BILL DYER: The interview

Howard Popeck:

Hmm. Oh dear – and oh well! This interview shown below this introduction was rejected for publication by a UK Audio magazine. This is pity, but that's life. The reasons are far from clear but somehow I guess that there was a historic clash of personalities between the interviewee and some anonymous 'suit' in the magazine's management hierarchy.

It’s the golden rule of publishing at work. This rule’s not printed anywhere – but all the staff of every publication are well aware of it. Quite simply – the golden rule is that the man with the gold makes the rules.

This begs the question – what is the purpose of the magazines? Well first and foremost, to make money. Okay, fair enough. But what then? You’d like to think, now wouldn’t you, that they are there to inform, to educate and to entertain? Err, yes of course you would. And so would I. However in this instance I would venture to suggest that the publication in question is both an extension of an over-ripe ego and more poignantly a public statement of the owner’s irrationality, paranoia, fear and of course negative predudice.

As for me, well of course I also exhibit predudice – both positive AND negative. But I'm not asking anyone to pay anything for the contents of this site. In contrast, the magazine charges real money each month. To me that means that ego should take a back seat.

So what can any of us make of this? Well for me I guess that the signs are becoming clear in that the days of the printed pages are numbered. Take online audio mags for example. They carry little advertising – and have no need to. They are written and edited by the enthusiasts rather than the faceless ‘suits’ – and they embody fresh thinking.

Fortunately I neither wear a suit nor earn my living from this industry. Thus I am immune to bribery, coercion and intimidation and happily can print whatever I like and whenever I like - notwithstanding the need for my comments to be legal, decent and truthful. Anyway, sermon over. Meanwhile, please read on …….

--------//---------

Iconoclast, rebel, audio anarchist and for some, the audio antichrist Mr. Bill Dyer is interviewed by Howard Popeck. HP, natural sceptical wanted to investigate some extraordinary claims by this pleasantly eccentric UK audio engineer. Having heard BD’s active Hailey loudspeakers his scepticism was confounded. HP believes Mr. Dyer is really on to something.

In one sentence, how would describe your ethos?

Simple. Most of the recorded audio programme material that is generally available is recorded in two channel mono, or biphonic not in stereo. The public are being mislead.

Hmm, a bit controversial?

Maybe. Actually no, it isn’t. Not at all. My views are far from controversial for thinkers who focus on recreating the original sound rather than just talking about it. But it’s true that I seem to contradict what most people in the sound industry accept as perfectly normal, that all recordings classed as stereo are in fact stereo. However my statement is true.

You like to ruffle a few feathers then?

What?

Rattle the cage?

Oh I see. No – I don’t go out of my way to do so. But there is so much blind ignorance, so much conjecture without objective analysis, so much nonsense that it’s time for someone to speak out. If not me, then who else?

Okay, fine. So what is stereo – in your language then?

The term stereo is derived from Greek meaning solid. It does not mean two channels. Many recordings of programme material may be labelled as stereo but is not true, stereo is also used to describe equipment which is completely meaningless. Did you or your readers know that nearly all pop/rock recordings are recorded in two channel monophonic, classical music recordings are normally recorded in stereophonic as well as some Jazz recordings.

What does that mean in practice?

To record stereo we have to capture the same phase difference that we perceive by our two ears. This phase information is the difference in level of the signal we hear when we listen to a sound emanating from a sound source. The very small difference in the distance that the signals have to travel to reach each of our ears is the delay that causes a slight phase change in the information that we hear.

This gives us the ability to locate the direction of the source of the sound very precisely in both the vertical, horizontal and lateral positions. We can locate an aircraft flying overhead, the position of a car approaching us.

It also allows us to pick out a particular instrument in an orchestra with great precession, and gives us the perception of spaciousness or ambiance.

More please!

Okay. We together with most animals we have an acute sense of direction, If our ancestors were not blessed with the ability to locate the precise direction from which a large carnivorous beast that was chasing them was coming from, or conversely if they were chasing a animal that would provide them with their next meal and they did not run towards it, Homo sapiens would have become extinct and gone the way of the Dodo.

Is that it?

No of course not! Look, ambiance or reverberation gives us the feeling of spaciousness we sense depending on the environment we are in. It varies depending on the size and the acoustic qualities of the room. If one is blindfolded and moved from room to room or outside the building, the different in the ambiance that we perceive is very obvious. Larger rooms have longer reverberation times. Ambiance is caused by the reflection of sound from the myriad of surfaces in the environment we are in.

Which means what exactly?

Concentrate please! The more lively the room, the greater the amount of ambiance. Unfortunately most recording studios have very absorbent surfaces to eliminate ambiance. Pop/rock studios are normally quite small and made as absorbent as possible. This is achieved by using literally tons of glass fibre. Anechoic chambers are rooms that are designed to have no ambiance at all, The ambiance of the recording venue can be captured by using stereo microphones. This ambiance will be reproduced if the loudspeakers are capable of reproducing stereo. Loudspeakers that cannot resolve the stereo signal will not reproduce ambiance. To experience real Ambisonics, the loudspeakers we use must be capable of reproducing stereo.

How does this relate to the recordings we use daily?

The following is a very simple explanation of the difference between two channel mono and stereo recording. Pop/rock music is recorded using mono microphones. For instance, there may be a number of microphones used to record a drum kit used in a pop recording, but they are single microphones that are designed to pickup mono signals from each instrument, processed through a mono microphone channel.

A panorama control (pan pot) which is a voltage level control device allows one to alter the voltage of a signal and send it to two mono outputs. The signal is positioned left, centre or right or anywhere in between….The output signal is still MONO emanating from two channels. To record STEREO, stereo microphone channels are required. All the microphones must be stereo pairs which means either two separate microphones or a single device with three or more capsules such as a Soundfield or other Ambisonic type of microphone setup.

In a nutshell then?

Forget nutshells please! But for brevity, panning a mono signal to Left Centre or Right cannot create stereo. The phase information cannot be created if it was not recorded in the first place!

You’ve used the phrase "cocktail party phenomena“ during lectures. What’s that all about?

Nice research Mr. Popeck! At a cocktail party where everyone seems to be talking at the same time, we can very easily listen to a voice of less amplitude, than that of the voices in the immediate vicinity. This is because we can use phase information to isolate and listen to the voice of the people we wish to hear. We have the ability to completely ignore the louder voices.

That is the same way that we isolate and listen to a particular instrument in a concert hall. frequency response is not of prime importance.

Okay, got that? Right, then moving on – the reproduction of true stereo manifests itself by creating the illusion of being at the venue where the sound is being produced. If the band is on a stage then the sound comes from above floor level. If the recording is of a thunder storm the sound seems to come from above the ceiling of the listening room and not from the speakers.

The reproduction of the sound of a large pipe organ should reproduce the long reverberation time of the large concert hall or the cathedral in which the recording was made, and not the ambiance of the listening room. Because stereo recordings are more natural, the reproduction of the human voice creates the illusion of the performer being made of flesh and blood and in the room with the listener. One gets a truly three-dimensional picture of the performer’s positions when recording: the flautists are sitting slightly to the left of the oboes: the cellists are to the right of the violas and in front of that double-basses and so on.

Hmm. Okay. I think I understand. So what about the so called ‘sweet spot’?

The other benefit with my approach is that loudspeaker positioning is not critical. There is no “Sweet spot”. None at all. Well …. none unless you feel that you must sit on the axis of the loudspeakers. You can place the speakers facing forward. The traditional 60° positioning is not necessary, because my loudspeakers reproduce ambiance. That’s it!

Any thoughts re UHJ Ambisonic?

You don’t look old enough to remember UHJ! Okay, in a nutshell – the pleasure of listening to a good stereo recording of music can give one a great deal of pleasure, only exceed by listening to an UHJ encoded Ambisonic recording reproduced through four loudspeakers reproducing four channel stereo complete with ambiance, there are other forms of Ambisonic recordings, but that is another story.

Any downsides?

Certainly. The down side is that the limitations of a mono sound source becomes immediately obvious and curtails ones pleasure by having to listen in mono. It would have given one much more pleasure if it had been recorded in stereo.

How do you view the relationship, if any, between phase and detail?

Loudspeakers that possess the resolution necessary to reproduce stereo will also reveal more of the finer detail contained in the original signal. Instruments that are masked by other loudspeakers are now clearly audible. Unfortunately many of the loudspeakers that are incapable of resolving stereo phase information will mask much of the detail as well.

Where does the blame lay?

Blame? I tell you where! To me it seems that the main culprits are complex passive crossovers.

Yes, but surely that’s axiomatic with the desire for a flat response?

The obsession of obtaining an extended and absolutely flat frequency response at all cost, using elaborate equalisation either analogue and more recently digital to the detriment of all other parameters, such as transient response, elimination of energy storage and quick settling time i.e. the ability to stop when the signal stops is the primary cause and of course the tremendous distortion produced by many loudspeakers.

More please!

Unfortunately many loudspeakers suffer from a lot of inertia. They have very poor rise times and correspondingly slow decay. This tendency to store energy and release it slowly produces the characteristic boomy bass that is produced by many sub bass loudspeakers. This distortion produces the masking of the high frequent instruments by the low frequency distortion. The simpler the recording and reproduction process is the truer the reproduction.

What about frequency and amplitude response?

The frequency and amplitude response should be sufficient to preserve the original balance, considering our comparative lack of sensitivity and our control over what we chose to hear when it comes to amplitude response.

Is that of paramount importance?

No – but it is comparatively easy to achieve and measure. If you have an inherently accurate loudspeaker, extending the frequency response may be of benefit, only if the other parameters are not degraded in the process. To design a loudspeaker that is accurate enough to reproduce stereo, the enclosure, electronic crossover, amplifiers and transducers must be very carefully designed to work together and compliment each other, the end result is that the loudspeaker will have a reasonably flat response and sufficiently wide bandwidth.

I think the readers will understand the theory. But what about the practical reality?

An A / B comparison between a pair of loudspeakers capable of reproducing stereo and other loudspeakers will reveal the huge differences between them. This is easily perceived even by the comparatively untrained listener. What surprises me is, that there are so few people who seem to be aware of the difference, I can’t recall having seen any mention of this made in loudspeaker or record reviews, some records which are mono are presumed to be stereo because of the label.

Are ears of more use that measuring audio equipment?

I think the question could have been phrased with more care, but anyway – to measure the performance of audio equipment the signals that are normally use are sine and square waves. These wave forms are artificially generated and do not occur in nature. They have very limited use when it comes to assessing the performance of audio equipment designed to reproduce programme material. These waveforms are useful if one needs to measure the maximum power that can be dissipated in a load, but this does not give a true indication of the level at which the amplifier will clip with music or speech.

Because?

Because continuous wave forms can be useful to measure frequency response and when designing active crossovers and tone controls. But when it comes to distortion measurements the information obtained is completely useless.

Useless?

That’s precisely what I said. Programme material consists of transient material, not continuous wave forms and that is why amplifier measurements can be very similar but the sound reproduction very different. The same is true of loudspeaker response curves by which I mean that the dynamic distortion produced by loudspeakers is very much higher than the static distortion. This distortion will effect any frequency response curves that are measured using sign waves or pink noise, because the distortion product will be added to the signal.

So …..?

To get a true indication of loudspeaker or electronic performance pulse testing should be employed.

So what led you to this conclusion?

In 1972 I had reason to find out why a very popular amplifier which had very good specifications sounded so awful. Using continuous waves did not reveal any defects, so I used a tone burst of 10kHz. I increased the length of the tone burst gradually with increments of 100 millisecond till I reached the point at which the negative feedback reduced the level of the input signal, this was 900 milliseconds. In programme terms this is an eternity.

Can you imagine that if the sound of a bass drum passes through the amplifier and 0.9 of a second later there is the sound of a cymbal, the cymbal will be modified by the negative feedback generated by the sound of the drum.

The output will bear no resemblance to the sound of a cymbal. It will be distorted out of all recognition. Yet this amplifier was used by many TV stations, Radio Broadcast and Recording Studios.

Thank you Mr. Dyer. Any last words?

You make it sound like a prisoner before imminent execution. Only kidding! My company, Digital Audio Systems Ltd market a range of High Definition loudspeakers that reproduce real stereo with width, depth and height . That’s it. and thank you.

Leave a Reply