sensitivity or frequency response for cabs?

[will33]

Here's one to start with... sensitivity. How should it be measured and please justify WHY it should be done that way.

Industry standard 1W/1M obviously, I say 20-20k because that is the accepted range of human hearing. Should qualify that by saying at birth. Some things fade a little with age.

 

[will33]

DukeLejeune has good points about on and off axis response. The off axis will always looks worse but as he said before, worse as compared to what? Company B's could likely be worse than that so compare/contrast and show why one's product is better. That would be possible if all these manufactureres would publish this stuff. If not, then one would be left with the option of purchasing one of every competing product and measuring them all. (I have a sneaking suspicion that info's out there somewhere too, just hard to get at). Room effects are room effects, they're going to change in every room, every time. Problem is the room you play this weekend as compared to next weekend are likely going to be vastly different so are you guys trying to correct for some sort of "average room", that doesn't make sense to me.

 

[DukeLeJeune]

I use an algebraic average on-axis response...

I think Eminence does essentially the same thing. From their website:

"Eminence sensitivities are expressed as the average output across the usable frequency [range] when applying 1W/1M into the nominal impedance. ie: 2.83V/8 ohms, 4V/16 ohms."

* * * *

Will33, I'm still thinking about this statement: "If somebody's idea of stellar tone is a fridge or an Eden, fine. If specs were out there, they could find out, on a technical level, what it is they like about it...."

Suppose I wanted to build a tone-clone of an Eden cab. If I get to see one or more frequency response curves, I can probably make an educated guess and fill in the blanks as to what they are doing (maybe not all the little details, but the big stuff). There's a pretty good chance they would have just handed me some of their most valuable trade secrets. Now if I can keep my manufacturing costs down, I can sell my Eden tone-clone for less than they do (less development cost to recoup), and even use their reputation to market my copies. So, I can see why Eden might think twice about publishing curves. Arguably as it now stands I could just go out and buy the cab and clone it the old-fashioned way, but a good published curve or two might save me from incurring even that expense.

 

[agedhorse]

Let me address a few of the points that have been raised and why it might not be in the customer's best interest or why it might be misleading...

1. 1W/1M sensitivity is fine but this is just measuring the small signal response and driver/cabinet response varies greatly (sometime way too greatly) with power. As power levels are increased, the transfer curve shifts due to the effcets of power compression which are only partly due to voice coil temp. There are other important mechanisms at play involving the magnetic circuit and position within the gap, saturation effects, non-linearity of the mechanical suspension components and the porting of the cabinet. These are a VERY important part of how a cabinet sounds when driven hard. Seeing that many folks here drive their cabinets well past the "RMS" ratings, this effect may be even more important to the real world sensitivity than the 1W/1M number.

2. Average sensitivity measurement bandwidth... it was said by several here that industry standard is 20Hz-20kHz. While that may be true of power amps, it's foolish to believe that this would be reasonable for speakers that have defined bandwidths or specific voicings. A subwoofer, for example has a design bandwidth of say 35Hz - 200Hz. If a true algebraic average of 20-20kHz was used, there are octaves of zero response and these weight the average SPL way down compared with say a box with a design bandwidth of 35-6kHz. So with the same exact test, the sub will measure many dB below the other cabinet, even though both might be identical over their design bandwidths. How many folks find useful output at 20kHz for a bass cabinet? Hopefully those that don't like tweeters and high end aren't offended by the presumption that a 20khz measurement bandwith makes their cabinets look worse by numerical comparison to one with a tweeter. The voicing of a bass speaker is generally not to 20kHz and few have anything meaningful beyond 6 or 7kHz. Why incude out of band response in the measurement averaging?

3. The space-loading of the cabinet changes the low end response of a cabinet, as does multi-cabinet coupling. What space loading do we as an industry use? Pro audio uses full-space and 1/2-space generally depending on the application. A flown speaker would generally be flown (duh) and the full-space numbers would be a better prediction of how the speaker will respond in situ. This would be inaccurate if the speaker was positioned on the ground or in a corner. I use 1/2-space for bass cabinets because that's the way the vast majority of players use them in. I have seen numbers that had to be 1/4-space or even 1/8-space because otherwise they would be well beyond the practical maximum efficiency of that class of loudspeaker.

4. The sensitivity of a typical bass speaker increases with frequency on axis and decreases as the position gets off axis. This is due to the beaming effects of the LF drivers and varies with driver size. This greatly affects the perceived voicing depending on position in the room.

5. Multiple driver cabinets... the measurement pickup location MUST be determined emperically due to interference patterns as waves sum and differ. Depending on where the mic is located, comb filtering will result and the resulting graph will be useless. The idea is to fine the position of minimum diffraction pattern.

These are just a few thoughts as I prepare for bed after 8 days in a row of very long days. These gigs are getting harder and harder and I'm glad I'm not out in the field as much as I used to be. Ugh.

 

[agedhorse]

Why not use the accepted standard that the rest of the loudspeaker industry has used since well before the electric bass was invented? It's not like there's any new ground that needs to be broken here. The notion that an entirely new method is necessary is just plain silly.

Because the established pro audio and consumer audio industry standards are preposterous for bass guitar. Using out of band measurements to generate average sensivity ratings skews the results differently based on different speaker voicings and null response bandwidth to the point that speakers that sound MUCH louder measure quieter that others that may have extended bandwith that is out of band for bass applications.

Publish measured SPL charts, axial and 45 degrees off-axis. Those who know how to interpret them can. Those who don't can learn, those who don't care can just ignore them. And you shouldn't have to go to any great lengths to meet this goal, as you certainly must have these on file for all your products already.

It's not those who really know, it's the ones who think they know and make false assumptions based on lack of understanding. God knows we have seen this here time and time again. You, in fact have been on the receiving end of some of this right?

 

[alexclaber]

Here's what I'm doing now with respect to specs (just putting the sheet together at the moment).

For the lows:
1. Low frequency half-space sensitivity at 100Hz
2. F3, F6, F10 all referenced to that
3. Minimum peak power it takes to reach 10% THD below 100Hz but above the tuning frequency
4. Max LF SPL based on sum of (1) and (3)

For the rest:
1. On-axis half-space sensitivity based on a weighted mean of the (A)200-400, (B)400-800, (C)800-1600 and (D)1600-3200 octave bands. Weighting is (8A+4B+2C+D)/15 = weighed mean.
2. Continuous thermal power handling (the usual quoted spec)
3. Max broadband SPL based on sum of (1) and (2)

Finally a 'usable frequency range' figure which is simply there so the cabs don't get written off by those that don't understand honest specs, and is based the range over which the cabs can produce giggable output (similar to -10dB points but also taking into account power handling and ability to extend response with EQ if power handling is high enough).

I don't know how Eminence comes up with their sensitivity number but I suspect it's a simple mean of response over the quoted band. However this approach is not logical as the energy content of music decreases with increasing frequency, hence my weighted mean approach.

 

[KJung]

Here's what I'm doing now with respect to specs (just putting the sheet together at the moment).

For the lows:
1. Low frequency half-space sensitivity at 100Hz
2. F3, F6, F10 all referenced to that
3. Minimum peak power it takes to reach 10% THD below 100Hz but above the tuning frequency
4. Max LF SPL based on sum of (1) and (3)

For the rest:
1. On-axis half-space sensitivity based on a weighted mean of the (A)200-400, (B)400-800, (C)800-1600 and (D)1600-3200 octave bands. Weighting is (8A+4B+2C+D)/15 = weighed mean.
2. Continuous thermal power handling (the usual quoted spec)
3. Max broadband SPL based on sum of (1) and (2)

Finally a 'usable frequency range' figure which is simply there so the cabs don't get written off by those that don't understand honest specs, and is based the range over which the cabs can produce giggable output (similar to -10dB points but also taking into account power handling and ability to extend response with EQ if power handling is high enough).

I don't know how Eminence comes up with their sensitivity number but I suspect it's a simple mean of response over the quoted band. However this approach is not logical as the energy content of music decreases with increasing frequency, hence my weighted mean approach.

Hey Alex! Hope you are doing well and that your company is growing! Good to see you on the site again!

 

[JimmyM]

i thought i'd just stay out of it for a while and let the big boys fight it out. but i'm gradually getting swayed back by the minimal specs folks. as we can see here, specs are often misinterpreted or over-emphasized as selling points. and let's face it...very few bass players are ever going to learn how to interpret specs to any meaningful degree. making full specs available to those who want them sounds reasonable, until you factor in how some dopey kids with no knowledge of what they mean thinks they know what they mean. then it costs the manufacturer some sales, and let's face it...companies are there to sell. i totally don't blame them for not emphasizing them if it costs them sales to do so. as for manufacturers taking the responsibility to educate consumers, that's all well and good till you see that the vast mjaority have zero interest in learning. and i can relate to that because to a large degree, i'm fairly uneducated on amp/speaker science, and till about 2-3 years ago, had zero interest in educating myself. glad i learned some about it, but i believe i'm very atypical.

 

[billfitzmaurice]

Arguably as it now stands I could just go out and buy the cab and clone it the old-fashioned way, but a good published curve or two might save me from incurring even that expense.

You could rent one at 10AM, run SPL and impedance sweeps and test the driver specs to find out everything you could possibly need to know to duplicate it, and have it back by 5PM.

Because the established pro audio and consumer audio industry standards are preposterous for bass guitar...

Baloney. If I didn't know better I'd say you're stalling, hoping that this will all go away.

specs are often misinterpreted or over-emphasized as selling points. and let's face it...very few bass players are ever going to learn how to interpret specs to any meaningful degree.

Learning how to interpret specs is like learning to sight read. It may seem daunting at first, but if you want to do it you can. You don't have to know how to sight read to get a gig, but if you want your name at the top of the will-call list you'll do it.

It boils down to honesty in advertising. For example, specs of 'Frequency +/- 3db 45-18K Hz Sensitivity 1W/1M 98 db' for a 1x12/tweeter are impossible to attain, but there it is, in print on the website of a well respected manufacturer. A perusal of said manufacturer's specs for every speaker they make reveals all of them to be fairy tales. And this manufacturer's deception, be it calculated or otherwise, is not the exception, it's the rule.
The fact of the matter is that measuring SPL is such a simple procedure today that whether the industry likes it or not it's going to happen, across the board, and soon. So they can keep dragging their heels or they can come clean now before they're involuntarily 'outed'.

 

[topcat2069]

i thought i'd just stay out of it for a while and let the big boys fight it out. but i'm gradually getting swayed back by the minimal specs folks. as we can see here, specs are often misinterpreted or over-emphasized as selling points. and let's face it...very few bass players are ever going to learn how to interpret specs to any meaningful degree. making full specs available to those who want them sounds reasonable, until you factor in how some dopey kids with no knowledge of what they mean thinks they know what they mean. then it costs the manufacturer some sales, and let's face it...companies are there to sell. i totally don't blame them for not emphasizing them if it costs them sales to do so. as for manufacturers taking the responsibility to educate consumers, that's all well and good till you see that the vast mjaority have zero interest in learning. and i can relate to that because to a large degree, i'm fairly uneducated on amp/speaker science, and till about 2-3 years ago, had zero interest in educating myself. glad i learned some about it, but i believe i'm very atypical.

Yeah.....

I jumped in early on the side of "it doesn't matter" and got bitch slapped! OUCH !!

as a recording engineer I do care about the response of my monitors, but as a Bass player, I'm more concerned with being in the pocket and playing the right notes. I'm not stupid , I just don't care... would AccuGroove cabs make me a better Bass player? would knowing and understanding what the specs of my speakers are make me better than practicing would ? Specs are important to the designers, as they should be ... I put my trust in what they design, but I'm a musician and in a band setting, playing for a crowd, it just doesn't matter [to me].

 

[jnewmark]

You could rent one at 10AM, run SPL and impedance sweeps and test the driver specs to find out everything you could possibly need to know to duplicate it, and have it back by 5PM.
Baloney. If I didn't know better I'd say you're stalling, hoping that this will all go away.
Learning how to interpret specs is like learning to sight read. It may seem daunting at first, but if you want to do it you can. You don't have to know how to sight read to get a gig, but if you want your name at the top of the will-call list you'll do it.

It boils down to honesty in advertising. For example, specs of 'Frequency +/- 3db 45-18K Hz Sensitivity 1W/1M 98 db' for a 1x12/tweeter are impossible to attain, but there it is, in print on the website of a well respected manufacturer. A perusal of said manufacturer's specs for every speaker they make reveals all of them to be fairy tales. And this manufacturer's deception, be it calculated or otherwise, is not the exception, it's the rule.
The fact of the matter is that measuring SPL is such a simple procedure today that whether the industry likes it or not it's going to happen, across the board, and soon. So they can keep dragging their heels or they can come clean now before they're involuntarily 'outed'.

It seems like you are describing the Genz Benz Neox 112T with those specs. Are you saying that GB is exaggerating or downright not telling the truth ?

 

[Queg]

Yeah.....

I jumped in early on the side of "it doesn't matter" and got bitch slapped! OUCH !!

as a recording engineer I do care about the response of my monitors, but as a Bass player, I'm more concerned with being in the pocket and playing the right notes. I'm not stupid , I just don't care... would AccuGroove cabs make me a better Bass player? would knowing and understanding what the specs of my speakers are make me better than practicing would ? Specs are important to the designers, as they should be ... I put my trust in what they design, but I'm a musician and in a band setting, playing for a crowd, it just doesn't matter [to me].

But as it currently stands, bass players have no way of knowing what a cab may sound like/what it's capable of until they hear it for themselves or glean enough consensus word-of-mouth experience from fellow players to make a decision. You or I may not be personally educated enough to use specs as a guideline, but at least it's a standard language and anyone who's sufficiently tech savvy can read them for us.

Here's a question for the big boys:
Currently on the amp forum there's a thread from someone who ordered up a well-respected 410 and was surprised that it was not as "full and deep in the lows" as they had expected. Experienced responses in the thread suggest he should have gone with another 410 model offered by same manufacturer. If the manufacturer was providing industry standard, accurate & detailed specs, could this consumer have had a greater potential to make the right choice to begin with?

 

[billfitzmaurice]

It seems like you are describing the Genz Benz Neox 112T with those specs. Are you saying that GB is exaggerating or downright not telling the truth ?

I'm saying that no 3.5 cu ft twelve inch woofer/tweeter cab is capable of 'Frequency +/- 3db 45-18K Hz Sensitivity 1W/1M 98 db'. As for being +/-3dB from 45-18kHz, that's a tall order from even high end 3-way hi-fi cabs, and those capable of it run around 87dB sensitivity. To get that flat a response with 98dB sensitivity one would need a very high quality 3 way PA cab, one a lot larger than 3.5 cubic feet.
Of course, there's always the possibility that I'm wrong. Just ask my wife. So anyone who wishes to prove me wrong, just post an SPL chart for a 3.5 cu ft 1x12/tweeter cab capable of this performance and I'll sit down to a serving of cold crow. Beats the wife's cooking anyway.

 

[rpsands]

Bill, I know it doesn't change the numbers completely, but according to Genz (agedhorse) he measures cabs in half-space. Doesn't that involve a 3db gain in the low end, which could get him closer from the realistic 12" bottom end of 95db?

Still doesn't account for the almost assured peaks and dips of >3db in the midrange, but closer I think.

 

[billfitzmaurice]

Bill, I know it doesn't change the numbers completely, but according to Genz (agedhorse) he measures cabs in half-space. Doesn't that involve a 3db gain in the low end, which could get him closer from the realistic 12" bottom end of 95db?

Still doesn't account for the almost assured peaks and dips of >3db in the midrange, but closer I think.

Half-space is the IEC standard. It's what you get for a result in WinISD and other box programs, and it's what used for driver data sheet SPL charts. So when a driver calcs out as 95dB sensitive that is a half-space result.

 

[kurotenshi]

Here's a question for the big boys:
Currently on the amp forum there's a thread from someone who ordered up a well-respected 410 and was surprised that it was not as "full and deep in the lows" as they had expected. Experienced responses in the thread suggest he should have gone with another 410 model offered by same manufacturer. If the manufacturer was providing industry standard, accurate & detailed specs, could this consumer have had a greater potential to make the right choice to begin with?

Now think about a person like me that lives in a country where, as happened to me a couple of moths ago, ordering two stomp boxes and paying shipping from germany was just a bit more expensive than ONE of the stomp boxes in the nearby store.
It just so happened that even the cabs that are in stock at the nearby store you cannot try out without giving the guys some degree of certainty that you'll buy something because they know people try things there and then buy online.

 

[KJung]

But as it currently stands, bass players have no way of knowing what a cab may sound like/what it's capable of until they hear it for themselves or glean enough consensus word-of-mouth experience from fellow players to make a decision. You or I may not be personally educated enough to use specs as a guideline, but at least it's a standard language and anyone who's sufficiently tech savvy can read them for us.

Here's a question for the big boys:
Currently on the amp forum there's a thread from someone who ordered up a well-respected 410 and was surprised that it was not as "full and deep in the lows" as they had expected. Experienced responses in the thread suggest he should have gone with another 410 model offered by same manufacturer. If the manufacturer was providing industry standard, accurate & detailed specs, could this consumer have had a greater potential to make the right choice to begin with?

I'm not one of the big boys, but my answer is 'no'. First, I don't think he bought the cab based on spec's, but rather recommendations (I participated in that thread, since I'm familiar with the cabs he is discussing).

Second, that cab is deep in the lows... IMO again, he is missing low mid punch, and mis-attributing that to a 'lack of low end'. It is hard for the layman (me included) to translate specs to tone.

Third, unless you include the rest of the input signal (including the player's technique and tone goals), specs would be relatively useless. He combined a cab with a relatively deep low end extension and polite low mid punch with... an amp that has relatively deep low end extension and polite low mid punch (and no low mid EQ control!). Guess what happened?!?!?!?... no low mid punch. That cab with a different head with a different voicing KILLS (I know this for a fact).

This is the problem. As Agedhorse says, you can't really apply 'stereo system' logic (where you input signal is constant, pure and compressed) to bass amplification.

Yes, if you had massively detailed information on every bass amp frequency distribution, every bass guitar frequency profile, and every cab, and you got an EE degree, the information might be useful enough to loosely predict what the entire system sounds like

 

[rpsands]

I don't need massively detailed info about every bass amp and every bass guitar and an EE degree to be able to ballpark how something will work from specs.

I'll give you a good example: Looking at the Jack 12 frequency charts and knowing that my bass and amp are pretty uncolored, I knew that I would need to boost in the upper bass to get the thumpiness that I wanted.

How did I know that? Because I used a bass amp with a parametric eq for a long time and got to know what frequencies get which sounds. Any idiot can do that.

Once you know how paramtric eq (or a good graphic) works you can apply that to a decent FR chart and make decent educated guesses.

 

[KJung]

Once you know how paramtric eq (or a good graphic) works you can apply that to a decent FR chart and make decent educated guesses.

Again, for the 1% or less of bassists who even know what a parametric EQ is, much less know how to use one or interpret the intervals, I agree

If I was a stock holder in a bass amp/speaker company that spent time providing that information, I would find out who was responsible and fire them for wasting corporate resources on something that would not sell any more boxes, nor make 99% of customers any more happy!

That being said, for the 1% or less that are intrested and can make sense of this stuff, that defines a niche, and guess what... there are a few small companies filling that niche... BillF, Alex, maybe Duke, etc. That's the way it works. And if the niche becomes big.. then you will see more of it!

 

[rpsands]

There's 1% who know what it is maybe; but probably 75% could understand it and understand specs given a little education and a drive by manufacturers to provide meaningful specs, and information about what they mean.

Most of this stuff is not that complex and can be distilled to plain English pretty easily.

There used to be a really good explanation of what each frequency band sounds like in plain english (e.g. subbass, bass, upper bass, lower mids, mids, upper mids, treble, high treble)