Importance of impulse response

[ernestcarl]

I’d like to see people post more complete IR data or even just a single sweep REW mdat file rather than just a picture of the FR magnitude graph most of the time. Yeah, context of the measurement be would appreciated as well.

 

[fpitas]

I’d like to see people post more complete IR data or even just a single sweep REW mdat file rather than just a picture of the FR magnitude graph most of the time. Yeah, context of the measurement be would appreciated as well.

That could aid those that are impulse buying

 

[ernestcarl]

That could aid those that are impulse buying

Well, I’d rather see the raw measurements than extremely detailed narrative descriptions of what equipment people own.

 

[fpitas]

Well, I’d rather see the raw measurements than extremely detailed narrative descriptions of what equipment people own.

No argument there.

 

[fineMen]

You can compare that to close to ideal timing of Neumann KH150:
KH150 post #786

Again, by my education I know how to set-up graphs to represent data. What I see here is irritating to say the least. The 'impulse' response, or deviation thereof, cannot be understood by the human vision. It is the same as with QR codes:

click me

What does it say? A lot, but not to Your eyes. It needs machinery, and so does the interpretation of an impulse response.

What is seen easily with the infamous 'impuls' is the deviation from an "ideal". So audiophools are easily tricked into thinking (?) that there is deviation, hence there is signal degradation. That IS true, yes, the signal is degraded. But that says nothing in regard to the information that the human hearing receives.

As I stated earlier, all the information is there as far it is relevant for the hearing, let alone the enjoyment of music, or even the information contained in reproduced human talk ... . It is just the amplitude spectrum over subsequent periods of time. Vulgo, the frequency response, amplitude, not phase.

The sad fact that even educated reviewers print 'impulse' graphs, and to double up the 'step' response too, doesn't mean it makes any sense at all. I personally think it is some b/s nobody ever thought about. Because the overly prudent engineers of the software used brought that in to just show off. Shi* happens ... , forgiven, but it remains shi*.

 

[617]

Here is timing measurement. It's certainly one of the worst I have ever seen. Look for example impulse envelope and step. Transient dynamics is quite totally lost because energy is distributed from edges to ca. 5 ms periods.
View attachment 242174

Without giving too much away, this is from some kind of 'professional' speaker? I've never seen a step response like this from a passive crossover.

 

[fineMen]

Without giving too much away, this is from some kind of 'professional' speaker? I've never seen a step response like this from a passive crossover.

One last comment to take home: if 'phase' was relevant, why is it, that white noise always sounds the same, given the amplitude spectrum is preserved? (Hint: Phase is chaotic, what if phase was ordered in one or the other way?)

Please discuss with Your neighbour and report. Thank You.

 

[Sokel]

If you can hear where your subwoofer is, you're either fooling yourself or it is creating distortion products several octaves into the midrange and treble.

Or you have it crossed way above 50Hz.

 

[gino1961]

i have the feeling that this is the plot that can tell more about the dynamic performance of a speaker ...
https://www.stereophile.com/images/917Kii3fig7.jpg

it's challenging to get a fast settling at around 1kHz where some ringing (?) can be seen ..
but above that the plot is very clean

 

[kimmosto]

I've never seen a step response like this from a passive crossover.

Passive 4-way with 4th order acoustic slopes could be quite close, but for example this is still better. Never listened this one but I believe it's better also subjectively.

 

[gnarly]

Well, I’d rather see the raw measurements than extremely detailed narrative descriptions of what equipment people own.

Here ya go...this is a tuning i did outdoors last week on a 5-way, ...... a 4-way synergy on top a double 18" bass-reflex sub.
(All kind of boat noise, lawn mowers, leaf blowers mixed in.......check out the noise floor in THD lol.)






Hey, how do you just attach a mdat ? Sure would be easier....

 

[ernestcarl]

Here ya go...this is a tuning i did outdoors last week on a 5-way, ...... a 4-way synergy on top a double 18" bass-reflex sub.
(All kind of boat noise, lawn mowers, leaf blowers mixed in.......check out the noise floor in THD lol.)


View attachment 242188View attachment 242189
View attachment 242190View attachment 242191
View attachment 242195View attachment 242196

Hey, how do you just attach a mdat ? Sure would be easier....

Yes, at the very least some time domain info should be included. One could have flat looking bass in a simple magnitude only curve, but extreme bass ringing very visible in the waterfall/decay views.

To attach a file, just (maximally) compress it to a zip format. Incidentally I found that applying FDW compresses the mdat to a much smaller file as well — so that can help too.

 

[thewas]

On-axis magnitude response is flat from 27 Hz to 24 kHz.
(Excess) Group delay is ~3.5 ms@100Hz by manufacturer. Time-windowed measurement shows a bit less of course.

Is there a GD plot available? Also would be interesting to see the vertical directivity unless it is a coaxial system.

 

[gnarly]

One last comment to take home: if 'phase' was relevant, why is it, that white noise always sounds the same, given the amplitude spectrum is preserved? (Hint: Phase is chaotic, what if phase was ordered in one or the other way?)

You do understand that phase is the relative timing relationship between frequencies, right?

Phase is quite ordered when sound originates from real production....a voice, an instrument, a fork dropping on a plate..
(I guess any sound, other than recordings playing on speakers, lol)

Fundamentals, harmonics, and timing relationships from real production are what they are, and necessarily emanate in relative sync.

Phase and timing gets a bit reordered when reproduced by most all speakers.
Hmmm, wonder if that could possibly have anything to do with why reproduction still can't capture the life-like sound of real production?

 

[dfuller]

Generally 'faster' means less bass, or that the room isn't causing certain bass frequencies to be inordinately loud. The easiest way to remedy bass/room problems is to simply take the bass out.

Bass is intrinsically slow - its slowness is what defines it as bass. How long does it take for a sound to have a pattern that can be called a 5000hz frequency? 1/5000 of a second is one period, so give it a two periods to even out and we're talking 1/2500 of a second. At 50hz it takes 1/25 of a second, which is 100 times slower.

This is a very crude illustration but I hope you can see my point. In reality our physiology requires different amounts of time to 'hear' a periodic pressure variation as a pitch. If I remember correctly, it takes even longer for bass to be recognized as such.

The most important effect this has is that bass will travel all around your room and interact with itself in all kinds of chaotic ways before you can even recognize it as a bass pitch. This is why bass is not localizable. If you can hear where your subwoofer is, you're either fooling yourself or it is creating distortion products several octaves into the midrange and treble.

Well, yes and no. It's very easy to have an incorrect port tuning that causes all kinds of weird bass overhang and very high group delay.

 

[ChargedCapacitor]

Just wanted to answer directly to OP:

Impulse response is a very useful tool to test and correct the speaker, as well as the source. For example, if you design a particular crossover, you should be able to predict how the impulse will look. You can also see these issues in the frequency response, but it's not always as obvious. So it can be used as a tool to ensure you have your crossover hooked up properly.

Also, the impulse is a great tool to see if you have issues with your source. Impulse is especially useful here, since you would not see these issues in the frequency response. For example, I had issues with the right channel of my source. These issues made the right channel sound less resolving and detailed than the left channel. However, when I measured the FR, everything looked good. When I looked at the impulse, and switched the channels so the same speaker played from left and right channel, the impulse was obviously worse on the right channel.

As far as how the speaker will sound, so long as the impulse response matches up with your predicted response and you see no glaring issues, it's pretty much impossible to determine or predict performance by a glance. Frequency response may not speak the final word in tonal / technical performance, but it certainly speaks the loudest.

 

[kimmosto]

Is there a GD plot available? Also would be interesting to see the vertical directivity unless it is a coaxial system.

Excess GD is my (time-windowed) measurement at 1m. Normal GD is manufacturer's data. Not too bad conflicts.

Directivity in vertical plane is also flawless.

 

[kimmosto]

Instant comparison to KH150 with as flawless timing as vented system with few ms latency can have

 

[fpitas]

Instant comparison to KH150 with as flawless timing as vented system with few ms latency can have
View attachment 242230

Yes, when I design vented I try to minimize the GD, even though it means a very soft LF "knee".

 

[kimmosto]

What I see here is irritating to say the least. The 'impulse' response, or deviation thereof, cannot be understood by the human vision.

This is slightly technical audio forum so I suggest to use few moments to study what step response and impulse response envelope (ETC) are and can tell you instead of repeating that you don't understand them or my output is kinda useless. It's not our problem and responsibility to filter or modify data in posts if you don't understand it.
Br, completely unscientific country boy