Speaker sensitivity numbers - how useful?

doc

Silver Supporting Member
Messages
7,778
I've been mulling over trying a few new speakers, and sensitivity is to my mind a major consideration - at least theoretically a speaker with a sensitivity of 100 should be significantly louder in the same amp than one at 97. The problem is I'm comparing sensitivity figures generated by different manufacturers. Lots of Eminence speakers tend to have higher reported sensitivities, while something like the Jensen Tornado has about a 97, and many Celestions are around 98. An Eminence Wizard is supposed to hit 102. Those who have compared several speakers in the same amp from different makers, have you noticed a significant difference in loudness consistent with the assigned sensitivity numbers?
 

IM4Tone

Member
Messages
3,769
Yes, I have noticed a significant difference between my three 2X12 cabs; at the same amp setting the Eminence is the loudest, the WGS and Celestion cabs are not significantly different from each other in volume to my ears. Whether or not they all follow their stated sensitivity or not, I never tried to check w/ an spl meter, but the difference of the Emi was easy to notice.

If your objective is more volume, or more headroom at the same volume, and you find an Emi that appeals to you tone-wise, I'd strongly recommend it.
 

wickedcookie

Member
Messages
1,186
I've been mulling over trying a few new speakers, and sensitivity is to my mind a major consideration - at least theoretically a speaker with a sensitivity of 100 should be significantly louder in the same amp than one at 97. The problem is I'm comparing sensitivity figures generated by different manufacturers. Lots of Eminence speakers tend to have higher reported sensitivities, while something like the Jensen Tornado has about a 97, and many Celestions are around 98. An Eminence Wizard is supposed to hit 102. Those who have compared several speakers in the same amp from different makers, have you noticed a significant difference in loudness consistent with the assigned sensitivity numbers?


They are marginally useful in comparing one speaker to another among the same manufacturer only.

There is no standard formula by which speaker sensitivity is calculated into a single SPL number.

There is also a lot of misinformation floating around about how SPL relates to perceived volume in our brains. On these very forums I've read everything from the notion that an increase of 3dB results in perceived volume doubling (not even close, unless we're going from something like 25 to 28 dB) to the notion that an increase of 3 dB is the smallest change perceptible to our brains (????). So take such claims with a grain of salt.

Here is a post by eigentone from the last speaker sensitivity thread a few days ago, he is spot on and I think from here on out I'm just going to quote him because it's much easier than trying to explain it anew each time the question is raised:

Too many musicians emphasize speaker "Sensitivity" as a very important metric. For example, they come to TGP and ask for a speaker recommendation for their amp -- then proceed to say "It must have a sensitivity of N". There's a lot more to sound and perception (=volume) than a sensitivity rating.


Sensitivity is not a standardized measurement and different manufacturers use different measurements to produce the sensitivity ratings they publish. Manufacturer A might evaluate a single frequency and Manufacturer B may evaluate several frequencies or a range of frequencies.


If you still think that sensitivity is somehow meaningful, read the frequency response graphs and notice how wildly speakers' frequency response varies. Guitar and bass speakers are not even close to flat. There are often significant peaks and valleys in that plot. If a manufacturer is measuring the output of a 1 kHz sine wave at 1m given a 1W input, they may get a reading of 97 dB. If they measure a whole tone or two in either direction, the result may be 5 dB higher or lower. So that speaker which has a sensitivity rating 5 dB higher might actually sound the same volume as the other speaker in as real life scenario -- because the second speaker emphasizes different frequencies.


Also, humans perceive sound different from microphones etc. So any "sensitivity" rating should be a range of frequencies which is weighted to match human perception of sound. This does not happen across the board.


Sensitivity can suggest whether one speaker is louder than another but it's often wrong or misleading because the two speakers don't have the same frequency response and measurements are not necessarily collected in the same way. It's best to ignore "Sensitivity" and try to focus more on frequency response graphs if you need an indication of a speaker's efficiency.


Sensitivity can be useful for QA, but you don't do that as a consumer.
 
Last edited:




Trending Topics

Top Bottom