B&C 21DS115

[Seek]

I know it's a touchy subject here but....

 

Sound quality?

I don't consider myself an expert on a par with an audiophile magazine reviewer or any such, but I spent some years gigging, some time behind the mixer, and did mobile DJ as well. So for whatever my subjective take on SQ might be worth, the DS has exceeded my expectations and even my hopes. It has been pushed hard, too, and whether sotto voce or firewalled it is solid and articulate. I have not regretted the purchase for a second.

[Infrasonic]

Is it a touchy subject?

Don't you like to touch... subjects?

 

Sound quality = subjective and that is frowned upon by some.

[lukeamdman]

I'm offended.  

[chrapladm]

I would love to compare the BMS 18 monster with the FP 18XL1800 or 18DS115. BUT for me the SW115 in 18 is the same price as the DS. The savings is when you buy the 21 version verse the SW152. Probably never be able to compare these drivers side by side. Thats why I generally buy BC for value and I am happy with them so far. Just would be nice to actually hear if these others ARE better or not.

[Ricci]

 

Sound quality = subjective and that is frowned upon by some.

 

 

It is subjective for sure. I have some opinions on it that I've slowly developed but even those are tainted with the measurement data. I spend a lot of time trying to look at the measurements and figure out why I think something sounds better. Whether that holds true for anyone else is debatable.

 

If you are asking whether the BMS sounds better than the DS series...I can't say. I haven't heard them head to head in the same type of cab with the same content in the same room. I'd expect them to sound more alike than different assuming the final frequency response is close.

[Ricci]

I would love to compare the BMS 18 monster with the FP 18XL1800 or 18DS115. BUT for me the SW115 in 18 is the same price as the DS. The savings is when you buy the 21 version verse the SW152. Probably never be able to compare these drivers side by side. Thats why I generally buy BC for value and I am happy with them so far. Just would be nice to actually hear if these others ARE better or not.

 

Off topic but what are you using these days? Seems like you always talk about building or trying different things and I just realized I don't have a clue what you actually use?

[Infrasonic]

It is subjective for sure. I have some opinions on it that I've slowly developed but even those are tainted with the measurement data. I spend a lot of time trying to look at the measurements and figure out why I think something sounds better. Whether that holds true for anyone else is debatable.

 

If you are asking whether the BMS sounds better than the DS series...I can't say. I haven't heard them head to head in the same type of cab with the same content in the same room. I'd expect them to sound more alike than different assuming the final frequency response is close.

 

Bolded = that is exactly the sort of thing I'd like to talk about here.

 

But, yes, on the 2nd part. Was wondering if you feel there is a notable SQ difference from all these top-tier bass drivers. It's something worth noting.

[mojave]

Sound Quality = (BL²/Res) / (LE/Res) 

1            Acoustic Elegance, TD18H+-4A Apollo
2            Acoustic Elegance, TD15M-4A Apollo
3            Rockford Fosgate, T3S2-19
4            Aurasound, NS18-994-A
5            B&C Speakers, 21-IPAL
6            BMS, 18N862 8ohm
7            B&C Speakers, 21SW152-4
8            18Sound, 21LW1400
9            JBL, 2242HPL
10           Faital, 18HP1060
11           MTX Audio, 9515-44
12           Dayton Audio, UM18-22
13           B&C Speakers, 21DS115-4
14           Dayton Audio, RS-18HO
15           TC Sounds, Pro 5100
16           TC Sounds, LMS-Ultra 5400
17           FocusWorks Audio, GUJ21V1
18           TC Sounds, LMS-5400
19           FocusWorks Audio, GUJ18V1

[neo2015]

I've modeled the B&C 18" ds115 - 8, and unless i've made some error, it seems that it only needs very small box to achieve a 0.707 alignment in a sealed box.

That's a major benefit when comparing with the other similar drivers.

[SME]

Mmmm sound quality.  I definitely noticed a big jump when I moved to my UH-21v1 drivers.  As soon as they started making noise, my wife said they sounded a lot cleaner.  I wanted to do a lot more testing and optimization before I reached any judgment myself, but I'm quite convinced now that they sound a lot better than what they replaced, even at very quiet playback levels.

 

The two main characteristics I noticed were a cleaner "darker" sounding tone.  Like if you imagine bass as having a "dark" sound, the sound from these subs is a much deeper black.  I suspect this may have to do with harmonic distortion levels.  The other thing I noticed was a significant improvement in intelligibility and detail retrieval.  I think this may have most to do with transient response distortion.  One other thing I noticed that surprises me (but could be placebo effect, I guess) is improvement to the sound quality of the low mids.  It's as though the noise floor there is somehow lower.

 

My hypothesis now-an-days is that most drivers behave non-linearly even at low playback levels, and this impacts accurate rendition of transient content.  This could be caused by some combination of inductance and/or flux modulation effects and possibly contributions from the suspension.  My thinking is that these effects can account for the perception that some woofers are "faster" than others.  Even though in a linear sense woofer speed depends only on upper end extension, the non-linear effects can impact "woofer speed" independent of the sine sweep response.

 

In general, transient response distortion would not be revealed at all by sine sweep measurements, so it's not really possible to judge the transient response linearity of drivers directly using the measurement data posted here.  However, some of the information can offer some hints.  Low nominal normalized inductance (Le/Re) is a strong positive sign, but even if Le is higher, it may still sound good if it's sufficiently linear.  This might be visible in the impedance curve, but I'm not exactly sure what to look for.

 

It would be interesting to try to devise some kind of measurement method to assess transient response distortion directly.  The trouble is deciding what kinds of test signals are most representative.  When it comes to linear response, a sine sweep measurement tells you pretty

much everything, but it tells you almost nothing about non-linear aspects of response.

[chrapladm]

Off topic but what are you using these days? Seems like you always talk about building or trying different things and I just realized I don't have a clue what you actually use?

I have a pair of 21SW152's ready to go in a Skhorn. Also in a few months am looking at possibly ordering a pair of 21DS115's for indoor subs.( Just using them in ported 14hz cabinets.) If plywood wasnt so expensive here I would have built a LOT of subs by now. BUT its around 70 or more dollars minimum for a sheet. SO I dont do much but read and then finalise what I am going to build. Have some TH50 clones that can use a few drivers also but I am thinking that the Skhorn will be more than enough for now so those will be put to the side.

 

Also the first thing I will be buying in the next month or so will be a AV700 from Denon. Or whatever its called. Once I get that I can start building my other speakers. I still need amps also. Its a very slow road for me as funds are projects are usually nonexistent.  So I tend to talk a lot and build very slowly. I do feel bad at times for taking so long to do build some of my wants.

 

Did some google searches and found my old posts about me and some of my sub builds. Years back and I am still almost starting my build.  Hopefully by Xmas this year I will at least have a 2.1 setup outdoors finished and a 4.2 indoors.

[Infrasonic]

 

Sound Quality = (BL²/Res) / (LE/Res) 

 

Nice list and I tend to agree with you on this one!

 

But... there has got to be more to it than these numbers. "Quality" is not just a fixed number of electro-mechanical properties.

 

Simple build quality and performance of moving parts has an effect on SQ, wouldn't we agree?

 

A TD18h+ sound great on paper but what if it had a loud suspension and made tons of vent noise (doesn't have a pole vent but...anyway)?

 

 

So... I agree but you know what I mean too. There has got to be more to it than that list alone.

[Ricci]

 

Sound Quality = (BL²/Res) / (LE/Res) 

1            Acoustic Elegance, TD18H+-4A Apollo
2            Acoustic Elegance, TD15M-4A Apollo
3            Rockford Fosgate, T3S2-19
4            Aurasound, NS18-994-A
5            B&C Speakers, 21-IPAL
6            BMS, 18N862 8ohm
7            B&C Speakers, 21SW152-4
8            18Sound, 21LW1400
9            JBL, 2242HPL
10           Faital, 18HP1060
11           MTX Audio, 9515-44
12           Dayton Audio, UM18-22
13           B&C Speakers, 21DS115-4
14           Dayton Audio, RS-18HO
15           TC Sounds, Pro 5100
16           TC Sounds, LMS-Ultra 5400
17           FocusWorks Audio, GUJ21V1
18           TC Sounds, LMS-5400
19           FocusWorks Audio, GUJ18V1

 

 

That's one way to look at specs and make a judgment.

 

The missing critical factor is displacement and power handling. All of my experience has taught me that systems with greater headroom sound better. IMHO it is one of the biggest factors to consider and this goes against the old-school audiophile or "SQ" wisdom that loud & big = Not as accurate . Some of the drivers high up on the list such as the JBL 2242, AE TD15M or Faital 18HP1060 have limited xmax for making bass and lower headroom unless using many drivers. They are better woofers or mid-woofers than they are <100Hz devices.

 

In an ideal world the signal coming out of the subwoofer and delivered to your ears / body would be exactly what is being sent into it electrically or alternatively exactly what you want that to be that fits your listening tastes. There are so many ways that it can be distorted. It can be smeared in time through room related nulls and resonances, system group delay, ringing, resonances etc. The frequency response can be skewed in a number of ways. By the room, the system design, EQ, signal shaping, etc. There is harmonic distortion. There are non harmonic distortions and noises: Air noise through the motor, port noise, cabinet buzzes or rattles, driver related taps or ticks or other mechanical sounds, objects buzzing or rattling in the environment etc. These types of noises from the driver or cabinet may not show up in a frequency response or distortion measurement. Also the signal delivered to you not only needs to match the shape of the electrical signal but also the amplitude. Dynamic distortions occur too. Amplifier clipping, thermal compression, loss of driver linearity due to high excursion, rounding off or squashing the peaks of the signal, port compression, etc. Dynamic distortions are a big deal to me.

 

Invariably drivers more accurately reproduce the signal and behave better when driven at 10% than they do 50%. By the time they are being driven at 75% (Let's call this xmax and starting to get some heat buildup in the coil) they have degraded significantly. The last 25% to the limits of the driver (xmech, extremely non linear behavior at the extremes and significant thermal effects due to coil heating) result in exponentially worse reproduction. The closer the driver is to being pushed towards it's full operational envelope the less accurate it will be. You will see increases in virtually ALL of the types of distortion above and more. Having more total headroom available will make improvements in all of the above. If you need to EQ the system to reach the desired FR you need the headroom to comfortably do that. You need the headroom to maintain that response shape, limit thermal shifting and track the dynamics of the signal at any playback level you might use. IMO having an extremely powerful system is item #1 towards good sound. I'm also of the opinion that the various types of distortions, added all together, become audible as sound signature differences between systems, earlier than would be expected from just looking at one aspect or the other alone.

 

Item #2 is getting the room/space and system to work well together. Outdoors is one thing, but in a building...The amount of distortion imparted on the sound wave by the time it hits your ears, cannot be overstated. Once you have plenty of headroom to easily cover your listening needs and avoid dynamic tracking, thermal compression, harmonic distortion and operational noise issues...The next big hurdle will be the room acoustics.

 

On a related note to the discussion about those minor intangible things not captured in a sine wave sweep measurement...The program I use for the burst tests captures both the drive signal and the captured acoustic signal from the speaker. Ideally they would look identical. Of course they never do. Some look downright atrociously distorted while still passing the distortion thresholds as a "clean" pass result. If I get time I'll post some examples.

[Ricci]

But... there has got to be more to it than these numbers. "Quality" is not just a fixed number of electro-mechanical properties.

 

Simple build quality and performance of moving parts has an effect on SQ, wouldn't we agree?

 

A TD18h+ sound great on paper but what if it had a loud suspension and made tons of vent noise (doesn't have a pole vent but...anyway)?

 

 

So... I agree but you know what I mean too. There has got to be more to it than that list alone.

 

Exactly...There are operational noises not captured in there. Overload characteristics and ruggedness. Also good old linear displacement. Not much good as a sub driver if it falls apart past 8mm excursion. Based on my goals I'd not want to be running it much past 5mm and that don't make much bass.

[shadyJ]

That's one way to look at specs and make a judgment.

 

The missing critical factor is displacement and power handling. All of my experience has taught me that systems with greater headroom sound better. IMHO it is one of the biggest factors to consider and this goes against the old-school audiophile or "SQ" wisdom that loud & big = Not as accurate . Some of the drivers high up on the list such as the JBL 2242, AE TD15M or Faital 18HP1060 have limited xmax for making bass and lower headroom unless using many drivers. They are better woofers or mid-woofers than they are <100Hz devices.

 

In an ideal world the signal coming out of the subwoofer and delivered to your ears / body would be exactly what is being sent into it electrically or alternatively exactly what you want that to be that fits your listening tastes. There are so many ways that it can be distorted. It can be smeared in time through room related nulls and resonances, system group delay, ringing, resonances etc. The frequency response can be skewed in a number of ways. By the room, the system design, EQ, signal shaping, etc. There is harmonic distortion. There are non harmonic distortions and noises: Air noise through the motor, port noise, cabinet buzzes or rattles, driver related taps or ticks or other mechanical sounds, objects buzzing or rattling in the environment etc. These types of noises from the driver or cabinet may not show up in a frequency response or distortion measurement. Also the signal delivered to you not only needs to match the shape of the electrical signal but also the amplitude. Dynamic distortions occur too. Amplifier clipping, thermal compression, loss of driver linearity due to high excursion, rounding off or squashing the peaks of the signal, port compression, etc. Dynamic distortions are a big deal to me.

 

Invariably drivers more accurately reproduce the signal and behave better when driven at 10% than they do 50%. By the time they are being driven at 75% (Let's call this xmax and starting to get some heat buildup in the coil) they have degraded significantly. The last 25% to the limits of the driver (xmech, extremely non linear behavior at the extremes and significant thermal effects due to coil heating) result in exponentially worse reproduction. The closer the driver is to being pushed towards it's full operational envelope the less accurate it will be. You will see increases in virtually ALL of the types of distortion above and more. Having more total headroom available will make improvements in all of the above. If you need to EQ the system to reach the desired FR you need the headroom to comfortably do that. You need the headroom to maintain that response shape, limit thermal shifting and track the dynamics of the signal at any playback level you might use. IMO having an extremely powerful system is item #1 towards good sound. I'm also of the opinion that the various types of distortions, added all together, become audible as sound signature differences between systems, earlier than would be expected from just looking at one aspect or the other alone.

 

Item #2 is getting the room/space and system to work well together. Outdoors is one thing, but in a building...The amount of distortion imparted on the sound wave by the time it hits your ears, cannot be overstated. Once you have plenty of headroom to easily cover your listening needs and avoid dynamic tracking, thermal compression, harmonic distortion and operational noise issues...The next big hurdle will be the room acoustics.

 

On a related note to the discussion about those minor intangible things not captured in a sine wave sweep measurement...The program I use for the burst tests captures both the drive signal and the captured acoustic signal from the speaker. Ideally they would look identical. Of course they never do. Some look downright atrociously distorted while still passing the distortion thresholds as a "clean" pass result. If I get time I'll post some examples.

Wait a minute, are you saying the subjective and ambiguous term 'sound quality' can't be absolutely defined by a single number?!

 

Heresy! 

 

[SME]

That's one way to look at specs and make a judgment.

 

The missing critical factor is displacement and power handling. All of my experience has taught me that systems with greater headroom sound better. IMHO it is one of the biggest factors to consider and this goes against the old-school audiophile or "SQ" wisdom that loud & big = Not as accurate . Some of the drivers high up on the list such as the JBL 2242, AE TD15M or Faital 18HP1060 have limited xmax for making bass and lower headroom unless using many drivers. They are better woofers or mid-woofers than they are <100Hz devices.

 

Obviously power handling and displacement are important factors when one is pushing the system toward its limits.  And from your description, using 50% of available travel for peaks is "pushing it".  I.e., maybe peaking at 25% is preferable.

 

But, what if one is not pushing things at all?  If operation is kept below 10% of peak excursion and 1% of power handling capability, presumably displacement capability and power handling are not important.  Are there differences in sound quality between drivers that are still apparent under these conditions?  Let's assume that the driver is otherwise "competently implemented" enough that there is no audible noise or buzzing under such conditions.

 

My impression is that such differences do exist and can be substantial, but I admit that it's hard to prove.

[shadyJ]

Obviously power handling and displacement are important factors when one is pushing the system toward its limits.  And from your description, using 50% of available travel for peaks is "pushing it".  I.e., maybe peaking at 25% is preferable.

 

But, what if one is not pushing things at all?  If operation is kept below 10% of peak excursion and 1% of power handling capability, presumably displacement capability and power handling are not important.  Are there differences in sound quality between drivers that are still apparent under these conditions?  Let's assume that the driver is otherwise "competently implemented" enough that there is no audible noise or buzzing under such conditions.

 

My impression is that such differences do exist and can be substantial, but I admit that it's hard to prove.

Look at the subject from the perspective of sound pressure waves. As long as the driver is turning the voltage signal into sound pressure waves at a 1:1 relationship, it is a perfect reproducer. Anything else is distortion, whether linear or nonlinear. If you have no linear distortion and you have no nonlinear distortion, you have perfect accuracy. Of course, that is the unattainable ideal, but a very good system can get close. With effort, you can get a very flat response that has insignificant distortion quantities. What difference does it make if that is achieved through a $5k super driver or a $30 buyout driver from parts express? A good driver, when it is operating at nominal levels, can have distortion quantities of 1% or less and so is inaudible. If the frequency response is flat, and it isn't making any mechanical noise like flutter noise, there is nothing else that can distinguish these units, so long as they have the same dynamic range. 

[Ricci]

Obviously power handling and displacement are important factors when one is pushing the system toward its limits.  And from your description, using 50% of available travel for peaks is "pushing it".  I.e., maybe peaking at 25% is preferable.

 

But, what if one is not pushing things at all?  If operation is kept below 10% of peak excursion and 1% of power handling capability, presumably displacement capability and power handling are not important.  Are there differences in sound quality between drivers that are still apparent under these conditions?  Let's assume that the driver is otherwise "competently implemented" enough that there is no audible noise or buzzing under such conditions.

 

My impression is that such differences do exist and can be substantial, but I admit that it's hard to prove.

 

 

Opinions vary on this.

 

Firstly are we talking a comparison between drivers or systems?

 

If differing systems are being compared, I agree there can be audible differences even when both are loafing at very low levels. For example comparing a small sealed cab to a huge tapped horn. It would be very difficult to get the 2 to match the same response shape and timing characteristics exactly. Even just the cab shape / size / difference in radiation points could be a factor.

 

Comparing just bass drivers would correctly be done with both in exactly the same cab, same environment, same electronics, same content, etc. I would even say that the nominal driver size should be the same. IOW 15 vs 15 or 18 vs 18 if we are getting really particular about it. If the cabs are different and the frequency response is unshaped and differs between the two, there could easily be an audible difference even at very low playback levels. But...If the only thing changed is the driver like outlined above AND the frequency response is equalized to be as close as possible, including say an octave above the low pass filter, in my estimation you would have an extremely difficult time correctly identifying which was which in a blind comparison. Provided both are operating well below what they are capable of. Notice that I said blind...Visual queue's can have real impact on what we hear and our enjoyment.

 

All that said we are talking about bass and bass takes a lot of displacement to reproduce. People also typically push bass harder than the rest of the frequency range. Sure when you are listening to some light music in the background and the peaks are 95dB you might be using 5% of the system and all is well. Sound and SPL is logarithmic and by the time you get to peaks requiring 115-120dB at the seats the subs are going to be well into the point where the sound has started degrading whether you can hear it or not. I guess what I'm saying is subwoofers are called to operate much closer to their limits more often than not so they are often well into the point where some of these distortions are starting to add up. Bass is addicting. No one I know has 4 18's to listen to elevator folk at 80dB all the time.

 

That's why people claim that more expensive more powerful drivers sound better. IMO they do. The cheaper driver starts to lose it earlier or in some case VERY early or perhaps never even has it to begin with and starts to degrade or change the sound. The more expensive / powerful driver is still at the loafing stage while the cheap one is hitting a wall. Adding more cheap drivers, for the extra headroom to match the single big money driver, could result in sound that's just as good or better in many cases. If you've got a ton of room and not much cash go with a ton of cost effective drivers. If you've got very little space but a fat wallet you will have to pay to get the most you can from the limited space. Some people have  a good amount of space and a healthy budget, then there are the poor souls who have no space available and not much money. It's a sliding scale from both ends of the spectrum.

[SME]

Look at the subject from the perspective of sound pressure waves. As long as the driver is turning the voltage signal into sound pressure waves at a 1:1 relationship, it is a perfect reproducer. Anything else is distortion, whether linear or nonlinear. If you have no linear distortion and you have no nonlinear distortion, you have perfect accuracy. Of course, that is the unattainable ideal, but a very good system can get close. With effort, you can get a very flat response that has insignificant distortion quantities. What difference does it make if that is achieved through a $5k super driver or a $30 buyout driver from parts express? A good driver, when it is operating at nominal levels, can have distortion quantities of 1% or less and so is inaudible. If the frequency response is flat, and it isn't making any mechanical noise like flutter noise, there is nothing else that can distinguish these units, so long as they have the same dynamic range. 

 

The reason that I distinguish between linear and non-linear is thus.  If all the distortion is linear (and noise free) and you make frequency response flat you will recreate a perfect sound.  However, that is not true for non-linear distortion.

 

The behavior of a linear system can be fully characterized by its impulse response or frequency (and phase) response.  Impulse response and frequency response are mathematically interchangeable.  Using this information, one can predict exactly what the system will do in response to any kind of signal.  These are the principles upon which measurement programs like REW work.  They are also the principles upon which most DSP including EQ are based.  All this stuff assumes linearity of the system in a variety of places, and it mostly works out.

 

But the ugly reality is that transducers *do* distort non-linearly.  They definitely distort non-linearly when pushed to their limits.  What's perhaps more debatable is whether the non-linear distortion in transducers is significant enough to have audible effects at very low levels.  And if so, whether some of these distortion mechanisms can be identified and characterized using sine sweep THD measurements.

 

As long as we are speaking about non-linear distortion we cannot be assured that sine sweep THD measurements are sufficient to predict distortion levels with complex signals.  The reality is that, without at least a working understanding of the underlying mechanism of distortion in the driver (i.e., BL vs. stroke or Le vs. current or whatever) we really can't say at all how non-linear distortion will alter a complex signal like music or sound effects.  The measured frequency response is largely useless.  In fact, if we could somehow measure it in a single instant without interfering with the operation of the driver, we would see frequency response changing constantly, depending on the signal going to the driver.  It may even be possible for a driver to behave chaotically, such as may be the case with cone break-up.  The behavior of a chaotic system cannot be predicted even if we do fully understand the underlying mechanisms.

 

So the important question in my mind is whether some kinds of non-linear distortion can have audible consequences even at very low signal levels.  If so, then that would suggest that you may not be able to match the sound quality of the hypothetical Uber driver no matter how many $30 buyouts you used.  Obviously *some* $30 drivers will have mechanical design problems and other issues that make them noisy when played at *any* level.  I think we can all agree on that much.  But of course, these kinds of problems will present themselves obviously with minimal free air testing.  I'm talking about non-linear distortion that may not be readily revealed using sine sweeps or sine wave test tones, but which might be a lot more obvious when e.g. kick drum or electric bass is being played, even at lower levels.

[shadyJ]

Ilkka Rissanen and Ed Mullen did a study of IMD in subwoofers and found that subs with low THD also have low IMD. You can read about it here, if you don't already know about it. It is not a surprising result really. 

[SME]

Opinions vary on this.

 

Indeed they do.  You raise some interesting points.

 

Firstly are we talking a comparison between drivers or systems?

 

Good question.  I am interested in driver differences, even though those differences are likely to manifest differently (potentially better or worse), depending on the cabinet design.  I find it more helpful to distinguish between linear and non-linear characteristics.

 

If differing systems are being compared, I agree there can be audible differences even when both are loafing at very low levels. For example comparing a small sealed cab to a huge tapped horn. It would be very difficult to get the 2 to match the same response shape and timing characteristics exactly. Even just the cab shape / size / difference in radiation points could be a factor.

 

They do have different bandwidths, in a linear sense.  However, you're almost certain to cross them over to something else on the upper end, and this will completely change the linear bandwidth of the system.  Ideally, the response in the transition will be smooth with minimal group delay.  A more band-limited tapped horn cabinet may be harder to blend than a sealed cabinet if the crossover falls near the edge of its bandwidth.  But in many circumstances, the room and boundary effects dominate and the blend must be hand-tuned for best results anyway.

 

With that said, linear bandwidth can have a substantial effect on non-linear distortion, depending on where the distortion enters the chain.  It makes intuitive sense to me that the limited bandwidth of a tapped horn may help to suppress higher frequency distortion components that may actually be very audible at extremely low levels due to a combination of Fletcher-Munson effect and reduced masking.

 

Comparing just bass drivers would correctly be done with both in exactly the same cab, same environment, same electronics, same content, etc. I would even say that the nominal driver size should be the same. IOW 15 vs 15 or 18 vs 18 if we are getting really particular about it. If the cabs are different and the frequency response is unshaped and differs between the two, there could easily be an audible difference even at very low playback levels. But...If the only thing changed is the driver like outlined above AND the frequency response is equalized to be as close as possible, including say an octave above the low pass filter, in my estimation you would have an extremely difficult time correctly identifying which was which in a blind comparison. Provided both are operating well below what they are capable of. Notice that I said blind...Visual queue's can have real impact on what we hear and our enjoyment.

 

Yeah.  Obviously proving audible differences is very hard.  And without a complete model of human perception, there's really no way to prove that differences are audible without testing real listeners in a way similar to what you describe.

 

All that said we are talking about bass and bass takes a lot of displacement to reproduce. People also typically push bass harder than the rest of the frequency range. Sure when you are listening to some light music in the background and the peaks are 95dB you might be using 5% of the system and all is well. Sound and SPL is logarithmic and by the time you get to peaks requiring 115-120dB at the seats the subs are going to be well into the point where the sound has started degrading whether you can hear it or not. I guess what I'm saying is subwoofers are called to operate much closer to their limits more often than not so they are often well into the point where some of these distortions are starting to add up. Bass is addicting. No one I know has 4 18's to listen to elevator folk at 80dB all the time.

 

What you say here is true for most people most of the time.  I may be a bit of an odd-ball with 4-21s who does the majority of his listening at or under 80 dB.  I mean, I do have a wife that I talk to quite a bit, and the system fills the whole house with sound.  I also listen a lot when she's in bed.  I will have to check out this "folk elevator" music you speak of though. 

 

What you call "light music" with 95 dB peaks could be a rather spirited level for typical loudness war stuff.  More dynamic recordings can peak in the 100s and break 110.  The 115-120 dB range (and higher) is pretty much limited to movies for me.  Even at 115 dB, my subs are at about 10% if it's at or above 20 Hz.  So my subs are practically idle for essentially *all* of my music listening.  In fact, it's rather funny just how loud I have to turn most things up just to get the amp "signal present" lights to blink.

 

Anyway, I don't mean to argue that expensive drivers have better sound quality than cheaper ones.  Price need not have much to do with it.  I mean, how much cost do shorting rings add to a particular design, on a unit-to-unit basis?  I assume it's a moderately small amount in most cases.  Like, you probably won't see shorting rings on a $30 driver, but there's a good chance you will on a $150 driver.  At the same time, the quality of the shorting-ring implementation is very important and probably has more to do with the skill of the engineer who designed it than the retail price.

[Ricci]

I would love to participate in a blind test group. Whether it be speakers, subs or even amplifiers. I think it would be a fun way to find out how good my listening skills really are or aren't. Something like a large AT screen setup to hide the equipment in a quiet outdoor setting would be great to eliminate room acoustics. Have a test group take notes and apply ratings on a 1-10 scale in a variety of subjective categories. Take measurements of each setup to match up with the listener notes and scores afterwards. It would be great if someone did a serious GTG just for this purpose.

 

 

What you call "light music" with 95 dB peaks could be a rather spirited level for typical loudness war stuff.  More dynamic recordings can peak in the 100s and break 110.  The 115-120 dB range (and higher) is pretty much limited to movies for me.  Even at 115 dB, my subs are at about 10% if it's at or above 20 Hz.  So my subs are practically idle for essentially *all* of my music listening.  In fact, it's rather funny just how loud I have to turn most things up just to get the amp "signal present" lights to blink.

 

 

I'm in the same boat. Sometimes I watch movies near REF with the subs 6dB hot just for fun or demoing for my friends and the bass is punishing, making me curious of how close to the limits I am. Then I look at my signal lights on my K20's and realize they are barely lighting the -15 light and the 19's might as well be taking a nap. Same deal with the ghetto-rich warehouse stereo with the MAUL's. I just now got the money together and got the drivers for the 2nd cab. I've been running one on a old 4Kw rated amp and it's already more than we use unless I'm being stupid to show off with bass tracks in which case the old Crest amp just waves the white flag. Once I get the second cab in there and the big amps that'll add another 12dB of headroom that I need because...Winning???

At least the output should be clean.

 

Scott,

You have to be in the same situation right? Are your signal lights even on most of the time? I can't see how they would be with the 77 JBL woofers or whatever madness you have going on.

[Infrasonic]

The JBL-based mains I believe have never flinched once at anything. Bad sounding recording have come in and gone out equally bad sounding but... that's not the same thing.

 

Yeah... I barely tickle those things.

 

The sub system is a bit different. I do a moderate amount of signal shaping and I like spirited levels, when I can.

 

Uhh...honestly, I still can't tell if it's just my house making awful noises or the sub system. Occasionally I'll hear a bad sound like suspension bottoming or something. I am WAY underpowered actually so I sort of doubt it. I think it's the house. Or maybe amp clipping? The lights never come on, ever. Gosh, even the signal lights barely light up. At least at "normal" levels.

 

That being said, there is still tons of headroom left over (especially >30hz). As far as I can tell, anyway.

[shadyJ]

I would love to participate in a blind test group. Whether it be speakers, subs or even amplifiers. I think it would be a fun way to find out how good my listening skills really are or aren't. Something like a large AT screen setup to hide the equipment in a quiet outdoor setting would be great to eliminate room acoustics. Have a test group take notes and apply ratings on a 1-10 scale in a variety of subjective categories. Take measurements of each setup to match up with the listener notes and scores afterwards. It would be great if someone did a serious GTG just for this purpose.

This has been something I have always wanted to do as well. A real qualitative subwoofer comparison would have to be done outside (or in a room as large as an indoor stadium). Also, as you say, the equipment would have to be hidden. That is the only real way to fairly compare subwoofers, at least with human ears. They would have to be compared using fast A/B switching between the subs as well, at least for complex signal content like regular movie or music content. The problem is, GTG's are all about having fun, not rigorous testing. I wouldn't want to ruin a GTG by insisting on rigorous adherence to a strict testing regimen. GTG's are usually about drinking a lot of beers and seeing how loud some speakers get, and that is fine, but this kind of testing wouldn't work for that kind of gathering.

 

I would want to see if these subs can be distinguished when they are operating well within their linear ranges. SME seems to think that there would be a difference. I don't think there would, but such a test, if conducted appropriately, would certainly serve as good evidence one way or the other. Neither SME or myself would be able to properly participate, because we already have predispositions that a blind test could not overcome. Still, I would love to hear it for myself.

 

I need to secure some kind of grant to do this with. 

[SME]

This has been something I have always wanted to do as well. A real qualitative subwoofer comparison would have to be done outside (or in a room as large as an indoor stadium). Also, as you say, the equipment would have to be hidden. That is the only real way to fairly compare subwoofers, at least with human ears. They would have to be compared using fast A/B switching between the subs as well, at least for complex signal content like regular movie or music content. The problem is, GTG's are all about having fun, not rigorous testing. I wouldn't want to ruin a GTG by insisting on rigorous adherence to a strict testing regimen. GTG's are usually about drinking a lot of beers and seeing how loud some speakers get, and that is fine, but this kind of testing wouldn't work for that kind of gathering.

 

I would want to see if these subs can be distinguished when they are operating well within their linear ranges. SME seems to think that there would be a difference. I don't think there would, but such a test, if conducted appropriately, would certainly serve as good evidence one way or the other. Neither SME or myself would be able to properly participate, because we already have predispositions that a blind test could not overcome. Still, I would love to hear it for myself.

 

I need to secure some kind of grant to do this with. 

 

Well technically, it shouldn't be a problem at all for *me* to participate.  Even though I might be biased to assume there *are* differences, if I can't reliably distinguish the two subs being tested, then my bias has no bearing on the outcome.  Your bias on the other hand could be a problem in that you could choose to ignore audible differences and always choose which sounds "best" at random.  But in reality, I wouldn't worry about that.  I think your input in such a test would still be very valuable.

 

BTW, I've heard that Harman did blind testing and found a correlation between intermodulation distortion (IMD) and listener preference.  I believe most of their tests are done at "reasonable" listening levels, but it's not clear whether those levels were high enough to push the bass systems to distort beyond their baseline levels.