WAJ on AUDIO - for truth in hifi / stereo / high-end audioHome
HiFi AMPS SOUND THE
Examining the Arguments from Quad & from The AudioCritic
reader recently sent me quite a significant portion of the writings
of Peter Aczel, founder of The AudioCritic magazine. What
a character? Audiophiles seem to either love him, or hate him, there's
no middle-ground with this individual - no fence-straddling, here.
Having read a few AudioCritic articles in the past, I felt I had a fairly good grasp of what the man and the mag were all about. I particularly liked his feature on the 'White Hats' and 'Black Hats' of the audio-world. (It's a pity Issue nos, 10, 11, and 14 could not have been included in the batch of PDFs the reader sent me, as these would have given greater insight into his philosophy on speakers - a prospect I was looking forward to). But then, now having over a dozen complete issues is quite significant. My thanks go out to that considerate reader, once again.
And, though his philosophy on speakers remains somewhat unclear to me, I'm now absolutely sure as to where he stands on the issue of the ("lack of") sonic differences between amplifiers of similar specs. This stance is allegedly similar to that of the late great Peter Walker (founder of the Acoustical Manufacturing Company Limited, a.k.a. Quad). And though I greatly respect both gentlemen, I find myself at odds with their fore-mentioned utterances on amps.
I'm especially shocked that Peter Walker could have made such a statement, considering the quality of his amps. I once owned a Quad 303, which equalled or bested almost every other amp I ever compared it to. The only amp which (I think) outperformed my little 303, in direct comparisons at the time, was an expensive custom-built "Tho-mas" tube amp I'd also owned. And, yes, both were similar in sound, but there were also slight differences between them. (I can see the gentleman's point, though, as it would have been very difficult to consistently differentiate between both these outstanding amps in an A/B comparison test, I'm sure. They were that similar in sonics).
But then, perhaps they should have told us that the very best amps will inevitably sound very similar to each-other. It may sound trivial but, to me, that's a much more credible statement than one which says that; all amps with similar specs; i.e. measuring alike in distortion-readings, input and output impedance, and etc, will sound virtually identical to each other. That's just not so, in my experience (That was me; paraphrasing Peter Aczel, by the way - but that's basically the gist of what both men are commonly reported to have said).
Virtually all my experiences with various amps throughout my audio-life (including the Quad and 'Tho-mas' mentioned) have led me to the opinions I've so far expressed. But perhaps most poignant is the result of a comparison a friend and I conducted with my UREI, Crown, Sherwood, and a Carver M500 amp I'd also owned at the time, around 10 years ago. These amps were compared against my friend's M400 Carver and a Peavey we had borrowed.
Only one of these amps was really of 'audiophile' quality, in my opinion. However, all these amps boasted similarly outstanding specs (as you'd expect of any half-way-decent solid-state amp). [The greatest difference was that the Crown's damping-factor was off the charts, compared to all the others.] In every aspect that matters, these amps fit the criteria which stipulate outstanding and similar specs and, therefore, they should have sounded identical, according to the fore-mentioned way of thinking. Right?
Nothing could have been further from the truth.
Oh yeah, both Carvers did sound identical - but they're both the same amp, really, virtually identical under the skin, so that doesn't really count. Apart from that, all these amps sounded quite different from each-other, with the Sherwood and the Carvers displaying the most similar characteristics (and similarly deplorable - if also typical of most moderately-priced hifi amps) with fairly atrocious highs and mid-mids, a complete lack of low-mids, and passable/commendable or believable bass; the best aspect of all these. The Peavey was similar in general characteristics, but disgustingly much worse in the mid-mids. The Crown was much better than those in most aspects, if only at first appearance, but far from ideal; being betrayed by a seemingly exaggerated but pristine high-end, immaculately clean and clinical mid-mids, a similar absolute lack of lo-mids, a remarkably tight though exaggerated mid-bass, and an overly tight (over-damped?) 'deep'-bass, with both the latter regions being also remarkably false and devoid of detail. The UREI was considerably more natural than the rest. Nothing outstanding - just natural in every way. (Much more exotic amps have also been bested by this amp, by the way, though not in direct A/B comparisons). I'll just leave it at that.
The point I'm making is that these amps, with very similar specs, were NOT identical, in sonics. And since this has been my experience throughout, I struggle to find the credibility in the arguments, along the lines outlined, so far as amps are concerned. Similar specs and measurements do not indicate similar sound, in my experience.
Nevertheless, I still maintain a healthy respect for both icons, despite my disagreement with some of their alleged views, as expressed.
Specifically regarding Peter Aczel, I believe he was/is a tremendous asset to the discipline he served - and still serves, surprisingly. In other words, he was right far more often than he was wrong. He still is more of an asset to our hobby than a liability. Without the slightest hesitation, I'd recommend that The AudioCritic (back issues and new blog) should be required reading for any burgeoning audiophile who wants to be guided by solid principles. My only caveat to such an audiophile is that one should always keep in mind that nobody is perfect, and even the greatest thinker can be wrong on a point or two. (And those who go on to discover Peter Aczel's 'Fourier expedition' should take note, yes, but perhaps not condemn the man totally for it, as there's too much to learn from this man to deprive one's self of his teachings merely because of a human indiscretion). Instead of 'nit-picking' the foibles, perhaps it's better to ascribe more importance to the 'big-picture'.
Taken in totality, and after gaining more insight into this stalwart, I'll admit that I'm even more of a fan than I was before. I'd estimate that I might be in agreement with perhaps 80% of his pronouncements. But, most importantly, I believe no-nonsense straight-talking characters like this are far too few in hifi today. Hifi journalism is now dominated by too many spineless, slithering, disingenuous "yes-men", pandering to the whims and fancies of powerful manufacturers, at the expense of the unsuspecting audiophile.
PART 2: Both Men Made Similar Statements - Yes - But; Based on Different CriteriaI decided to look further into what both Peter Walker and Peter Aczel are reported to have said about the sonic similarity of amplifiers.
For reasons which will become obvious, I'll concentrate initially on Peter Walker. But first, I'll repeat a segment of what I'd said as it is obviously in accord with what Peter Walker had actually meant, in my view.
In my research since first addressing this topic in part 1, I've found that Peter Walker had made several statements on the subject. Taken as a whole, it has become obvious to me that the gentleman did not really differ from my views at all - not even by an iota. What was misleading to me, and I'm sure to others, is that over the years some have taken the man's words out of context. Also, in a scenario where an individual had made several statements relating to one subject, it is unfair and potentially misleading (as is the case here) to highlight any one part, or shortened-version, of these statements as representative of that individual's definitive stance, especially where the shortened-version is bound to be distorted, owing to the lack of the supporting context (as is also the case here).
This is what many have done by saying that Peter Walker asserted that; All competently-designed amplifiers sound the same, for instance - to give just one example of an erroneous over-simplified and fore-shortened version of the man's actual statements - one of the many versions which would have led anyone astray. (There are many so-called 'competently-designed' amps I wouldn't even install in a chicken-coop - I have far too much respect for fowls to subject them to such punishment).
But first, in erroneously equating Peter Walker's stance on the subject to that of Peter Aczel, and refuting both, I'd said this in part 1: This stance (that of Peter Aczel) is allegedly similar to that of the late great Peter Walker. And though I greatly respect both gentlemen, I find myself at odds with their fore-mentioned utterances on amps. I'm especially shocked that Peter Walker could have made such a statement, considering the quality of his amps. I once owned a Quad 303, which equalled or bested almost every other amp I ever compared it to. The only amp which (I think) outperformed my little 303, in direct comparisons at the time, was an expensive custom-built "Tho-mas" tube amp I'd also owned. And, yes, both were similar in sound, but there were also slight differences between them. (I can see the gentleman's point, though, as it would have been very difficult to consistently differentiate between both these outstanding amps in an A/B comparison test, I'm sure. They were that similar in sonics).
On closer inspection, the gentleman differed not one scintilla from what I'd said above, regarding the potential difficulty in differentiating between the very similar sonics of exceptional amplifiers. Note where I'd said that both of my very best amps, at the time, were similar in sound. And note also where I'd said of Peter Walker and Peter Aczel; ...perhaps they should have told us that the very best amps will inevitably sound very similar to each-other But this is exactly what Peter Walker was trying to tell us, all along. Idiots have misinterpreted him, over the years, and latter-day idiots like me have allowed ourselves to be misled by the fore-mentioned fools.
Peter Walker's only fault here is that he didn't express himself in simplified language which idiots can more readily understand. For example, I'm guessing there would have been less confusion if he'd said something like; The more accurate an amplifier is, then the more similar its sound will be, when compared to other similarly accurate amplifiers - with all else being equal.
That's basically what the Quad stalwart was saying. And I agree in totality. This makes a whole lotta sense, since 'accuracy' denotes one standard/one sound. Therefore, those amps which closely approach that 'one standard' WILL sound alike - inevitably - and without doubt!
[See some of Peter Walker's relevant statements, linked below:
At paragraph 8 of the document; 'Quad Amps 1002', he tells us, "A competently designed amplifier of top quality should pass such a comparison test with no detectable differences." From what preceded - and from what followed - in that piece, it's obvious he was referring to a comparison of such amps, including his own Quads, against the standard; i.e. the closest equivalent of the impossible-to-achieve; 'straight-wire with gain'. (Obviously, I cannot illustrate the full context here - please click on the link). In other words, this meant that a comparison of the output of such an amp would have been similar to the input of said amp. And the logical extension to that, in his estimation, is that any and/or all amps which closely approach this ideal should sound similar. (Apologies; for the quality of those docs, by the way. Perhaps saving the image, then enlarging, could help).
His statement, in full context, differs dramatically from what those who've corrupted and fore-shortened his statements have come up with. He was specifically referring only to all those amps which closely approach the ideal; 'straight-wire, with gain'. This is the main criterion. Shortening and over-simplifying the gentleman's statement to say; All competently-designed amplifiers sound the same, is grossly misleading, and bordering on the dishonest, in such cases where the man's meaning has been deliberately twisted. They fail to include the man's definition of; "competently-designed amplifiers" - which is; amps closely approaching 'straight-wire with gain', or perfection, in actual performance. And, in so doing, they've painted a highly distorted picture of the man's stance. (Nor was he touting specs, per se).
And, for those who seek to imply otherwise, Peter Walker distinctly qualifies his statements on the subject of similar sonics in amps by telling us; "All this is not to say that all amplifiers are the same." Now, what could be clearer than that? This is a blatant and explicit contradiction of statements from those who try to convince us that Peter Walker ever claimed otherwise. (For confirmation, refer to paragraph 2 of the document labeled; Quad Comparative p.01). ]
However, though I believe he meant something similar, Peter Aczel's statement seems to place a heavy dependence on 'respectable' specs as the main criterion determining that all amps will sound the same. Here's one of his statements to that effect; "As I have said, and written, innumerable times, any two amplifiers with high input impedance, low output impedance, flat frequency response, and sufficiently low distortion and noise will sound exactly the same at matched levels if not clipped.'"
Here, I disagree!
I disagree with such a statement, based on my experiences, some of which were cited above, regarding those amp-comparisons I was involved in. (Even today, I doubt whether our most sophisticated labs are able to measure for all aspects which cause differences in the sound of amps, etc. And too many similarly-measuring amps have been demonstrated to differ significantly in sonics, over too many years).
For instance, all the amps cited in my own comparison (described previously) fit all the likes of Peter Aczel's criteria - to a 'T'. Yet, except for the two virtually-identical Carvers, they all sounded different from each-other - and, in some cases; very much so. (As mentioned, only one of these amps was 'audiophile' quality. I don't doubt that another amp of similar quality would have sounded similar - as did my Quad and 'Tho-mas', in the past).
Whilst Peter Walker's stance is based on the main criterion of an amp closely approaching the theoretical ideal; 'straight-wire, with gain', Peter Aczel's main criteria, for amps which are allegedly bound to sounding the same, are obviously the amps' specs and distortion-measurements. How many of us have experienced where components, with virtually identical specs and distortion-measurements, sound distinctly different from eachother?
Peter Walker's stance above is inordinately more credible, in my view. [Yes, it's true that he did, also, mention specs and distortion, and many other things. And yes, he did believe that there is a correlation between specs and sound-quality - as most of us do, albeit; to a limited degree, in many cases. However - and this is critically important - he did stipulate that perhaps the relevant measurements and specs, which would explain/illustrate differences between amps, aren't usually published - re; 1st paragraph, Quad Comparative p.01, linked above. (That may well be so, in limited instances, but people like me, however, contend that some of the relevant measurements for this haven't yet been discovered, let alone published). If Peter Walker was ever more extreme than that and, inevitably, contradicting his own statements in the cited documents, then I'm not aware of it. I can only go by what I've seen of his utterances. The bottom-line is that he based his relevant statement on "competently-designed amplifiers" (not merely on distortion figures, specs, etc). And his definition of a "competently-designed amplifier" is one which is designed to closely approach the ideal state of; a 'straight-wire, with gain'.]
But perhaps we should also look further into exactly what the AudioCritic had said. Perhaps there's a hidden clause somewhere justifying such a stance as his.
PART 3: Whither the Measurements?
It seems many audiophiles would stake their lives on measurements. But my experience hints at the distinct probability that we (including our most brilliant scientists) are not measuring every single relevant aspect of our systems' and components' performance. And even in those aspects of performances that we do measure, all is not being revealed to us. Let's look at a case in which a certain type of amp (recently introduced, and enthusiastically touted) is outperformed by some which measure even worse than the newly-touted amp-alternative.
Class-T amps have been heralded in some circles as the wave of the future and certainly competetive with conventional types, especially when their impressive measurements and specs are taken into account. Yet a former acquaintance of mine found that his T-amp was comprehensively out-performed by his 1950s 12 watt tube amps driving the same high-efficiency Altec speakers the T-amp couldn't get the best out of.
Also, in wrapping-up his review of a similar T-amp, here, this 6moons reviewer compared it to some expensive low-powered tube amps (which measure nowhere near as good as the T-amp) and had this to say; "..the SoT acquitted itself quite well. It is not as good as those amplifiers either in sound quality or ability to play music but it is not so far off the pace as to be completely embarrassed by the comparison." One tester even prefered it to a much more expensive tube amp, though my focus here is; not which amp was better, but whether there was any discernible difference in the sound of the amps.
Seems there were differences, after all.
Impressive performance from the T-amp nonetheless, though it still proves the point: Its measurements may not be as good as the typical solid-state amp, yet said measurementnts and specs are eons better than those of the other ancient tube designs which generally bested it in performance.
They are certainly very good bargains, for their sizes and prices. But neither do I believe those early T-amps are capable of being the equal of the better, or best, conventional soild-state amps.
Peter Aczel himself had the opportunity to address this very issue in his own review of a T-amp, and even promised an A/B/X comparison to a conventional solid-stater. Problems, however delayed that comparison, and the much-anticipated follow-up hasn't so far materialized, 8 years later. One is inclined to believe, however, that despite his complaints of very-slightly inferior specs and distortion measurements, at the very end of the frequency-extremes, and when over-driven, respectively (but still eons better than any low-powered tube-amp) he must have had some indication of the T-amp's sound, in relation to other amps he'd used, during the time he'd had it long before its break-down.
Yet, he omitted to supply us with any such opinion, other than to say; "..I thought I heard a few subtle, momentary sounds I didn’t like..."
But then, even that seems unbelievable! And, pending A/B/X comparisons notwithstanding, why not just elaborate on what was already heard, especially since there's a hint of a difference from the normally-expected sonic characteristics? Yet, I really can't see how even such a difference (as in; "..subtle...sounds I didn't like") could have been possible since, despite minor quibbles, this T-amp SHOULD have sounded EXACTLY the SAME as ALL others which have manifested reasonably similar measurements.
Unfortunately, it seems elaboration on the already-heard 'subtle' differences, in this particular case, might just have to wait a few more years for that A/B/X comparison.
However, let's now look at something as rudimentary and as common-place as frequency-response measurements. Let's also recall one of Peter Aczel's pronouncements, where he tells us; "As I have said, and written, innumerable times, any two amplifiers with high input impedance, low output impedance, flat frequency response, and sufficiently low distortion and noise will sound exactly the same at matched levels if not clipped."
And let's also look at what we found when my friend and I compared those amps I'd mentioned: All of those amps boasted ruler-flat frequency-response curves. Yet, of the six compared, four of them displayed less than exemplary trebles; relatively dull, somewhat fuzzy, and apparently curtailed (though all measured flat to 20khz). Of the two remaining, the UREI was passable, nothing exceptional. And the Crown was undeniably outstanding in the trebles, relative to all the others - pristine and prominent - perhaps too much so, as sibilants were also enhanced (despite nothing in the measurements revealing these differences). And, do recall that all the amps, but especially the Crown, also displayed other sonic features which are not apparent in the measurements.
Even more glaring discrepancies were found at the lower-midrange (my favorite stomping-ground in assessing equipment) where the pretenders in hifi (and they constitute the vast majority) are ruthlessly revealed. Indeed, of the six amps in that comparo, only one provided natural and realistic low-mids. The rest seemed to eliminate the region between 160 and 320 hz completely, and absolutely.With that being said, I certainly wouldn't trivialize Peter Aczel's stance, however. Though I believe the criteria on which he bases his stance are wrong, there's also some measure of merit to his arguments. I'll link you now to issue # 16 of The AudioCritic. At page 31 of the mag (page 25 of the PDF) there's an article of his on "Basic Issues of Equipment Reviewing and Critical Listening". And in the 2nd paragraph of the segment entiled; "What sounds the same?" he seeks to qualify the specific circumstances under which amplifiers will sound the same, as articulated in previous statements of his, and similar to my quotation of his remark above.
are those qualifications? Power amplifiers must
have high input impedance, low output impedance, no frequency-
response anomalies, and be at all times operated
within their voltage and current capabilities in order to
sound the same."
I still can't say I agree with that (I'm more in line with Peter Walker, here) but much of his surrounding arguments do hold merit, in my view. I must say I'm in agreement with much of the thinking which inspires his stance. For instance, I've seen elsewhere where he asserts that amplifiers are much less important in the audio-chain than they are usually thought or touted to be. In agreeing I'd suggest (again) that once a certain level of performance is achieved (let's say; that of a Quad 303) then there's not much more performance to be gained from even an amp costing upwards of $100,000 - perhaps very little, but certainly not worth very much more than the cost of that Quad. (Blame it on the 'law' of diminishing returns). In other words; any amp which is as good as the inexpensive 303 will be difficult to differentiate from that mega-buck amp in any blind-test and, therefore, will be good enough for inclusion in ANY state of the art system. (Incidentally, this is also why the likes of Stereophile refuse to subscribe to the concept of A/B 'blind-tests'. How else would they be able to assist in selling - and to justify - mega-buck amps, and etcetera?)
So then, I guess you could say my personal stance is midway between that of those like Peter Aczel and that of the card-carrying subjectivist. Where I differ with The AudioCritic is that I recognize that there are also many (crap) amps which measure just as good as the likes of a 303, but which are nowhere near as good. I have to go by what I've experienced, myself - time and time again. Therefore, I cannot ascribe as much importance to the currently available measurements (saying, basically; that ALL amps which measure alike sound alike) as those like Perter Aczel do. I stop somewhat short of that extreme!
Nevertheless, I also agree with his view that speakers are THE most important components in the audio-chain (see my own views, here). And, though I may have neglected to mention the listening-room in my own revision of the hierarchy of the audio-chain (never thought of it, really) I'm also in agreement with his view that the room is pretty-much as important as the speaker-system - inextricably linked, in fact. (Admittedly, and specifically, it's not that I never considered the room, it's just that I never considered the room as part of the audio-chain, per se, nor as part of the speaker-system, until I read his views on the subject; pure indisputable logic).
The AudioCritic (the man, himself) is no audio-idiot. It's just a pity that he ascribes a wee-bit too much importance to the measurements, in this instance
Indeed, there's enough evidence to not only suggest, but to shout to all who'll listen to the fact that measurements do not illustrate all facets of an audio-component's performance. If Peter Aczel's statement above were true, then it'd be a waste of time to listen to amps before purchase - all we'd need to do is to look at the specs.
Moreover, there would be virtually no need for amp-reviews, either.
PART 4: Conclusion
The relevant difference in the utterances of both Peters, in my view, is that Peter Walker based his stance mainly on the actual real-world quality of amplifiers, with the criterion that said amps closely approach the ultimate standard; 'straight-wire, with gain' (i.e. output closely equates to input - with gain). Such amps, therefore, and by definition, are absolutely bound to sound similar.
There lies the flaw in Peter Aczel's argument, in my opinion, as countless amps with similar measurements & specs can be easily (and frequently have been) proven to differ in sonics.
Peter Walker's arguments, however, are impossible to dispute. This is because, in emphasizing the requisite "top quality" of the amp(s) the sonic quality of such amps is implicit. Top (sonic) quality implies close adherence to one standard - accuracy. Therefore, ALL "top quality" amps which, by his own definition, approach this one standard (or level of accuracy) are logically BOUND to sound alike (as accuracy - one standard - equates to one sound).
So, though both gentlemen wind up saying close to the same thing (as it appears on the surface) the main criteria on which they base their arguments are substantially different. In recognizing the differences in these criteria, at the base of both arguments, I'd suggest (again) that Peter Aczel is wrong - on this one - and Peter Walker is right.
But then (again) these are only my humble opinions!