Speaker "Speed"

[voodooless]

I know on ASR, they are so big on measurements and basically saying if you can’t measure it it’s not real, but this seems like a pretty fundamental example for how not everything can be measured.

Not agreeing on what to measure != not able to measure it.

 

[Tell]

Isn't "speed" all about frequency response? I mean the higher the frequency the faster the driver have to move, and if it can't move that fast it won't play that frequency. Or am I missing something?

 

[Keith_W]

First, define precisely what you mean by "speed." I have a hunch that if you are able to define it precisely, then you will also find there is a measurement for it. On the other hand, if you can't define it precisely, there probably is no measurement for it.

Hit the nail on the head.

In broad terms, audio can be describe in amplitude, time, and space. Therefore, any subjective description that references amplitude or time will have objective measurements, provided that one can agree what the subjective terms mean. Examples of such descriptions - "bright", "sibilant", "dull", "bloated" are all amplitude related phenomena. "Attack", "tightness", "dynamics", "veiled", "reverberant", "dry", and "clarity" are time related phenomena. "Soundstage" and "imaging" relate to space. These are useful terms and have meaning provided that the meaning is agreed upon by everyone. This is opposed to subjective descriptions like "organic" or "musical" or "clinical" which have no meaning.

I take "speed" to refer to a time related phenomenon. So depending on what he is asking, the answer will be found in any one of the time displays - impulse, step, group delay, phase, etc. I myself use the terms "slow" and "fast" to describe good time alignment and lack of ringing, and my typical test for this is a drum transient, preferably preceded and followed by silence. Listening to Chinese or Japanese drums is ideal. These are massive drums, and listening to them on a full range and time aligned system is incredibly satisfying. The thwack on the skin with the immediate BOOM of the drum is what I call "fast". OTOH if the system is poorly time aligned, the bass freqs follow a few ms later and it sounds "slow".

 

[ppataki]

Just purely theoretically speaking: can we say that there are things that we cannot measure yet but maybe in the future we will have means to measure them?
What I mean is that we currently have a set of graphs (freq. response, phase response, impulse response, step response, decay, etc. etc.) available to scrutinize the measurements - but maybe in the future we will have other/additional ways to interpret measurements?
Or am I very sci-fi here?

 

[Blockader]

I also asked this question here:
https://www.audiosciencereview.com/forum/index.php?threads/measuring-transient-response.17219

But no conclusion so far, still not sure how to measure transient response

Group delay measurements(in anechoic chamber) provide the transient response, which is the delay it takes for the driver to respond to the input signal at each frequency. This represents the speed of the driver.

If a system has a smooth on-axis response, a consistent distortion profile (distortion indicates either non-linearity or misuse of DSP), and good extension at both the low and high ends—especially the high end—it's considered 'fast.'

Crossover filters can introduce group delay issues around the crossover frequency. However, unless the crossover is poorly designed or has an order higher than eighth, these delays are not audible, even around 3000 Hz, where human hearing is most sensitive to phase problems.

Group delay of KH 120 II.

The bump around 45hz is happening due to lack of extension of the system below 47~hz. With subwoofers, that line is going to be flat down to 20-25hz, depending on the extension of the subwoofer of course.

 

[ppataki]

group delay measurements give you the *transient response*.

Yes, I think I also came to the same conclusion since the time I posted the above
So agreed!

 

[Speedskater]

If you name two (tested) loudspeakers, one with 'speed' and one without.
We could examine what measured parameters are very different.

 

[Blockader]

Just purely theoretically speaking: can we say that there are things that we cannot measure yet but maybe in the future we will have means to measure them?
What I mean is that we currently have a set of graphs (freq. response, phase response, impulse response, step response, decay, etc. etc.) available to scrutinize the measurements - but maybe in the future we will have other/additional ways to interpret measurements?
Or am I very sci-fi here?

Measuring and determining what to measure are not the biggest challenges. The greater difficulty, I believe, lies in correlating these measurements with psychoacoustics and user preferences, which is the most significant challenge in sound reproduction and audio science research.

We can evaluate the usefulness of reflections at different incidence angles using the IACC, but typically, we treat all reflections equally when estimating the in-room response of speakers. We can measure the subjective preference for soundstage envelopment of speakers with LEV, yet we do not factor in the directivity width of speakers into preference scores or differentiate preference scores based on various directivity characteristics.

Furthermore, while we can measure the audibility threshold of group delay per frequency, to my knowledge, no research has yet measured the preference scores for group delay per frequency. Similarly, we can measure intermodular distortion in speakers, but there appears to be no research on preference scores for different levels of IMD except Geddes.(And his conclusion was that IMD has positive correlation with user preference)

We are also capable of measuring the vertical directivity of speakers, but the benefits of ceiling reflections remain unclear. As you can see, we can measure many aspects, but a handful of studies cannot capture the entire picture. A broader collective effort is required to map the correlation between these measurements and psychoacoustical facts and preferences effectively.

It is valuable to make educated guesses about areas not yet covered by research, using subjective impressions etc. However, I believe it is not beneficial for the community to make subjective, bold claims based solely on what their '32-bit ears' have heard.

 

[DonH56]

I have found audiophiles mean different things when they say "speed" with respect to speakers.

Often enough when they say it "sounds faster" it can be duplicated by applying a mild boost in the 2~3 kHz octave. That is, "speed" implies greater upper midrange response, but if you turn it up too much then they are "bright". For subwoofers, boosting the 100~200 Hz octave or so usually results in listeners saying it has more "speed". So a "brighter", or more "thumpy", sound is subjectively perceived as "faster".

From a more technical point of view, I equate "speed" to frequency and short-term transient response. Steady-state frequency response, e.g. wider bandwidth, means a "faster" speaker, though that is not always desirable since a driver typically has its best performance (distortion, dispersion, etc.) over a limited bandwidth. It makes the most sense to consider a speaker system rather than a single driver in that case. For transient response, a "fast" speaker responds quickly but the real goal is on the decaying edge -- a "fast" speaker (particularly subwoofer/woofer) will "stop" quickly without "ringing" or having a long envelope on decay (after the sound stops) that muddies the sound. Decades ago I found that distinguished some of the best (sub)woofers and systems from the more mediocre; it was not how fast they ramped up, since multiple drivers spanned the frequency range, but how well the cones stopped so the snap of a drumstick or sharp crack or thump of a drum or cymbal decayed naturally without the driver contributing additional sonic signature to the decay.

Interestingly, at least to me, constant group delay (providing the best pulse response) did not seem to matter all that much when comparing speakers many years ago. There are so many other variables that it tended to get buried. Critical for a radar system, desirable (perhaps) but not so critical for a speaker system, at least best I recall from old listening tests.

All of this is quite measurable, as is most anything audio, including things like imaging and so forth that purely subjective folk tend to dismiss as "unmeasurable".

 

[Looneybomber]

Isn't "speed" all about frequency response? I mean the higher the frequency the faster the driver have to move, and if it can't move that fast it won't play that frequency. Or am I missing something?

You’re correct and there’s a person called Fourier (Fourier Transform) that agrees with you. An impulse (infinitesimally short blip) has infinite frequency response.

 

[antcollinet]

Most people dramatically over estimate the speed of the attack phase of musical instruments - even percussion. Typically cymbals are used as an example. You can download a cymbal clip here:

Free Cymbal Sound Effects | MP3 Download | FesliyanStudios

Free sound effects of cymbals including hits, crashes, crescendos, swells, scrapes and more! Highest HD quality MP3 downloads available.

www.fesliyanstudios.com

I've used : Crash-Cymbal-Hit-B
Opening in Audacity, it looks like this (one channel). The total attack phase (from start to peak volume) is about 0.17 seconds.

Zooming in to the attack phase, you can see it starts just before 0.14 seconds (actually around 0.139) The initial strike lasts until about 0.147 (around 8 thousandths of a second), before the lower frequency shhhh sound starts. The lower frequency shh fundamental is in the region of 500 to 700Hz

Then zooming in further to the start of the attack (The strike). See how low the initial amplitude is - and how slowly it ramps up. And the frequency of those first few cycles we are looking at is about 8.7kHz (the file sample rate is 48kHz) - less than half the usual minimum speaker bandwidth.

I don't think the average speaker is going to struggle to keep up here. Remember with a 20Khz sine wave, it is starting, getting to full speed, and stopping again before going in the opposite direction 40,000 times a second.

 

[Multicore]

Are there any metrics for the transient response or "speed" of a speaker? Subjectively, certain ribbon speakers I've heard have reproduced cymbals and strings with a faster attack than otherwise comparable tweeters.

I've seen it being described on other forums as the ability to accurately follow a square wave, but I've no idea if this is an accurate definition. I'm also not sure if the relevant metrics are already present in Amir's measurements.

Can you give a sentence or paragraph of text in which "speed" of a speaker comes up?

 

[Tell]

You’re correct and there’s a person called Fourier (Fourier Transform) that agrees with you. An impulse (infinitesimally short blip) has infinite frequency response.

Yeah I know about that guy, he seem to have been a smart man!

You can look at this from the other side of the frequency spectrum as well btw, some people think a speaker plays "fast bass" when it plays little or no of the absolute lowest octaves.

 

[maynardewm]

First, define precisely what you mean by "speed." I have a hunch that if you are able to define it precisely, then you will also find there is a measurement for it. On the other hand, if you can't define it precisely, there probably is no measurement for it.

Sure, I can try. I think it would be measuring the time it takes for a speaker diaphragm to go from at rest to move back and forth once. As the most basic form of the test, which could probably be measured with a high speed camera.

 

[Jim Taylor]

I think it would be measuring the time it takes for a speaker diaphragm to go from at rest to move back and forth once.

Isn't that the very definition of one cycle of frequency (Hertz)? Would not the speaker's output of the same frequency that was used to drive it prove - ipso facto - that it was capable of the required speed?

No camera required.

I believe that @antcollinet and @DonH56 have given you excellent explanations. It's worth your while to peruse them.

Jim

 

[mhardy6647]

Yeah... any discussion of loudspeaker (or driver) speed pretty much disconnects my brain at the stem.

 

[TankTop]

Are we talking about the same "speed" that usually is referred to in 'high-quality' subwoofers, like "fast" drivers?

If so, this is the ability of a driver to stop its impulse. Usually not a problem with small drivers due to a relatively low mass, but certainly something to take into consideration with big drivers. For simplification only, I would refer to "small" woofers as ≤ 10" in size and to "big" woofers as > 10" in size.

Rythmik Audio, for example, call their solution 'Direct Servo', which is nothing but a "stop motor": an additional sense coil that gets its feedback through the amplifier without delay, to stop any given movement. The effect could be measured via group delay; and will manifest in magnitude of distortion.

====
Edit: A similar phenomenon could also be the reason in lower quality tweeters.

I would think larger drivers in an apposed configuration would also have a faster transient response. Possibly why the Perlisten D215s measures so well as well as many other subs with this design

 

[dfuller]

The bump around 45hz is happening due to lack of extension of the system below 47~hz.

It's the port tuning, really. Ports are resonators.

 

[beeface]

I just measured my speakers for speed. 0km/h.

 

[tmtomh]

In my experience one of the most common cognitive difficulties in understanding audio reproduction systems is grasping that frequency is a measure of speed.

In that spirit, I think @DonH56 makes a very good point that the perception of speed is often about midrange/upper-midrange "snap" and low-end "tightness."