Analog Electronic Components for Too Much and Just Enough Money [Issue No. 11 Winter/Spring 1988]

Two unique products for the ultimate signal path and five other really good ones are not a bad haul for any reviewer. Here there is even the entirely new experience, in one instance, of electronic engineering as a satiric art form.

We had enough philosophy on audio amplification in Issue No. 10 to last us and our readers awhile, so we shall plunge right into our reviews this time, but not before musing wistfully about one thing.

You cannot have more power supply per cubic inch for high-quality audio amplifiers than is possible with the Carver magnetic-field technology. Furthermore, a magnetic field power supply can be designed to deliver any amount of current into any kind of load, even if the existing Carver models are limited in that respect. Another thing you cannot have, at least not to our knowledge, is a more accurate and stable stage of amplification than the JE-990 discrete operational amplifier. Would it be too much to ask for a power amplifier with an all-990 audio signal path and a magnetic field power supply? It would be very compact, not terribly expensive and awesome in performance. The 990 circuitry is in the public domain, and the Carver technology is available under license, probably for a smallish royalty. Any takers? Boulder Modular Preamplifier (interim report) Boulder Amplifiers, a division of Silver Lake Research, 4850 Sterling Drive, Boulder, CO 80301. MS11 Phonograph Pre amplifier, MS21 Selector Switch, MS32 Output Controller, MSO01 Power Supply, $2744.00 the system. Tested samples on loan from manufacturer.

We received this entirely new and eagerly awaited all 990 preamp/control system a bit too late for completing the tests we need before publishing a full-fledged review, which you can expect in Issue No. 12. Here we just want to report that the equipment exists and that our initial impressions of it are very favorable.

The packaging of the Boulder "front end" is in very professional-looking separate modules, each 14" deep and occupying just under 5" of horizontal shelf space. The MS21 Selector Switch and MS32 Output Controller modules, although separate in construction, are generally delivered as an integrated double module; the MS11 Phono graph Preamplifier module is always separate and can be placed right next to the turntable. The MS01 Power Supply module can power up to four amplification modules through computer-type connector cables.

We have not even listened to, let alone tested, the "MS11, which was delivered with a built-in Deane Jensen step-up transformer for MC cartridges, thereby creating an apples-and-oranges comparison problem vis-a-vis the active pre-preamp stages we have been using lately. Boulder will also come out with an active MC pre-preamp option; we hope to be able to report on both versions. What we have been using with a great deal of satisfaction is the MS21 cum-MS32 line-level control unit. The best way to describe it is as an upstream extension of the Boulder 500 power amplifier; it has the same 990 sound or rather lack of sound as the voltage-gain stage of the 500 and ends all discussions regarding the desirability of a passive line stage because it is equally inaudible but provides the few dB of gain needed to augment the output of CD players, etc. Various neat little touches like individual left- and right-channel polarity inversion, stereo channel reversal and suchlike goodies, all with indicator lights, also distinguish the unit from the run-of the-mill, as does the dead silence in the absence of a signal, regardless of switch and control positions. Balanced output is an important professional option.

The price may seem steep for what is basically still "just a preamp" (of course, so is the SP11 at twice the price), but the quality of construction and parts is extremely high, and the modular format, while more costly in terms of metalwork for the separate chassis, allows you to buy only what you need and specify the options you can use. As the Boulder literature a bit facetiously points out, "With modularity, obsolescence is not part of the design. For instance, when CD's are no longer available and you need a Phono

Module after all ... well, use your own imagination." (Are Jeff Nelson and Randy Gill closet digitophobes?) Speaking of Boulder's prices, Randy took us to task for wishing out loud in print that the $2875 Boulder 500 were available for around $1995. He feels that the amplifier is an extraordinary bargain and asserts that any number of well-known high-end manufacturers would not hesitate to charge $6000 for a product built from a comparable list of parts. Randy, we believe you, but to prove your point you should get competitive bids from Bill Johnson and, say, Dave Hafler.

Carver "'Silver Seven"

Carver Corporation, P.O. Box 1237, Lynnwood, WA 98046.

"Silver Seven" vacuum-tube monophonic power amplifier, $8750.00 ($17,500 the pair). Tested samples on loan from manufacturer.

Imagine you are canvassing the farthest-out fringes of the high-end audio cult and asking the freakiest, tweakiest and richest equipment fetishists what they would consider to be the absolute ultimate power amplifier, regardless of cost or practicality, if someone were willing to make it for them. Almost certainly they would ask for a vacuum-tube design-more "musical" than solid state, right?>-but much more powerful than any currently available, maybe 500 watts per side, monophonic of course, made with a whole forest of 6550's (the most expensive output tubes), bigger transformers than the world has ever seen, various cultist brand polypropylene capacitors, Van den Hul silver cable throughout-remember, these tweakos seriously believe in all that stuff.

Now, one amplifier designer who definitely does not believe in all that stuff is Bob Carver. As any reader of our Issue No. 10 surely knows, Bob is totally convinced that he can achieve exactly the same audible results with transistors and his magnetic-field power supply, at a comfortable three figure price. We were completely flabbergasted, therefore, when we found out about the Carver "Silver Seven," which is exactly the power amplifier hypothesized above but very much a reality-and priced at seventeen and a half kilobucks the pair! For a moment we thought that Bill Johnson or possibly Dan D' Agostino had started to do business in a Bob Carver rubber mask. We quickly cornered Bob and demanded an explanation. His reply to us, not verbatim but in essence, went something like this: "No, intellectually I don't believe in this engineering approach, it seems to me quite unnecessary, but I had so much fun doing it this way. Why should only the esoteric amplifier designers have this kind of fun? Not that I took the Silver Seven project lightly-the amplifier is as good as I know how to make, as good as anybody knows how to make, in fact better. The tweaky features can't possibly hurt--I've never been able to hear one iota of difference between copper and silver wire for example, no matter how hard I tried, but silver is at least as good, so if somebody believes it's better that's okay with me. I'm easy. More important is that I know I can t-mod a regular Carver solid state amplifier to sound indistinguishable from the Silver Seven, and I'm going to do it because that's what I really want to sell. But the Silver Seven is the kind of amplifier I used to dream about when I first got involved in audio, and now I've got one!" Thus spake Bob, approximately. We detect a trace of ambivalence in his posture; sure, he now has his Fantasy Island amplifier and his "ultimate" transfer function, but he is also parodying the engineering style of the lunatic fringe as a satirical commentary on the high-end audio business.

The satire is no less poignant for being essentially harm less. The net benefit to all of us will be the Carver M-4.0t solid-state power amplifier, the Silver Seven clone Bob is currently working on. At $799 or $850 or whatever (we are told it will be in that general price bracket), it promises to be an interesting item indeed.

As for the Silver Seven itself, it is almost porno graphic in its visual impact, the sexy tube look carried to orgiastic extremes. Each channel is on two separate chassis, power supply and audio, so that a complete stereo amplifier comes in four monstrously heavy units. These have a very high-gloss black finish with discreet silver accents and are suspended on special isolation feet, which rest on marble slabs cut to the size of the chassis. No kidding. There are 15 (count them) huge "milk bottles" lined up in three rows on each audio chassis; 14 of these are output tubes, 7 up and 7 down in push-pull. In front of them are three smaller tubes and a special heat sink for the latter. Each audio channel has not one but two giant output transformers, the rich man's way to buy deep bass and extended highs without the usual trade-off between transformer size and leakage inductance.

Even bigger is the power transformer on each power-supply chassis, but the sexiest feature there in our opinion is the slanted panel with two 1950's-style round meters for volt age and current. Absolutely sinful-there ought to be a law against it.

Measuring the Silver Seven is made somewhat complicated by the fact that it has three output transformer taps for various load impedances and three bias settings that affect not only the class of operation but also the output impedance. That means nine possible transfer functions and therefore nine times the usual amount of work on the lab bench. (Lots of luck with the t-mod, Bob!) We stopped quite a bit short of the whole nine yards, sampling only a few typical lash-ups. The results were sufficiently revealing to satisfy our curiosity. The official power rating of the amplifier is 475 watts into 8 ohms from 20 Hz to 20 kHz with less than 1% THD. At the 4-to-8-ohm tap with the middle bias setting, we measured less than 0.25% THD at that power from 27 Hz to 15 kHz. Only at the frequency extremes does the distortion rise anywhere near the specs.

Clipping at 1 kHz under the same conditions is seen at 547 watts with 0.1% THD; power bandwidth referred to 475 watts with 1% THD is 17 Hz to 28 kHz. The small-signal (1 watt) frequency response has its -3 dB points at 1 Hz and 125 kHz, and there is no transformer peak whatsoever. THD at that signal level is in the 0.02% vicinity at all frequencies except the lowest. Square waves look as good as we have ever seen, without the slightest evidence of transformer coupling, and super stable with capacitive loads. At the 1 to-2-ohm tap with a 1-ohm load, the picture is equally impressive. Square waves still look great. Clipping with a 1-kHz CW input occurs at 625 watts, with 22 Hz at 600 watts, with 15 kHz at 529 watts. Using a signal with a lower duty cycle, we measured even much higher 1-ohm clipping wattages. This is not a load-sensitive amplifier.

Our listening tests resulted in the rather unexpected finding that the Silver Seven sounds uncannily like a big brother of the MESA/Baron M180 reviewed in our last issue. That, of course, is very high praise, as we had never before heard a tube amplifier as good as the modified M180, and the Carver has the same utterly convincing, suavely musical quality but with considerably greater authority, solidity and dynamic impact on account of its higher power.

Especially with inefficient or very low-impedance speakers, the higher attainable SPL and superior low-frequency sock put the Silver Seven into an even more exalted category.

For the $16,200 difference in price, the highly religious community to which the Carver amplifier is addressed may in addition be deeply moved by an extra measure of that in effable midrange liquidity, by new epiphanies of imaging... but we must stop being so nasty. Let us simply say that both the M180 and the Silver Seven are correctly designed, the latter with about 4 dB more power, and correctness is not a highly variable quantity. Again we have to note that the sound is not particularly "tubey," very neutral in fact, although the focus is ever so slightly softer and the subjective balance a little warmer than with the ultrahigh definition but less powerful Boulder 500. A lovely sound by any criterion and, of course, an adult toy beyond compare.

One more comment. Although much more expensive than the Audio Research M300, the Carver "Silver Seven" is not overpriced in the same sense. It is merely overdesigned and unaffordable-not the same thing when you think about it.

Citation 21

Harman/Kardon Incorporated, a Harman International Company, 240 Crossways Park West, Woodbury, Long Island, NY 11797. Citation 21 Control Preamplifier, $549.00. Tested sample on loan from manufacturer.

Citation is a name fraught with early-1960's nostalgia to your Editor and his generation. Harman/Kardon originally launched the late Stew Hegeman's cost-effective state-of-the art tube designs under the Citation name, and as you know we still think very highly of the Citation I preamp after all these years. We also recall, but not as fondly, a solid-state Citation preamp designed by Stew a number of years later, still in the iron age of transistors. In the early 1980's, Harman/Kardon resurrected the name in an attempt to market a price-no-object high-end line designed by Matti Otala. These electronic birds of paradise, the Citation XX power amp and Citation XXI preamp, never flew-wrong product at the wrong price at the wrong time from the wrong source-but they were extremely high-tech and made an impressive statement. The current Citation line is closer to the original in concept, offering advanced engineering and quality construction at reasonable prices slightly above the Harman/Kardon scale. In the case of the Citation 21, the designation seems to echo that of the Otala super preamp, although there is no real resemblance, the new unit being simpler, more practical, more versatile. It is definitely a child of recent progress, illustrating how much preamp you can buy in 1987-88 for a little over $500 when a good company tries hard.

We like the understated black look of the Citation 21, accented by tiny green LED's placed inside the unit's small (but big enough) push buttons. There is a large number of the latter to implement the unusually wide range of control functions-yes, even defeatable bass, treble and loudness controls are included, and may the purist gods have mercy on Harman/Kardon's soul-indeed, we cannot think of any thing omitted except such far-out features as polarity inversion or stereo channel reversal. A nice little bonus is the CD Direct switch, which bypasses all active and passive stages to permit direct connection of a CD player to the power amplifier without unplugging and reconnecting any cables. We could live with this kind of input, output and control flexibility for a long time and never feel deprived.

The circuit-board layout, signal routing and grounding used in the Citation 21 show a great deal of thought and genuine awareness of what can happen to a fragile signal before it reaches the output jack. The circuit philosophy is ultrawideband, as you would expect of the company that made bandwidth a marketing issue, and the RIAA equalization is effected passively in the open loop with further fine tuning in the feedback loop, resulting in superb accuracy.

Our measurements revealed +0.0 dB equalization error from 40 Hz to 20 kHz (i.e., literally none), with an apparent boost of perhaps 0.3 dB at 20 Hz which may have been an artifact of our test setup. Overall, that betters the manufacturer's specs and also beats our September 1987 sample of the more than three times costlier Krell PAM-5 (at least in that one respect). In general, all our measurements corroborated the specs, which of course are those of an extremely high-performance, no-compromise unit and therefore quite unnecessary to nitpick on individually. The only design detail we disagree with is that the flat-gain MC amp, which can be switched in ahead of the equalized phono stage, has an input impedance of only 56 ohms. We prefer to see several hundred ohms here for a variety of reasons; loading the cartridge down is the easy way out for cartridge and circuit designer alike. A small quibble, since the MC sound of the Citation 21 is highly respectable by any standard.

As we explained in Issue No. 10, we see a gradual convergence toward a single standard of sound quality in this type of audio component, and it would therefore be inappropriate for us to emulate the onanistic descriptive style of some of the underground reviewers in an attempt to define the sonic profile of the preamp. We tried some rather quick and not entirely conclusive ABX comparisons instead; in the case of the MC and MM stages we ran into an apples and-oranges dilemma because of input impedance and RIAA equalizaton differences that were more significant than any actual amplification differences we might have heard. The line-level stage of the Citation 21, however, acquitted itself quite nobly against all comers, including the new Boulder super preamp. Our statistical sample was much too small, in the number of trials as well as the number of listeners, to even tempt us to jump to a sweeping conclusion, but we can certainly state that the 21 is not "blown away" by the very best, regardless of price and notwithstanding the fondest hopes of the high-endniks. When switched into and out of the tape loop of the Boulder, the line-level signal path of the Citation seems to add a barely audible layer of smearing or veiling, but this is a horrendously severe test, in effect comparing A against A+B, rather than A against B, and the effect is truly minuscule. We also noticed when trying to adjust the volume control of the Citation for unity gain in this test that the pot did not track quite as precisely as the expensive P&G in the Boulder. Once again-no free lunch.

Our net impression of the Citation 21 is very much on the positive side-a well-engineered, handsome, versatile piece of equipment at a surprisingly affordable price. We think Stew Hegeman would have approved of it.

Citation 22 Harman/Kardon Incorporated, a Harman International Company, 240 Crossways Park West, Woodbury, Long Island, NY 11797. Citation 22 High-Voltage/High-Current Power Amplifier, $999.00. Tested sample on loan from manufacturer.

This one is a genuine bargain-two very nicely built 200-watt mono power amplifiers on one chassis, sharing only the line cord, loaded with features, for less money than many a pricey moving-coil cartridge.

We know the kind of engineering thinking that went into the design, having had some interesting discussions with designers Richard Miller and Marty Zanfino, both of whom are hardened veterans of the Matti Otala era at Harman/Kardon and have thoroughly digested the Citation XX experience. This is the kind of amplifier design you would expect to emerge from that background: 250 kHz small signal bandwidth, only 12 dB of overall negative feedback, high instantaneous current capability, symmetrical circuitry and layout, discrete solid-state devices in the signal path. A special feature is a high-voltage operating mode, available via a rear-panel switch, which allows an 8-ohm speaker to draw as much power from the amplifier as a 4-ohm one, thus making the power specs the same into either load. The idea is that an additional high-voltage rail is cheap whereas high current capability is expensive, so why not let some one who has already paid for the latter get the full benefit of it regardless of the higher impedance of his speaker? Rich 30 and Marty are obviously using the old noodle. Bridged mono operation into 8 ohms is another switchable option, with 400 watts output available in that mode.

In the July 1987 issue of Stereo Review, that favorite bogeyman of the high-endniks, Julian Hirsch, published a very thorough test report on the Citation 22, with accurate laboratory measurements now fully confirmed by our own.

As our older readers know, we hate to reinvent the wheel and then talk about it; suffice it here, therefore, that the amplifier meets its specs comfortably but is not quite as straight-wire-like on the lab bench as the Boulder 500, which was the only other high-powered, solid-state, true voltage-source power amplifier we had available for logical comparison (at more than three times the price, to be sure).

We were very happy, even so, with the Citation's bench performance; we found no eyebrow-raising anomalies.

We also used the Boulder 500 as the foil for the Citation 22 in a brief ABX listening test. Levels were carefully matched within +/- 0.1 dB. As long as we knew which was which, we could have sworn that the Boulder sounded more solid and authoritative, with a deeper soundstage and a warmer character, the Citation giving an ever so slightly thinner impression by comparison. Whenever X was select ed, however, neither your Editor nor a very capable associate could reliably distinguish the two. We suspect that if we kept recycling the same 20 seconds of some particularly revealing piece of music and worked at it for hours, we could zero in on the difference, which is obviously very small or possibly nonexistent. Again, our statistical sample was far too small to yield conclusive results, other than the unquestionable fact that the Citation 22 is no slouch against high-priced competition. (Maybe better than the old XX?) There is one thing about the amplifier that we heartily disliked, indeed resented. That is the perversely "different" speaker terminal hardware Harman/Kardon is forcing on the user. It requires unscrewing a so-called end post, stripping the end of the speaker wire, spreading the bare wire strands radially around a gold-plated hole at the back of the end post, then screwing the latter back into the terminal-four times over to complete the two red and two black connections! If you have fancy banana plugs soldered to your fancy speaker wires, hey, cut them off. We ended up preparing banana-jack adapters, but we can assure you that the Hungarian maledictions that ensued in the process would have scorched the ears of a sergeant of the hussars. Incidentally, Julian Hirsch complained about the same thing. Maybe we are all getting old, deaf and ill-tempered.

That aside, the Citation 22 has our full endorsement.

Citation 23

Harman Kardon Incorporated, a Harman International Company, 240 Crossways Park West, Woodbury, Long Island, NY 11797. Citation 23 Active Tracking Tuner, $599.00. Tested sample on loan from manufacturer.

This is not really a review. We do not review tuners for two reasons: we have no RF laboratory and we have some serious doubts about today's FM stereo broadcasts as a high-fidelity medium. As we wrote more than eight years ago, "Why would you want to buy expensive new shoes when all the streets in town are unpaved and muddy?" The Citation 23 tuner is so nice, however, that the least we can do is to acknowledge its existence as part of the new Citation line and briefly report our experience with it.

The claim to fame here is exceptionally high rejection of FM adjacent-channel interference without the usual trade offs in stereo separation and THD. Theoretically, there should be no such thing as an adjacent channel in a given broadcast area, since FCC frequency assignments to stations are supposed to be at 400 kHz intervals (alternate channels), leaving a vacant 200 kHz channel between any two local stations. The FCC definition of "area" collapses, however, in the suburban sprawl between our cities, so that the real world spacing of stations from the point of view of a typical residential antenna is indeed at 200 kHz (i.e., 0.2 MHz) intervals. Frankly, it has never happened to us that we were just dying to hear a station blanketed by a more powerful neighbor only 0.2 MHz away, but the Citation 23 could have satisfied such a craving. It is, in effect, two FM tuners in one: a very good conventional tuner with the usual broad band-tuned IF filters and an entirely new kind inspired by aerospace communications technology, without passive filters of any kind. The latter takes over when Fine Tuning is selected and uses a phase-locked loop (PLL) circuit to lock onto the FM carrier and track its modulations within very precise limits, reading only the signal necessary for good stereo separation and low distortion. If you insist on all the gory details, there is an excellent technical brochure available from Harman/Kardon on the entire Citation line and also a highly competent review of the tuner by Leonard Feldman in the January 1988 issue of Audio. Unlike us, Len is a good RF man, and in this context we must voice the same confidence in him as we accorded Julian Hirsch above. These guys know how to measure.

More interesting to us was listening to the Citation 23, with a high-quality live broadcast of an orchestra as our program source, well captured by our Dennesen two-element indoor antenna (which is all we need for the garbage even the better stations put on the air 99% of the time). It was distinctly our impression that the Citation gave us deeper, tighter, more detailed bass and a sweeter, more musical top end than a number of not very recent but respectable tuners we had kicking around the lab and a larger number we had gone through over the years. We realize that this is not a particularly precise and authoritative statement, but then FM stereo is not a particularly precise and authoritative medium. We could certainly live happily with such a tuner; we even like its styling and controls, which match those of the Citation 21 preamp. One minor complaint: the never lit "FM" and "AM?" status indicators on the display panel should have been removed, not just disabled, when their functions were reassigned to LED's on the selector buttons.

Oh yes, the Citation 23 also incorporates an excel lent and relatively wideband AM tuner.

Citation 24 Harman/Kardon Incorporated, a Harman International Company, 240 Crossways Park West, Woodbury, Long Island, NY 11797. Citation 24 High-Voltage/High-Current Power Amplifier, $649.00. Tested sample on loan from manufacturer.

The best way to describe this power amplifier is as half a Citation 22 at two thirds the price. For that reason, it is not nearly as good a value, although it still includes the extra high-voltage rail and is just as nicely made; we doubt, however, that our type of reader would opt for it merely to save $350.

Two Citation 24's in the bridged mono mode would still be slightly bettered by a single Citation 22 for $299 less; the smaller amplifier has that much less power supply (one power transformer for the two channels, etc.) and not quite the damping factor, not quite the waveform perfection, not quite anything of the bigger one. The whole thing is an attempt to mimic the flagship amplifier at half power for marketing purposes. We are even willing to concede that the attempt is successful-barely successful-electronically and sonically, but the excitement is simply not there. We never order the 7-ounce steak, either.

Hafler XL-280

The David Hafler Company, 5910 Crescent Boulevard, Pennsauken, NJ 08109. XL-280 "Excelinear" Power Amplifier, $600.00 wired ($525.00 kit). Tested sample on loan from manufacturer.

This is not exactly news; various reviewers and editorial letter writers have been having fun with Dave Hafler's straight-wire differential test (SWDT) and the "Excelinear" tweak on the XL-280 for well over a year now. The reason for this brief report is that the Hafler people, after reading in Issue No. 10 about our involvement in the various and notorious Carver null tests, sent us an XL-280 and one of their XL-10 passive switchboxes "to play with." Contrary to their tacit expectations, we found a serious fly in the ointment.

The X1L-280 does not have a flat frequency response.

Why this was never mentioned in the voluminous writings on the subject is a total mystery to us. The fact is that above the audio range the response starts a gradual rise to a peak of almost 7 dB somewhere between 360 and 370 kHz, with a Q of approximately 2, in the worst-case position of the Excelinear variable-capacitor trimmer. In the best-case position of the trimmer the peak is still in excess of 3 dB but at a somewhat higher frequency. The square-wave response of the amplifier always shows a severe leading edge spike as a result, so that one could never in a technical sense talk about straight-wire-like behavior, since straight wires pass square waves unaltered, n' est-ce pas? We know exactly how this came about, but that does not mean we approve of the situation. What we are dealing with here is a bit of Marketing Man's Engineering. The marketing man wants to be able to say that his amplifier is more like a Straight Wire with Gain than anyone else's. He will never tell you that those words are a popular and convenient figure of speech rather than an exact electrical definition of the perfect amplifier. The scientific truth is that the perfect amplifier is better represented by a low-pass filter with gain, the filter having flat response up to a certain high frequency and rolling off with a controlled slope above that point. The delay through such a filter is a law of nature and a fact of life, not something undesirable to be gotten rid of for the sake of music. The peculiar response profile of the XL.-280 is an attempt to reduce the delay that shows up as an I/O difference in the Hafler SWDT setup to less than that of a theoretically perfect amplifier-just to look more like the popular icon of the Straight Wire with Gain. Marketing triumphant over physics.

We are in no way suggesting that the Excelinear gimmickry affects the audio range and thus degrades the audible performance of the amplifier. The XL-280 is a well engineered MOS FET power amplifier in all other ways; its 200-plus clean watts per channel into 4 ohms and very fine sound could easily sell it without any hotshot legerdemain.

For our quickie ABX listening tests we set the variable capacitor in each channel to yield the smallest possible high-frequency peak, not the best SWDT null. The results were much the same as in the case of the Citation 22; there appears to be no such thing as a high-powered, solid-state, true-voltage-source power amplifier that is "blown away" by any other of the same breed, unless there is some major foul-up somewhere. The similarities within the category are so much greater than the barely (if at all) perceptible differences that it will take us a lot more work than we have done on the subject so far before we can commit ourselves to any definite preference. The facile A-is-better-in-the-midrange than-B kind of audio journalism seems downright irresponsible to us at this point. We do prefer the XL-280 to the Citation 24, however, simply because the Hafler offers more clean output for less money, although it may not be quite as beautifully built-American kit design versus Asian consumer packaging.

--------

[adapted from TAC, Issue No. 11 Winter/Spring 1988]

---------

Also see:

Top-of-the-Line Digital Components: CD and Beyond

Cartridge, Arm and Turntable vs. the Groove: Who's Winning? [1977]

Various audio and high-fidelity magazines