Speaker Sensitivity: Advertised vs Measured

[Xulonn]

I don't like shading to accommodate colorblindness

Like about 10% of American males, I have poor red-green color differentiation ability - but I have no problem reading any of the example graphs you proposed. And I do like the alphabetical presentation.

 

[Robbo99999]

I'm fine with whatever, though I don't like shading to accommodate colorblindness and I like going alphabetical so brand trends can be seen and specific models are more easily discoverable.

So,

A:
View attachment 68951
Or
B:
View attachment 68952

I'd go with version A.

It's not worth dedicating the whole thread to discussion of graph choice though, so I think some people are getting a little fixated about it.

 

[MZKM]

Added the average difference by company;

 

[Jon AA]

Because some people might want to know the absolute sensitivity in SPL, rather than just the difference.

I agree with that. While the difference between advertised and measured might be interesting to some, particularly those who might want to brand-bash, it's not a performance measurement that's useful to anybody.

The absolute measured sensitivity, however, is a performance measurement that is pretty important to a lot of people. A graph people could look at to see the actual sensitivity of all speakers measured could be highly useful.

 

[Rebelhifi]

I found this from the Klipsch website. It looks like they account for the directivity of the horns farther way and work backwards (they add room boundary but this is not the full difference).

Klipsch measures sensitivity on home loudspeakers in the following manner:
1. We place the speaker to be tested in our anechoic chamber and do a free space measurement (no boundary gain or room gain) at a distance of 3 meters. This distance is chosen in order to be in the far field of the speaker. The test signal used is wide band pink noise.
2. Starting with this result in dB, we add 9 dB to obtain a one-meter equivalency, as the industry standard for expressing sensitivity is one meter. The 9 dB difference is all inverse-square gain. Move half as far away and gain 6 dB etc.
3. To this we add 4 dB for room gain and boundary gain to translate the measurement to a typical listening environment. Speakers are not operated in free space in any normal application. An increase in sound pressure comes from proximity to nearby walls, floor and ceiling. Theoretically, a maximum of 18 dB increase is available through corner placement but that is rarely the position chosen for full range loudspeakers; and the increase is also frequency dependent, being prominent at low frequencies. Additional measurable increase comes from room gain wherein the room is pressurized by low frequency information. Again, this is frequency dependent impacting only the low end of the spectrum. We have verified the 4 dB figure we use in numerous empirical measurements and believe it to be quite accurate.
Our KPT-904 professional theater speaker was mentioned above. It should be noted that the KPT-904 is a model designed to be placed behind the screen at a movie theater and as such does not benefit from as much boundary reinforcement as in a typical home installation. The 4 dB room gain figure is not applied to the sensitivity measurement for such models.

 

[MZKM]

I found this from the Klipsch website. It looks like they account for the directivity of the horns farther way and work backwards (they add room boundary but this is not the full difference).

Klipsch measures sensitivity on home loudspeakers in the following manner:
1. We place the speaker to be tested in our anechoic chamber and do a free space measurement (no boundary gain or room gain) at a distance of 3 meters. This distance is chosen in order to be in the far field of the speaker. The test signal used is wide band pink noise.
2. Starting with this result in dB, we add 9 dB to obtain a one-meter equivalency, as the industry standard for expressing sensitivity is one meter. The 9 dB difference is all inverse-square gain. Move half as far away and gain 6 dB etc.
3. To this we add 4 dB for room gain and boundary gain to translate the measurement to a typical listening environment. Speakers are not operated in free space in any normal application. An increase in sound pressure comes from proximity to nearby walls, floor and ceiling. Theoretically, a maximum of 18 dB increase is available through corner placement but that is rarely the position chosen for full range loudspeakers; and the increase is also frequency dependent, being prominent at low frequencies. Additional measurable increase comes from room gain wherein the room is pressurized by low frequency information. Again, this is frequency dependent impacting only the low end of the spectrum. We have verified the 4 dB figure we use in numerous empirical measurements and believe it to be quite accurate.
Our KPT-904 professional theater speaker was mentioned above. It should be noted that the KPT-904 is a model designed to be placed behind the screen at a movie theater and as such does not benefit from as much boundary reinforcement as in a typical home installation. The 4 dB room gain figure is not applied to the sensitivity measurement for such models.

They doesn‘t state what frequency range is used, since Klipsch usually has boosted treble, they likely run the sensitivity on that.

If that claims they are doing +4dB compared to anechoic, than the 96dB rated RP-600M should be 92dB somewhere in it’s range. Giving them 1dB of leeway, there should be a region where it averages 91dB, looking at its Spin:

It hits 91dB at ~800Hz, but the only region where that is the average is ~ 10kHz-18kHz.

Also, it’s not 9dB for 3m but ~9.5dB, so unless that comment was written in error, that’s another 1/2 decibel that can be used for beneficial (for them) rounding.

 

[LTig]

A

I'm fine with whatever, though I don't like shading to accommodate colorblindness and I like going alphabetical so brand trends can be seen and specific models are more easily discoverable.

So,

A:
View attachment 68951
Or
B:
View attachment 68952

A.

 

[carlosmante]

I'd go with version A.

It's not worth dedicating the whole thread to discussion of graph choice though, so I think some people are getting a little fixated about it.

Right presentation of Data is a very important aspect of any Scientific Research.

 

[mhardy6647]

mmmm -- a table would've worked just fine for me.

 

[amirm]

3. To this we add 4 dB for room gain and boundary gain to translate the measurement to a typical listening environment. Speakers are not operated in free space in any normal application. An increase in sound pressure comes from proximity to nearby walls, floor and ceiling. Theoretically, a maximum of 18 dB increase is available through corner placement but that is rarely the position chosen for full range loudspeakers; and the increase is also frequency dependent, being prominent at low frequencies. Additional measurable increase comes from room gain wherein the room is pressurized by low frequency information. Again, this is frequency dependent impacting only the low end of the spectrum. We have verified the 4 dB figure we use in numerous empirical measurements and believe it to be quite accurate.

Thanks for posting this. What they are saying applies to every speaker, not just their designs. To the extent they are the only ones doing this, they are inflating their spec relative to other manufacturers.

 

[MZKM]

Thanks for posting this. What they are saying applies to every speaker, not just their designs. To the extent they are the only ones doing this, they are inflating their spec relative to other manufacturers.

I think they are especially deceiving in that the Heritage line says it’s an in-room spec but none of their other lines do.

I own a Klipsch sub, and instead of a CEA-2010 spec like some sub companies use, mine is using 1/8 space loading, which by searching online is anywhere from a 9dB to 18dB boost.

 

[stevenswall]

I can only straw many this argument: "Speaker sensitivity is an important aspect that can sway potential buyers towards or away from a specific model or brand."

What is the logical/reasoned cause of this swaying people considering amps are cheap and powerful, and vacuum tube amps that do cost much more for more power aren't really in keeping with the spirit of ASR?

Right now I see it as a spec that makes people say, "oh look, this sounds better and might expose more of the gritral texturalial details of my megamacro dynamicros," and then it plays louder and they say "See! I can hear the difference!"

Isn't sensitivity just loudness? Aren't there reasonable multi-hundred watt amps to be had these days?

Perhaps there is some dynamic compression element I'm missing, but I haven't seen that measured or noticed it like I've noticed mastering compression.

 

[MZKM]

Isn't sensitivity just loudness? Aren't there reasonable multi-hundred watt amps to be had these days?

Perhaps there is some dynamic compression

Important for home theater, but can be as large of a factor for much with high dynamic range like orchestral/classical.

Most speakers I see can handle no more than ~100W (+20dB), some budget ones can only handle 50W or even less.

Movies call for 105dB peaks at the listening seat.

I sit 12ft away, which is an ~8dB loss, but you have say 2dB of room gain for a bookshelf, so make that -6dB from the base sensitivity, if you have 85dB speakers, that makes it now 79dB, so to get +25dB you would need 400W, and that’s just assuming 8ohm, it we are talking 4ohm then that’s 800W. Again, that’s even if the speaker can handle that, which none do.

 

[stevenswall]

Important for home theater, but can be as large of a factor for much with high dynamic range like orchestral/classical.

Most speakers I see can handle no more than ~100W (+20dB), some budget ones can only handle 50W or even less.

Movies call for 105dB peaks at the listening seat.

I sit 12ft away, which is an ~8dB loss, but you have say 2dB of room gain for a bookshelf, so make that -6dB from the base sensitivity, if you have 85dB speakers, that makes it now 79dB, so to get +25dB you would need 400W, and that’s just assuming 8ohm, it we are talking 4ohm then that’s 800W. Again, that’s even if the speaker can handle that, which none do.

$650 with a Crown XLS 2502 that does 775 wpc at 4 ohms. I can see the cost argument to get peak loudness with many channels, just wanted to know if there was anything inherently better about a higher sensitivity speaker as it seems like an odd purchase factor and dated design goal with moderns amps having so much power for so little.

 

[Sancus]

$650 with a Crown XLS 2502 that does 775 wpc at 4 ohms. I can see the cost argument to get peak loudness with many channels, just wanted to know if there was anything inherently better about a higher sensitivity speaker as it seems like an odd purchase factor and dated design goal with moderns amps having so much power for so little.

I think of it as one way to achieve high SPL output/"dynamic" capability, or whatever you want to call it. The other way, of course, is with lots of power and normal sensitivity. However as MKZM pointed out, not many normal medium to low sensitivity speakers can actually handle the power they would need to get crazy loud. For example, the F208, a huge floor stander, recommends 50-350W on its spec. I'm sure it can handle somewhat more... but 800W? And that's already a pretty large and heavy speaker.

If you want to get really loud for longer listening distances, many of the designs seem to switch to things like compression drivers and huge/multiple woofers which you see in the Genelec main monitors or the JBL M2, they have higher sensitivity but also can handle a lot of power.

 

[MZKM]

$650 with a Crown XLS 2502 that does 775 wpc at 4 ohms. I can see the cost argument to get peak loudness with many channels, just wanted to know if there was anything inherently better about a higher sensitivity speaker as it seems like an odd purchase factor and dated design goal with moderns amps having so much power for so little.

No quality differences, just the ability to get loud enough. And again, a lot of drivers aren’t rated for more than 100W.

 

[richard12511]

$650 with a Crown XLS 2502 that does 775 wpc at 4 ohms. I can see the cost argument to get peak loudness with many channels, just wanted to know if there was anything inherently better about a higher sensitivity speaker as it seems like an odd purchase factor and dated design goal with moderns amps having so much power for so little.

Most speakers actually lack the ability to get to reference levels at 10+ feet. If you like movie theater sound, or you like to listen loud, or you like classical music, most speakers actually can't handle the dynamic range required. For example, only the Genelec "Main Monitors" satisfy all those requirements.

 

[Zedly]

I can only straw many this argument: "Speaker sensitivity is an important aspect that can sway potential buyers towards or away from a specific model or brand."

What is the logical/reasoned cause of this swaying people considering amps are cheap and powerful, and vacuum tube amps that do cost much more for more power aren't really in keeping with the spirit of ASR?

For me sensitivity was a factor because I am using a 5.2.4 Atmos setup, so I have 9 speakers to drive. With that many speakers and channels, amplification is not so cheap any more. I ended up buying an Outlaw 7000x (130W into 8 ohms, all 7 channels driven.) for just under $1000. A Monolith with 7x200W is $1600. At those prices, with that many speakers, I didn't feel I could ignore sensitivity.

 

[Sancus]

Most speakers actually lack the ability to get to reference levels at 10+ feet. If you like movie theater sound, or you like to listen loud, or you like classical music, most speakers actually can't handle the dynamic range required. For example, only the Genelec "Main Monitors" satisfy all those requirements.

I think reference level itself is frequently misunderstood/misapplied though. For example, your starting point for it is 115dB in the LFE channel, which is not going to happen without large, expensive(>$1000) subwoofers. One reason huge DIY subs are popular in the HT community, probably. Without that capability, worrying about the 105dB requirement in the other channels is pointless.

Classical is probably less bass heavy than the average spectrum of pop music, so then the focus would shift to capability in the 100hz-1khz range or something like that, but I haven't seen a study on the average spectrum of classical, that information would be great to have.

 

[Rebelhifi]

For my theater, I went with Klipsch Reference Premier due to the sensitivity and dialogue intelligibility of horns for movies (I have found dynamics and dialogue the most important elements of my theater). So, sensitivity does sell. I did investigate the real sensitivity of 91 db (Stereophile, Sound and Vision as it was typically higher than others in their publication -- 2 to 3 db higher than standard dome tweeter systems) and went with Crown XLS2502 amps (to get the technical 105 db potential -- never listen that loud.). At 12 ft, I am starting at 85db (with 2 db of room gain), the power levels are 128 watts 8 ohm / 256 watt 4 ohm @ 105 db. While dynamic, great for movies, and live concerts (I use electrostatic speakers for critical listening), my issue is the slight noise level (hiss at the speakers -- can't hear it with my projector going). Additionally, I could not hear the hiss with my previous Infinity Speakers of less sensitivity.

Note: I have owned many speakers -- Klipsch, Martin Logan, Boston Acoustics, Infinity, Advent, Acoustic Research ****** I guess I am partial to American brands -- just noticed.