Sean: Although I think you know that I would not buy a piece of gear based on specs, and I am sure you wouldn't either, I do want to join you in commending Stereophile for being just about last man standing when it comes to publishing original test results. In fact, I think they should expand the practice: Test the gear written up in the columns, not just the reviews; Add more speaker tests to help quantify dynamic and frequency response vs. distortion + noise; And like JA finally got around to doing this month - but seemingly only because it involved his beloved LP12 - start including measurements when it comes to analog gear. Measurements are a useful check on both gear quality and reviewer accuity.
Having said that though, I haven't got any problems with the specs of any of the pieces mentioned. My only problem with the Pass amp, for instance, is if the true measurements deviate greatly from the manufacturer's claimed measurements - which it seems as though they might - but not with the unit's measured capabilities per se. Yes, it does seem faintly ridiculous to me that an amp the XA160's size, weight, type, and price shouldn't offer more output grunt and linearity, but I don't think you can extrapolate from that into some kind of accelerating trend toward a 'personalized fidelity' paradigm within the industry. And complaining about output impedance in an amp with no global feedback seems a little disingenuous to me. Maybe something was sub-optimal with the test sample anyway, and I'll withold judgement (on this narrow issue) until Pass responds in the magazine.
As for the BAT preamp, it would have been nice had JA included a measurement of its response somewhere in between unity gain and max output. Absent that, I have to infer from his comments that the measured behavior really did not bother him, and that he really did think this piece was well-engineered for use within normal operating parameters. That there might be some slight change in the location of the upper -.5dB point (into average loads) dependent on volume setting does not, to me, seem to be all that unusual, or to disqualify a piece from consideration as a high fidelity instrument. In fact, it seems quite possible to me that such a volume control might well be engineered to deliver superior transparency through the heart of the volume range - no doubt what BAT would contend. And again, I'm not sure what you're driving at with remarks about I/O impedances - the BAT's input impedance was unmeasurably high, and the output impedance seems reasonable for an all-tube design that eschews cathode-followers, and will work fine with plenty of amps.
With the AH! CD player, you're probably right, it certainly does appear as if the response has been slightly tailored to yield a small dip through the 'brightness' range. But hey, customers interested in a heavily modded, tubed version of a mass-market player might find this attractive. And though the rest of the results may not have been stellar, they were all OK, and they're probably of no audible consequence or any great sin at this price.
More to the point, neither you nor I have heard these pieces - the reviewers did. And again, I can see no real basis, just from this anecdotal evidence, to back up your assertion that 'new gear is getting worse' in general. |
But Sean...
OK, you've got me on the price of the CD player. I should have said that its specs were nothing to get upset about at any price.
I'm not being cute when I say that. Outside of its quite mild frequency response deviations, I wouldn't be shocked if Stereophile's published specs on that unit don't actually constitute an overriding determinant which correlates well with its perceived sound quality. And that response deviation is of a subjectively not-undesirable character when looked at from the standpoint of the mastered sound of many disks, and will in any case likely be swamped by the non-linearities present in the speaker/room response.
But basically, I actually agree with you that this type of euphonic tailoring is not admirable, and that reviewers often seem to go in for components that make average recordings sound 'nice', rather than revealing what's truly there. But I can also understand why others might enjoy this kind of design - after all, music listening is about personal pleasure, and who am I to argue with the preferences of another?
The real core of my argument is not that the AH! player's specs are worthy of a spirited defense, but rather that its *sound* (whatever that may be) is worthy of being judged - as is the sound of any component - primarily against the *sound* of other players in its price range. I probably wouldn't choose this player, because like you I happen to care a bit about uneditorialized frequency response, but that doesn't mean that the unit might not excell in some other parameters at its price point. I don't know however, and neither do you, 'cause we haven't heard it.
Your position in the last sentence of your second paragraph makes it clear that you consider competently-measuring SS gear to constitute some kind of standard of 'neutrality'. But not all 'neutrality' is created equal. I'm sure you don't require a lecture from me on the differences between steady-state test measurements and far more complex music signals.
So there's the 'neutrality' of maintaining flat amplitude response into varying speaker loads by designing for low amplifier output impedance, and then there's another 'neutrality' of designing for lowest TIMD and timing errors by eliminating the global feedback that enables SS amps to appear more linear under static test conditions.
I am not one to propone that what audiophiles hear can never be measured - I do believe that for any audible effect, there exists some discoverable technical property that can be correlated with what the ear perceives. That's in theory only though. In life our measurements are imperfect in design and execution, and the real music problem too complex for current measurement practice to shed the degree of light that would be required for correlative certainty regarding all audible phenomena.
On top of which, I still think this month's issue of Stereophile in no way 'proves' your contention that the overall direction of the industry is heading toward increasingly non-linear devices. That would be a false sylogism - an unscientific supposition, in other words, for which both the evidence and the reasoning are flawed.
I'm particularly surprised that you don't give Nelson Pass, a designer you've often stated your admiration for, more of the benefit of the doubt regarding this new amp. I'm not saying I know he's 'innocent' - again, neither you nor I have heard the thing - but I would be inclined to assume that he hears something positive about the fidelity of this design approach that he couldn't attain through his previous designs, at least until conclusively demonstrated otherwise. And again: even if you or I didn't like it, why care? All that matters in the end is that whoever buys it likes it. It's a big enough market for different interpretations of reality (and that's all we've got) to flourish without doing harm to one another. IMO. |
Onhwy61 makes what I suspect is a very valid point about evolving audiophile perceptions regarding the desirability of 'accuracy'. It is of course not a new observation, and whether it's one that is causally, or only tangetially, related to the point of Sean's original post is open to debate. But I don't want to come across, with my above statements supporting individual preference and the marketplace, as seeming to be totally cavalier about this legitimate issue.
I would generally support the contention that during hi-fi's 'golden age' - despite the state of the art then not being as 'accurate' as is possible today (yeah I know, some would argue that point...) - the *goal* of achieving literal 'high fidelity' to the signal encoded on the recording medium was more enthusiastically embraced by the audio community as being the Platonic ideal the hobby ultimately aspired to. Today, it does sometimes seem that such an ideal is currently regarded, by at least a sizeable proportion of self-described audiophiles, as being distinctly subordinate - if not outright antithetical - to the goal of pursuing the 'Absolute Sound'.
Although I'm all for the recreation of the 'Absolute Sound' as our overarching dream for in-home music reproduction, it is alas an essentially impossible one. As I see it, the problem in attempting to simulate such a perception through the introduction of deliberate non-linearities into the replay system is that such an approach devalues and is often in direct conflict with the necessary (and uncontrollable by the listener) imperatives represented by the signal that is recorded on the software. To me, we *should* be pretty much completely dependent on the recording and software to capture and encode that 'Absolute Sound', and call upon the replay system only to try and transmit that as best we can. But by the same token, it is hard to argue with the position that our replay systems will *always* be non-linear - and in significantly arbitrary fashion (and the same goes for recording systems) - so *why not* attempt to tailor the reproduced sound to render recordings (some, if not all) in the most subjectively pleasing light? After all, isn't this hobby all about personal enjoyment?
Well, yes it is - but: I personally cannot fully enjoy music reproduction that I consider - or the evidence shows (and such evidence could be measured, OR HEARD, and maybe not able to be measured) - to strive for something other than hewing fairly closely to the ideal of 'accurate' reconstruction of the recorded signal. But I realize that's just my preference. I think it's a logical preference, one that I'm most comfortable with, but not one that's shared by all audiophiles. So I'm not going to get upset when gear manufacturers make gear for that segment of our market which has had enough of the pursuit of 'accuracy' and gets more pleasure out of trying create a convincing and pleasing illusion with their systems, even if they know that might deviate from what's strictly on the disk, and despite the problems I myself find with this type of philosophy. In an imperfect world we have to pick our poisons, and if the marketplace didn't naturally dictate that diversity was the right solution, then we wouldn't have the proliferation of firms in this business that we do. I for one can gracefully accept that fact, and still keep my opinions. |
Y'know Sean, I actually had not yet thoroughly read Chip Stern's review of the AH! player when I responded above. But upon reading it carefully, not just skimming, it seems that quite a few of his comments on this unit's sound are not the kind of impressions I had implied I would have expected based on the measured results. To me, this supports one of my main contentions, which is that there's not necessarily a lot that can be directly inferred about the sound of electronic gear from the way it measures, within limits anyway.
Also, rereading the Pass review and tests, I have come to realize that really the only things about this amp that should raise eyebrows, outside of aesthetic and economic factors, are these: 1) the current capability shortfall vs. the claimed spec, and 2) the fact that apparently no caution against pairing it with a low-impedance, low-efficiency speaker - particularly in large rooms or for high volume levels - seems to have been mentioned by Pass. The shortfall in rated output (claimed 160W vs. measured 118W at 1% THD, I believe) is trivial in practical terms (at least for the specified 8 ohm load). And I was wrong above to suggest that the amp might show any linearity problems when operated within its margins. So it's only the 'grunt' question and matters of system-matching that are open to question in my mind (again, size and price notwithstanding, since these are issues that are matters of personal taste and wherewithal, unrelated to sound or engineering quality). Looked at that way, it doesn't sound crazy to me, for the right applications, that an amp featuring only two gain stages, no global feedback, just one pair of output devices, and pure class-A operation should yield only modest power for its size and weight (along with a high-than-normal output impedance). Given an appropriate load (impedance that is relatively high, flat, and non-reactive, coupled with above-average efficiency), such an amp might well sound better in some ways than more conventional designs. (Too bad JA no longer seems to be including the data graphs he once made using the Paul Miller system which showed THD vs. output power at varying loads driven by a more 'music-like' test signal. I wonder where this amp would really begin to hard-clip into 8 ohms - possibly much higher than the 1% point, considering Fremer's audition results which indicated no audible strain). None of the foregoing is meant to excuse Pass' overrating of the peak current capability (a claimed 12A vs. a measured 3.8A), or his coyness about power delivery into 4 ohm loads (not spec'ed, but which he seems to imply - incorrectly - is the same as with 8 ohms), but tubed amps with similar 'limitations' (I'm thinking especially of OTL designs) and similar (or higher) pricing are successfully matched with appropriate speakers. Maybe Pass himself will bring to market a speaker ideally mated to his new amp... |
Aboldor: Your linked post from NP sounds like a good answer for customers, but only in part. He still fails to address the questions I listed above, namely the realistic current-capability and speaker-matching issues. Maybe his "Manufacturer's Response" in the mag will be more expansive. It can inferred however from his words that the company was apparently completely satisfied with the sound of the review samples as-submitted.
Drubin: The Rushmore is an active speaker, with its own built-in amplification.
Sean: Then there's always the XA-200... |
Onhwy61: That's basically what I've been saying all along regarding the amp itself (read my above posts). My 'complaint' (if you want to call it that - it's not like I would have been buying a pair of these tomorrow anyway, no matter what ;^) has mostly to do with what seems like a somewhat unrealistic and incomplete product *description*, given the amp's not-unexpected limitations in areas we both acknowledge as reasonable for its apparent design brief. I have no trouble, based on what I know so far and the company's history, in believing Fremer's conclusion that it's nonetheless a very fine-sounding product (and could maybe perform even better given optimally-matched speakers). In other words, the amp seems to me like a niche product (within the larger niche of products in its more exclusive price range), despite the image we commonly hold of large, heavy, costly solid-state monoblocks being that they should all drive 2-ohm pigs to arena volumes at 20Hz. I am defending the diverse approach as being fundamentally not silly or worthy of ridicule. I suspect the amp represents an honest engineering effort with good reasons behind it, and let the marketplace decide. |
Ultraviolet: I am not sure I know what you mean when you say that most live music is 'etched' and 'bright'. I have to assume you are talking about the sound of PA systems, not live acoustic music, but even so, none of the PA's I've heard sound 'etched', as in overly emphasizing the leading edges of transients, or especially in overly pinpointing images. I've heard some PA's sound 'bright', as in too hot in certain ranges, but not confined only to the 'brightness' range. But they are just as likely to sound tubby or muddy and indistinct as a dominant flaw.
Live acoustic music, on the other hand - and as the only meaningful barometer of reproduction 'accuracy' - can never really be accused of sounding 'overly bright' or 'etched': it is what it is. The performance space acoustics or audience vantage point may introduce perceived response anomolies at the listening position (which generally attenuate, not accentuate, treble power), but whatever 'brightness' comes out of a horn's bell or off a violin's bow cannot be argued with.
I'm also at a loss to understand your assertion that tube amplified systems, including the ones you've set up, sound 'very inaccurate' as a rule. This will come as news to most audiophiles. Although I've never owned SS amps that would contend for state-of-the-art, my tube amps have impressed as being more realistic overall in their handling of many musical attributes. I've certainly heard SS amps that sound just as good or even better in many areas than the tubed amps I currently own, but I think competent designs from either camp can undoubtedly sound pretty close to real music, and are not 'very inaccurate' or even all that different in many respects.
I am a musician too, a guitarist, and one of the things that I think biases me toward tube amps - besides of course the well-known ability of tubed guitar amps to generate pleasing distortion characteristics - is that I've never played through a SS guitar amp that sounded as 'alive' as a decent tubed amp. Even the best ones I've tried all give somewhat of a 'dead' feel, in terms of the relationship between what you play and what you hear which combine to create the elusive quality of superior 'touch', when compared to the 'lively' feel of good tubes. This observation may or may not really have a direct bearing on the 'accuracy' of high-fi amplifier types - it just helps explain my personal interest with tube amplification in general. But I can tell you for sure that I never would pursue this route in my stereo system if the results sounded anything remotely resembling 'very inaccurate'. |
Ultraviolet: Now it is my turn to clarify myself. My comments about tubed guitar amps were not intended to be an extension of my comments about what live acoustic instruments sound like. I was addressing two different things in one post. The point about tubed guitar amps was just included as a subjective observation on the species of tubed amps generally, but I inserted a disclaimer about this observation's applicability as far as hi-fi is concerned. Any electric guitarist will tell you that the response of their amp is an intimately felt quality, that has a great deal of bearing on how you actually play. The upshot is that I suspect that at least part of my interest in tubed hi-fi amplification has to do with this guitar-centric 'behavior reinforcement' I experience every time I plug in my instrument, and I recognize that it is not necessarily an entirely valid paradigm when transferred to the hi-fi world (although it could be).
As an aside, I think that when it comes to reproducing natural instrumental timbres (assuming a recording even attempts to capture the 'absolute sound' at all - see Onhwy61's post), the way in which microphone technique is employed to capture the original event is at least as significant a determinant of whether a given reproduction will sound too 'bright' or 'etched' (or too 'warm'-ed over) as is the type of electronics used in the playback system, and often more so.
[I'll also mention that I personally prefer a playback system having a 'neutral' (as in relatively flat) treble power response, which is part of why I chose Thiel speakers. So don't let the fact that I like tube amps (mine are VTL) fool you into thinking I must prefer euphonically 'warm' or rolled-off sonics. I emphatically do not - there's other qualities about tubed amps that one can gravitate towards besides tonal balance.] |
Trelja: What leads you to conclude that amps in general have been "taking a dive in quality"?
As for the linked AA thread, I think I'll skip it. I don't know if I would agree with any of the arguments made in it or not, but I just don't care enough about audio writing anymore to bother debating the subject. Even when it's done well, the thrill is gone for me. I consume one of the audio rags in an evening, and then hunger for something meaningful and stimulating to read. I admit it took several years, but I'm finally worn out and feel like I've read it all before. Most of all, it's boring, because audio is an intrinsically trivial and limited subject. Even if the field were written about with unimpeachable integrity always, I would still have little interest in reading about it anymore. Maybe it's just me ; I went through the same thing with rags devoted to several other specialty hobbies over the years since I was a teenager. In the end, either you do a thing or you don't, but endlessly reading about it eventually loses its diversionary appeal. I suspect that's why I glommed onto Audiogon's forum: It got me off more to write about audio for a while than to read about it, but even that's getting a bit old. In the frame of mind I'm in now, it's easy to read magazine reviews and simply focus on everything wrong I find with most of them, but that's just cheap mental masturbation. If the Stereophile subscription weren't practically being given away, I would let it lapse (as I have TAS - too expensive for what I get out of it), and next time maybe I will anyway. I do have several specific criticisms of the mag, but in reality, if they didn't put out a product of sufficiently high quality, I couldn't read it at all, and I still do... |
Trelja: Tubes are go, thanks for asking :-)
About the amp thing: The reasons I ask about your downward trend observation are, A) Because isolated anecdotes do not a trend necessarily make, B) Because lateral movement or running in place don't = "taking a dive", C) Because being overpriced for the performance also does not = "taking a dive" in quality. I also have to agree with Onhwy61 that ability to drive 2 ohms (or to measure particularly impressively in any bench tests) is often unrelated to sound quality in many applications, and do not agree that being SS places this obligation upon a design, regardless of price.
As for your desire that the implementation of a technology should improve over time, that's hard to argue with in most workaday instances, but in audio (despite the fact that just such a phenomenon has arguably occurred in many aspects of the field) I think the 'boutique' nature of the industry/market demands and rewards a continual flow of new products, whether they are significantly improved or not. To wish for the only practical alternative would be unrealistic in my view. (And the NAD...well, let's just say I started out in this nonsense with a circa '86 2200, and have my strong doubts about yours keeping pace sonically with stuff like the Hovland, Pass, or other contemporary overpriced audio jewelry, though I can't attest to it directly...) |
I believe it was in the most recent Stereophile (same as the Radia review) that Michael Fremer gave the Theta Enterprise amps a conflicted but ultimately lukewarm pass. The head of that company, Neil Sinclair (whose digital gear I own and admire), then came back with what I thought was a blatantly disingenuous rebuttal in the manufacturer's response (despite the fact that the reviewer pointedly and repeatedly praised their flagship Citadel amps), attempting to paint Fremer as an unreconstructed tubophile with an agenda. Apparently Sinclair found it easy to overlook Fremer's reference being SS, as well as his bestowing recent raves upon the Halo JC-1 and MF Kw amps. (And for Sean and Joe, also overlooking JA's self-described "mixed bag" of measurements.)
I can think of at least two reasons why truly negative reviews hardly ever appear. One is that there probably are hardly any products that deserve to be totally panned, but more fundamentally, I think most reviewers (correctly) take the position they cannot in good conscience come across as being completely authoritative or definitive in their criticism just because something fails to ignite in their system, to their ears. Of course, the same reasoning of moderation and qualification ought to apply to 'raves' as well as flops, but many reviewers ignore this sensible proviso (not necessarily citing Fremer here, but Bolin probably qualifies). Combine the reticence to pan with the propensity to drool, and you have the recipe for reviews in general seeming too liberal with their praise, and often just plain unbalanced. Thus (one of the reasons for) the continual suspicions of undue manufacturer influence, and for anything less than a head-over flip often being perceived by manufacturers as essentially a 'dis'. Underlying this phenomenon seems to be the unspoken assumption that most readers would rather thrill to breathless hosannas than digest thoughtful, realistic assessments. |
Just a point of defining our baselines for discussion (to which rebuttals are welcome): I noticed some responses above seeming to imply that maybe companies like those taken to task here for making products which don't measure the way some people think they ought to for the money, did not devote sufficient engineering time and resources before releasing half-baked gear. Although this might happen from time to time (and although many products do get 'upgraded' after their introductions), I'm much more inclined to give companies like Hovland, Pass, or Legacy the benefit of the doubt (for good or for ill), inasmuch as I tend to assume that their products' performance, both audible and measured, is entirely the intentional result of deliberate design choices made by competent and careful professionals, whose reputations and future livelihoods are understood to be on the line with every product introduction.
Ultimately, you can't go very far for long selling only the equivalent of fancy faceplates and colored lights in any business, and I don't think a quick buck is why most designers get into audio (possibly excepting some in the cable field), whatever one might conclude about their performance priorities or degrees of native talent. There's probably room to argue about assessments of the prevailing level of basic design competence and/or the depth of the talent pool relative to past eras, but to me it's overly cynical to suspect a widespread, shoddy disregard for anything besides short-term profits in such a risky, tough, and fanatical business as audiophiledom. In short, I'm willing to grant from the start that most designers believe wholeheartedly in their products, and strive to make what in their opinions are the very best they can for any given design brief and price point. To sometimes disagree with the results is one thing, but it's another to suggest that those responsible either don't know enough, or even worse, really know better. Ours is a marketplace that encourages diversity, and that includes appearances, prices, and meausured performance as well as sonics ; if that seems to be in conflict with 'objective' notions of accuracy and value, then it was probably ever thus. |
While I admit that I more admire, concerning this one narrow issue of specsmanship, companies inclined to understate measured performance rather than overstate it, there is gear which I am happy with in my own system from companies that I believe do indulge in a little 'optimism' in this area. Is this 'fraud'? If it is, I can't say I care very much as long as a component nevertheless gives all the sonic enjoyment I could ask of it.
I feel it is fairly routine, for instance, for electronic gear tested in Stereophile to show S/N figures that are worse than claimed, or I/O impedance figures bearing only a passing resemblance to spec. On the other hand, amp power figures into various loads and at differing distortion levels often seem to be as frequently understated as overstated, and *both* directions of disparity could be, perhaps ironically, attributed as much to marketing considerations as anything else. When it comes to speakers and frequency response, tolerances criticized above must be taken into account with the knowledge that the way in which test results are obtained has a major impact on reported figures, and that JA's test regimen is quite limited in certain crucial respects that could cause it to diverge significantly from methods which a speaker manufacturer might deem most representative for modeling real-world performance.
Anyway, I still say none of this actually establishes that new gear is 'getting worse', or is in general more at odds with its claimed specs than was older gear. |
I don't even think it's a matter of which do you trust, your ears or the specs, because often the two have remarkably little relation to each other. I don't think this is because measurable performance has no bearing at all on perceived performance in theory, but because A) the ways in which gear usually gets measured often have little resemblance to how it's actually used for listening to music within a system context using human ears, and B) there aren't necessarily measurements which can be made to directly quantify many observable sonic characteristics, but only by inference and extension, which are no substitute for actual auditioning. I suppose this means that in order to demonstrate that "new gear is getting worse", one would have to show that it actually *sounds* 'worse' - and this of course is not objectively possible to prove (and I'm not even sure anybody is contending that this is the case). |
