Interview with Quad's Peter Walker in 1978

[DSJR]

Actually Marco, my samples of II's and 303 actually sound pretty darned similar - into the BC2's anyway. I stick with the Quads downstairs because of the lovely warm glow from the KT66's, the warm glow from the 303 power light isn't quite the same somehow....


Mind you, I'm seeing and hearing the old bridged D-60's in a whole new light nowadays with my sooper-dooper new mega-foo interconnects

[Marco]

Lol - fair enough, Dave. Any 303s and Quad IIs I've heard sound markedly different, in the way you'd expect valve and SS gear to do. For starters, 303s, to my ears, don't sound quite as flat and lifeless, or indeed, 'pipe & slippers'!

Are those Mark Grant "mega-foo" interconnects?

Marco.

[Reid Malenfant]

Dave, you may note that there is a capacitor on the output of the 303 & in all honesty it really isn't very large in the capacitance stakes Ok, so the 303 may well have a bit more "damping factor" or should we call this output impedance as it varies over frequency, but in all honesty both amps lack a great deal of control that others take for granted

What is that cap 2,000uf? In order to be more out of the picture it should have been 50,000uf as the bass response rolls off below 40Hz even with an 8 ohm load Might be ok with a sealed box with say a 50Hz resonance but show it a reflex with a tuning frequency lower than 40Hz & you might as well be a hippo wallowing

[Marco]

Dave, you may note that there is a capacitor on the output of the 303 & in all honesty it really isn't very large in the capacitance stakes Ok, so the 303 may well have a bit more "damping factor" or should we call this output impedance as it varies over frequency, but in all honesty both amps lack a great deal of control that others take for granted

Indeed! And especially in the case of the Quad IIs, no way are they "technically faithful to the input", as if that's how music is supposed to sound, I'm Angelina Jolie in a pink tutu!!

Marco.

[colinB]

From what i understand the engineers do listening tests but they are listening to their own tastes in music which is why some componenets are known for being good with certain music and not others.

[Marco]

Indeed, Colin. Of those who actually listen to the equipment they build, it will be voiced using the designer's taste in music as the benchmark, hence indeed why some kit excels more with one genre of music than others.

I prefer equipment though which has been designed and voiced to excel with all types of music - and it can be done!

Marco.

[Marco]

Some further thoughts on this, Dave:

Actually Marco, *my* Quad 303 and *my* II's actually sound pretty darned similar - into the BC2's anyway.

Yes, but haven't your Quad IIs been modified by Glenn Croft? I'm sure that I've seen you mention that they're been "Crofted". And if so, that's likely why they sound punchier, and somewhat 'livelier', and more akin to the sonic presentation of the 303.

Have you ever actually compared stock Quad IIs to the 303?

If you have, and you actually think that they sound "pretty darned similar", then our respective experiences are rather different!

Marco.

[Marco]

I'll get to it later

Marco.

[DSJR]

The problem for me is, all the other Quad II's in stock form I've heard have been with the truly atrocious 22 preamp, which in standard form smothered fine details with a huge thick blanket - unbelievable! At least "my" II's have a chance to breathe..

Now, and I hope Mark Grant will read this (and I know Alex and Dave W will back me up here on this), my "problems" with my HD1500 cables (neutrik pro phono to 1/4" jack, so can't be easily reversed) were to do with a certain perceived "stridency" in the sound. I think I can say now that it's NOT this cable, but an additive factor of the heavy-duty CD to preamp interconnect which still had a slight possible "enhancement" of upper-range details, coupled with a preamp that doesn't sound in the least bit "warm and valvey," despite the wonderful atmosphere it brings to suitably recorded music, a pair of lean and possibly "snappy" amps (damping factor of at least 400 even in bridged mode...) and the BC2's, which are crisper in balance than either the BC1's as I remember them, or Alex's rather lovely SP1's for that matter.

The first thing I did was to resurrect the supplied Micro-Seiki interconnect, re-fitting their phono's at the amp end. This gently toned things from CD down a touch. The second thing I've done is to make up some fresh cables with the Van Damme mic cable Dave gave me tp go between pre and power amps. The gauge of the conductors looks similar to the Pro-patch which works so well as a tonearm to amp interconnect, but the outer jacket is a double-thickness flexible "stuff" which makes it look a bit posher yet keeps it flexible.

The end result of the above is that the sound retains the atmosphere with the D-60's/Croft combination, yet is just a touch more "civilised" than with the MG cables. Had I Alex's SP1's (Anthony, I haven't forgotten ), the MG cable would be fine, as Alex's samples were at our last get-together, but for the more wayward BC2's , I'm delighted with what I have at present - Siouxsie's recordings haven't sounded as good since I heard them on my mates old bolt-up Naim and 'Brik active system..


Mark, wasn't the 303 designed to sonically mimic the II's in many ways? I believe the output caps were the largest they could get back then (any bigger would have been too large and probably too expensive then as well. Mine have Rifa caps as replacements, but I think the values are pretty stock from memory. Since they're now screw-connecting, they'll be easy to replace with summat bigger in value one day..

[RobHolt]

Hi Rob,



Lol – I’m glad that you have for a change! Don’t be scared to upset the 'resident faithful'. We like lively debates here on AoS!



I’m sure that they did, but it doesn’t mean that such listening influenced the design process, which is what was of concern to me.



I suspect that any mastering would’ve been done using the solid-state Quad studio or broadcast amps of the day, and not Quad IIs. Heaven forbid, for the sake of accuracy, if it was the latter!

I have no beef with the 303 or 405, as they are different designs.

However, as far as I’m concerned, you cannot build a 'musically accurate' amplifier (i.e. that whose sound is genuinely musically convincing, since the end goal, after all, is to reproduce realistic and believable sounding instruments and voices, through a hi-fi system, by faithfully replicating, as closely as possible, the musical nuances of the source audio signal), without LISTENING to equipment as well as measuring it, during the design process.

It’s an amplifier’s ability to accurately reproduce those all-important nuances and subtleties, and not simply gloss over them, which to our ears, makes it musically convincing, and separates it from one which is merely accurate, up to a point, at reproducing the source music signal. Quite simply, our ears will readily detect the absence of those nuances, which test apparatus will likely have missed (as I do not believe that we can currently measure ALL that we can genuinely hear in audio), and thus will cause us to feel that the reproduced sound isn't 'real'.

In my view, relevant measurements aside, and taken as a given, only when equipment is then judiciously voiced by ear, by someone who genuinely understands music, and all its inherent complexities, does one faithfully capture those all-important musical nuances (depending on how good the ears are of the designer in question), which I have referred to above, quite simply because the technical limitations of our test equipment results in us currently being unable to measure all that needs measuring with music signals, in order to ascertain, with the sound of equipment, what is actually real (or rather as close to real as possible) and what is merely a competent copy of the source music signal.

As an aside, and I’m sorry to say it, but it pains me to read some of the absolutist stuff that Serge Auckland writes on pfm on this subject. He’s a nice enough chap, undoubtedly, but severely deluded if he genuinely believes that any audio equipment is *genuinely* sonically transparent. If only it were! Incidentally, Serge is most welcome to join and defend his views; in fact I would invite him to do precisely that, as it would no doubt make for an interesting debate, albeit it would most likely end in us agreeing to disagree.

I believe that there is no such thing as (genuinely) sonically transparent audio equipment. In my view, ALL audio equipment imparts its own sonic signature, subtly or otherwise on proceedings, by the very nature of the signal having to pass though a multitude of electronic components, wire and/or PCBs, all of which to some degree affect its integrity, and thus colours the music reproduced by the equipment in question, although none of this is likely to show up on our current test equipment. To be blunt, you’re in cloud cuckoo-land if you think that your favourite amp (or whatever) is sonically transparent...

I mean, let's get real, how can both Quad IIs and, say, the Quad 303 be considered as sonically 'transparent' or 'accurate' when they both sound different?? Yet both were presumably designed by Peter Walker to be "technically faithful to the input", such was his mantra, (and I could cite many examples of the same with equipment from other manufacturers). What's going on there, then?

The answer, of course, is that neither amp qualifies as being genuinely transparent, as every amplifier made to date on this planet, valve or transistor, imparts its own sonic signature on reproduced music, in varying degrees of subtlety. Therefore, when assembling our hi-fi systems, all we can ultimately do is choose our preferred form of coloration, whether it be valve or transistor-derived, and enjoy our music through it on that basis. Genuine transparency in audio equipment is but a pipe dream!



But a music signal is a very different thing from a noise or test tone signal.

Just because an amplifier ensures that what appears at the output matches that on the input in all but magnitude at audible frequencies, doesn’t automatically mean that it will succeed at accurately reproducing the source music signal.

The goal with music reproduction via audio equipment, after all, is to accurately reproduce MUSIC (not noise or test tones), which is probably why Quad IIs fundamentally fail at the former, and what happens when you don’t actually listen to the equipment you build!



The problem is that "technically faithful to the input" can, in reality, mean a very different thing to being (genuinely) faithful to the source music signal.....



The way I do it is embrace the fact that no piece of audio equipment or system can ever be truly 100% musically transparent or accurate, and simply choose components and build a system, which to my ears, best replicates the sound of real instruments and voices, based on my (extensive) experience of listening to them live. Therefore, from my listening experience to date, I intrinsically *know* how things should sound - and I'm far from alone in being able to do this.

Live, un-amplified, music is the benchmark I use to test my system against, and so I need no 'measurement apparatus', other than my brain and the God-given organs attached to the sides of my head (known as ears) in order to assess whether, on that basis, my system sounds musically convincing.



Well, if you’ve re-inflated it, I’ve probably just gone and burst it!!

Over to you now, mate....

Marco.

Lots to tackle there but many of the arguments overlap so I'll bullet point my response:

- Quad II sounding the same as a Quad 303 (or 405 or any other)

We cannot really look at amplifier sound and performance without considering the load conditions. Poor old Peter Walker was often misquoted and never actually said that all amplifiers sound the same. He actually claimed that all 'competent' amplifiers would sound the same where proper consideration had been afforded to the interface at each end.
What does that mean?
Well, the Quad II is a 1950s amplifier and was manufactured into the 1960s.
Like many valve amplifiers of the day (and a number today) it had relatively high output impedance and low power - some 15 watts on a good day if you were lucky.
Output impedance directly impacts the frequency response of the loudspeaker system where that system does not have a perfectly flat impedance characteristic. Very few loudspeaker systems do - planar systems being probably the only real example. In all other cases, a high amplifier output impedance will distort the power transfer characteristics and alter the system frequency response. It is one of the primary reasons why valve amplifiers sound different to solid state and it is very obvious with measurement.
The biggest change occurs where the impedance is at its most reactive - typically around the loudspeaker system bass resonance (and port area) and also around the crossover point(s). In these areas, the response is altered and often by several dB. The overall effect is to pull the frequency response into resembling the impedance plot as output impedance increases.

So what's that got to do with a Quad II sounding like a 405?
Everything - because the Quad II was designed to drive 1950s and 1960s loudspeakers which were typically of higher overall impedance than those in later decades, and generally less reactive. High overall impedance minimises the effect of high output impedance. So, for example, a Quad II will sound very like a 405 or 303 into original 16 Ohm LS3/5a, but quite different into a modern 4-6 Ohm design.

The plot below shows the result of experiments we made on a Tannoy LGM by driving from a low output impedance typical of SS and then from a 5 Ohm impedance. Note how the response and therefore the sound changes:





What you can see are deviations in excess of 6dB in the bass and 2-3 dB higher up the range. On a 16 Ohm loudspeaker system with a less reactive impedance characteristic these figures would reduce markedly, as would the sonic differences.

- Music is complex while test tones are simple and don't truly test the amplifier.

The amplifier simply responds to changes in voltage and current. Music signals can be defined by changes in voltage and current. A music signal is no more difficult to reproduce than a handful of test tones.
Humans perceive music as complex because we separate the individual strands both audibly and visually if we look at the waveforms, but the amplifier doesn't - it has no power of perception and only sees voltage and current.
On a slightly different point, asking an amplifier to reproduce 20khz at full output is a pretty tough test, and tougher than most music signals since you are asking the amplifier to swing fully and rapidly to doth DC rails. Might look simple and easy to humans but that's tough for many amplifiers.

Returning to Quad, they made the null test a feature of their testing and would invite competitors to the factory and ask them to submit their amplifiers to this test. Quad were able to demonstrate that you could null the amplifier output by injecting a portion back into the input in anti-phase. Now, if the amplifier were materially changing the signal passing through it in any way, it wouldn't null. But it did, indicating that signal complexity wasn't causing any difficulty over and above the lab test signals.

- The general objective v subjective debate and 'what we like'.

Objectivists tend to be labelled as white lab coat wearing boffins who use test instruments to asses equipment and want everything to have a flat response and no distortion. I am certainly not among that group, nor would I ever insist that someone must use a particular bit of kit because it measures well. For me this confuses two very distinct issues.
For me, to objectively assess two amplifiers is to compare them under blind conditions in order to discover, in as unbiased way as possible, any differences that might exist. Having done so and having found a real difference, having a preference is fine and will always remain subjective!
If an amplifier doesn't have a flat response and it has some slightly odd distortion on the signal, that's also fine if the listener accepts these things for what they are and enjoys the result. For me the whole objective thing is about discovering and explaining why things happen in a rational way. What I or anyone else might like is a different matter.

Anyway I've rambled long enough tonight but I hope that was useful.

Rob.