signal path, in series with the load you hear, in this case the tweeter). This also told us that the series resistor, which remains in the tweeter's signal path at all attenuator switch settings, is surely also degrading the fidelity of the 968's tweeter. The 968 tweeter would surely sound even much better yet, than it already does at the highest level attenuator setting, when Hyperion changes this series resistor for one with better sonic quality. Note that it is to the credit of the 968 tweeter driver, and its superb transparency, speed, and clean purity, that it so clearly reveals the sonic shortcomings in the crossover resistors feeding it.
It will be easy for Hyperion to improve these resistors. In the meantime, a workaround is for you to simply set the tweeter level control at its highest setting. We found the 968 to be very listenable at this tweeter setting, in part because this silk dome tweeter sweetens recordings so deftly. But if you find it too bright, you can simply add some more stuffed furniture, drapes, wall hangings, etc. to your listening room, to absorb more of the high frequency energy radiated by the tweeter.
Poor Quality Bass
The third sonic weakness of the Hyperion HPS-968 concerns the quality of its bass. The quantity of the 968's bass is excellent, as is its low frequency extension and power. But the 968's bass quality is downright poor. How can the bass quality be so poor, when the woofer drivers themselves are so good? The problem is simply the tuning of the bass port or vent. The 968's (rearward facing) vent emits a loud, long lasting, bass boom overhang on every bass transient. To make matters worse, it is a one note boom. Thus, regardless of the bass frequency of the input signal, the loud, long lasting boom emitted in response by the 968 vent stays at the same boring pitch. Incidentally, the 968 exhibited this awful vent boom not just in small rooms (which induce or exaggerate boominess), but also in our huge castle ballroom, which is very kind to the bass quality of loudspeakers, allowing them to put out their very best.
This problem is simple in origin (the bass alignment of the system simply needs to be retuned). But the manifestations and ramifications of this problem are many, and are compound felonies, obscuring the real music (or sound effects) signal in many pernicious ways. First, the sound of the original real bass transient is obscured because the 968 vent boom is so loud. Second, the natural tail decay of the original real bass transient is obscured because the 968 vent boom overhang lasts so long in time. This vent boom overhang also obscures all the other musical information (even in the midranges and trebles) that occurred in the time right after the original bass transient fell silent, and this obscuration naturally degrades the transparency of the 968 system, compromising the superb transparency that the 968 drivers intrinsically are capable of. Third, the pitch and tune played by the original real bass transient (e.g. a plucked jazz bass going up and down the scale) is obscured because the loud, dominating 968 vent boom is always at the same one-note pitch.
To double check ourselves, that the 968's bass quality really was this poor, and that we were not imagining things, we deliberately conducted a special experiment. We grafted another loudspeaker's woofer section under the 968's head unit, to form a hybrid loudspeaker system, a 968 with a different woofer system. And we specifically chose a woofer section that we had independently previously assessed as having very good quality bass, in this case the woofer section of the Von Schweikert VR-4jr. This special hybrid loudspeaker system was outstanding. Its bass quality was night and day different, and night and day superior, to that of the whole 968. The 968 has bigger woofers and a bigger bass cabinet than the VR-4jr, so it can play bass louder in quantity, and perhaps slightly lower in frequency extension. But in bass quality, in accurately reproducing the true sound of real live bass, the smaller Von Schweikert totally dusted the larger Hyperion 968. In fact, the contrast in bass quality was so dramatic that, when we substituted the 968's own bass section back in, to once again form a complete 968, its bass quality not only confirmed our earlier judgement of a poor rating, but actually sounded even worse in quality than it had sounded on its own before, since we had just experienced, in direct sonic contrast, such superior quality bass from the Von Schweikert. Incidentally, this special experiment's hybrid loudspeaker system really sounded great, in every way. Its bass frequencies had the excellent quality of the Von Schweikert, while the middle and upper frequencies had the superior sound of the Hyperion 968's midrange/tweeter head unit, which is even more transparent, faster, and more accurately coherent (with the midrange driver's polarity corrected by inverting its input terminal wiring) than the VR-4jr's midrange/tweeter head unit.
Oddly, other reviewers have praised the 968's bass highly, and without reservation or qualification. One must surmise that these reviewers know how to listen only for bass quantity, and don't know how to listen for bass quality, nor do they know the sound of real live bass transients (as a reference) well enough to hear anything amiss - nor do they know the physics of loudspeakers well enough to know that factors determining bass quality are distinct from factors determining bass quantity, so both are important to assess.
What are the factors determining the 968's poor bass quality? Simple. The 968's vent tuning design, for the chosen bass vent alignment, has produced a vent frequency response whose shape is a sharp, narrow, high Q peak. The laws of physics tell us that any device or any filter (e.g. a bass vent), with this sharp, narrow, high Q peaked frequency response shape, will ring violently, both loudly and long, as actually heard in the time domain, with a long lasting overhang. Moreover, the periodicity of that long ringing overhang is a fixed, unchanging pattern, and that periodicity is interpreted by the listening human ear/brain as being a one-note boom at a single frequency, the frequency corresponding to the periodicity of that fixed ringing pattern. The obnoxiously loud, long lasting, fixed frequency one-note boom, resulting from this sharp, narrow, high Q peaked frequency response of the bass vent, will be excited (and triggered to go off) by every bass transient in the program material that has energy within the frequency band covered by this sharp, narrow, high Q peak.
Now, which bass transients in the program material would have energy within this narrow frequency band, to trigger the obnoxious one-note boom? The correct answer, which very few loudspeaker designers seem to know, is virtually every bass transient. You see, it so happens that, by definition, every sonic transient (musical or sound effect) contains a very wide spread of frequencies that is also infinitely dense spectrally, and which usually extends all the way down to (but not including) DC (zero Hz). For example, even a transient pluck of the very highest string on a violin actually contains bass energy virtually down to DC. Certainly a plucked jazz bass, going up and down the scale contains significant energy, extending down to DC and infinitely dense spectrally, thus surely lying within that sharp, narrow, high Q peak of the 968 vent's frequency response, and thus triggering the 968 vent's obnoxiously loud, long lasting, one-note boom, for every note played up and down the scale on that plucked jazz bass. Thus, this simple factor, a bass vent tuning design with a sharp, narrow, high Q peak, both predicts and fully explains the poor quality bass we in fact hear from the 968.
Why did the 968 engineer design this frequency response shape into his bass vent, since it produces such poor bass quality from the whole system? Many loudspeaker designers, perhaps driven by their marketing departments, make the mistake of trying for maximum quantity of bass, since that's what makes salesworthy advertising claims, and they might not even know what good and accurate bass quality means or sounds like. To get the maximum quantity of bass, these engineers choose a bass alignment and vent tuning that gives them maximally flat frequency response to the lowest possible frequency. This generally means that they need the biggest possible amplitude boost from their vent/port, and, simply speaking, to get this they want the vent's frequency response to have the highest, sharpest peak, which also in turn means the narrowest, high Q peaked frequency response, exactly the shape that produces the worst quality bass, as discussed above.
Is there a better vent design choice that produces good quality bass, and can the 968's designers easily fix their poor quality bass by simply changing to this better vent design choice? Yes, and yes. The best sounding bass vent design, which produces excellent quality bass (tight, firm, well defined, with very little overhang), has a frequency response shape that is just the opposite of the sharp, narrow, high Q peak that produces such obnoxiously bad bass quality. Instead of being narrow, the frequency response hump of this best sounding vent is wide and broad. Instead of having a sharp, high Q peak, the frequency response of this best sounding vent actually has a flat plateau on top. This best sounding vent design extends the bass response of the system woofers way down, about as low as the sharp, narrow, high Q peak vent design does. The only aspect in which there is a slight tradeoff, with this best sounding bass vent design, is in the system bass response, as extended downward by the vent, being dead flat. Instead of being boosted to dead flat, the system's bass frequency response is boosted by this best sounding vent design to a shelf, a plateau that is flat but sits a few dB below dead flat system response. However, as we have discussed extensively in previous IAR articles on this subject (beginning with IAR issue 34), human hearing is relatively insensitive at low bass frequencies, so you'll hardly be able to even hear, and certainly won't be bothered by, the fact that the extended low frequency bass response of a system with this best sounding vent design happens to be on a flat shelf that's a few dB less loud than it ideally should be. But you certainly will hear, and will appreciate, the far, far better, more accurate bass quality that this vent design will give you, on all bass transients.
Until Hyperion improves their vent tuning design in the 968, is there anything you can do to improve the poor quality bass from this loudspeaker? Only a little. Take the small protective cloth that the 968 head unit came wrapped in, fold it over once to get merely a double thickness, and slide that double thickness of cloth into the rear slot of the 968's bass cabinet, across the full width of the slot. This will add resistive damping to the high Q, long ringing overhang of that vent's obnoxious one-note boom. This resistive damping impressively reduces the temporal duration of each one-note boom, and slightly reduces the loudness of this obnoxious boom. However, mere resistance cannot change the reactive tuning of a bass system, so this resistive damping does nothing to alleviate the obnoxiously fixed, one-note periodicity of the ringing overhang (that requires vent retuning and redesign to fix). Incidentally, if you overstuff and thus overdamp this vent, obstructing too much of the flow, then the whole system's low bass power and extension will suffer, since the overdamped vent can't contribute enough amplitude boost.
So, there you have it. The Hyperion HPS-968 has outstanding promise to be a truly great loudspeaker system, and at a bargain price for this performance and quality. With three quick and easy design fixes by the manufacturer, it could fulfill this promise, and would then fully earn its place in your listening room that it does not yet quite deserve.
Back to Table of Contents