PRaT and the four pillars for system evaluation
PRaT. Please excuse me. For those readers unfamiliar with this odd term PRaT is an acronym, shorthand for pace, rhythm and timing, and it makes an interesting case study for those of us who fancy ourselves as amateur psychologists, and who also have more than a passing interest in the power of words.
Whenever humans bond together over a shared interest or type of work they use coded language exclusive to the group. Sometimes it’s purely a matter of convenience, but often, using acronyms, and words that outside of the group might have a different meaning, is also a way of excluding outsiders and reinforcing group identity.
Sometimes a coded reference can seem to take on a life of its own and become a meme. The internet is more efficient than word of mouth and so memes now propagate with greater rapidity. It has certainly helped our acronym PRaT to become common currency. What also helped was the adoption of PRaT by one or two canny audio industry marketeers. Over time it became both a shorthand reference to a particular quality supposedly heard in some audio systems, and at the same time a dog-whistle for audiophiles who, by hearing it and repeating it, became unwitting advertisers for those brands.
Pace, rhythm and timing had its genesis in work by speaker designer, latterly audio reviewer and publisher, Martin Colloms (below). His book, High Performance Loudspeakers, now in its seventh edition, is perhaps the most comprehensive and deeply researched volume in its field, an essential for anyone who calls themselves a speaker designer or wishes to be a better-informed by-stander.
During the ‘70s Colloms had become interested in the relative sonic characteristics of American and British loudspeakers, noting that those designed for the US market more often than not suffered from a bloated and slow low end. In chapter five of High Performance Loudspeakers he deals with this in depth, noting that group delay in the lower octaves can result in bass being delivered to listeners’ ears later than the higher frequencies so that the timing of musical events is corrupted.
He observes how the propulsive effect of bass has a profound impact on our perception of the rhythm and meter of music, and if timing is off, then things just don’t sound right or musically credible and engaging. The book’s subsequent discussion of the technical steps required to achieve good timing is typically detailed and thorough, highly informed by Colloms’ experience of over 200 commercial designs.
Stop there and we might think that timing is all that matters. However, if we read the impressive work from cover to cover we understand very well that timing is only one of the key aspects of audio equipment performance. Despite this, it was seized on in isolation, the acronym got appropriated by the marketeers, repeated ad lib by a generation of magazine readers, and now we are where we are with PRaT part of the lingua franca of the audio industry.
Timing in context
As Colloms shows us in his book, timing is just one among a number of key musical qualities, all of which are important and all of which need to be addressed by designers if audio kit is to perform as we want it to and get us as close as possible to the once live musical event. Beethoven seems to have understood the key musical pillars too; building the towering achievement of his lifetime’s work not on timing, but on timing and dynamic expression, dynamic agility and tonal detail. Actually, pick any composer. Timing, dynamic expression, dynamic agility and tonal detail are the building blocks that they have all used, are using, and will use in the future. The four pillars are therefore what we should pay attention to when we evaluate any audio component or system.
I am going to deal with timing first, but only because it’s what got us into this whole subject in the first place. Let’s be clear: focusing on timing alone is musically illiterate. It takes one element out of the essential context of all the others and elevates it to totemic status, almost as if it is the only thing that matters. And, as Colloms showed, timing is a technical competence which a speaker (or other audio component) either has or hasn’t. Pace and rhythm on the other hand are two subjective dependencies that cannot exist accurately without timing competence.
Only if our audio component is timing-coherent might we perceive the actual pace and rhythm that the musicians created when they performed together. Timing can only be judged good when the component or system is letting us hear each note in its entirety, from beginning to end and in the correct relationship to those before and after it. We should hear an accurate reproduction of what the musicians did at the recording session and what is laid down on the master tape. A high noise floor, inadequate bandwidth or intrusive colouration might prevent us from hearing the start of a gradually building transient, or the long decay at the end. Notes might be presented late or early due to phase errors, speaker group delay or, in the case of a turntable, pitch instability. The result might still strike us as good fun, enjoyable even, but we cannot say that the timing is ‘good’. We are listening via an effects box, not a high-end audio system.
It can be instructive to take a friend or a relative to their very first live acoustic music performance. Ask them what they notice about the music and usually with very little prompting they remark first and foremost how physical it feels, how the cello, the piano or the singer’s voice could be felt as a compressing, vibrating sensation in their chest. They are describing another of the four key musical pillars. We might call it dynamic expression, or dynamic power, but whatever words we use it is this musical quality that more than any other tells our brain that we are in the presence of a live event.
When we listen to a wide range of audio products we hear evidence that not all audio designers recognise the key role played by dynamic expression, or understand how to make their products achieve it. We might typically hear impressive detail, but overall a flat sound that fails to convince our brain to go along with the illusion that we are hearing something close to live music. A key pillar – dynamic expression – is muted or entirely absent.
This is not a matter of loudness, but of air moving power; the two are quite different. Consider an extreme example. Take the music we hear in a lift. It might be detailed enough that we recognise the song, yet the sound is somehow not engaging. In a similar manner, many audio systems paint a detailed 2D sound picture on the front wall of the listening room but also failed to engage us. Contrast that with the relatively rare systems that can create a 3D aural illusion that even at low volume settings compresses the air in the room and somehow makes us feel that we are almost in the presence of a live performance. Dynamic expression is about how the system transfers the sonic energy on the recording; yes, the trouser-flapping bass, but also the sharp breath that a flute player uses to produce a stab of sound from the instrument, or the sonically weighted ‘phuddd’ that a good system delivers when a vocalist’s breath hits a closely-placed microphone. If the air in our listening room is compressed by this energy, our brain more easily buys into the live illusion.
This is the term that describes the ability of an audio component to match the changes of amplitude captured on the recording. Often people assume it’s a reference to the speed of change, and in part it is. They also often assume it’s about the big dynamic swings, and indeed it is. But it’s also just as much about the really tiny changes in amplitude, the subtle stuff encoded in the margin between the noise level and the louder, more explicit musical information. We know this pillar is properly present when we play a familiar track on a new component and notice sounds that we have not noticed before; a performer shifting in their chair, the oh-so-gentle tinkle of a triangle, and so on. Composers (most of them anyway) employ the full range of amplitude available from the instruments they are writing for. An audio system should show us this.
It’s blindingly obvious that when we listen to music, along with timing, dynamic expression and dynamic agility we are listening also to the tonality created when instruments are struck, plucked or blown. A single note from a cello, for example middle C at 261.6 Hz, is attended by harmonics, overtones of the primary or fundamental note. The way the fundamental note is, literally, accompanied by its harmonics, is what gives the note the complex tonal character that we hear; making a cello sound like a cello, a piano a piano, and so on. Drums are tuned to particular frequencies and they too emit harmonic overtones, although listening to many audio systems we’d hardly know it. So notes, at least those from acoustic instruments, are not mono-tones. A system with tonal competence will allow us to hear the complexity that lies behind the apparent simplicity, and appreciate better what the composer or performers wanted us to experience.
One of the conceits of the audio industry is that we need golden ears or training to be able to discern good audio reproduction from bad. Many years ago, a friend of mine was auditioning speakers at a hi-fi dealer’s shop. He did not like the sound produced by the heavily-promoted major brand, and said so. In barely concealed irritation the shop assistant snapped at him “…but how are you listening?”. The question was loaded with contempt. How could he be discerning? He was only a customer, not an audio professional. My friend calmly replied “With my ears.”
Most of us, even if we have not been to a concert or listened to the radio for decades, have a memory of how various instruments sound. If we occasionally top up that memory by attending live musical events, and we have a working pair of ears, then we have all that is required to be a discerning listener and make informed judgements about how an audio system is performing.
One of the traps here is that the live experiences really need to be purely acoustic, rather than of electronic, or amplified acoustic instruments. That’s not to imply criticism, or suggest a hierarchy of artistic value, but simply to observe that music derived from electronic sources is not a reliable reference because it can be manipulated sonically even at the point of performance. In contrast a violin or human voice delivered purely acoustically is what it is; we can bank the memory in confidence and compare it with what our audio system is producing.
So there you have it. Ears, audio memory, dynamic expression, tonal veracity, dynamic agility and timing. We’re good to go.