Perception of Pace/Rhythm/Timing (PRaT) -- genetic? Options

[hollowman]

A somewhat "controversial" topic in audio gear subjective reviewing is the reproduction of Pace/Rhythm/Timing, sometimes abbreviated PRaT. (Sometimes, PRAT with a cap. 'A' may denote A=acceleration). The topic also includes an interdependent subjective metric, dynamics.

Perhaps the seminal introduction to the phenomenon is the Nov. 1992 article in Stereophile, Pace, Rhythm, & Dynamics by Martin Colloms . I noted that this topic is "controversial" in that it not often reported in gear/equipment reviews -- especially in forum- or message-board-based reviews ... where most folks stick to the common (and, IMO, repetitive) BMTS (bass, mids, treble, soundstage). From there, some may additionally describe, e.g., a headphone's 'dynamic' or 'speed' qualities. But when asked to comment on PRaT, the same reviewer responds with puzzlement, smug ridicule, or ignorance.

Pace/Rhythm/Timing -- or PRaT -- is something I'm particularly sensitive too. It is one of the first attributes that I notice the presence or absence of in audio gear, like headphones, all else held equal. I even notice it in slow music, like adagio or lento movements (the Colloms article above goes into this a bit).

Because of the apparent "underreporting" of the "PRaT" phenomenon as important audiological subjective metrics -- and even some hostility as to its importance or even its existence -- I'm wondering how much its perception is genetic. This may be like genetic sensitivity to perfect pitch:
http://perfectpitch.ucsf.edu/study/
Or, more simply, the PTC genetic bitterness test from high-school biology class?

This post has been edited by hollowman: Jul 4 2011, 16:42

[drewfx]

A simple, non-judgmental question:

If a system can reproduce, without audible distortion, a sine wave oscillating 15 thousand (or more) times per second, why in the world would one believe the same system couldn't reproduce, without audible distortion, musical timings that are probably in the range of 50-100 times greater?

This post has been edited by drewfx: Jul 5 2011, 16:43

[benski]

A simple, non-judgmental question:

If a system can reproduce, without audible distortion, a sine wave oscillating 15 thousand (or more) times per second, why in the world would one believe the same system couldn't reproduce, without audible distortion, musical timings that are probably in the range of 50-100 times greater?

I'm going to play devil's advocate here. Two major issues with amplifier and speaker design that could explain this are slew rate and servo feedback. A typical drum hit involves a huge transient that can be effected by the amplifier's slew rate as well as the natural inertia of the woofer's speaker cone. In addition, the "bounce back" of the cone after the initial transient can cause the speaker's motion to diverge heavily from the audio waveform. Self-powered monitors can fix this with servo feedback motors or other mechanisms (flyback transformer?) to give better motion control. I know this is a major problem in electronic injection automotive systems where the injector behaves basically as a miniature subwoofer and the timing of the system can get ruined by the voltage effects of the magnetic driver moving on its own due to inertia and elasticity.

Whether or not this manifests audibly in music equipment is subject to debate, but it does give a plausible rationale for prat effects.

And, again, I'm playing devil's advocate and not actually claiming that speaker elasticity and inertia is going to cause an audible effect except on very bad equipment.

[DonP]

[
I'm going to play devil's advocate here. Two major issues with amplifier and speaker design that could explain this are slew rate and servo feedback. A typical drum hit involves a huge transient that can be effected by the amplifier's slew rate as well as the natural inertia of the woofer's speaker cone. In addition, the "bounce back" of the cone after the initial transient can cause the speaker's motion to diverge heavily from the audio waveform. Self-powered monitors can fix this with servo feedback motors or other mechanisms (flyback transformer?) to give better motion control. I know this is a major problem in electronic injection automotive systems where the injector behaves basically as a miniature subwoofer and the timing of the system can get ruined by the voltage effects of the magnetic driver moving on its own due to inertia and elasticity.

Whether or not this manifests audibly in music equipment is subject to debate, but it does give a plausible rationale for prat effects.

And, again, I'm playing devil's advocate and not actually claiming that speaker elasticity and inertia is going to cause an audible effect except on very bad equipment.

In a rational design, a subwoofer (or woofer) isn't supposed to reproduce frequencies outside it's useful range, so with that drum hit, the low frequencies go to the woofer, and the higher frequencies go to the higher frequency drivers. Also recall that a bandwidth limited low frequency signal (like goes to a woofer) mathematically won't be able to precisely represent when the drum hit occurs. That information lies in the higher frequencies.

Faults in how the speaker as a whole handles that (say, a time difference between drivers) or an amp's inability to drive the speaker should be handled by conventional measurements without having to drag fuzzily defined terms into it.

This post has been edited by DonP: Jul 5 2011, 21:54

[greynol]

fuzzily defined terms into it.

Thank you!

This is in part why I linked the other thread. As a musician, I'm somewhat disturbed seeing these well defined and accepted terms hijacked by the audiophile community in order to further their dialog of subjective woo.

[Woodinville]

This is in part why I linked the other thread. As a musician, I'm somewhat disturbed seeing these well defined and accepted terms hijacked by the audiophile community in order to further their dialog of subjective woo.

Especially since pace, rhythm and timing can be conveyed in a 4kHz bandwidth with a 20dB SNR.

[lbstyling]

This is in part why I linked the other thread. As a musician, I'm somewhat disturbed seeing these well defined and accepted terms hijacked by the audiophile community in order to further their dialog of subjective woo.

Especially since pace, rhythm and timing can be conveyed in a 4kHz bandwidth with a 20dB SNR.

Hi,
I joined to post on this.
Prat is something I've taken a interest in for many years, and I have asked many people what is responsible for it. Good prat is reported to be particularly noticeable by its ability to reproduce fast paced music with accuracy, particularly in the bass range. The closest I got to a possible answer is speaker related.

Option 1.
A raised FR around 60hz giving a louder and therefore more prominent band around a kick drum impact


Option 2
Low inductance drivers either for there range (ie a pro sub as opposed to a long throw consumer option) or outright ( as in designing a speaker that rolls off at about 50hz but has a raised response at 60 and will likely include low inductance as a driver feature by default as the design doesn't seek to play low anyway which is what you add inductance for (to extend the LF extension.
Inductance effects the delay of the movement of the driver to the signal, so although a driver will either produce the signal flat or not, it may not start and stop the signal very close to the correct time, amplifier feedback is also related to this (damping factor)

The perception that electronics (amp etc) are responsible for perceived prat is just because the speakers the demoed amp is usually played with is actually the reason for the sound. Good current delivery always helps reinforce the low end too.


A discussion with the designer of 'kudos' speakers confirmed the rather 'British' tuning technique.