hard pick, or a hit on a triangle. The direct live sound in the near field, as recorded by a close up microphone, sounds like a "ting," with a hard "t" attack. The Plinius sound makes this attack softer and sweeter, so it sounds like a "ding," which is just like it would sound live when heard from the typical distance of an audience seat. And then some tube amplifiers that are too rounded and dull in the trebles would alter this "ting" too far, changing it into a "bing" (note how the "b" sound is even duller and more rounded than the "d" sound, which in turn is softer than the "t" attack sound).
The Plinius sound is immensely enjoyable, relaxing, and seductive. Time after time, when comparing the Plinius to other amplifiers on many diverse cuts of music recordings, we found ourselves seduced into simply enjoying the music, and listening longer to the same cut when we were listening through the Plinius than when we were listening through another amplifier. That in itself is probably the ultimate testimonial, right there. That's what you want for the sound of any system you hope to enjoy, be it for music or for home theatre: a sound that you just want to keep listening to.
In the interest of forthright objectivity, we have to note that, as seductive and subjectively enjoyable as the Plinius sound is in creating a musically natural sound, it is not as objectively accurate to the input signal as some of the best competing amplifiers. Objectively, it is, for example, easy to hear that the best of the competing amplifiers have better articulation and delineation of individual treble details than the Plinius (for instance, one of our lab reference amplifiers, the superb Clayton, individuates upper frequency musical details better than the Plinius with its gentle defocus). So the Plinius sound is ideal for your enjoyment of very natural and realistic musical and voice sound, for both film soundtracks and music recordings. But for home theatre it will not give you the sharpest zap and sizzle on those machine gun sound effects in those action flicks, nor for stereo or surround music listening will it give you the ultimate in sharply etched analytical detail. Of course, most music lovers do not want to be constantly assaulted by every analytical detail that a typically too close recording microphone picks up, and instead they prefer to hear the musically natural sound that they hear from live music when sitting at their usual audience distance in a concert hall, the same musically natural sound that the Plinius sound delivers so well. Likewise, after you have outgrown the juvenile phase of home theatre, where you were naively impressed by every action flick zap sounding as sharp as possible, and when you have matured to films that feature realistic music and dialogue, then you'll be joining the majority who want all this music and voice to sound as naturally realistic as possible, which again the Plinius sound delivers so well.
Special Advantages For Home Theatre
For use in music systems, the Plinius sound already has an advantage over other solid state amplifiers, because it executes so well its subtle tubelike transformation of too closely miked recordings into a more natural, more enjoyable semblance of the live music sound you actually heard from your concert hall seat. Then, when applied to home theatre systems, this Plinius advantage actually grows and becomes of even greater benefit.
How? Consider the following. Music is very important to most films, helping to move the story along and helping to magnify the effect upon you of the other elements of film language: the dialogue, the cinematography, the actors' facial expressions, etc. Indeed, music can be so important that it is sometimes the only element responsible for changing the course of the whole story plot and for effecting radical turnabouts in your perceptions, attitudes, and emotions. As a simplified example, consider a still cinema shot of a tranquil woodsy landscape with no characters in sight, and tranquil music playing; suddenly the music turns menacing and suspenseful, and just as suddenly you know that the whole plot has turned, with a likely murderer hiding behind one of those trees.
But movie music does its crucial work amidst a curious paradox. It is most effective when you don't hear it. It may be a key ingredient of the film, and it may be crucial to manipulating you the audience, but it is supposed to do all this important work subliminally, without you being aware that you are being manipulated, or even being aware that the music per se is even present at all. Your conscious attention is supposed to remain focused on the film, the story, the characters, etc. And (in most cases) if the music is doing its job well, you're not supposed to consciously notice the music changing, the music changing you, nor even that there is music playing at all. If you consciously notice the manipulator and the manipulation, then you won't be as emotionally moved.
Now, many solid state amplifiers, with their typical glare in the upper frequencies, impose artifices on music, especially on massed strings that are so often used for background music. When the massed violins acquire an artificial screech, what should have been subliminal background music grabs your conscious attention, thereby undermining the whole crucial mission of music that is supposed to move you without your consciously noticing. To make matters worse, this artificial screech also distracts you from the film, offendingly annoys you, fatigues you instead of relaxing you, and ruins the whole suspension of disbelief and magic transport to the film venue, since you are now painfully aware that you are merely in a room with an artificial reproduction system screeching at you. Here's where the Plinius sound comes to the rescue. Its musically natural reproduction assures that massed violins don't screech or offend or sound artificial. The music on the soundtrack stays where it should and does what it should. If the music's intended role is subliminal background, then it does not call undue conscious attention to itself by sounding artificial. If on the other hand the music's role is active, as in a concert video, then the Plinius sound still reveals the music as it should sound, as you would expect live music to sound, from a typical audience seat.
Furthermore, it's no secret that the equipment chain employed to prepare and deliver film soundtracks to you can be of poorer sonic quality than that employed to master stereo music recordings. The film soundtrack chain might well employ more and poorer IC chips, compressors, and reverb units, all of which could contribute to music sounding less natural and more artificial (especially in solid state ways) on film soundtracks than on high quality stereo or surround music recordings. Thus, even a hypothetically perfectly neutral solid state amplifier would reveal these ugly, distracting, and fatiguing artifices contaminating the music (and also the naturalness of dialogue) on many film soundtracks. But the Plinius sound would gently work to counteract these artifices imposed on film soundtracks, especially where inferior IC chips impose solid state artifices on voices and music by leaning out their natural warmth and adding upper frequency glare. We saw above how the Plinius sound gently corrected the artifices of typical too close miking, to produce a natural sound that more closely reproduces the sound of live music you actually hear from a more distant typical concert hall seat. In much the same way, the Plinius sound (with its rich warmth and polite upper midrange counteracting the leanness and upper midrange glare from IC chips) can likewise gently correct for the artifices imposed by inferior solid state circuitry earlier in the signal chain, since these artifices are generally similar to the sonic "artifices" imposed by too close miking. Thus, the Plinius sound can be of extra special advantage in a home theatre application, helping to compensate for the limitations in many film soundtracks, and thereby helping them to sound more natural, more realistic, more enjoyable, and more effective in emotionally moving you to a more rewarding cinema experience.
How Plinius Gets This Unique Sound
How does Plinius get such a musically natural, tubelike sound out of solid state circuitry? And why don't other solid state amplifiers sound this way?
To start with, there is no such thing as an intrinsically perfect amplifier. All amplifiers commit some degree of nonlinear error, which technically is a distortion, and they do so even at quiet music levels. For the remainder of our discussion, let's focus on ordinary listening levels, not on loud levels where you are pushing the amplifier into higher levels of overload or overdrive distortions. At ordinary listening levels, the amount of distortion in most amplifiers (single ended tube amplifiers excepted) is made miniscule, and thus it does not sound to us like distortion per se (certainly not like the obvious distortion you hear when pushing an amplifier into overload or clipping). Instead, this miniscule residual distortion affects what is technically called an amplifier's transfer characteristic, which you can think of as being the characteristic that an amplifier imposes on a music signal in transferring it through its circuitry, from input to output. Thus, we hear this transfer characteristic (from the miniscule residual distortion) as affecting an amplifier's sonic personality. Obviously, the nature or pattern of this transfer characteristic, i.e. the nature or pattern of the miniscule residual distortions, can have a large effect upon the sonic personality of an amplifier. In most solid state power amplifiers, this pattern has a certain kind of nature, which is largely responsible for the artificial sonic personalities of these amplifiers, with hard glare, sterile leanness, and fatiguingly bright etchiness. In Plinius solid state electronics, this pattern has been deliberately engineered to be different, thereby giving Plinius electronics a different sonic personality, one that is much more musically natural.
Most solid state amplifiers have that artificial sounding pattern in their transfer characteristic because of the way they employ negative feedback, and because of the way they try to reduce total overall distortion, without paying particular attention to the pattern of the residual distortion that does remain. Now, most solid state amplifiers need to employ large amounts of negative feedback, because solid state amplifying devices are inherently quite nonlinear (distorting), and therefore need significant amounts of negative feedback to linearize their amplifying behavior. Incidentally, this is unlike tubes, which are more linear (especially triodes), and therefore don't need as much negative feedback. This large amount of negative feedback works at all listening levels, even quiet levels, so it is always shaping the pattern of the small residual distortions, the pattern of the amplifier's transfer characteristic, and hence the sonic personality of the amplifier. Admittedly, this typically large amount of negative feedback does do some good work for the amplifier, like widening its bandwidth, reducing its output source impedance, and reducing the overall total distortion level.
But negative feedback also does some bad things. One bad thing, of particular relevance here, is that negative feedback actually raises the distortion level of higher order distortion byproducts, to a higher (worse) level than they would have been without negative feedback. Thus, even while reducing the total overall level of distortion, negative feedback changes the pattern of the small residual distortion that it leaves behind, thereby changing the sonic personality of the amplifier (and doing so for all listening levels, including quiet or ordinary levels). There is a general consensus that the hard glare and sterile artificiality of typical solid state sound is largely due to the predominance of these higher order (especially the odd higher order) distortion byproducts, which large amounts of negative feedback actually add to the music you hear. Even though negative feedback adds only small amounts of these higher order distortion byproducts to solid state circuits, our human ear/brain is very sensitive to detecting and being annoyed by even these small amounts in the sonic personality of the amplifier, because the higher order distortions are not present in live music, so they sound very amusical, which makes them stick out like a sore thumb even in small amounts. That's especially true for intermodulation distortion byproducts, which are not even harmonically related to the original musical sounds on the recording, and which therefore sound doubly ugly when they are higher order intermodulation byproducts added by these solid state circuits.
On the other hand, the human ear/brain is very tolerant of lower order distortion byproducts (say second and compressive third), even in moderately large amounts. Indeed, air itself is a slightly nonlinear medium for transmission of sound, and our ear mechanism is also slightly nonlinear, so we are accustomed to hearing some low order distortion even when we listen to live music (especially through a typically long distance of air), and we accept it as being musical, because it is part of what we actually hear as our live music reference.
That's why many low feedback tube amplifiers can sound so musically natural. Yes, their overall measured distortion might be somewhat high, higher than the total measured distortion of most solid state amplifiers, but the composition of their distortion is very different from the composition of distortion in most solid state amplifiers. In the low feedback tube amplifiers, the composition is primarily low order, say second order and compressive third order. That's the same composition that we already are accustomed to hearing as part of the sound of live music, as added by sound transmission through a long distance of air (to our typically distant audience seat) and our hearing mechanism. In fact, it wouldn't be surprising if, by deliberately adding some second order and compressive third order distortion to a typically closely miked recording, you would obtain a sound that more closely resembled what you know as the sound of live music as heard at a distance in a concert hall. On the other hand, in typical solid state amplifiers, the composition of the added distortion is predominantly higher order (since the higher feedback and the circuit design reduces [even effectively eliminates] the lower orders of distortion, but actually creates new amounts of higher order distortion)--which makes the sound artificial and amusical. Even though those newly created additions of higher order distortion are small in amount, they are detectable and annoying precisely because they sound so artificial and amusical, and are alien to our live music experience.
Furthermore, the topology of most solid state amplifiers sets out to deliberately eliminate even orders of distortion. Engineers are trained to treat any and all forms of distortion as the enemy, and not only to reduce total overall distortion, but also to ferret out and reduce each type of distortion individually, without paying particular attention to the global picture of what distortions they leave behind. They are taught that any design trick that can reduce any kind of distortion must be inherently good, so they should use it, without paying particular attention to the nature or pattern of the small residual distortion they leave behind, to the transfer characteristic they leave behind, to the sonic personality they leave behind. With solid state circuits, it's relatively easy to design the circuitry so that it has complementary symmetry. This circuit symmetry holds the theoretical promise of canceling out all asymmetrical distortions, i.e. all even order distortions, right off the bat, without even needing to attack them via negative feedback. This means that only odd order distortions are left behind, as needing to be corrected via negative feedback. But this actually creates several new sonic problems. Let's take a minute, over the next few paragraphs, to look at these sonic problems.
First, negative feedback's creation of higher odd order distortions, coupled with the elimination of the even order byproducts by symmetric solid state circuit topology, creates gaps in the natural progression of distortion byproducts. You are left hearing only the odd order byproducts, since all the even order ones have been omitted. That sounds musically unnatural because our live music experience is rich in even order, asymmetric distortions, especially the low 2nd order distortion. Also, the general consensus is that even order distortions contribute to the natural warmth we hear from live music, so stripping them away and leaving only odd order distortions should make the music sound lean, cold, and sterile--just like solid state does in fact often sound.
Second, because symmetric solid state circuit design eliminates the even order distortions, and then must rely totally on negative feedback to reduce the odd order distortions, this means that the higher order distortions which negative feedback freshly creates will be all of odd order, with none of even order. Thus, typical solid state amplifiers create new, worse levels of higher order distortions only of odd order (5th, 7th, 9th, etc.), and these are generally recognized as the ugliest, most artificial sounding byproducts (again, especially when they appear at frequencies not harmonically related to the music, due to intermodulation).
Third, these higher odd order distortions sound even more unnaturally artificial because they are usually expansive rather than compressive in other solid state amplifiers. What do we mean by this? Even order distortions are essentially asymmetric, affecting the top half of the music waveform differently than they do the bottom half, whereas odd order distortions are essentially symmetric, affecting both the top half and bottom half similarly. But what then do the odd order distortions do wrong to the music waveform, if they affect both halves similarly? They can compress the peaks to lower than they should be (which is called compressive odd order distortion), or they can expand the peaks to higher than they should be (which is called expansive odd order distortion). A slight compression of music's sharp transient peaks would of course make them sound slightly softer, whereas an expansion of these sharp peaks would make them sound sharper or harder, i.e. artificially hard and sharp. Now, it turns out that natural phenomena associated with hearing live music (transmission through air and our hearing mechanism) are compressive in nature, not expansive. And it turns out that some audio phenomena are likewise compressive in nature, not expansive, for example triode tubes and analog tape (which explains why they sound so musically natural, even when their measured distortion is quite high). On the other hand, it has been suggested that expansive odd order distortion sounds very unnatural, adding artificial hardness and glare to music. Thus, solid state circuits whose feedback generates higher odd orders of distortion that are expansive in nature rather than compressive would thereby sound artificially hard and glary, just as many solid state amplifiers in fact do.
Fourth, the actual solid state devices used in these symmetrical circuit topologies are in fact not truly symmetrical. An NPN transistor does not in fact behave symmetrically to a PNP transistor. Thus, these well intentioned efforts at canceling even order distortions do not work perfectly, leaving behind some messy, complex asymmetries, which probably sound amusically ugly.
The Plinius engineers wisely recognize all the above factors. They take cognizance of the fact that an amplifier's sonic personality is shaped by the pattern of its transfer characteristic, even at (indeed especially at) ordinary listening levels. They are aware (more so than the engineers at other solid state manufacturers) that, if music is to sound natural, then the pattern of the amplifier's transfer characteristic should follow a natural progression, with mostly low order byproducts and progressively lesser amounts of progressively higher order byproducts in its composition, even in the miniscule amounts present at ordinary listening levels. This is the pattern we hear from live
(Continued on page 71)