• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Speaker Sensitivity: Advertised vs Measured

MZKM

Major Contributor
Forum Donor
Joined
Dec 1, 2018
Messages
2,426
Likes
5,178
Location
Land O’ Lakes, Florida
#1
Speaker sensitivity is an important aspect that can sway potential buyers towards or away from a specific model or brand. Thus, it gives an incentive for manufacturers to fudge their numbers; this fudging can be accomplished via the frequency range, weighting, in-room estimates, etc.

I decided to analyze all the passive speakers Amir has measured thus far, which is 36 as of writing this (some models did not have advertised sensitivity). For the calculation of the measured sensitivity, the average, non-weighted SPL from 300Hz-3kHz of the on-axis was used, this parameter was chosen as it is what SoundStageNetwork/NRC uses for their measurements (except they use it on a listening window). John Atkinson of Stereophile calculates sensitivity by feeding a "20kHz-bandwidth noise at a standard level, capture the output waveform...and apply B-weighting to the 1/10-octave-smoothed power response.".

Here are the results:

Screen Shot 2020-06-15 at 1.02.44 PM.png
Screen Shot 2020-06-15 at 1.14.06 PM.png
Screen Shot 2020-06-15 at 1.19.51 PM.png
Screen Shot 2020-06-15 at 1.00.42 PM.png

As we see, 50% of the speakers measured thus far deviate no more than 1/2 a decibel from their advertised sensitivity. The CEA-2034 standard states that a ±1.5dB variance between measurement setups is considered "good", and using that parameter rises the conformity to 75%.

The brand with the worst results is by far Klipsch. For their two models measured, their relative levels of differences are far from identical; this occurrence could be attributed to the different parameters Klipsch uses and differences in measurement technique (John Atkinson has measured the RP-600M as well and his parameters and measurement setup resulted in a difference of -6.4dB; he has measured 5 Klipsch models total and the average difference is -3.4dB).
As one of the most popular Hi-Fi / home theater brands, it is disheartening to see the use of parameters which result in them appearing to have much higher sensitivity; it should be of note that for Klipsch's current Hertiage line (Heresy IV, Cornwall IV, Klipshorn AK6, La Scala AL5, and Forte III ), they all state "Sensitivity in average listening room", yet none of their other model lines have/state this parameter.

The three brands with the best results are Ascend Acoustics, ELAC, and Pioneer. While the Dayton Audio & Polk speakers did match their specs exactly, only having 1 speaker measured for each cannot be used to form a conclusive statement.

While parameters used may be different, it is good to see the majority of the speakers measured thus far meet their sensitivity spec.

All interactive charts/data, along with some others, are available here, and updated when an applicable speaker is reviewed.

EDIT: Update, I asked some companies on their rating/testing parameters.
 
Last edited:

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
1,174
Likes
1,305
#2
The Klipsch discrepancy is especially problematic imo, as I often see them recommended to newbies based on that sensitivity. Go to a random audio forum, ask for advice on what to buy, say you like to listen loud, and that you only have an avr; chances are you're gonna get a lot of Klipsch recommendations.
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
1,174
Likes
1,305
#3
Seems like most manufacturers are fairly honest :) Looks like most "round up" ;), but I don't see that as a big deal. Some of those Harman ones are kinda pushing those rounding boundaries, though.
 

Music1969

Major Contributor
Joined
Feb 19, 2018
Messages
1,045
Likes
328
#5
The brand with the worst results is by far Klipsch. For their two models measured, their relative levels of differences are far from identical; this occurrence could be attributed to the different parameters Klipsch uses (John Atkinson has measured the RP-600M as well and his parameters with his measurement setup resulted in a difference of -6.4dB; he has measured 5 Klipsch models total and the average difference is -3.4dB) .
As one of the most popular Hi-Fi / home theater brands, it is disheartening to see the use of parameters which result in them appearing to have much higher sensitivity; it should be of note that for Klipsch's current Hertiage line (Heresy IV, Cornwall IV, Klipshorn AK6, La Scala AL5, and Forte III ), they all state "Sensitivity in average listening room", yet none of their other model lines have/state this parameter.
As soon as I saw the thread title I thought Klipsch, even before opening.

Someone should forward this to them for comment.

Stereophile made a similar comment in measuring one of their speakers.
 

Blumlein 88

Major Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
9,874
Likes
13,360
#6
I wish I had some good data on the various ESL's I've owned. Acoustats claimed like Soundlabs to be 88 db/watt/meter. In fact I'd say more like 78 or 80 is for real with a heavy lean toward 78 db.

Stereophile did measure 101 db/2.83v/meter which was lower than the claimed 105 db for the K-horns, but still awfully high.
 

ta240

Active Member
Joined
Nov 7, 2019
Messages
291
Likes
445
#7
The Klipsch discrepancy is especially problematic imo, as I often see them recommended to newbies based on that sensitivity. Go to a random audio forum, ask for advice on what to buy, say you like to listen loud, and that you only have an avr; chances are you're gonna get a lot of Klipsch recommendations.
Exactly, I've seen those recommended for low power amps so many times.
 
Joined
Dec 3, 2019
Messages
73
Likes
49
#8
It should also be noted that achieviing higher linearity can result in a overall lower sensitivity. This can be a design choice in products such as Klipsch where a flatter response may be sacrificed to achieve a higher sensitivity rating
 

Kvalsvoll

Active Member
Joined
Apr 25, 2019
Messages
242
Likes
425
Location
Norway
#9
What this shows is that the majority of manufacturers actually manage to measure and present a reasonably accurate specification, as long as the numbers presented has no important relevance to marketing.
 

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
4,975
Likes
10,952
Location
Gold Coast, Queensland, Australia
#11
The brand with the worst results is by far Klipsch.
I would consider the Harman group (JBL/Infinity/Revel etc) speakers is likely the consistently worst offender. All their speakers appear to be less sensitive than claimed and, with 31% of the total tested speaker cohort being theirs, I think the stage is set.
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
1,174
Likes
1,305
#12
I would consider the Harman group (JBL/Infinity/Revel etc) speakers is likely the consistently worst offender. All their speakers appear to be less sensitive than claimed and, with 31% of the total tested speaker cohort being theirs, I think the stage is set.
They seem to be the second worst.

Nothing touches that 8.3 difference, though. That could mean the difference between a pretty efficient(hifi) speaker, and a terribly inefficient one.
 

stunta

Addicted to Fun and Learning
Forum Donor
Joined
Jan 1, 2018
Messages
958
Likes
1,046
Location
Boston, MA
#13
I would consider the Harman group (JBL/Infinity/Revel etc) speakers is likely the consistently worst offender. All their speakers appear to be less sensitive than claimed and, with 31% of the total tested speaker cohort being theirs, I think the stage is set.
It's not about the number of speakers from one group is it? It should be about the variance. There is also selection bias given Harman's speakers are tested more than others. And then there is test system variance between manufacturer and Klippel. Is it fair to conclude Harman is the worst?
 

phoenixdogfan

Addicted to Fun and Learning
Joined
Nov 6, 2018
Messages
747
Likes
641
#14
The first chart shouldn't be a line chart.
I disagree. Among other things it also shows the actual sensitivity of all the speakers tested relative to one another. Looking at that, I can see that LS 50s are one of the very least efficient speakers among everything that's been measured to date. Small wonder everyone places so much emphasis on amplifier matching. A point of audiophile wisdom that turns out to have corroboration in measured performance.
 

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
4,975
Likes
10,952
Location
Gold Coast, Queensland, Australia
#15
It's not about the number of speakers from one group is it? It should be about the variance. There is also selection bias given Harman's speakers are tested more than others. And then there is test system variance between manufacturer and Klippel. Is it fair to conclude Harman is the worst?
I'm having a bit of fun for sure, but as with all of Amir's testing, patterns start to emerge. That is the great value for me. I applaud this new thread by @MZKM. I hope he continues it. :)

100% of Harman speakers appear below rated sensitivity and represent over a third (35%) of all the speakers tested.

The two Klipsch are way below spec, but only two have been tested. Not a good start for them either, especially with Klipsch partially known for their legendary sensitivity from the past.

I think sensitivities sell. They always have, except when they are too high or too low. Audiophiles have gotten used to bookshelf speakers around 85-87dB, and floorstanders 88-91dB. Outliers below 85dB are chalked up as inefficient and 'amplifier hungry' and numbers into the 90s or even over 100dB are associated with less "refined" loudspeakers. Consider sensitivities at 1W were always referenced to the nominal impedance, but with many speakers being low impedance and low efficiency, 2.83V @4R is really 2W, not 1W like it is for an 8R nominal speaker.
 

PenguinMusic

Active Member
Joined
Dec 20, 2019
Messages
276
Likes
69
#18
Hi guys,

What I can see is that here, most manufacturers give close to accurate sensitivity for their speakers.
And that is good...

As far as I am concerned, I really hope that my brand of speakers (Linn from Scotland) doesn't "cheat" on sensitivity.
If that would be so, it would be a disaster as they are advertised to have low sensitivity already.
That makes them already quite difficult to drive on paper !
And they indeed require a power amp that delivers pretty much power to deliver...

Regards.
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
3,846
Likes
2,336
Location
Tarragona (Spain)
#19


One data is wrong in the graph, at least with KEF Q100 5.25" coaxial speaker. KEF says 86 dB/W/m but true is almost 85 dB/W/m. (84.5-85).

KEf LS50, 85 dB/W/m -> true 83 dB/W/m.

Have measurements been made exactly at 1 m or 3 ft (as Stereophile)?
 
OP
M

MZKM

Major Contributor
Forum Donor
Joined
Dec 1, 2018
Messages
2,426
Likes
5,178
Location
Land O’ Lakes, Florida
Thread Starter #20
One data is wrong in the graph, at least with KEF Q100 5.25" coaxial speaker. KEF says 86 dB/W/m but true is almost 85 dB/W/m. (84.5-85).
KEF says 86dB @ 2.83V/m, which is the same parameter as the speakers measured by the Klippel, and it got 85.7dB.
 
Top Bottom