• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Step Response: Does It Really Matter?

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
Below are step responses from a number of different high end / audiophile speakers:

Lsfig11.jpg

Sfpfig08.jpg

atc_scm19_step_response.jpg


1112DT-SM55fig7.jpg

Mlpfig6.jpg


Clearly, they're all over the map and highly variable.

Some look nearly "textbook" good, while others look pretty bad from a "textbook" POV.

Yet many of the speakers with "bad" step responses get positive subjective reviews.

If there is such variability, and low correlation with subjective preferences, do step responses really matter very much?

Is optimizing for superior time-alignment actually an unimportant design goal?
 
OP
watchnerd

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
I don't pay attention to them. But JA does and hence the reason he measures them in stereophile. Maybe we can j_j to comment on it. :)

@j_j Is there much evidence that time coherence and/or phase matter much?

BTW, that last one is a Martin Login Prodigy, I imagine @RayDunzl 's look similar (along with my ML hybrids, too, one of which got busted in the move).
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
BTW, that last one is a Martin Login Prodigy, I imagine @RayDunzl 's look similar (along with my ML hybrids, too, one of which got busted in the move).

Yes.

Raw step ML reQuest (20 years old) at the listening position:

upload_2017-10-15_4-45-20.png


Prodigy:
Mlpfig6.jpg
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
JBL LSR 308 at listening post (trace inverted somewhere, raw it is inverted from this)

upload_2017-10-15_5-11-57.png


atc_scm19_step_response.jpg



Interesting little "wiggle" on both after the tweeter contribution.
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
JBL LSR 308 (green) and ML reQuest (black) with AcourateDRC and 6144 tap DSP. ML has cheap subs, may contribute to the early rise.

upload_2017-10-15_5-2-47.png


Is a "textbook" step better?

I don't turn off the DSP, so, maybe.

They sound more "like" each other when correction is applied, except spatially. Indistinguishable (to my deaf ears) for casual listening.
 

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,180
Location
UK
The step response can be direct combined sound of a woofer & mid & tweeter (say) in an anechoic situation, or can be a composite of DSP'ed driver(s) and room reflections at the listening position in a real room - the DSP effectively would have to complement the room reflections at the listening position. Or it could the combined output of a phased array e.g. floor-to-ceiling identical drivers with curve and 'shading'.

In a real room, the first sounds close to a step source playing in a room; the second sounds like a step source in an anechoic chamber - even though you are sitting some distance from the speakers in a room. You would have to hold your head very still at the precise position where the step is valid. The third would sound close to a step response in an anechoic chamber or fairly 'dead' room - its purpose is to dominate strongly over room reflections.

The reason why the first sounds only "close to" a real step source playing in a room is because it isn't a point source. The step is a composite of the output of multiple drivers and only forms at a single point (or maybe an arc or something like that) in space. Room reflections will not be those of a true step. It is, however, "close to" a point source if the drivers are small and close together.

What happens to the second one at other points in space, and how it sounds, is anyone's guess...

The philosophical point is: do you want to hear the sound of a step source playing in your room, or do you want to hear the sound of a step source in an anechoic chamber? Do you want to hear reflections of the room, or only what is in the recording? Do you think that 'headphone sound' without having to wear headphones is the ultimate goal? This will define how you tackle the problem. If you don't ask those questions, the result could be very confusing: taking steps to achieve what appear to be better measurements, while simultaneously driving you away from your implicit goal.
 
Last edited:

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,180
Location
UK
A perfect step response could be obtained from a 15" woofer paired with a 19mm tweeter at some point in space in an anechoic chamber (or using gated measurements in a real room). If there was a standard position, say 1m from the tweeter at tweeter height, at which measurements are usually made, the system could be tuned for perfection at that point - a bit like a VW engine. It would sound terrible at all other points, and a in a real room, though.
 

j_j

Major Contributor
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
2,267
Likes
4,758
Location
My kitchen or my listening room.
@j_j Is there much evidence that time coherence and/or phase matter much?

That is a bloody complicated question, to say the least. First, time coherence is a dual of phase. Remember that a time delay is exactly a phase shift of 2 pi f t, where f is frequency and t is the time delay. So a time delay is exactly equal to a line in phase shift, and the slope of the line is the time delay.

But does it matter? Sometimes. Phase shifts above about pi/6 or so inside of one ERB (or critical band) (n.b. ERB is a measure of the cochlear filter bandwidth, so are critical bands. ERB's are more accurate until you get down to very low frequencies, where there is some serious dispute, and a lot of confusion between frequency and time issues) can be audible. (we're talking monaural here) That's because they can change the firing time of the inner hair cells in the cochlea.

Phase shift between two far-removed frequencies is not audible.

Time delay, however, or phase inversion (aka how many crossovers work) CAN be audible if it creates a change in the signal envelope. This is where digital crossovers, among other things, excel. But this is a complicated issue, and I have to shy off being too precise here, sorry, at least for the time being.

Between ears, about 5 MICROSECONDS (Yes, milliseconds was a brain fart!) is the shortest credible ITD. This, however, depends enormously on frequency and the signal structure. At low frequencies, time delay (or phase shift) can be audible. Between 500 and 2kHz, give or take, it's much less audible. Above that, it depends extremely on the envelope of the signal. Again, I must shy off being too, too specific for the time being.

Neither of these directly answers questions about step response. Remember, first, that step response and impulse response are mathematically related. So you can talk about either meaningfully. And the answer to 'does that matter' is more complicated than I feel writing about at 3AM. Sorry.
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
A musical note has (at least) four parameters associated with its envelope, and any number of variations within those.

Poor step response would muddy the attack, or attack and transition to decay (and or sustain), in my opinion, and observed by my hobbyist experimentation.

There would likely be little to no audibility during the sustain, decay, and release stages of the envelope of a note, rarities excepted.

So, good step response contributes to the articulation of the sound of drum beats, guitar plucks, piano notes, some vocalizations, etc.

ar·tic·u·la·tion
ärˌtikyəˈlāSH(ə)n/
noun
1.
the formation of clear and distinct sounds in speech. "the articulation of vowels and consonants"

2. MUSIC - clarity in the production of successive notes.
 
Last edited:
OP
watchnerd

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
JBL LSR 308 (green) and ML reQuest (black) with AcourateDRC and 6144 tap DSP. ML has cheap subs, may contribute to the early rise.

View attachment 9352

Is a "textbook" step better?

I don't turn off the DSP, so, maybe.

They sound more "like" each other when correction is applied, except spatially. Indistinguishable (to my deaf ears) for casual listening.

Okay, that's an unexpected result...

I get how a DSP crossover can improve step response by delaying certain driver inputs to make the whole more time coherent, but how does 'global' DSP address this....by delaying certain frequencies above, say, the high-pass filter?
 
OP
watchnerd

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
A musical note has (at least) four parameters associated with its envelope, and any number of variations within those.

Poor step response would muddy the attack, or attack and transition to decay (and or sustain), in my opinion, and observed by my hobbyist experimentation.

There would likely be little to no audibility during the sustain, decay, and release stages of the envelope of a note, rarities excepted.

So, good step response contributes to the articulation of the sound of drum beats, guitar plucks, piano notes, some vocalizations, etc.

ar·tic·u·la·tion
ärˌtikyəˈlāSH(ə)n/
noun
1.
the formation of clear and distinct sounds in speech. "the articulation of vowels and consonants"

2. MUSIC - clarity in the production of successive notes.

This all intuitively makes sense from a pure physics point of view.

But it doesn't seem to map to subjective experiences. There doesn't seem to be a correlation between superior step response and blatantly superior subjective evaluations.

To quote John Atkinson's digest version (on Stereophile) of his AES paper:

"Again, this is an aspect of loudspeaker behavior that has proved controversial. One school of thought holds that it is very important to perceived quality; another, which includes almost all loudspeaker engineers, finds it unimportant. Floyd Toole, now with Harman International but then with Canada's National Research Council, in his summary of research at the NRC into loudspeaker performance that is described in two classic 1986 papers [32, 33], concluded thusly: "The advocates of accurate waveform reproduction, implying both accurate amplitude and phase responses, are in a particularly awkward situation. In spite of the considerable engineering appeal of this concept, practical tests have yielded little evidence of listener sensitivity to this factor...the limited results lend support for the popular view that the effects of phase are clearly subordinate to amplitude response."

This is also my view. Of the 350 or so loudspeakers I have measured, there is no correlation between whether or not they are time-coherent and whether or not they are recommended by a Stereophile reviewer. However, I feel that if other factors have been optimized—on-axis response, off-axis dispersion, absence of resonance-related problems, and good linearity—like a little bit of chicken soup, time coherence (hence minimal acoustic phase error) cannot hurt. In my admittedly anecdotal experience, a speaker that is time-coherent (on the listening axis) does have a small edge when it comes to presenting a stereo soundstage, in terms of image focus and image depth. But time coherence does not compensate for coloration, poor presentation of instrumental timbres, a perverse frequency balance, or high levels of nonlinear distortion."


Read more at https://www.stereophile.com/content/measuring-loudspeakers-part-two-page-4#cjIq8Dt6R1pdm6S9.99

There could be a couple of issues, however, with his methodology:

1. JA seems to be measuring step response at 1m, which, as @Cosmik points out doesn't really mean jack for time coherence at the listening position

2. If other decay resonances aren't controlled first, it doesn't matter, i.e. if the driver has a high-Q resonance that rings like a triangle, or the box has woody resonances, their bad effects on decay, rise time, and energy storage over-rule good step response
 
OP
watchnerd

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
The philosophical point is: do you want to hear the sound of a step source playing in your room, or do you want to hear the sound of a step source in an anechoic chamber?

Maybe neither?

When I listen to live music, I can't imagine there is perfect phase/time coherence at the listening position given all the reflections.

Similarly, the microphone recording the event is not in an anechoic environment and has a step response of it's own.

I guess, ideally, I'd want to know the step response of the microphone and have that applied to the speakers at the same distance the microphone was from the instrument (assuming mono).

Do you want to hear reflections of the room, or only what is in the recording?

The recording.

Do you think that 'headphone sound' without having to wear headphones is the ultimate goal?

Partially...but with crosstalk and somatic components.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
I get how a DSP crossover can improve step response by delaying certain driver inputs to make the whole more time coherent, but how does 'global' DSP address this....by delaying certain frequencies above, say, the high-pass filter?

"Global DSP" - to me A.K.A."room correction" - modifies the signal sent to the speakers adjusting frequency response and phase...

Here's the preamp output:
1 - no correction (flat sweep result)
2 - with correction - modification to swee level across the range

upload_2017-10-15_11-53-39.png


Big dips sent in the bass due to the main woofer + subs being about 9dB hot in addition to the slight hotness of the woofer in the raw signal.

But more to the point:
Phase (time) adjustment...
1 - raw
2 - corrected

upload_2017-10-15_11-54-50.png



The bass gets pulled forward in time, or the highs delayed, or both... Hard for me to say. The FIR peak is in the center of 6144 taps if that helps the analysis.

My cross on the mains is 180Hz (sealed) and the subs (23Hz ported) are running in parallel with the same mains signal, but start rolling off above 100Hz or so, invisible here.

Raw and "corrected" impulse:

upload_2017-10-15_12-0-3.png


upload_2017-10-15_12-0-41.png
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
The above is for the MartinLogans with subs

Here, signal correction for the JBL LSR 308 measured at the fpreamp output:

Phase:

upload_2017-10-15_12-5-41.png


Impulse:

upload_2017-10-15_12-6-14.png
 
OP
watchnerd

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
"

The bass gets pulled forward in time, or the highs delayed, or both... Hard for me to say. The FIR peak is in the center of 6144 taps if that helps the analysis.

Given the quicker rise time of tweeters / electrostatic panels, doesn't it have to be the highs are delayed?
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
Given the quicker rise time of tweeters / electrostatic panels, doesn't it have to be the highs are delayed?

Relatively, yes. But I should think FIR can slide things earlier or later however the filter is designed.

Hey! Wait a minute... This was Step Response Inquiry:

Raw

upload_2017-10-15_12-12-45.png


Martin Logan + subs correction:

upload_2017-10-15_12-13-14.png


JBL LSR 308 corrrection. (might be inverted)

upload_2017-10-15_12-13-54.png
 
OP
watchnerd

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
That is a bloody complicated question, to say the least. First, time coherence is a dual of phase. Remember that a time delay is exactly a phase shift of 2 pi f t, where f is frequency and t is the time delay. So a time delay is exactly equal to a line in phase shift, and the slope of the line is the time delay.

But does it matter? Sometimes. Phase shifts above about pi/6 or so inside of one ERB (or critical band) (n.b. ERB is a measure of the cochlear filter bandwidth, so are critical bands. ERB's are more accurate until you get down to very low frequencies, where there is some serious dispute, and a lot of confusion between frequency and time issues) can be audible. (we're talking monaural here) That's because they can change the firing time of the inner hair cells in the cochlea.

Phase shift between two far-removed frequencies is not audible.

Time delay, however, or phase inversion (aka how many crossovers work) CAN be audible if it creates a change in the signal envelope. This is where digital crossovers, among other things, excel. But this is a complicated issue, and I have to shy off being too precise here, sorry, at least for the time being.

Between ears, about 5 milliseconds is the shortest credible ITD. This, however, depends enormously on frequency and the signal structure. At low frequencies, time delay (or phase shift) can be audible. Between 500 and 2kHz, give or take, it's much less audible. Above that, it depends extremely on the envelope of the signal. Again, I must shy off being too, too specific for the time being.

Neither of these directly answers questions about step response. Remember, first, that step response and impulse response are mathematically related. So you can talk about either meaningfully. And the answer to 'does that matter' is more complicated than I feel writing about at 3AM. Sorry.

Thanks for the explanation. I hadn't even thought about the cochlear filter angle.

This makes me think that all the MQA stuff about addressing "time smear" in the recording process can't possibly be audible...
 
Top Bottom