Current Trends in multi thousand dollar speakers

Cdc: While i personally believe it takes "big power" to do bass correctly with most "audiophile approved" type speakers, this is only a very small percentage of the big picture. None the less, i put my money where my mouth is and have 1000+ wpc driving my dual subs in my main system, 1200 wpc driving my HT mains with 1600 watts driving just the subs in that system, a dedicated amp rated at 100 wpc driving the subs in my bedroom system, 800 wpc driving the speakers in my office system, etc...

As you mentioned, dynamic compression, rise and fall times, ringing, phase shift, frequency response linearity, loading characteristics ( Z aka "impedance" at resonance, etc... are all other important factors. Given that vented designs typically have poorer rise and fall times with increased ringing, and vents introduce multiple phase shifts rather then the one that sealed designs offer, and typically have peaks at resonance whereas low Q sealed and stuffed designs don't, and vented designs typically have a higher impedance, which translates into less amplifier control and ability to load power than a low Q sealed design does, etc... Obviously, you can see where i'm going with this without me continuing any further.

As to your question about "playing with the damping of vent", this can be done. This is much the approach of a stuffed Transmission line and a Vario-Vent or "Aperiodic" design. Both offer many of the benefits found in sealed designs and may even surpass sealed designs in some areas. These are both somewhat "tricky" to design though. Some of the old Dynaco speakers were "Vario-Vents" and produced very good bass output AND definition for their size. That is, so long as you kept the spl range within reason.

As far as the JMLab's speakers go, i thought that the they sounded "barrel-chested". I think that much of this could be solved by adding damping material into the cabinet. The reason that it sounds like a "barrel" is that you literally have a large open wooden chamber resonating away. By adding damping material, you reduce both the excitation of the cabinet itself and absorb some of the energy that would otherwise be bouncing around in the cabinet causing a "bass echo" or "ringing".

In terms of "tighter bass" sounding "lean", that is a common thought. That's why i've stated that people are just TOO used to "bass slop". When they hear what tightly controlled bass with minimal overshoot and ringing sounds like, the articulation and definition is readily apparent but they still can't get used to NOT hearing all of the overhang. Most equate "bass slop" with "bass weight", which isn't quite the same thing.

As to your specific questions, here goes:

1) As to your comment about Musical Fidelity gear, most that i've heard sounds noticeably lean. Haven't heard all of their models though, so can't answer that specific part of the equation.

As far as i'm concerned, you shoot for speed and neutrality throughout the entire system. If you try to go with any other approach, you're back to "complimentary colourations".

2) Properly designed speakers are designed to work with specific room boundary effects taken into account. If a manufacturer doesn't provide basic suggestions as to where the speakers should be placed, chances are, the speakers are going to sound VERY different from room to room. Since obtaining neutral response in a consistent manner should be the goal of most "audiophiles", speaker design and placement become key factors in what we hear and total system performance.

Other than that, most designs already suffer from bloated bass as it is. Adding further low frequency reinforcement by placing the speakers closer to a corner, rear wall or both, etc.. will only make the presentation sound thicker and slower. With that in mind, some high-tech speakers with a lot of research put into them are designed to be placed up against the wall. If you pull them out from the wall, you'll have too lean of a presentation.

3) It is not so much the roll-off that makes a speaker hard to set-up, it the "Q" of the peak at resonance. Q covers both the amplitude and bandwidth at resonance. Speakers with a lower peak at resonance will be easier to place because the bass sounds less "one notey". Speakers with a narrower bandwidth peak will also be easier to place because they have less chance of exciting the various multiple nodes spread over a wider frequency range that exist in every room.

What the shallower roll-off of a sealed design offers is greater usable output below resonance. Once a vent is done, it drops like a rock. Anything below the point of resonance on a vented system unloads ( minimizes damping ) on both the vent and the driver. Excursion of the driver increases and the bass becomes quite muddied due to the uncontrolled extension. Better vented designs, like that of the Merlin's, use some type of high pass filtering to minimize these problems.

As far as Thiel's go, the placement problems with them may be two-fold. First of all, they use passive radiators. Passive's are slower than ports in terms of transient response. Slower response sounds muddier, making it harder to find a position in the room where the "mud" is reduced AND nodes aren't excited. Couple this with treble response that may require slightly different speaker positioning to work optimally and you've got yourself a lot of work ahead of you finding a good compromise between the two.

This is why i said that properly designed speakers will take into account room reinforcement. The designer should be able to give you an idea of where the speaker will work best in your room. While every room is a little different and may require some trial and error, those suggestions should get you pretty much in the ballpark. After all, they are familiar with the bass peaks, roll-off rate and can factor in room reinforcement from there. On top of that, they should have taken into account these factors when designing the rest of the speaker too, so the mid / treble radiation pattern should be designed to compliment optimum bass reproduction.

As i've mentioned before, very few speakers take all of this into account. If they did, you would be seeing a lot more "oddly" shaped speakers with acoustic treatment on the baffles. Reducing diffraction of the baffle itself is one thing, but taking into account room interaction with that baffle is one step ahead of most designs. Sean
>

Isn't it utopia (no pun) to expect full-range response from a commercial passive, dynamic speaker??

Ultimately, CAN there be such a thing as a passive, dynamic, FULL-range speaker within physical & electrical & COMMERCIAL reason???
Let's look at rule of the thumb parametres of a "good" speaker -- as we've discussed it 'till now.

Take good sealed box with a Fs say of 60 Hz (not bad!). Say we equalise it (we're making happy assumptions here, and considering an anechoic room) to reach 30Hz (i.e., we need to equalise ~12db/octave). Already, that one octave will require an extra 16x as many W to reach the same spl as the 60 (as in W=P^2/rho).
Now, the original 60 was quite impressive anyway, and probably required some equalisation in the first place (say, baffle step, etc) so there is some extra energy used there too. Say a 6db/octave loss from 120Hz -- so we need 4x the watts from 120 to 60.

In other words, for 1W at 120Hz, we'll use ~65W for our 30Hz. Looks OK? Think about 2W at 120 requiring ~130W for the 30Hz. And so on...

The results in W availability look staggering. Hardly atractive in a commercial product...

Or are my numbers wrong (which they could be:))?. Or does in-room compensation save the day -- by how much??? +2-3db?? Still, that doesn't save us from having to opt for Sean's kWatt amplification -- and Sean's only using that for his BASS drivers... other amps(s) drive the rest of the speaker...

No wonder some people in the past used Onken for the low bass... No wonder, bi-amping is in the air!

The "thickening" in response when one tries to get more bass warmth by taking advantage of walls' reinforcement is more due to the automatic "revoicing" of the upper-bass/low-mids region due to the change from in frequency of the 2pi/4pi transition. MOST speakers sound muddy and congested when wall/floor mounted if they're designed to work in free space. It's interesting to note that the JM Labs and PSB surrounds seem to be their normal 2 way monitors stuffed into a different cabinet, without modification of the crossover, and hence these speakers sounded quite wrong when wall-mounted, yet very nice on stands. OTOH the Boston VR-MX surrounds, for example, sound very well balanced when wall-mounted, being voiced that way. Indeed some speakers do require some front wall support for proper bass response, but there's a pont where the boundary support frequency "lift" band goes up as the distance goes down, until when wall-mounted, the speaker usually sounds congested in the lower mids.
Sean, great post, but again, if a properly designed vented cabinet uses a nice low Fs, and the driver's happily within its excursion limits, what's wrong with a 24/dB drop below, let's say 30-40Hz? Of course "sealed" is easier to get right, but the total bass power response always requires a bigger cabinet.

Ernie, I'm piping in here:

what's wrong with a 24/dB drop below, let's say 30-40Hz (...) if a properly designed vented cabinet uses a nice low Fs, and the driver's happily within its excursion limits

It doesn't sound impossible -- BUT:
Problem seems, it's not very easy to do and COSTLY. The cabinet would be very large (and/) OR, the "clarity" of the bass reproduction, as in how many bass "notes" actually come across, could (still) suffer...
More to the point the COST of the appropriate driver(s) could easily skyrocket the retail price of the speaker. Example: an average Lambda woofer (they have good /reasonably priced woofers) costs $~280ea; say the manufacturer buys these at $150/pair --

So? We've just added $1500 to the retail price of the speaker for an improvement which, while laudable for US here, is UNlikely to:
a) move opinion-"leaders" (i.e. reviewers)
b) be appreciated & impress many potential buyers, when the speaker is auditioned within the confines of a dealer's (or the buyer's) average listening room and the response anomalies most rooms produce in the bass

So, it's not likely to skyrocket sales... which is what manufacturers need in the first place (I'm not saying that's a bad thing:))
Cheers!

Greg: Room gain is very difficult to calculate and will somewhat vary with size of the room and speaker placement within the room. This is why good engineers design a speaker for specific placement within a room. Bare in mind that i'm not just talking about bass loading characteristics either. That way, they can factor in specific low frequency reinforcement factors AND dispersion characeristics consistent with the placement that the speakers will be used in.

While many people don't take such things into consideration when making their purchases and can't figure out why their speakers don't sound good where they have them placed in their room, they may also be the same type of people that buy a 4 wheel drive SUV to drive around in the city and / or take long trips on the highway with. While one can argue "vehicles are a matter of personal preference", one can also say "mis-application of a product outside of its' intended use will deliver less than optimum results". One can use such a product and be happy with it, but that doesn't mean that something else that was more suitable for the task at hand wouldn't have worked even better.

This is why i've stressed factoring in room size and placement in varous threads. Great speakers in the wrong room and / or great speakers that are improperly placed soon become "bad speakers" that they can't get rid of fast enough. Too many folks end up buying speakers that were designed to operate in a different environment than how they want to use them and then end up paying the price.

Ernie: I would suggest that you read Vance Dickason's Loudspeaker Design Cookbook. Your understanding of cabinet size and bass extension / output needs some help. I don't mean this to sound rude, as i was under many of the same false assumptions when i first started reading / learning about speaker design myself*. The fact that there are loudspeaker designers / manufacturers that don't fully understand many of the variables involved speaks of how complex of an issue this really is.

Part of the extension / total output / lower F3 on a vented system comes from the fact that, with such a high amplitude peak at resonance, the output, which is frequency related, has to fall off quite a bit before going below the average output level.

For sake of an example, let's assume we have a speaker that averages 88 dB's across the band. Due to the undamped oscillation at resonance of the driver / vent combo, there is a +5 dB peak in the low frequency region. As we've seen, this high of a figure on a vented design is not unrealistic and is possibly even conservative on some designs. This means that at the point of resonance, the output of the speaker is at 93 dB's. It is only after the lowest tuning frequency of the vent occurs that the speaker begins to roll off. That means that roll-off starts AT the peak.

Starting off with such a high peak means that we now have to lose 5 dB's of output just to get back to the reference output level of 88 dB's. In order to find our actual F3 ( -3dB point ) of this design, we've still got to lose yet another 3 dB's of output. This would give us a total of an -8 dB drop before we actually made it to the frequency where F3 is measured.

As you can see, since roll-off occurs at the same rate that frequency is lowered, building in a bigger peak at resonance automatically gives you more extension. That is, as it looks on paper using a -3dB reference point. Other than that, one can introduce such a peak into a sealed system with an EQ and still enjoy the better transient response of the sealed box. If you want to mention the phase shifts that the EQ brings with it as being detrimental, i'd like to mention the phase shifts that the vent brings with it.

As to your question about "what's wrong with a 24 dB slope", that answer has to do with linearity, transient response and ringing. The sharper the slope, the poorer the transient response, the more ringing and the less phase accuracy. This is why most "high end" audio gear strives for wider bandwidth with gradual roll-off. Better out of band performance equates to better & faster in-band performance. Same goes for speakers.

The question here is does one want more total output with a lower F3 at the expense of added bass bloat aka "lack of control", reduced linearity, poorer transient response, etc.. or is a slightly higher F3 with improved transient response and linearity more desirable? Since F3 only tells a small percentage of the picture, take a look at this. For sake of clarity, these figures are based on the accepted principles that vented systems fall at a rate of 24 dB's an octave and sealed designs fall at a rate of 12 dB's an octave:

-3 dB's on a 40 Hz vented system
-12 dB's at 30 Hz
-24 dB's at 20 Hz.

-3 dB's on a 50 Hz sealed system
-12 dB's at 25 Hz
-24 dB's at 12.5 Hz

While the initial -3 dB point on the vented design looks better on a spec sheet, it actually offers LESS extension when all is said and done. This is true even though the sealed design looks "weak" at an honest 50 Hz rating. On top of that, the vented design has all of the other problems i.e. reduced transient response, increased ringing, more phase errors, out of phase leakage from the port into the upper bass / lower midrange region, etc... You have to deal with all of these sonic drawbacks as tis type of design tries to get you the limited extension that if does offer.

This is not to mention that the vented system is completely "unloaded" or "undamped" below 40 Hz. Any low frequency notes fed into the speaker below that point will only result in massive excursion with the potential for damage / increased distortion. If you've got vented speakers and play vinyl, DO NOT try to play massively warped records at high volume. This will cause MASSIVE "woofer flutter", possibly resulting in permanent damage.

As a side note, it is quite possible to design a bass hump / resonant peak into a sealed design. By increasing the Q, you get such a peak. Like the vent, the transient response is reduced and ringing is increased, but unlike the vent, the slope remains shallower and phase integrity is not damaged quite as badly. As such, it is possible to match the F3 of a vented design with a sealed design, but the benefits of improved transient response are partially negated.

Vented designs DO look "more impressive" on paper, but that is only because the spec's that we use to quantify the performance of speakers aren't quite as sophisticated as those that we use for electronics. On top of that, there is no such thing as electronics / room interphase like there is with speaker / room interphase.

Vents vs sealed boil down to quantity vs quality. Vents will almost always give you more total output, but at great expense. Sean
>

*To be honest, i was under the impression that vented systems allowed smaller cabinet size and greater extension for many, many years. I could never understand why someone would want to buy a bigger cabinet that didn't play as low and took more power to operate. That's because i was reading spec's but didn't know how to interpret them, nor had i actually listened to a lot of sealed designs at that point in time.

The reason why the sealed cabinets were "bigger" was because they were properly designed and tuned for extension. The vented cabinets were tuned for more apparent bass and higher output i.e. the illusion of "deeper" bass via "more" bass. This is how most "Pro" speakers are also designed and you can see further evidence of this by reading Bill Fitzmaurice's articles in Audio Xpress.