tuning processors are two things that immediately come to mind. If the source becomes that flawed, there is less pressure to make distinguished sounding playback equipment, and the ability to hear why one would bother becomes more difficult and less important as well. In an ironic twist, the entire process (from music creation to playback) becomes a negative feedback loop with an end result that is far less than the sum of the parts and, even more depressing, far less than what we are capable of considering all the available resources that truly advance the state of the art and not just the state of the marketing.
Listen clearly, choose carefully.
© Cadence Magazine 2006