There are several aspects that still remain a mistery for the non-tech audiophile crowd. For example, the way cables can affect the musical performance. I know this is a dead-old issue but I still get many e-mail messages from audiophiles who ask some technical explaination on this topic. Can you shed some (technical) light on the topic?
Cables are by all means a piece of electronic equipment. It has electrical characteristic which can change the input waveform and, therefore, is a transfer function.
For example, resistance would attenuate, the capacitance would roll off the treble and at the same time add a *pole* to the output stage of the amplifier, sometimes causing instability.
There are also skin effect (though allegedly insignificant in audible frequency), leakage current (concerning the properties of the dielectric material) and things like electro-magnetic interference to consider.
Based on the above argument, I think I can conclude that cables DO make a difference and that it is a *black art* based entirely on science and technology.
The only problem is, of course, technology is still very much limited nowadays.
Another old debate is whether or not the electrical measures can describe the way an HiFi component sounds.
For some components, for example, CD players, it seems difficult to establish a close relationship between measures and listening results.
Which is you point of view?
This is a very good question. Let me explain it by an example again.
Take the frequency sweep measurement for example. It does tell if the piece of equipment in concern is tonally balanced or not but it certainly does not describe the phase nor the harmonic structure of it because it is only measuring the amplitude.
It is like, listening to a flat response of a violin without knowing the subsequent harmonics, then why don't you listen to the sine wave of that note?
The main point, and it is a point I totally agree with a famous loudspeaker designer, is that we have to find the correct measurement method, or the correct combination of measurements to accurately describe the perceptible sound quality.
This, of course, is out of my reach and I suppose some engineers and psychoacoustic researchers will have to pick it up from here!
During the last decades, solid state HiFi amplifiers have been improved with new sophisticated components and designs but it seems that a large part of the audiophile crowd prefers tube-based amplifiers.
According to you, do tubes have some sonic advantage?
Yes, silicon technology (now more than silicon, actually) has improved but most of the design topology is unchanged. Most importantly, the MOSFETs still follow a square law and the BJTs exponential, neither of them is linear.
Engineers cheat by using small signal models in designing amplifiers and IC which works like a wonder at low voltage swing.
But with a high voltage swing at the output stage of a power amplifier, the assumptions of the model break down. The *gain* is no longer linear and that leads to harmonic distortions.
This is quite embarrassing (a laugh) because I do specialize in semiconductor devices... Back to the tubes.
Although the current driving capability is lower, valves are significantly more linear when properly biased; the I-V curve of the famous 300B is nearly a straight line. Think of it, then valves can be thought of as being more *hi-fidelity* than solid state devices in this particular application of audio, and there is little wonder why a lot of people, including myself, are predisposed to the sound of valves.
Another sad truth is that analogue solid state components are now optimized for digital application and that leads to many other designing problems.
For loudspeakers it seems that the usual measurements of frequency response, time response and in-off-axis response don't tell much about the REAL performance of the loudspeaker.
Which kind of measures do you suggest to fill this gap?
I cannot consider myself an expert in instrumentation nor even know much about psychoacoustic, so I think I will give this a 'technical' pass.
However, I think the best thing to do now is to have more reviews in magazines, so that readers can appreciate roughly what a piece of equipment sounds like.
This, of course, depend on the ability and experience of the reviewers.
How do you consider the new digital media, DVD, SuperCD etc?
I would say DVD or the 24-bit, 96kHz standard is the way to go. Higher standard, of course, would be easy to set but difficult to produce economically with the present technology. Technically, the 24/96 standard would lead to smaller quantisation error, larger bandwidth and improvement in many other area like spatial resolution (the *stereo feeling*).
It has already been proved that the aforementioned standard DO sound better than the current flawed CD standard. This standard is taking off already but it will take some time to mature, just like any technology in any field other than audio.
My secret desire is a Krell or Mark Levison high-end DVD player which can be infinitely upgradable and has a very properly designed analogue output stage...
As a technical reviewer and electronic engineer how do you consider the plain simple listening tests where it is the ear of the reviewer that judges everything?
I think my answers above have already clarified my point slightly.
Technology nowadays is not high enough to give a direct correlation between electrical characteristics and the subjective sound quality that human hear. So, the only way to go is to rely on the trained ears of the good reviewers, but let me emphasize the word: good, rather than the word *reviewer*.
To elaborate the meaning of good would take ages, so to simplify, the *good reviewer* should know what the truth is (like the sound of, say, a real violin) and can decide how true and faithful the presentation of a piece of equipment is to the originals.
How do you consider the lively DIY and tweaking scene?
Why not?! I am a seasoned tweaker myself! Cost constraint does prevent the manufacturers from using truly high grade components sometimes, in order to stay ahead of the competition.
A well-known Managing Director of a British hifi company has once said to me that a pound in the production can lead to a 10 pounds increase in the selling price (sorry about the currency!).
However, there is nothing wrong when we can get a 10 pounds improvement by swapping a 1 pound component, right? By all mean, high fidelity does mean *high accuracy*, and without highly specified components, this idea can never be achieved.
Courtesy Leo Lam for TNT.
Copyright © 1998 Lucio Cadeddu