Who reviews the reviewer? On TIC, we answer that age-old question by asking our highly qualified, always opinionated and never bashful panelist David "Anonymous" to analyze the Shootout results. When Jim fires off a Shootout, it might come bouncing back at him from an unexpected angle. Duck, Jim!
I try to do each shootout without prejudice. This means I try not to look at schematics, examine things under the hood or draw ANY conclusions as to the front end, the analog stage or materials used. This is not always possible as I already know some tuners' "innards." The direction I choose, or it chooses me, is sound quality ONLY (with a good antenna system). That is, hopefully, with no preconceived thoughts of how the tuner got there from the 75 ohm input to the analog outs. I think David factors in how good any certain tuner SHOULD be after dissecting it. Not only is that good but really needed for a different point of view. A more scientific one, as it were. That is the only caveat I would want everyone to be aware of. At least that's my take, that David and some others factor in what a tuner should be because of engineering and build quality, not just what it IS when the sound reaches their ears.
All of the tuners I discuss in Ricochets are listened to via a dedicated Antenna Performance Systems APS-13 on a rotor. Cabling is quad shielded RG-6 coax with a shield ground to lower noise and measured low VSWR. Levels are matched to within .01 dB on a 7-digit true rms HP meter. The fixed output of each tuner is always used for comparisons (see comment below under "Soapbox").
The system is described under my profile on the Our Staff page, and additional details are presented here. All electronics is Boulder-based due to its neutrality, high input impedance (200 K ohms to avoid loading poorer tuners' outputs too heavily), and low noise and distortion. Spectral and other Boulder gear is available when needed. Volume control tracking is assured via a custom .01% discrete Vishay series attenuator. Separation is tested to be greater than 100 dB at or below 15 kHz so that the image of the tuner is not altered. Please note that some "respected" preamps have less separation at 15 kHz than the best tuners and stations, which is not too swift! Input RCAs are all WBT, assuring superb contact. All interconnect cables are professional, super-shielded, semi-rigid Transformation Audio Systems products with locking RCA collets, perimeter grounds and active shield bias. AC cables are the LAT AC2 throughout. Speaker cables are professional Transformation Audio Systems reference low (2 ohm) impedance, two-gauge silver designs. Speakers are usually a 5-way custom design designed for recording studios, but Snell A references, Quad ESL989s or Sony SSM9ESDs make appearances.
All equipment is run off six dedicated 20-amp circuits with twisted 6-gauge conductors run off the opposite phase as the major current draw items in the rest of the house. Sensitive gear is run off a PS Power Station and all products are surge-protected via a Brickwall surge suppressor. Vibration- damping products are generally Transformation Audio Systems constrained layer pods, but products from EAR composites are also employed.
The system resides in a 42' x 17' x 13' room with many angles and significant sonic treatment. Speakers fall on the solid, windowless 17' x 13' wall and are backed by a 2x6 wall reinforced with double-layer flakeboard externally and cross-braced for low resonance. The floor is tongue-and-groove 1" flakeboard glued and screwed (every 4") to TGI joists below. Listening is done just below the acoustic center of the tweeter and mid at about 43" above the floor which is covered by felt-based pad and wool Berber carpet for its acoustic properties. The floor in the area of the speakers is two layers of flakeboard with a constrained layer in between, covered with another 1" of oak tongue-and-groove flooring for rigidity and lack of resonance. Every effort has been made to absorb early reflections and diffuse the reflected sound in the room.
Tests are run knowing the type of tuners as well as double-blind. Any lack of correlation between the open and double-blind results is questioned for validity and repeatability. As Jim mentions, I do look at the tuners, their schematics and construction quality to understand where the designers' biases are. However, this is factored out because during the double-blind tests, I only know that one of the tuners is a Sansui TU-X1. I have someone else install the tuner to keep bias from creeping in. While I am definitely not without bias, I know and admit where those biases fall. I do admit that I have a strong bias toward good electronics design practices and may not take cheaper, lower-model tuners as seriously as maybe I should. However, in my opinion, real tuners have 5-gang or better front ends and this alone rules out many lesser players! All tuners are compared against my reference Sansui TU-X1. Both a SME/SME and Transfiguration Temper Supreme Mk V analog front-end and a Custom Philips CDM9PRO based drive and Boulder DAC are available for comparison with records and CDs (can be valuable when the playlist of your favorite station is known). One listening chair is placed centered between the speakers about 15' away (essentially far field listening).
Tests can also be verified via Sennheiser HD600 headphones with the fabulous Headroom Max preamp with stepped attenuator. If you don't know about the Headroom guys, are on a budget and love good sound, check them out! This gets around the acoustic issues associated with all rooms and is a valuable tool for judging tonal balance, dynamics and other non-image related audio properties.
In case you think I dove off the shallow end of the pool or drank too deeply from the "pond" near my house, I know that some of this is overkill, but it makes me happy and assures me that the items in my control are not seriously compromised! Now if the #@%! stations and source material were better, life would be better!!!!
Now a moment of SOAPBOX talk from David: Note that if you are using the fixed output of one tuner and the variable output of another to match levels, you are likely biasing the listening results. If you are not matching levels, then you probably are reaching at least some incorrect conclusions. Usually, the channel tracking of the pots is terrible and seriously damages stereo balance and overall imaging. Furthermore, the impedance is sometimes higher in the variable output, making it more susceptible to loading problems (can be a real issue with a low-input impedance preamp - lots are 10 K or 20 K ohms!). Finally, some tuners like the Technics ST-9030 have filters added in the variable output that increase treble rolloff, but improve the birdie problems. So make sure that you know what you are comparing so you do not reach incorrect conclusions. Many listening tests that I have seen or heard about are really evaluating the quality of the pot and frequency response of the tuner/preamp combination. Tests are also done in the lab to check important parameters that are seldom or never checked (like RFIM and IF symmetry and phase). If the tuners are not in alignment, then you are definitely not reaching valid conclusions!
David's top three tuners: the TU-X1, L-02T and F-26
I am increasingly of the belief that the Sansui TU-X1, Kenwood L-02T and Pioneer Series 20 F-26 are really in a class by themselves. I hear much bigger differences between stations overall than between these tuners! However, none of these tuners sound identical and each has different merits and weaknesses. In my system, with my FM market, I rank the TU-X1 #1, the F-26 #2 and the L-02T #3 (on a sonic basis). Each can make a claim under various listening/RF market conditions as the best production tuner made. There may be others, but none that I have seen or heard yet. All three embarrass all of the more recent digital tuners that I have compared to them. There is a gap between these three and the next best tuner that I have heard and measured.
I noticed that I had to reposition my speakers to get the best out of each tuner. For example, the TU-X1 has so much stage width (separation) that placing speakers too far apart yields a hole in the middle and diffuse sound (as Jim notes in his review). This effect can also be a problem when trying to optimize a system for CD versus LP. The TU-X1 has better frequency extremes (both measurable and audible) but will not show those advantages if the speakers are bandwidth limited, placed so that they have ragged frequency response, or in a system optimized for LP instead of CD. The L-02T appears to favor smaller speakers, because many smaller designs don't have the extreme bass extension and to a lesser degree extreme treble smoothness of larger (read this as 3 - 5 way, not physical size) speakers. Placement of the speakers can be further apart with the L-02T. The bandwidth compression scheme of the L-02T's IF appears to hide or at least diminish some of the low-quality nasties so prevalent in broadcast radio these days, but also makes it prone to certain types of external interference. The F-26 is not as sensitive as either of the other two tuners, but possesses an ease and naturalness in the mid-treble (especially on an excellent classical station) that is not always present in the TU-X1 & L-02T. Overall, I would place the F-26 a little behind the TU-X1 and L-02T under most (but not all) listening situations. Note that the F-26 stages a comeback if you have a station that requires the narrow IF for decent reception due to a crowded dial, multipath, etc.
I must mention that one has to be careful with the F-26. There is a circuit that selects between the wide and narrow IFs. It is linked to the touch sensor and if that control is set too low (either on the back panel or the internal pot) oscillation between the wide and narrow IF/detectors can result. It can be bad enough to kill a downstream preamp under some situations. So make sure this circuit is working before it gets connected or your preamp may die like mine did!
Overall, I like the TU-X1 best because of its extension on the top and bottom, because it is the second most immune to WWV beats against the IF frequency, and because it has a very "big" dynamic sound. (I reference WWV because of its enormous power. It is a significant factor over much of the western US.) I would liken it to symphonic music whereas the L-02T appears to lean toward chamber music. The F-26 is approximately halfway in-between the character of the other two tuners and may be the most neutral in terms of scale. I am soliciting feedback from other parties that have all three tuners to see if they agree with these findings. Bottom line, if I were in a FM market with lots of multipath or reliant on DXing to get an adequate number of stations, then the L-02T and TU-X1 excel. If I was looking for the last bit of sound quality or in a market where there are lots of closely spaced stations, I would lean toward the TU-X1 and the F-26 especially if the system is optimized for CD. If I had to pick one tuner as being capable to handle anything that I can throw at it, the TU-X1 wins by a nose. Clearly, all three tuners are truly excellent and are capable of superb results. If you told me that I could only take one tuner to the proverbial desert island, it would depend on the proximity to a station and would either be the TU-X1 or L-02T! [More on the L-02T here; more on the TU-X1 and the L-02T here.]
[Editor's Note: A jerk in an Internet discussion group who had never used an L-02T carelessly claimed that Jim was "biased" toward it.] In regard to being biased toward the L-02T or the TU-X1, hey, it is as Voltaire said: "The best is the enemy of the good"! The other factor is many people are tied to their belief that they have the best and when anyone challenges this, then the other person is the one that is biased. I find it interesting that tuner collectors with more experience and more tuners are more open to being wrong or further enlightenment than those that have owned few tuners, but love the ones they have. Enlightenment is time-consuming and expensive, and requires that you are willing to be proven wrong!
Another factor is that many collectors have brand or appearance biases that prevent them from trying certain tuners. I know of at least three heavyweight collectors that do not own a mint TU-X1 because it is big, black, looks like an old 4-channel receiver, and is made by "SAN-SEWAGE" (I used to be one of them). Another factor is that many collectors want to dislike the tuners that you have to pay dearly for... and there I can't blame them! Finally, there are those collectors that are stockpiling huge numbers of classic tuners that don't want to tell you that their favorite is a L-02T or TU-X1. The L-02T is rarer than the TU-X1 in the U.S., but there are still a significant number turning up for sale (albeit at rather dear prices). Asia is still a relatively untapped source for both tuners!
In regard to Jim's testing methodology, it is valid given the constraints that he describes. I find it hard to agree with a few of his ratings myself - like the Luxman T-117 at #3 (too high in my opinion), the Yamaha CT-7000 at #14 (too low in my opinion) [more on the CT-7000 here], or the Technics ST-9038 at #5 (way too high in my opinion), but that doesn't mean that what he heard with his units in his system isn't correct! Inter-component synergy is a very funny thing and it is poorly controlled. In addition to the easy phenomena like output and input impedances and AC/DC coupling, there are differences in grounding, shielding and other factors that will definitely influence the resulting system sound. I have seen cases where an otherwise average design locked in perfectly with certain other gear. Bottom line, Jim is providing an invaluable service to TIC readers and his results may save them thousands of dollars in search of the perfect tuner for their system and market. Certainly I know of people who spent tens of thousands before they found the TU-X1 or the L-02T! I hope that Jim doesn't take some of these people personally, because their opinions are worth what we paid for them ;-).
Jim responds: I don't disagree, entirely, with David about the T-117 and CT-7000. Also my tube system with mini-monitors can't go as low in the bass as his system can.
Analysis of L-02T compared to L-01T, L-1000T and KT-917
I spent a weekend (literally) analyzing the L-02T. It is no wonder I was having trouble following the signal flow looking at the guts of my tuner. This is easily the most complex tuner that I have ever encountered. It has the most fully realized front end of any of the Kenwoods. While it is similar to both the L-01T and L-1000T in some aspects, the lack of tuning capacitor sections in local mode and the RFIM in distant mode are both issues for the L-01T. The L-1000T's front end is made mediocre by the varactors in the front end. The L-02T's front end even outperforms the KT-917's because of the balanced JFET mixer instead of the diode mixer. The JFETs allow the oscillator to be run at a lower voltage and this solves the radiation problem that the KT-917 can have if you run multiple tuners in one house (essentially the KT-917 can broadcast 10.7 MHz because of the drive requirements for the diode bridge). Practically speaking, the KT-917 is close to as good as the L-02T, but one has to be careful of its role as a broadcast unit! [More on the KT-917 here.] The local oscillator of the L-02T is very nicely done. It uses similar semiconductors to the top Pioneers and Yamahas except that they double up on the 2SC710 to lower noise by about 3 dB. Overall, the L-02T's front end is one of my favorites.
The L-02T's IF is enough to make you cry. It is explained in theory, but comparing the merits of the additive/subtractive mixer and VCO is extremely difficult compared with a standard IF configuration. After a lot of adjustments of my L-02T and a ton of measurements, I am convinced that when aligned correctly, this IF is among the best (along with the TU-X1 and F-26). It is inelegant from an engineering viewpoint, because it takes so much circuitry to implement and it should be shielded better than it is for best performance. However, this is one of the great IFs out there. None of the other Kenwoods has an IF that is close to the L-02T's, in my opinion. That is part of why the traditional Kenwood traits of lack of bandwidth (600T) and/or lack of life and richness (KT-917 and L-01T) are not issues with the L-02T. The other reason is the detector.
The detector is an excellent PLL type. I find this ironic, because Kenwood was such a champion of pulse counters and they abandoned them for the better performing PLL in their ultimate (in my opinion) tuner. I think the ratio detector in the TU-X1 is better, because of instability and noise in the PLL. A double PLL, fully shielded, would have improved this tuner further. However, overall, the L-02T wins among Kenwoods again in this stage.
The MPLX is the one of only two areas of weakness in the L-02T. On the positive side, this means that one can get even better sound by running the detector out to a separate MPLX and audio stage. When I did this over the weekend it ran very close to the TU-X1 and F-26, with all three using the same external multiplex unit. I wish that the L-02T used the MPLX from the L-01T or KT-917 (very similar to each other and based on the HA11223W). The KT-917 wins this stage, but the L-02T is no slouch either. I just wish they had done a better job. So the L-02T is in third place for the MPLX stage among Kenwoods.
Like virtually all tuners of interest, the audio stage sucks! When one combines poor op-amps (or simple discrete circuits) with low-value electrolytic coupling caps, low drive current and poor filtration, you get what was designed in, e.g., bad sound. This is another argument for implementing an external MPLX/Audio. The L-02T is too expensive to modify and damage the value, but fortunately you don't have to because virtually all of the compromises are after the detector.
The power supply is typical Kenwood, i.e., the best in the industry for tuners. It is amazingly large and performs to an excellent level. However, it is going to be a big pain in the butt to replace all of these aging electrolytics with Elna Cerafines and Black Gates.
As with almost all Kenwoods except the 600T, shielding, cabling and layout are not done too well. Kenwood could have learned a lot from these practices within the Yamaha CT-7000. One reason for this is the size and complexity of each of the boards. However, any top-notch tuner should have a detector shield as the CT-7000 and the F-26 have incorporated.
Overall, the L-02T is clearly one of the great tuners. I would caution any prospective owner that few have the ability to align this correctly (not sure I have it totally right and I still can't optimize the IF because it requires a distortion level of .001% from the RF generator!). However, the tuner did sound better than prior to adjustment (Jim - I owe you one there - I almost returned this tuner to its previous owner!). Overall, I think that this one will stay in my tuner collection and is prized with the F-26 and TU-X1 as the best production tuners that I am aware of. The Technics ST-9700 may join this list, but I am still working on it. My gut says that it will be more competitive with the CT-7000 than with the top three, which kills me because of the effort that I have invested in it.
Now I know why so many collectors that own an L-02T is tight-lipped about how they like it! It is likely because they want another one and they are exceptionally rare. I am trying to cultivate a Japanese source of these and will let you guys know if I am successful (right after I get a second one ;-).
There is an interesting note in the margin of the Kenwood L-01T schematic in the service manual. It shows that there were 75 tuners made with a decent quality pulse count detector and that the bulk of the units have a seriously compromised detector and therefore sound. Jim's comments are indicative of the problematic range (most) of the L-01Ts. The much better group of 75, have sound fully comparable (according to one quality listener) to that of the KT-917. If you are able to find an L-01T in this range of 75, you have THE rarest Kenwood and a darn good tuner. If not, you are better off with the KT-917. In regard to the KT-917, it is a fabulous tuner from a RF perspective, but I can't agree with Jim's ranking as #6 (at least not in my market with my system). When I listen to the KT-917, I hear a tuner that is lacking in timbre and detail. I hear excellent bass, but a slightly edgy treble (birdies or other nastiness the usual culprit). I hear a wide, shallow soundstage with a degree of dynamic compression. I do not like the IF, detector, layout, shielding or cabling for a tuner of this price range at all (a change of previous thoughts on this tuner)! The power supply and front end are among the best in any tuner. The KT-917 would be a candidate for improved IF, detector while retaining the excellent MPLX. Overall, I would place the KT-917 and the L-01T approximately in 15th place from a sonic standpoint. Both tuners can be improved by gutting the audio stage and starting over! I do not recommend this to the casual enthusiast because of the design knowledge and layout expertise that is required.
As an aside, I have not found any pulse count or digital detector that equals any of the better quads, ratios or PLL detector types. It is interesting that Kenwood abandoned the pulse counters for their ultimate tuner (the L-02T). Some, like Jim Bongiorno, have been a "voice in the woods" in regard to the problems with pulse count detectors. While I apparently have some differences of opinion on the Sumo Charlie that can be traced to the Mitsumi front-end, the lack of correct alignment and a poor standard of construction, I happen to agree with many of Jim B's opinions particularly in regard to detectors. The problems are there for the measuring, so to speak, if one has the time, understanding and equipment. I have seen all three of the better types of detectors (quad, ratio and PLL) result in exceptionally low distortion (.01-.03%) when done correctly. While some of the latter Accuphase units have made better design choices in regard to pulse counters, only one of their tuners cracks my top 10 based on sound quality. Therefore, as a general rule, if your taste goes more for performance and can tolerate the periodic alignment, stay away from the pulse counters as a group, as none of them recovers the signal faithfully.
Wow, an F-91 above the TX-9500II and the TX-9800. Here I have to differ with Jim's viewpoint substantially, but I will note a positive Pioneer bias on my part. While the F-91 is not the worst of the latter Pioneer F series (that dubious honor is reserved for the F-93, F-99X or F-90, all of which suck air in various ways) it falls behind all of the better classic Pioneers in my opinion (see below). In my system, the F-91 shows good MPLX performance and decent sound, but highly flawed implementation in the front-end and IF. The presence of a strong interference source 50 miles away literally prevents it from operating correctly part of the time. The sound is two-dimensional, albeit wide in terms of sound stage. In regard to spectral balance it is cool, thin and threadbare in my system. Please note that my system has substantially lower noise and distortion than is achievable by any tube gear which may be part of the reason for the difference. Overall, I find this tuner to be attractive, quiet, but very forgettable.
My own personal pecking order of Pioneer tuners is: #1: F-26, #2: F-28, #3: TX-9500II, #4: TX-9100, #5: TX-9500, #6: TX-9800 (but only in another market - the lack of shielding makes it unusable here), and then the rest.
I read Jim's shootout of the TX-9500II and find it interesting that it did better in RF terms than the TX-9800 (which has the same front end without the shield). I believe this supports a point that I have been making about the importance of shielding. It is interesting how well the both the TX-9500II and the TX-9800 did in Jim's RF tests compared to the superb L-02T, which has the best RF performance that I have ever seen in a production tuner. It is also interesting that TX-9100s sell for so much less than Accuphase T-100s when they are approximately equal when taken in total, and that the TX-9500 seems to get so little respect when it betters so many popular Kenwoods.
I can't agree with Jim's placement of the TX-9800 above the TX-9500II, overall. I have owned two TX-9500IIs that I recall (so many tuners, so much tuner affliction ;-) ), and both have outperformed in an audio sense all of the TX-9800s that I have owned. I have likely spent more time listening and analyzing each because I am a fan of Pioneer tuners. Jim's comments suggest to me that this TX-9500II may have an alignment issue. Normally, while the TX-9500II clearly doesn't have the depth of the best, the bass up to the midrange has always been rich and full. I agree with Jim's comments in regard to the output Pioneer IC - it makes a fair amount of difference compared with the stock version. Jim, I'm hoping that you get a chance to hear the F-28 and F-26, because they are in a class by themselves among Pioneers.
The F-26 is one of the top tuners in my opinion and the F-28, while not perfect, is a really good balance between price and performance (one of the best in my experience). I have described the sound of the F-26 elsewhere; the F-28 is not as good but has a similar sound that is slightly thinner, more two-dimensional and colder in balance. The F-28 has a modified version of the classic TX-9x00 5-gang front end. It uses a varactor for the LO to form the weird equivalent of a 6-gang tuner. The performance at its usual selling price of around $275-400 is exceptional and can be easily improved with a little work. Both of the Series 20 tuners are superb and place in my top 10 when it comes to sound quality. The F-26 is differentiated by a truly superior RF and IF, and by its overall higher construction quality.
The CT-7000 is a conundrum for many tuner enthusiasts. It is a striking-looking tuner with an elegance to it that few other tuners convey. It also represents the high-water mark for classic tuner construction (bettering the superb construction of the Accuphase T-100, McIntosh MR 78, Pioneer Series 20 F-26, etc.). In particular, the separate boards, shields, and shielded, impedance-matched interstage cabling for each stage of the tuner are exemplary and convey performance advantages in areas where interfering signals are present. The quality of the components used in this tuner is also above the grade typically found in most tuners of the day. The CT-7000 is also unusually well-balanced overall and receives very good marks for both its RF and audio qualities. This tuner deserves to have the power supply caps, RF/AF connectors and few other parts (audio coupling caps) replaced with modern high-quality replacements. Overall, I consider this one of the better tuners ever made (this is a change in my position on the tuner based on two exceptional and reworked samples). The tuner's display can also be improved with some of the high-intensity blue LEDs (or clear with the colored cap over it) as a replacement for the original bulbs which are a pain to replace. A final touch would be a detachable AC cord with an internal CORCOM. It is a must to align this tuner correctly! Finally, the CT-7000 is one of the best for headphone users both because it has a jack and internal amp and because its frequency balance is complementary to some of the best headphones. In fact, add a pair of Sennheiser HD600s at their current low price to this tuner and you have an excellent-sounding FM system.
However, that being said, the tuner should be better than it is. I would have preferred more selectivity in the front end. I would have also liked to see LC filters in the wide IF as well as more selective ceramic filters in the narrow IF. It would have been nice to see the output filtration set a touch higher and the bass response flatter and better controlled. Overall, I agree with Jim's listening comments, but have a few points to add. I would rate the CT-7000 higher than Jim did, I would place it between 4th and 9th out of all tuners (on sonics) depending on how one reacts to its weaknesses. The tuner has a mid-bass thickness and is slightly polite in the treble. Its image depth is excellent. The image width and specificity is slightly off the pace, but the naturalness and timbre of the tuner helps to draw you in anyway. The deepest bass is not bad, but sounds attenuated because of the plump mid-bass. The tuner is much quieter than one would expect given the specifications which were more a function of the test gear of the day and not of the tuner itself. This is an excellent tuner for local stations that have exceptional sound and don't have too many interfering stations in proximity. Lacking in selectivity, this tuner is a poor choice for the NYC market or for DXers.
In conclusion, I believe that the CT-7000 is a "must-own" for serious collectors, provided that they are in a market that is neither DX-reliant nor too crowded with signals.
I can't say that I am surprised about the outcome of the KT-815 Shootout. It is a good little tuner, but one that performs roughly at its price level rather than above its price level, unlike the Pioneer TX-9800, TX-9500II and Series 20 F-28. I am more a fan of individual tuners rather than any one brand. If I have a bias (gee, do ya think ;-)), it is for attractive tuners that also sound great. Sadly, there appears to be nearly an inverse relationship between my sense of style and the better tuners! I really dislike the looks of the TU-X1, for example - it reminds me of a '70s vintage 4-channel receiver (apologies if anyone likes that look).
SAE Mark VI:
The SAE Mark VI... where do I start? When I think of both my experience with this tuner and that of a friend of mine, it brings a slightly different image to mind than Jim paints! I think of turning it on late at night, like Jim does, but instead of wine... I think of being clad in a Nomex fire suit with a quality Halon extinguisher at the ready!! I prefer listening to this tuner at night so that the first signs of flame are readily apparent. While I do not disagree with Jim's overall assessment of this tuner's sound quality, I wonder why anyone would chance a fire for a tuner that is neither as good nor as cheap as many on the list above it?? The Nixies are interesting, but I would prefer a tuner that I was fairly sure would never cause me to read my homeowner's insurance disclosures. I would recommend avoiding this one!
The Denon TU-850 is a nicely made, solid analog tuner that can be had cheap from the right source. I find that it is a nice-sounding tuner, albeit one that is lacking at both frequency extremes and is less three-dimensional than the tuners that are in my top 20. It is one of the more consistent and reliable older tuners in my experience. It qualifies as a decent tuner using my definition of at least 5 capacitor gangs and dual IF bandwidths. Its strength is that it is a very quiet tuner with a decent balance between RF and AF performance. All three examples that I have owned have had excellent stage width, if not specificity. The looks of the tuner depend on how you react to the almost "communications equipment" looks of the unit. I imagine that this tuner will stay working longer than some of its generation because of the parts and construction choices. This is a tuner that will outperform most of what is available new today and will certainly be working longer! However, in terms of classic tuners, the lack of extension (top and bottom) and the lack of depth and image specificity make an otherwise excellent tuner fall to a lower position. I would image that the build quality and the RF performance are worthy of some analog updates by a qualified party, but then why not start out with the Pioneer TX-9x00 series and get better results?
Nikko Gamma V:
The Nikko Gamma V is interesting to me for a variety of reasons. It represents a lesson to many that a decent linear phase LC filter IF, combined with a decent detector and HA11223 MPLX, can sound so good. I agree with Jim that this one is a little bright, but that can be traced to the audio output stage where the usual ICs and caps are screwing things up. Note that the Gamma V is not too different from most tuners in that regard! As a 6-"gang" tuner, it qualifies for the coveted status of a "real tuner" in this reviewer's opinion. It is built very well for its day as several others have correctly pointed out. I think that this is a clear value tuner for people who want (limited) presets and a digital readout. However, I am not a big fan of the Gamma V's RF front-end stage. It is prone to RF intermodulation, has increased noise compared to the best tuners, and has some lesser front-end problems too.
In terms of sound, the tuner has very good bass and is overall tonally quite decent, except for the bright and slightly thin highs. The Gamma V has a very wide soundstage, but is lacking in depth and the elusive "draw you into the music" quality of the best tuners. I definitely like tuners like the Accuphase T-107/T-11/T-108 and the Pioneer F-28 better, but the Nikko is a good tuner for the money. I would recommend it for people who don't need to DX or sort out stations that occupy adjacent channels (e.g., in cities with good stations spaced out decently). The Gamma V sells fairly cheaply relative to its build quality, power supply quality and overall sound quality. I also find that Gamma Vs are reliable when not abused, and have shown less unit-to-unit variability than many.
McIntosh MR 78:
The MR 78 is an interesting tuner and is one that causes people to form polarized opinions. I have owned two, both from the later series that had better output filtration. I have also heard a Richard Modafferi modified unit and while it was better, I found it more of a DXer's tuner than a tuner for the best sound.
On the plus side, it is one of the best-built, best-shielded units out there. It has a well-designed and executed power supply, too. The shielding is not just a quality or appreciation factor for those of us who live near a strong interference source. It is important to note that IF selectivity is NOT the only factor in receiving clean stations in a packed RF market. This tuner has a very good RF front end that is relatively low in RFIM [RF intermodulation - Editor]. Also, this is a tuner that meets its published specifications in my opinion. However, an ongoing issue that I have with the industry is that too few graphs and scope/spectrum analyzer photos are shown to allow the technical enthusiast to look past the specs. Further, specs like RFIM and a complete quieting curve should be shown, as they tell a lot about a tuner. For those who operate totally on manufacturers' specifications, one should be aware that many tuners that I have measured not only failed to meet specs, but still failed after they were optimally aligned! Comparing specification sheets is a lesson in futility! There are some that have great specs (for the ones they choose to publish) and are basically useless in the presence of a strong interference source! Enough of the technical soapbox for now....
In terms of pros and cons elsewhere, there are several factors to consider. I am a fan of multiple IFs and believe that it is impossible to build a tuner that will suit most markets without them. However, the problem is typically that a tuner either has lots of selectivity in all IF modes or (more frequently) inadequate selectivity in all modes. I know of tuners that fall into both camps. Many have taken to selecting a tuner with great wide IF sound and then substituting GDE ceramic filters to bump the narrow selectivity. This is relatively straightforward, PROVIDING you have the technical background and the proper equipment and can select the best filters from a fairly large lot. In my experience, the production tolerances are quite large, so if you buy one or two and use them, chances are you are making your tuner worse. I would have loved to hear this tuner if the alternate channel selectivity steps were 30 dB, 80 dB and 120 dB instead of the quoted 55 dB, >90 dB and >>90 dB. It is my opinion and experience that it is very hard to get an IF with ideal gain/phase performance if it is too narrow. Too few companies have employed group delay equalizers in at least their wide mode to optimize sound quality. I believe this is where the MR 78 loses some sound quality. I also believe that, depending on other design decisions like IF/limiter circuitry, dual or triple detectors matching the requirements placed on them by the IF characteristics can provide performance improvements. There are a few tuners that do this to advantage.
Another area where I think the MR 78 loses a little sound quality is in the MPLX section. Although in fairness, one must remember or read the tuner thread on significant tuners, that the best ICs for this function were still years away. There is little question that the LA3450 and the HA11223 are capable of better performance. Again, I would have loved to see this as a factory or Modafferi update.
The audio section falls short of what is possible now, but again this is not a recent unit! One has to appreciate what Modafferi and Mac accomplished in 1972 when most tuners were junk!!! We owe a dept of gratitude for the performance level and innovative thinking that pushed Kenwood to do the L-02T, Sansui to do the TU-X1 and Pioneer to do the F-26. The MR 78 is a non-trivial tuner and could be one of the best with a few updates. It is also capable (in fact more capable than most) in its ability to receive stations in the presence of interference or interfering adjacent stations! People should be glad that this tuner exists. It virtually forced the other manufacturers to do better because their designs of the day (and many since) pale in comparison.
That being said, the MR 78 is both measurably and audibly behind other tuners optimized for maximum sound quality. Take a look at 1 kHz THD, the phase characteristics of the IF and the performance of the detector and MPLX, and a few flaws show up. The fact that this is true should not lessen the pride of ownership! It simply reflects its age, the ICs of the day and the designer's focus. I hear sound that is relatively extended on the bottom and a touch less so on the top. It does not have the separation or depth information of the best and can sound a little thin compared to the best.
However, I can't agree with placing this tuner in 36th place! There is zero question in my mind that it will outperform many tuners on the list above it. While I don't disagree with Jim's sonic assessment, one of the problem of these tests is that the focus is always on the top 5 or 3. It would be an interesting "true-up" to return to a few of the lower tuners and compare them against one another. This is much easier to say than to do, but I suspect that the accuracy of the comparison is such that some lower tuners might move up and others might move down if one did this. It also depends on the listener's priorities and system. That is why I say there is no World's Best Tuner - there are compromises in every tuner that I have seen, heard or measured. How those compromises match your FM market, your system and your preferences is EVERYTHING! I have little doubt that in some markets the MR 78 would fall in one of the top 2 or 3 places! I would definitely put the MR 78 above the Magnum Dynalab MD-108 (currently 9th) or the FT-101A (13th) and its lower-rated siblings (24th and 33rd), higher than the Pioneer F-90 (11th) or any of the other F series tuners (F-99X, F-91 and F-93) and above most (but not all) Kenwood tuners!
From a RF perspective, there is little doubt that the MR 78 is one of the best. From an audio perspective, it commits acts of omission rather than commission, so it never sounds bad, or noisy. Many tuners don't pass this basic test!!! It just doesn't have the draw-you-in, super low distortion, phase linear sound of a few of the best audio tuners. On my own list of top tuners where RF, AF and construction practices are all considered, it places in about 10th place overall. If one focuses on RF and construction, then it is in the top 4 or 5. If one focuses on sound only, then it slides back down to 12th or so. Also worth noting, I believe that this is a tuner that with some care and update will outlast the user! There are not many tuners that I am comfortable saying that about.
So in summary, while I don't disagree with Jim's sonic assessment of this tuner, I am compelled by its other virtues to place it higher, personally. If I had to go to a desert island and take the MR-78, the Sequerra Model 1, or one of the Magnum Dynalabs, I would definitely take the MR 78 (even if the proverbial desert island was local to only a single clean station)! In terms of a historically significant tuner, this is maybe THE most significant because it forced everyone else to do much better! My thanks to Richard Modafferi and the Mac design team!!!
The Revox B760 and I have a tolerate-hate relationship. Let me digress. Circa early 1979, a defense contractor comes to me and asks if there are any consumer FM tuners that look good to me for observing lower-cost receiver construction and design techniques. I recommend the Revox and 4 others (thank goodness for the 4 others)! The company engineers proceed with testing the Revox and find that it has a bad mixer design that results in extremely bad RF intermodulation. The tuner was also found to have a poor capture ratio and the poorest quieting of any of the five. Since it was the most expensive of the group, I took some heat for this personally. Also, I have owned several B760s, including two bought new from a failing Chicago audio store. They were quickly outperformed by my first "real" tuner. So I am admitting a negative bias on this tuner, albeit one formed based on measurements from some of the world's best RF engineers! Please note that these comments are not inconsistent with Jim's shootout DX results.
In terms of sound, I don't have the same level of praise as Jim does for this tuner. While it has good bass, a rich midrange and a polite high end, it has some flaws too. The image projects well forward of the speakers, even on recordings known to have a significantly deeper soundstage. This is accompanied by a lack of the crushing dynamics of the top tuners like the Sansui TU-X1, Pioneer F-26 and Kenwood L-02T - probably due to the B760's high noise floor. During quiet passages, noise is evident on all but the finest stations. I find the B760 mostly pleasant to listen to except that it won't pick up some stations and is noisy. To my ear, it is cut from the same cloth as several warm-sounding tuners like the Sequerra Model 1 and poorer examples of the Yamaha CT-7000. The CT-7000 makes an interesting comparison because it is usually warmer sounding than the Revox, but has far fewer RF anomalies and much better quieting (both steepness of curve and maximum quieting). I also find that while it is pleasant, the Revox has a more rolled-off high end than the better examples of the CT-7000 (however this requires careful alignment of the discrete MPLX decoder). On an audio basis, the B760 doesn't make my top 20! However, it is important to note that I am particularly biased against noisy tuners. I am also biased against tuners that can't receive so many otherwise serviceable stations. In my market, this tuner brings in 13 fewer listenable stations than the TU-X1.
On the construction front, the B760 is a tuner that has a lot of parts... make that a lot of old and aging parts. Some would argue that so does the Kenwood L-02T, but there at least it results in fabulous performance! I always like the application of the KISS ("Keep It Simple, Stupid") principle as long as it is not taken to the point of diminished performance. I worry about the long-term ability to keep the B760 running (although I have first-hand experience with Revox that was exceptionally positive). When I combine RF and AF performance with construction and tally the tuner's overall ranking, it fails to make my top 30 tuners. So while there are similarities between Jim's review and mine, I would only rate this an OK tuner. I know of tuners that routinely sell for less than $200 US that will outperform this one on every front, and therefore I do not recommend it.
So how is the TU-919 designed? It is quite different from the TU-X1. It is a 5-gang tuner and those that know me well understand that I think that this places it in the second class after the 6-9 gang group. The absence of the extra gangs does make the tuner quite sensitive, but more susceptible to RF intermodulation and spurs. It uses GDT ceramic filters instead of LC filters and a group delay equalizer like the TU-X1. However, the TU-919 does a fairly decent job and shows good selection practices for the ceramic filters. It loses a noticeable amount of sound quality because of the relatively narrow "wide" IF setting. This practice always incurs a sonic penalty. The 919 has a good MPLX stage, but has the normal problems with the analog stage. While appearance is a matter of subjective preference, I really like the looks of this tuner better than the TU-X1. But then, I am beginning to believe that all of the great tuners are relatively homely in various ways. The 919's internal shielding is approximately equivalent to that of the TU-X1, and both are off the pace of the best in this regard. Until recently, the TU-919 technically was a better buy for most users than the TU-X1 at prevailing selling prices. However, TU-919s had been moving up in price range, even before the recent $1,000+ eBay sale, and can be outperformed by tuners that are significantly cheaper (more on this in the future).
In terms of the TU-919's RF performance, the ability to pull most local stations, immunity to interference, and RFIM all suffer compared with the TU-X1. Only on widely spaced, distant stations (where extra sensitivity is required) did the TU-919 beat the TU-X1. In terms of sonic performance, I find a much larger difference than Jim reports in his system. I hear some thinness, a lack of dynamics and a little more noise on higher-frequency signals than with the TU-X1. I also find that the TU-919 has a slightly more distant perspective with a little less image specificity. The elusive "you are there" quality is not as obvious with the TU-919. I like the sound of the TU-919, but it is clearly inferior to a perfectly functioning TU-X1. The TU-919 always has a few more extraneous noises than the TU-X1. Overall, the TU-919 is neither my first or second favorite Sansui tuner; those honors remain with the TU-X1 and TU-9900, respectively. Finally, the TU-719 is not as good overall as the TU-919.
To sum up, the TU-919 is a great tuner at $300, but at double that price (or more), it blends in with the pack. The TU-9900 shares more of the design philosophy of the TU-X1 and I believe that it is still a good buy.
The Technics ST-9030... wow, do I have a long history with this tuner! I remember the early ads (circa 1978) from Technics that stated that "audiophiles (or was it perfectionists?) would feel better if this cost over $1,000." Well, it didn't, and in shows in spots and not so much in others. In terms of value, there are a lot of these tuners out there, so don't pay $300 for one. I have bought several perfect ones for less than $100 and sometimes less than $50!
Let's start with the good news. This is a pretty good tuner and it still doesn't cost an arm and a leg on the used market (although many tuners are increasing in price). Many still operate perfectly and depending on whether they were rack mounted and cared for, many are still in good shape physically. It is a historically significant tuner because of its 8-gang tuning cap and a balanced mixer. This alone puts it in a pretty elite group. The tuner has one of my favorite front ends, which unfortunately is not matched by equal excellence elsewhere. The layout and shielding of the front end is first-rate, too. I agree with parts of Jim's review and not as much with others (could be unit-to-unit variation again). I had one ST-9030 that had a mid-bass hump like Jim reports that is reminiscent of other tuners like the CT-7000 et al. I had another that seemed to go all the way down and didn't have a hump. Note that the ST-9030's parts quality was not the equal of some of the top-dollar units from that era, and that may account for the variation. Also, the coupling caps used vary significantly in actual value, as they do in many tuners. I find the sound of the stock ST-9030 to be pleasant, warm, unfatiguing, but ultimately forgettable. I am glad that the tuner didn't have a pulse count detector because most from that era resulted in a tuner of average to poor sonics. This is a tuner with tremendous modification potential from an outboard MPLX all the way to a front end onward keelhauling! How successful you will be depends on the level of surgery and the talents of the "doctor."
On the negative side, with a number of changes, this tuner could have been one of the really great ones ever. But in Technics' defense, it would not have been in the same price point if the problems were corrected. The layout (other than the front end) is sub-optimal, as are the shielding, cabling and MPLX. While the use of two detectors for the dual IF is interesting, one really good one would have been cheaper and would have allowed funds to improve other areas. In terms of the dual IF, I am glad to see LCs for sound and ceramics for selectivity. This makes sense, but it would have been nice to see some group delay EQ to tweak it for better performance. Now to the sonic problem of the tuner. The frequency response, distortion, and lack of immunity to MPLX nastiness pull the rest of the tuner down into the forgettable bucket. Use of the detector output and a really first-rate MPLX/audio stage makes this a darn good tuner and competitive with or better than many more expensive units. On the sonic front, lack of depth, lack of involvement and that "you are there" quality and lack of image specificity all hurt this tuner in stock form.
Overall, I would not place this tuner so far down the list because of its merits. Remember if the front end sucks, you can't get the signal back with great parts and design further down the chain!! This is the undoing of many tuners! There are a number of tuners that Jim listed above the ST-9030 that I would not concur with; for example, I would place all of the following tuners lower than the ST-9030:
Pioneer F-99X, Pioneer F-90, Kenwood L-07TII, Magnum Dynalab FT-101A Etude, Technics ST-8077, Sumo Charlie, anything from Revox!, Magnum Dynalab FT-11 New, SAE Mark VI or any other model, Phase Linear Model 5000 Series 2, and possibly others....
However, Jim raises an excellent point with his summary "inoffensive but bland." I would group the ST-9030 with another that I have a long history with, the Yamaha T-2, as the "cutoff tuner between the keepers and the also-rans." So is it a good tuner for cheap, a great tuner for modification into a real contender or a good tuner to sell while their price is inflated... well, that depends on you, your location, your talents and your budget. In my opinion, it is all of these things, but to different people. It is, however, a tuner that is historically significant. Are there cheaper tuners that are much better, yes there are! But you will have to wait for the scoop there.
My views on the Accuphase T-109V have resulted in me being called crazy by detractors. I was early in my praise of this tuner as an excellent modern tuner. Many still say it was unlistenable, which it is not and neither is the T-109. It is easily the best tuner produced anywhere in the world in the last ten years! It is clearly and decisively better-sounding than some tuners that I used to think were among the best of all time (like the Kenwood KT-917, Sansui TU-9900, Pioneer F-28, etc). For those that don't or can't admit when they were wrong, this is an example where additional study with measurements can keep a person honest. The golden-eared subjectivists out there should really balance their approach with measurements to see what traps they are falling into.... So where does the T-109V stand in my current stable of tuners? Well, it doesn't, but I'll explain that more later.
The Accuphase T-109V is a very significant tuner. It is the best modern-day tuner by a fairly wide margin, and I'll take it all day long against the current group of production tuners like Magnum Dynalab, etc. It is a significant tuner historically for a number of reasons. It is one of the two finest tuners from this respected manufacturer, it has balanced outputs and a remote control, it is well-made compared to most tuners, old or new, and it may well be the last great tuner from Accuphase or, for that matter, any vendor. It is now discontinued because of the lack of availability of the output module. Also, if you like the gold champagne look, so popular in Japan, it is a striking tuner. Since the tuner is new, the service life should be longer than for many of the classics which are beginning to have parts go bad. Accuphase tuners have stayed working for unusually long periods of time, anyway, as a testimony to the construction and parts quality used by this manufacturer. So all in all, this is a tuner that people with very substantial collections should have in their stable of tuners.
So how does this tuner sound? Deceptively good. It doesn't do anything significantly wrong and is blessed with some of the better RF performance of any recent tuner. Is it better than the T-109? Yes, it is. Is it the best Accuphase tuner of all time? Well, that is a judgment call, but I personally like the T-108/T-11 vintage tuners slightly better. In many systems, the T-107 can equal the T-109 and come within a decent margin of the T-109V. I disagree with the large difference in rating between the T-109 and T-109V by Jim. Based on several samples of each, they are closer than Jim's rating would have you believe, and the circuitry is quite similar between the two. I think that the T-108/T-11/T-107 tuners are better values for the budget collector because they sell for substantially less than a T-109V. With the great review by Jim, the prices on the T-109V are only going to go up.
Back to the sound. This tuner has a purity about it that is enchanting. It also possesses the very solid bass that Jim mentions in his review. It does not equal the stage presentation of my TU-X1s, nor does it have the station-grabbing power and immunity to RFIM, overload, and the time standard (WWV), or the other characteristics that are lost using varactors. I would not recommend it to people that live in very crowded FM markets like the Northeast U.S., for people near the time standard in Colorado (WWV), or for people in areas with substantial multipath issues (mountains or downtown areas dominated by tall buildings). These people are better served by a TU-X1, L-02T, F-26 or other quality tuning capacitor-based FM tuner. The T-109V has easily the best pulse count detector of all time, but it is measureably and audibly inferior to a really first-class ratio or quad detector. The IF is excellent, but the use of GDT ceramic filters is no substitute for quality LCs with a Group Delay Equalizer (GDE) like the Sansui TU-X1 and others use. Don't get me wrong, GDT ceramics have their place in narrow IFs and can also be used to advantage with wider passbands for wide IF tracks. They are also very easy to use and readily available compared with top quality LC types.
Overall, the sound of the T-109V is very pleasant and very clean, but it doesn't have the "you are there" liveness or ability to draw you into the music of the best tuners that I own. I still prefer the TU-X1 to both the L-02T and the T-109V. Adding another TU-X1 recently just added another data point in the camp that says it is a better tuner than the T-109V or the L-02T. The presence of a great ratio detector and multiple GDE LC IF stages really does yield benefits. The IF chosen by the designers of the L-02T was abandoned by experts at Harris and Motorola years ago in favor of top quality GDE LC standard type IFs. Even the best pulse counter detectors lose information that the best "old technology" ratios and quads can extract, so both the L-02T and the T-109V give up sound ground to the TU-X1 in the IF and the detector, respectively. The T-109V has an advantage over many classic tuners because of the high level of integration (low parts count), especially compared to the L-02T. The T-109V utilizes a relatively high-quality output stage, which helps the sound against the T-109V and other tuners, as well. So where did this tuner end up in my collection? Well, in the recent housecleaning that I did, I sold it in favor of several better tuners. Now, I have joined the heretics on the other side by saying that while this tuner is excellent, it is not the best ever or even in the top five.
The T-109V also doesn't measure the best of any tuner that I have owned. Several that I have owned were capable of better performance when realigned (however, it is true that virtually all production tuners benefit from realignment). It is one reason why some collectors pay big dollars for the designer of a tuner to tweak the adjustments for best performance. A cautionary note is important here because many tuner alignment centers do not have the quality of equipment necessary to do the job well, so if you get tuners realigned, find out how long they spend on each tuner, what they do, and what gear they use. While the Sound Technology gear gets rave reviews, there were better units available from Japan. As many manufacturers have found, you really need to look interstage at the signal rather than just take a Device Under Test (DUT) approach to get the best results. Finally, one must measure more parameters than are spec'd by most manufacturers to get the best performance.
So is the Accuphase T-109V at least the best varactor-tuned tuner in the world (since electronic tuning became the norm instead of weighted tuning knobs, string and variable capacitors)? Well, no, it doesn't achieve that honor, either. It may have the best combination of construction quality, sound, measurements, and durability/longevity of any varactor tuner, though. So where does this tuner come out on my alltime list? Well, several tuners that I have found recently have posted surprisingly excellent results, so that neither the T-109V or T-109 (or, for that matter, any of the Accuphase tuners) fall within the top five. As for the identities of these previously unknown tuners of surprising excellence, I am still not ready to tell all. My next Ricochet will start down the path of explaining the evolution of a group of tuners that is surprisingly excellent and will begin to tell why most of the group equals or betters the best Accuphase tuners. So am I still a fan of the Accuphase marque and the T-109V in specific, yes I am, but I am reducing my collection to distill down to the five best tuners that I have ever found. I have four out of five and will have to wait for the fifth because it is the rarest of all FM tuners and was never produced commercially....
Jim, while I do not agree across the board, I still think that you have done an amazing job trying to sort the whole tuner question out. If people read your comments it can save them thousands. I think that if you ever go back and do a championship tiebreaker, it could open up a whole new can of worms. Everyone wants to know THE best tuner. Well, as we have all learned, it isn't that easy. With variability from tuner sample to tuner sample, different requirements for different RF environments and local vs. distance issues, one could debate this endlessly. If you add convenience features, aesthetics and other attributes, it gets almost impossible. I tend to think that the approach of a group of excellent tuners as you have done is the best.
Sansui TU-X1 vs. Kenwood L-02T:
The Sansui TU-X1... pause, drum roll... yes, it really is as good as they say! At this moment in time, I have owned five (maybe six in another 32 minutes and 24 seconds) of these heavy, utilitarian-looking tuners, and have heard another three of them. While I am not generally a fan of this brand as a whole, Sansui did build a few really terrific tuners. The TU-X1 tuners have less unit-to-unit variation than most tuners out there, but I have heard some that are substandard. I have heard a total of three that had inadequate quieting and sounded like they had a combination of alignment problems and perhaps some parts mortality issues. Alignment helped, but did not correct all of the measurable or audible problems.
I must surmise that the Sansui TU-X1 evaluated by Jim is one of those that needs some work. I base that on the fact that the only times I have heard any diffuse sound from a TU-X1, it was due to the low-level noise smearing the intrinsic focus of the tuner. I have two L-02Ts to compare to my three current TU-X1s, so I am quite sure that I have a good idea of the sound of the L-02T. While the L-02T is an exceptional tuner, it is now down to fourth or fifth place in my personal rankings (sorry, I had to whet your appetite for future Ricochets). What I hear with the TU-X1 is excellence and a lack of emphasis on any one range, from the deepest bass to the highest treble. The tuner portrays scale better than almost all other tuners in that the image changes dramatically from chamber music to symphony. The TU-X1 is exceptionally gifted at drawing the essence and emotion from the music. It is very easy to get caught up in its "you are there" sound quality. I hear a tuner that has a slightly forward presentation, but with awesome depth when it is present in the material. Playback of some known excellent recordings, courtesy of some radio station friends, was quite interesting because I could compare them to the CD and LP presentations of the same music. A few live broadcasts also helped to fill in information about the sonic differences.
Based on Jim's comments, I used seven different speakers to re-evaluate the TU-X1. Those speakers are the Sony SSM9ESDs, the Snell A references, the Quad 989s, the Quad 988s, the JMlab Grand Utopia Bes, the JMlab Micro Utopia Bes and, last but not least, a pair of custom four-way studio speakers with carbon fiber enclosures. This test was the most time-consuming that I have ever done, because I obviously had to borrow some of the speakers and move them all (my darn back is still killing me)! Anyway, I am here to report that the image focus of the TU-X1 on my Quad 989s is every bit as good as it is on the two JMlab speakers. The TU-X1s bettered the L-02Ts on every one of these speakers with solid state bipolar, FET or tube devices between the source and the speaker. I tried numerous stations to make sure that the sound of the station was not the dominant factor in deciding between the tuners. I made sure that the load impedance of the preamp was not impacting the sound of either tuner, and matched levels to .01 dB. You really have to match levels and check loading because there is the possibility of making significant judgment errors without such practices.
So what did I hear after all of that trauma? I keep coming back to the fact that the TU-X1 is more dynamic, has much better bass and slightly better treble, is more involving, and appears to recover more of the music's harmonic overtones than the L-02T. I believe that the combination of overall design excellence coupled with particular distinction in the IF and detector accounts for the sound quality. I am shocked that the TU-X1 sometimes sells for half of what recent L-02Ts have fetched. Does the TU-X1 equal the L-02T in terms of RF performance? Well, yes and no. I am able to retrieve some stations that are far more distant with the TU-X1 than I can with the L-02T. On the other hand, sometimes the RF performance of the L-02T is better at separating two closely spaced weak signals. Both tuners are better than most of the stations they can monitor, but I continue to believe that the TU-X1's sonic performance clearly exceeds that of the L-02T.
I also worry about the reliability of the L-02T because of the tremendous parts count and the manner in which the boards are stuffed into the chassis. The TU-X1 has lots more space, which will prevent overheating of the parts. You also get a really fine AM radio in the TU-X1 for "free." I recommend that anyone with a TU-X1 use some inexpensive air-conditioner filter foam to prevent intrusion of dust into the tuner through the cooling vents (actually this is a good practice for any tuner with vents).
In terms of value, it is hard for me to call either of these tuners a particularly good value or a particularly bad value. Both are getting increasingly expensive due to their relative rarity, but neither is enough better than some of the cheaper tuners to warrant the expense unless you are a rabid tuner collector (OK, ALL OF YOU TUNER COLLECTORS THAT HAVE $10K OR MORE IN FM TUNERS, TAKE A STEP FORWARD). I know of guys that have eight or more great tuners and only one CD playback system (names withheld to protect the guilty ;-) )! Enough has been said about the design of the two tuners that I will not rehash all of that here again.
I have not found any tuner from Kenwood, Pioneer or Accuphase that will match a properly operating Sansui TU-X1. So are there any tuners better than the Sansui TU-X1, in my opinion? YES. How many? I am not going to say just yet, but I will say that at least one limited-edition custom tuner has bettered the mighty TU-X1. So it can be outperformed by an audible margin....
Magnum Dynalab and Fanfare:
Into the abyss we dive to handle Magnum Dynalab and Fanfare, of all choices. Let's start with the easier choice, that being the Fanfare FT-1A. This is a bad tuner when you cut to the core of it (please note that I am not known for pulling punches). While I am glad to see that there are manufacturers still making tuners, Fanfare owes more to the potential owner than is delivered here. The sound is thin, bright and two-dimensional, i.e., I agree with Jim but am perhaps a harsher critic of the unit. I began to wonder if the de-emphasis was wrong, but didn't like the tuner well enough to take the time to measure it. The folks at The Audio Critic didn't like it much better. There are so many tuners that outperform it. It is billed as a high-end tuner, but take an old Pioneer (TX-9x00, TX-8x00, all the way down to TX-5x00) tuner and you get better sound, better specifications and much better value! So save yourself time and money and look elsewhere.
Now on to MD. Most people are totally polarized on these tuners (love or hate). Let's start with the good points! It is nice that a company is still building component tuners. It is also nice that people who want to buy new with a warranty have that choice. HOWEVER, MD could offer more performance for the money, as I have said elsewhere. The FT-11 (new or old version) is not a very big step above the Fanfare tuner. The freebie tuners in most modern Yamaha and Denon surround-sound receivers give the FT-11 a significant run for its money, making it a very poor value. I find the sound to be two-dimensional and largely image inspecific! The tonal balance is threadbare and not particularly extended on either end. The problems are sins of omission rather than coloration, but again, a nice little 20-year-old Pioneer will send this tuner running from the sound room. Another tuner to be avoided....
Moving up the MD food chain, we come to the ubiquitous FT-101 and the many variants of the FT-101A. These are clearly better tuners, as Jim states in his reviews (and as would be expected for more money). I have experience with half a dozen of these over a period of eight or more years. These tuners have decent sound, are more extended than the FT-11 (old or new), and look very professional. I like the looks enough to have bought one along the way. However, the front end is incapable of dealing with signals that the cheaper Philips AH6731 can receive all day long. To get more specific, the MD front end doesn't like hot adjacent or alternate channels! It is reasonably sensitive, but that is actually fairly easy to do. The trick (or one of them anyway) is to provide sensitivity, selectivity, rejection and overload margin in the front end. The IF and detector appear to be the root cause of the sonic weaknesses and the unit-to-unit variability. The sound can be described as lacking in harmonics (or richness and warmth, as Jim put it). It also did not produce a convincing sense of depth or image specificity, which Jim didn't comment on. This is really apparent against the most image-specific tuners (like the Kenwood L-02T or Pioneer F-26 which, in my opinion, has better image specificity than the L-02T). HOWEVER, the FT-101/101A is an expensive tuner!! It should be better than decent! Just look at the list of tuners above it in Jim's rankings and compare the price of some of the used units. I would want a better value than this, too! I would love to see this tuner reworked in the IF and detector, with added internal shields, and see what it could do. The changes wouldn't even cost that much if MD did them.
I agree with the better audio stage comments made by Jim. I would add that depending on the impedances involved (and whether a dual is required), the OPA2604 is joined by the OPA2134, the OPA627, the AD797 and others in providing good sonic results. The basic OPA134 (sounds just like its 2134/4134 cousins) op-amp is found in some of the Mark Levinson reference gear that many like. Yes, I know that the 5532 is quiet and reasonably low in distortion, BUT there are better parts and this is an AUDIOPHILE tuner! Anyway, I would still like to see MD improve this tuner - it is quite decent and would move well up the list if they bought a few used tuners with group delay equalized LC IFs and good classic detectors and studied what was done there!
OK, the MD-108. First, this isn't one of the world's best tuners as some have claimed. This is one of the best-LOOKING tuners (from my perspective, anyway) and is clearly the best of the MDs. The sound is very good except in the bass and the imaging is quite good, too. Depth is hard to assess because it has a more forward sound than I like. The low midrange (read this as where low string instruments live) is lacking in harmonics. I could take issue with many of the parts choices beyond the MIT caps that Jim mentioned. Overall, the problem again is one of value! This thing costs enough to buy a Sansui TU-X1, a Kenwood L-02T AND a Pioneer F-26, if you are careful!!!! I would take that set of tuners all day long while looking at a picture of the MD-108, if you need the visual fix. Unfortunately, there are also tuners that sell for $200-500 that are better tuners than the MD-108. For me, this one is also hard to recommend.
I have a love/hate relationship with this tuner. Take a very good front end with superb RFIM performance (in local mode), add a decent IF using some of the same devices as the world's best tuners, a very good detector and a very good MPLX, wrap it in a sexy black anodized aluminum extrusion and you have a very interesting tuner. They even shut down the digital circuitry after tuning has been completed and use LEDs instead of a fluorescent display to minimize noise. So much promise.... I have heard a few of these that were very special-sounding, too, coming close to the best but falling short on the bass and treble extension and imaging precision. HOWEVER, it is ridiculously hard to find one that is working right now! For some reason, this tuner is among the worst when it comes to luck of the draw when purchasing one. I know of people who have bought a half-dozen to get a good one and over a dozen to get a great one. One of the variabilities seems to be component aging, and another is the routing of the inter-board cabling, but there also appear to be changes in alignment (more than most tuners). So here we have a tuner that at its best can border on greatness, but is a high-risk proposition even for those like me that really like it otherwise. It is still a great tuner for those with technical expertise and time on their hands or if it is bought at the right price. However, once again the average Pioneer TX-9x00 is more consistent and represents a better value for most. I also find that it is very hard to find a T-2 that has no appreciable mars on the main front/top extrusion. I am guessing that the reason that it placed low in Jim's Shootout is equal parts lack of precision and that it wasn't a perfect sample. The drift comments and the RF performance are indicative of a less-than-perfect sample (of which there are many)!
This one doesn't surprise me! The Philips AH6731 is a very nice tuner, as both Jim and Bob have pointed out. It starts out with a very decent architecture, e.g., 5 gangs, 24 LC elements, good overall build quality (American made) and impressive internal use of shielding. From a technical viewpoint, my concerns are that I would have liked it better if it had a 6 or more gang front end (although sensitivity versus selectivity balance in the front end was done intelligently, like the Pioneers). I also disagree with the use of spring clips as in the Kenwood 600T, because they can be problematic in a tuner of this age. The AH6731 doesn't benefit from some of the better ICs that came along 1-3 years after it was designed. However, for the money, this is a nicely designed tuner.
From a sonic standpoint, I am slightly less enamored with the tuner than Jim, but still positively impressed. I noted a slightly narrower and less deep soundstage than the "Big 3" (Sansui TU-X1, Pioneer F-26 and Kenwood L-02T). I also noted a slightly "polite" (to use the British term) bass end. However, overall, the sound was remarkably well-balanced. The noise level was just a touch off the pace of the Big 3, but was relatively low. I did not find that the AH6731 had quite the naturalness and "draw you into the music" quality of the Big 3, either. However, in terms of sound per dollar, it joins a relatively small list of the best in this regard (Pioneer F-28, Pioneer TX-9500II, Tandberg 3001A). I suspect that like the mega-tuners, the top bang-for-the-buck tuners will increase significantly in price over time (perhaps more so than the mega-tuners like the Big 3).
Operationally, I found the AH6731 to be straightforward. In terms of RF performance, it was again off the pace set by the Big 3, only more so than in the audio department. It pulled in 8 fewer stations than either the TU-X1 or the L-02T. It also pulled in some marginal stations with generally poorer quality than the TU-X1 or L-02T. The AH6731's internal parts quality was variable. Parts were nowhere near the standard set by the Pioneer F-26 or a handful of the other best tuners in this regard. Internal layout was good and, as noted above, the shielding was very good (if not at the same level as a Yamaha CT-7000). Please note that I am using a very high standard to compare this tuner against, and that all of the other tuners it went up against were much more expensive originally, except for the Pioneer F-28.
In terms of weighing this tuner's overall merits, I would say that those that favor sound quality over RF performance in a relatively low-priced tuner will like this unit. If one favors RF performance, both the Tandberg 3001A and the Pioneer F-28 will outperform the AH6731. The Tandberg also outperforms it on an absolute basis, as well. So here we have a tuner that will sonically outperform several much lauded and very expensive tuners and still provides very respectable RF performance for a relatively low price. I absolutely agree with Jim's rating of this tuner above the Luxman, Sansui and Technics tuners that he has listed a few slots below it. However, for the things that I value in terms of the very finest tuners (refinement, draw you into the music qualities, no frequency-related flaws or birdies) or the things that many high-end tuner collectors value (rarity, historical significance, construction, performance, etc.), I would place a number of tuners above it, including the Tandberg 3001A and Pioneer F-28. I do not feel that the Philips is quite as close to two of the Big 3 as Jim appears to feel at this point - but then, he hasn't heard the Pioneer F-26 and F-28, the Tandberg 3001A, and a few others that have made appearances in my system. However, in fairness to Jim (and me), it is hard to draw significant conclusions based on a sample of one tuner. Variability, alignment, and general care all enter into this, so Jim's sample may surpass the one that I had access to. Also, one again has to consider the systems involved and potential component synergy. Overall, the AH6731 is a very fine tuner at a bargain price, but not in my personal Top Five.
The Yamaha T-70 found its way to me because a friend no longer used it. Since I never turn down a free tuner or one that I can play with, torture and modify without fear of losing any money, I couldn't resist! I must admit that this plain-looking, simple varactor-tuned digital tuner didn't light my candle much, that is until I gave it a fair try.
Let me cut to the chase, this cheap little tuner is extremely good! For those who want to do FM on a budget, you would be very hard pressed to find a better tuner for the money! It reminds me of the NAD 4130 in that it sounds better and lower in distortion on overmodulated stations than many more expensive tuners. However, it is a better tuner than the little NAD. It represents the first digital/varactor tuner that Yamaha did right (which is good because few companies did much right in their first digital tuner). Let's talk about the sound.... It tonally is quite neutral from top to bottom. It lacks a bit of extreme low bass and doesn't have the low-to-mid midrange richness of the big boys, but it is really good tonally. It is quite extended in the treble as well. In terms of noise, it is quieter than many of the older famous tuners (not too surprising because lower noise figure FETs came along after many of the great tuners). The image is reasonably specific, possessing decent depth but not as much width as the best. The dynamics are limited slightly by the lack of extreme bass and midrange warmth. BUT, the overall freedom from distortion, noise and other garbage is truly amazing for a cheapie tuner!
I found the user interface to be OK, nothing exceptional but nothing really bad, either. It is nice for those who are space constrained or have to hide tuners from the wife (you know who you are but the names are withheld to protect the guilty ;-)). In terms of the RF performance, my biggest caveat is use of this tuner in a tough urban area where tuners with more effective gangs and ones that use tuning capacitors rule supreme. However, for most locations, this little tuner does very respectably. Many will chose to ignore it because it is not as pretty, impressive, sexy or famous as other tuners, but it will kick the $#^! out of most of the tuners below #14 in the shootouts in my opinion! In fact, I plan on keeping at least one in my permanent collection because I prefer its group of strengths and weaknesses to such tuners as the Luxman T-117 or the Sansui TU-717. After trying one of these tuners, I had to see if it was an accident or whether this "gift" tuner was representative of the T-70 breed. After looking at four more of these little gems, I can report that they are quite consistent unless they have been abused. The magic of this tuner appears to be intelligent design tradeoffs. As an engineer, I admire this because super-priced tuners had better be great. It is much harder to balance all of the design decisions to arrive at an affordable tuner that outperforms many more expensive models! I think the ratio detector and the gain/phase EQ and selection of good GDE ceramic filters have a lot to do with the overall success of the product. Modifying the analog output stage and power supply can result in a cheapie tuner that is able to place in the top 10 without effort!
For those who like the full technical story, this tuner uses a decent, but unexceptional front-end design. However, the varactors are better-selected than in most digital tuners. The IF has 5 GDE ceramic filters (3 of which are bypassed in the local mode). Therefore, unlike some past Yamaha tuners like the T-2, the local/distant mode is really a wide/narrow IF selector and should have been labeled as such. The first IF amp is discrete transistors, while the second and third each use a uPC577. These are the same IF amps used to advantage in the Yamaha T-2 tuner. As I have written before, the use of a gain/phase EQ stage in the IF is an advantage and one that is used in this tuner. The detector is a well-designed ratio unit. The multiplex is a Sanyo LA3380 pilot-canceling PLL type. This is an improvement over the one used in the T-2 and is also used in a fancier implementation in the Yamaha T-80. The power supply and the analog stage are undistinguished, but easy to modify. Overall, it's the balance of performance compromises that makes this an excellent tuner. I would rather have a T-70 than most of the M-D tuners (sorry guys).
So for tuner fans on a budget, here is a dream tuner that for now can be had as cheap as $15! The good news is that there are a ton of these tuners out there, so even if this review drives prices up, there are plenty to go around and I can't believe that they will ever get truly expensive. I currently own 5 and paid less than $150 total for them.
So, now that I have gone on about how great this tuner is, what is the downside? Well, there are not that many negatives, actually. I would have liked it better with more effective gangs, especially if they were realized with variable capacitors. I would have liked it better with a lower IM front end with a balanced mixer, or with LC filters and more selectivity. An analog multiplier MPLX would have been nice, along with a better power supply and better shielding. However, many of these comments could also be directed at far more expensive tuners with big reputations!! Like I said earlier, the magic is in the engineering tradeoffs and the designers at Yamaha can be proud of this effort!
This was the most difficult tuner to evaluate that I have ever had in my system. Because of the architecture, it is like evaluating two tuners, one that has standard analog output and another with an SPDIF digital output. I really appreciated the opportunity to take an extended look at the tuner on loan from a very serious collector.
The tuner's front end has double-tuned varactors in front of a microprocessor-controlled attenuator. This is followed by a single dual-gate MOSFET used as the only RF amplifier. Use of a single RF amplifier is, in the view of this tuner enthusiast, the correct topology for an optimum balance of sensitivity, RFIM, overload and a variety of other factors. This single RF amplifier is followed by another pair of varactors. This precedes a balanced mixer which is fed by a new Direct Digital Synthesis Local Oscillator. This circuit features a DSP-based sine wave generator fed by a quartz oscillator. The front end continues the basic architecture adopted by Accuphase since the introduction of the T-11 and T-108 tuners a number of years ago. It is an excellent front end and is one of the better front ends that I have seen, although not quite to the performance standard of the Kenwood L-02T which benefits from the use of tuning capacitors instead of varactors, and an additional tuned stage between the RF amplifier and the balanced mixer. The MOSFET used for the RF amp in the Accuphase T-1000 is, however, quieter than the one used in the L-02T or that of any vintage tuner.
The T-1000 IF uses the complement of the TA7060, TA7302 (x2) and a TA7310 for amplification. [Our contributor Jovit comments: "The TA7060 and TA7302 are both used as IF amplifiers but the TA7310 actually contains a VCO, mixer and a buffer. The TA7310 was used to downconvert the 10.7 MHz IF to 2.5 MHz for the Accuphase DGL/pulse count detector." -Editor] The selectivity elements are a mix of linear phase ceramic filters and LC filters. The IF appears to be designed for low distortion, low noise, and the best overall sound quality rather than ultimate selectivity. As mentioned above, the IF appears to be designed for best sound and does not have multiple IF substitutions to allow wide, normal and narrow pass bands. This is one of the design decisions that I do not understand. It prevents the tuner from being an optimal design for DXers (those FM listeners that are located far from their stations) or an optimal design for those FM listeners that have substantial multipath problems (whether caused by mountains or large buildings in a dense urban area). I suspect that the reason for the single IF has to do with matching the characteristic of the DSP-based multiplex decoder with the IF. Although I don't know this for certain, it may have required additional DSP processing power in the multiplex decoder to implement a multiple bandwidth IF stage. Given good reception characteristics with a wide bandwidth IF stage, the T-1000 is capable of phenomenal sound from this IF stage. However, there are other tuners bettered suited to DXing and/or multipath elimination like the Kenwood L-02T, Sansui TU-X1 and Accuphase T-109V for multipath and some of the better Yamaha and Onkyo designs for DXing.
The detector appears to carry over essentially unaltered from the T-109V and is, in the opinion of this tuner enthusiast, the best pulse count detector ever produced in an FM stereo tuner. It is radically better than some well-thought-of early efforts like the Kenwood KT-917. If you like the sound of the detector in the T-109V, you will like this detector just as much. The circuit complement is 5 x 74HC04, which are high-speed CMOS inverters fed to an exclusive-OR gate followed by low-pass filter. [Jovit comments: "The T-1000 DGL implementation consists of 4 x 74HC04 (hex inverter) and 1 x 74HC86 (quad XOR) if you look closely at the brochure's picture of the AGL section on the 2nd page. The 74HC86 is at the bottom. The simplified schematic indicated 24 inverters were used for the delay section. Therefore the DGL would need exactly 4 74HC04s to accomplish this." -Editor]
The multiplex decoder breaks some new ground compared to the T-109V. It first converts the output of the detector to digital via an AKM 5385, 24-bit, 192 kHz A/D converter. This is a part touted by some as a mastering grade A/D converter. It is more than good enough to fully exceed the frequency, distortion and noise potential of the FM Stereo media. In fact, it has -103 dB S/N+D, -114 dB S/N and 114 dB dynamic range. The T-1000 then uses a TI digital signal processor to synthesize and then cancel the pilot tone. The DSP is followed by an AKM AK4114, 24-bit, 192 kHz Digital Audio Interface Receiver (DAIR). While jitter is not specified for this part, most of the better DAIRs have reduced jitter tremendously compared with a few years ago. Finally, D/A duties are conducted by an Analog Devices AD1853 D/A converter. This part has -104 dB S/N + D, -117 dB S/N (48 kHz stereo), and 116 dB dynamic range (48 kHz Stereo). The output op-amps are the unexceptional JRC5532 parts. I would have liked to have seen better op-amps such as the AD797, OPA627 or at least a OP2134. While the D/A converter parts are in vogue right now and are one of the better parts available in the multi-delta sigma type, I would have personally liked it better if the PCM1704 R2 style DAC had been used. Others will no doubt disagree, but in my view the PCM1704 is the best single IC DAC that I have tested. More about DACs later. Some analog purists will say that they don't want an A/D converter, a DSP or a D/A in their tuner. My comment to them is that this is their loss. With the bandwidth of FM being strongly limited by the 19 kHz pilot carrier, the 48 kHz sampling rate of the digital components should be more than wide enough to prevent information loss. In fact, listening tests (see below) seem to indicate fidelity gains NOT losses.
The power supply regulators are the ubiquitous 78xx/79xx series. I would have liked to have seen the use of a better regulator such as one of the low-noise Linear Technology parts or, better still, discrete low noise regulators. Overall construction quality is to a very high standard which is certainly much better than most of the popular super tuners of the '80s and '90s. In typical Accuphase fashion, the specifications are guaranteed and, according to this reviewer's measurements, exceeded in practice. I wish more tuner manufacturers would guarantee their specifications and actually meet or exceed them, as Accuphase has made a practice of for more than 20 years. This is a very strong argument for buying an Accuphase tuner, in my view. Another reason is the fact that unlike most of the older supertuners (L-02T, TU-X1 and the like), one is not faced with aging electrolytic capacitors that need replacing, aging and low-quality interstage coupling capacitors, and very likely the need for alignment. The last point is becoming non-trivial because fewer and fewer labs have high-quality RF generators that are kept in certified alignment, so finding a lab that can align one of the older supertuners is becoming harder all the time. When one considers the ever increasing cost of the L-02T and TU-X1 tuners, this tuner is a much better value for everyone except those that have to live with appreciable multipath or are DXing. Add all of this up and the T-1000 is a ticket to problem-free FM Stereo enjoyment.
If the reader can't tell by now, I like the Accuphase T-1000 very much even with a few quibbles about the design choices (especially the IF which I hope that Accuphase corrects in a Mark 2 version - please, please). If the IF were changed in a future version to incorporate multiple IF bandwidths, it would easily be even better. While I would have liked to have had the benefit of multiple IF bandwidths, I found that all of the best stations in my locale were received with exceptional sound quality. I do not do a great deal of DXing or listening to stations with substandard signals, so the lack of multiple IFs didn't matter much to me in practice. However, the sound quality of this tuner, which exceeds ALL other FM tuners that I have had in my home, DOES MATTER! So in answer to the question that many will be wondering, yes, this tuner is better sounding than the T-109V, L-02T and TU-X1 given a decent-quality, relatively interference-free, FM station.
The sound of the T-1000 was characterized by an authority and solidity not found in other FM tuners. I attribute this to the new multiplex decoder combined with many other excellent stages that were carried over from the excellent T-109V. The bass was unsurpassed in my experience with FM tuners, and this also underscores the solidity of the sound. In the midrange, I heard ambiance, overtones and other subtleties that no other tuner comes close to replicating. The treble may appear to some to be lacking in extension, but it actually is very detailed and clear. The lack of birdies and other FM artifacts is probably the reason why the sound can appear to be soft on the high end. In my opinion, it is just cleaner than any other tuner that I have ever heard. The overall sound is fantastic via the analog outputs, which was the basis of all of the preceding comments.
However, the real treat is with a top-quality external DAC. If the user has a DAC that has one of the better chipsets in it (in my view these would be the unequalled UltraAnalog 20400A, DAC20, or one of the better implementations of the Burr-Brown PCM1704PK DACs) combined with a top-quality discrete analog stage (or at least one of the top-performing op-mps), one can further improve on the sound. I was able to listen to this tuner with the Assemblage 3.1 Platinum (2 x PCM1704PK), Boulder 2020 (5x PCM1704pk), Euphonix Model 2 (Third Generation UA implementation), Mark Levinson 30.6 (2x PCM1704pk), Parasound DAC2000Ultra UA20400A), and the Spectral SDR2000Pro (2x UA DAC20). The sound via any of these great DACs takes FM to a level that I had not imagined was possible. While the additional cost of one of these DACs makes the T-1000 a very expensive proposition, it is not an additive cost if one already possesses one of these DACs. There were no interface problems with any of the DACs listed above, just plug and play. The external DACs took the sound up to an even higher level of refinement. Overall, the T-1000 is the best-sounding FM tuner by a very significant margin that I have ever heard and this includes some very rare custom tuners that I have spent time with. If you are a collector of fine FM tuners, and one who is not on a restrictive budget, you MUST own the Accuphase T-1000 tuner! I have heard ambient information, instrument harmonics and other real musical detail that I would not have believed possible. In direct comparison with some very fine supertuners, several listeners called them "unlistenable" by comparison. While this may be a bit of an overstatement, it is indicative of the very substantial sonic advantage that this tuner has over any other that I have heard or measured. While this is a very expensive FM Stereo tuner, it represents a much better value than a 20-year-old supertuner or one of the megabuck tuners where the manufacturer is using tubes in the analog stage to obtain a boutique sound.
So is this the world's finest tuner? Well, for pure sound quality on the better stations (those not subject to egregious levels of multipath and not extremely weak signals (a la DX)), the answer is a resounding YES. For a downtown New York or equivalent urban setting with really bad multipath, the answer is no. In such a situation, one is better off with one of the best tuning capacitor-based supertuners like the L-02T or the TU-X1. For a DXing fan, you would be better off with a tuner like a Yamaha T-85 or Onkyo T-9090. So there you have it folks, this is one heck of tuner and clearly up there with the world's best in several very important areas. If absolute sound quality is your holy grail, then this is the FM Stereo tuner of your dreams!
Kenwood L-01T, KT-917 and Pulse Count Detectors:
[1/25/09: Like finding a great old master recording in a record company's vault... this material was put on the shelf in July 2002 and has never seen the light of day, until now. -Editor]
The TU-X1 is dead, long live the Kenwood?
Eric, please don't post or forward to the other tuner enthusiasts. I want to buy some other tuners based upon this first, then maybe we'll post it. I stayed up all night working on this, so I don't want everyone benefiting from it before I have time to. I've gotta find me a pair of L-01Ts, ASAP! So, as I said in the subject line, let's kill the TU-X1 and crown the Kenwood L-01T king!
About 2:15 AM, I made some interesting measurements (I borrowed a $100K+ HP Spectrum Analyzer from work). It appears that contrary to "conventional wisdom" that the pulse count detectors are a problem for maximum fidelity, actually, a more accurate statement would be that most consumer tuners (as opposed to military or the very best communications FM tuners) do not implement quality pulse count detectors!
Further, the new Blaupunkt "digiceiver" technology does not solve the problem at all! In fact, the digiceivers actually cause another class of problems that are due to the A/D and D/A limitations as well as some really weird digital artifacts from the DSPs they are using. While the technology works well in a car (where the ambient noise hides a myriad of sins), it doesn't work nearly as well for very high quality home tuners.
Most pulse-count FM demodulators are used because on the surface they are intrinsically linear without tuned circuits. Further, the lack of tuned circuits and the use of highly reliable digital circuits makes the longevity excellent. As you may know, the output of such a demodulator contains a pulse train whose duty cycle changes in proportion to the FM modulation (roughly akin to delta modulation). To recover this modulation, this pulse train must be passed through a low-pass filter to remove the pulses while retaining the recovered composite baseband. Since the typical pulse frequency varies by who designed the circuit and how they are implementing it (e.g., double conversion (Kenwood) or the simpler time delay variant (Accuphase)), I will illustrate the problem with an example. For example, if one has a pulse train around 700 kHz and the composite baseband extends to 100 kHz, the filter must be reasonably aggressive to provide flat, linear-phase response to 100 kHz while simultaneously suppressing the 700 kHz and associated FM sideband components below the noise level of the demodulator. Noise and adjacencies can cause interference just above the composite baseband region. For example, with a 700 kHz IF, an upper second adjacency can cause a 300 kHz [700 kHz (IF) - 400 kHz (difference between monitored channel and second adjacency)] component to appear in the baseband. This can add to intermodulation distortion of this stage. It is therefore desirable to use a filter which has a cutoff frequency as low as feasible given the 0 - 100 kHz linear-phase requirements. It is further desirable to eliminate coupling caps to ensure low-frequency phase linearity. One must also eliminate DC offset that is a problem in a number of these pulse counters, including the KT-917. The use of phase equalization permits the filter to be substantially more selective without compromising constant group delay performance in the passband. If the cutoff frequency of the filter is increased in an attempt to place the 0 - 53 kHz stereo baseband within the linear-phase region of the filter, without phase equalization, the noise rejection will suffer accordingly. One can place a much more selective filter in place and say goodbye to the SCA channel, but this can cause phase problems in the baseband. All of this places some significant requirements on the filter that are not met in any of the tuners that I have measured or done circuit analysis on to date. Those requirements are an active filter implemented with an amplifier stage with substantial bandwidth and phase margin to implement a group-delay equalized filter stage. This requires the use of FETs, bipolars with very high Ft specs or really kick-ass high-speed op-amps! Most of the best parts for this application didn't exist when the classic Kenwood, Pioneer, Accuphase and other brand pulse counters were designed. The problem is generally analogous to a DAC I/V stage. Building an expensive and elaborate discrete circuit reduced the desirability of the pulse counter approach, so most used cheap (and inadequate) op-amps for the job. Secondly, none of the pulse counters that I have seen in home tuners utilize group-delay equalizers to allow the proper balance between bandwidth and phase linearity, on one hand, and filtering on the other. Therefore, they compromise noise, bandwidth/phase, or both. Thirdly, most firms used very small coupling capacitors to reduce the capacitors' size and cost which compromise bass phase linearity in this stage or, worse, had substantial DC offset messing circuit performance up. So we have a condition where instead of delivering the "detector holy grail" we have pulse counters with treble and bass limitations, increased noise and distortion, and/or DC offset problems. Using complex test signals that are easy to analyze in the presence of interference proves this premise in the lab. I will go through the detailed math to ensure that I didn't miss anything. However, early results are very compelling and match listening results (my stock CT-7000s and TU-X1s sound better than my stock Kenwood KT-917 or my Accuphases (until this morning :) )). I think this generally holds true among those that have heard all of these together in a fair test.
I have often wondered why the Kenwood and later Accuphase tuners don't match the fidelity of the very best tuners using a super wide band ratio detector (a/k/a Accuphase T-100 (1.2 MHz), Yamaha CT-7000 (wider) or Sansui TU-X1 (wider still)). This is even true of my super-modified Accuphase T-107 and T-109s (until this morning). I ripped the pulse counters out of my T-107 and T-109 and reworked them (and while tired this AM) I was astonished at the results. I can't wait to do the same to my KT-917 and the mint one on the way! Meanwhile I really want two L-01Ts!
It is amazing to me that, given the superiority of the Kenwood L-01T and KT-917 front ends, they don't kill the sound of the T-100, TU-X1 and CT-7000. These pulse counter problems could explain why, but actually only partially! The good news is that the op-amps/simple transistor circuits used in most of these pulse counters (and the too-small coupling capacitors or DC offset problems) can be replaced/fixed with better parts/design with startling results even without designing a group-delay equalizer (although that would be the hot ticket). It could be that a KT-917 modified in this way will kick the TU-X1's butt! I am willing to bet the price of two L-01Ts that it will be the new king when the pulse counter is fixed!
Now you see why I don't want to go public yet: everyone and their Mom (or Dad) is searching for a perfect TU-X1 while the L-01T (and maybe even the KT-917) are probably much better tuners with this flaw fixed. The reason that I say maybe for the KT-917 is that the IF of the TU-X1 is better than the KT-917's. A KT-917 front end, a TU-X1 IF, TU-X1 detector and a properly designed MPLX and output stage would be a really killer tuner (but one could do better still if Magnum Dynalab or someone would really try :)). The reason that the TU-X1's IF is better is that it implements Sansui's patented group delay equalizer with 12-stage LC filter for Wide and a SAW filter (inherently quite phase linear) for narrow. However, all of the group-delay equalizer IF patents have expired, so one could easily do this today without risk of patent suit. Note that Peter in his mod review on TIC mentioned that he replaced the IF in the KT-917 with a custom SAW IF and that he replaced the pulse counter in his KT-917, so he may have understood some of these problems! However....
The Kenwood L-01T has a better front end than the KT-917 and has a very good IF that is more phase linear in the Wide mode than that of the KT-917! Note that it is impossible with ceramic filters to do as well as the IF in the TU-X1! Ceramic filters are cheap, easy to use and require no adjustment, but they don't have the phase performance of the GD-LC or even the SAW type filters. Filters like the very best military stuff can be as much as 18-stage LC filters, perfectly aligned, group-delay equalized and permanently potted in epoxy! Anyway, the pulse counters are not perfect as implemented and will have to be modified to put them in the same sonic league as the TU-X1. I believe that the L-01T will be better than the KT-917 because of its IF implementation provided that the detector is fixed. So there you have it, the L-01T could be king without a lot of work!! Or a new front end on the TU-X1 is another much harder route :). The front end of the TU-X1 has more than 20 dB more RFIM than several tuners and 30 dB more than the best. As Yoda (almost) said in the second (or third) Star Wars movie, there is another (Skywalker) Supertuner. I'll let you know once I have researched more. It may be better than all of the others once the few circuit problems are fixed! I gotta get some sleep - the brain core melted sometime around 7:00 AM! However, someone may have to sand the smile off my face.
I called a designer that I think very highly of this AM and his response to all of this was, so you figured out pulse counters... good for you! He went on to say that three more big "discoveries" and I would be capable of beating his design (god, those guys were good). He is going to let me earn it though....
Best regards - David
3/6/11: I am just back from 2011 CES in Las Vegas. Some great new components have been unveiled or further improved. It is possible to build a system that was unimaginable just a few years ago. From speakers like the breathtaking MBL101Xtreme and 101E/2, to more conventional speakers like the Magico V3 and V2, there are speakers breaking new ground on the sonic excellence front independent of your listening preferences. In the area of preamps and amps, things are perhaps even better with units like the unbelievably complex
Soulution 700/710/720 to the simpler, but no less spectacular, Spectral
DMC-15SS, Spectral DMC30SS, DMA360S2, and DMA260. Exceptional CD players, DACs
and/or music servers from Boulder (1021) MSB (CD IV and DACs), Resolution
(Cantata) and others keep the digital arena very interesting. Turntables,
Cartridges and other components have been joined by truly superb new components.
So as I asked during my last voyage out of the "tuner retirement village," where
does that leave the FM Stereo enthusiast? As I said before, HDFM, Internet/4G
radio, Sirius and XM are all still challenges to traditional FM Stereo.
Personally, I hope that FM Stereo fights a long fight and eventually finds a
resurgence much akin to what happened with vinyl as many rediscovered what a
great record could sound like on a top drawer turntable and tonearm with a MC
I have helped to build systems for myself, and others, that resolve differences that I could not appreciate a couple of years ago. While components like those listed above are impressive, so too are advancements in affordable headphone-based audio for those with more modest budgets. Plug your favorite tuner (or CD player) into a Schiit Asgard headphone amp (yes, that really is their name, they are really nice people who design great gear, but have a sense of humor that is refreshing in the audio business) and select the 250 ohm Beyer DT-880 Pro version headphone or the AKG K701 (if you are more of a midrange junkie) and be prepared to hear things that you didn't used to be able to hear at any sane price. If you use the web to buy the phones and the amp, they will set you back about $500 plus shipping, and I defy someone to find better sound for the money. I did the only rational thing, I bought several of each!
So while prowling the Las Vegas Convention Center (The Zoo for show veterans), the Venetian, and assorted other venues, what news for the dedicated FM Stereo fan? Well, thankfully in spite of Pioneer, Kenwood, and many other once great brands nearly or completely disappearing from the audio scene, Accuphase is back with a new super tuner (the T-1100) for the bleeding edge of FM Stereo. Accuphase may have read my review of the T-1000 because a significant part of the improvement comes in the IF department. However, the already amazing performance of the stereo decoder was also improved and not just by a little. Try 60+ dB separation from 20 Hz to 15 kHz and nearly 80 dB at its maximum. I am quite certain that this is substantially better than most stations and better than any tuner that I have ever seen or measured! In fact, it is nearly a 20 dB improvement at the top end and sounds that way on great stations.
I find that the new T-1100 tuner is more attractive (at least to my sensibilities) by placing the display on top and controls underneath making it appear more balanced. It still has the massive champagne-gold colored, aluminum faceplate and spectacular Accuphase build quality. Is it worth the substantial increase in price? Well it depends on you, the stations that you have access to, and the depth of your wallet for audio toys. Given the shift in the value of the dollar to the yen, the performance, and the quality of construction, the $6,500 price that I was told about at CES does not seem totally out of line. However, one must realize that we are way, and I mean WAY, up the diminishing returns curve here! We are talking about an assault by some excellent engineers at the limits of what is possible in an FM Stereo tuner. A quick look at the guts shows more than a few design, parts and layout changes. Power supplies have been improved, more DSP wizardry is obvious, and every technical specification of the prior T-1000 was maintained and some improved dramatically. I did not have time to analyze the circuit in detail. Such an effort may have been frustrating if I tried because of the amount of processing going on inside the TI and AD DSPs and other VLSI devices.
So for you hardcore RF guys, is this the "end all" tuner for downtown multipath or extreme fringe areas? 'Well, I am not sure' is the honest answer. Since I only had the tuner in two locations with two antennas, all I can say that selectivity is definitely improved with six Finite Impulse Response, DSP implemented, Digital IF Filter, bandwidths of 50 kHz, 75, 100, 150, 250 and 500 kHz. It performed flawlessly in both locations and with both antennas that I was able to use it with. So unlike the T-1000 where even in my location I could see that the old RF greats still had an edge for tough reception problems, they didn't have the same edge against the T-1100. Will this hold true for downtown NYC, or 200 miles from a station with a huge array antenna on a rotor? Well one answer is I hope so, but perhaps someone in one or both of those types of locations can report in and let us know.
For the sound quality junkies, this tuner further adds to Accuphase's considerable reputation by once again pushing back the limits of what I thought was possible. They maintained both single-ended and balanced analog outputs, added a nice polarity switch capability for the balanced outputs, and thankfully maintained the ability to output the digital stream into a reference-grade external DAC. I continue to believe that reference-grade outboard DACs are still the way to go rather than using the internal DACs. The sound, which was spooky clear before in the T-1000, had even more authority on the bottom and in terms of dynamics, while maintaining the delicacy and imaging that so astounded me in the T-1000. The separation appeared to me to further improve sound staging compared with the T-1000. Meanwhile, the noise floor appears to be even quieter. I would guess that this is likely as a result of being able to balance audio and RF performance in the IF and further abating stereo decoder artifacts. The T-1100 is simply in every audio parameter the best sounding tuner that I have heard by a large margin. Listening to a great station, on headphones, at significant volume, the quieting of this tuner when there is no content is really pretty amazing. Using a good headphone set-up also reveals important sonic cues that are not always apparent in speaker/room based systems even if the system is really exceptional. This tuner also has better attack and decay of notes and proper harmonic structure than any other than I have heard. You think a Kenwood or a Sansui can do violins? Then you better listen to this tuner! If the violins don't sound spooky real, then it is probably a bad recording and/or playback system at the station. If you have a really great classical station, then you are in for a treat. The sound quality is what one friend described as "otherworldly." What's more, given the parts and the amount of work being done with Digital Signal Processing, I imagine that this tuner will hold its performance longer than many, if not all, of the old greats.
So for those who already own the T-1000, should they sell their tuner and trade up or just outright buy a T-1100? Well again that depends on you, your wallet, and the stations/reception situation that you are in. If you have the money, love FM Stereo, and have enough great stations within reception range to make the investment worthwhile, my answer would be an overwhelming ABSOLUTELY! However, at this price range, while not unreasonable given currency fluctuations as I said above, the user is within reason to expect the best tuner ever made for their situation. If you live in an extreme fringe or deep urban area, I would recommend borrowing one from a dealer, to determine whether it is state of the art in your location. I strongly encourage you to do so if you are REALLY serious about FM. I have already told several friends with deep pockets and/or large tuner collections that they should tee up, or at least sell part of their aging collection to get a T-1100. Should people with a stack of L-02T, TU-X1, CT-7000, et al tuners get one? They must if they want to say that they have all of the greatest tuners of all time. Going back to the T-109 and the T-109V, is illustrative of the level of improvement for these last two great Accuphase tuners. While the T-109 and the T-109V are both excellent tuners, they pale in comparison to the sound of the T-1100 when listened to into a great DAC. I know some enthusiasts who have bought a LOT of tuners, Really Large Collections, and now have an Accuphase T-1100 or T-1000 paired with a Kenwood L-02T or a TU-X1 and NO other tuners. They are also very happy with what they hear.
As another listener so poignantly stated, "Imagine if we had this tuner back in the 1970s when there were so many great stations!"
Whether you are personally an Accuphase fan or not, tuner enthusiasts owe this firm a debt of gratitude because they are nearly the only company left who has displayed any interest in furthering the format. Most of the so-called separate tuners are the same cheap junk put into most current surround receivers with a separate box and supply. Accuphase has invested some real time and money here and I wish them success with this tuner.
That's it folks, the old King is dead, long live the new King!