AVS Forum banner
Status
Not open for further replies.

SVS B4 Subwoofer Product Review!!!

14K views 151 replies 34 participants last post by  javry 
#1 ·
#52 ·
Quote:
Originally posted by Vetinari
One of my projects this weekend is to add some of those plastic discs to the bottom to assist in sliding the sub across carpet. I'll let you know how it goes...


Brian
Hey Brian,

Please do let me know how this goes. A red subwoofer huh?

Javry
 
#53 ·

Originally posted by DaveN


Quote:
Originally posted by javry
I can always bring my K-2 with me but you'll have to fight to keep me away from the volume control. LOL


Dave
yeah....you and that darned volume control. U da monsta!!! Just don't forget the beer. Don't bring the K-2 though....cause if you do, I'll probably want one.... :(

Javry
 
#54 ·
I personally like the heli scene in the matrix when the choper hits the building.


daredevil organ smashing...movie sux though:)

pearl harbor fight scenes

I still like the THX intro with the world blowing to ****..always cool:)
 
#55 ·
TV-


Did anyone ever figure out what was up with the frequency response curve which is posted? The maximum output numbers cited are also more in line with a 90-92dB/1W/1m sensitivity. Were you able to duplicate these measurements? I am somewhat curious if there were any proximity issues in measuring at 1m off the side.
 
#56 ·
>>>TV-


Did anyone ever figure out what was up with the frequency response curve which is posted?115dBs...which I do agree with in many instances). That would imply what...sens. around 106 or so?



I wish he would have done more testing on it, but his methods and my methods are so different, no one should expect our data to absolutely parallel anyway. Different PCs, different mics, different measuring distances, different input signals(for FR for example) and probably a lot of different setting in the Soundtechnology software itself.


Tom V.

SVS
 
#57 ·
Quote:
Originally posted by john doran
Titan A.E. really is a nutty DVD for bass - it's the one i use to demonstrate my own B4+.

Man you aren't kidding. I just watched it for the first time. I don't have a B4+ but it sure as hell gave my 25-31+ a workout.
 
#59 ·
Quote:
Originally posted by goneten
I can't believe that no-one has mentioned MR(Reloaded) as a bass test disc.Staggering amounts of low bass.


Regards
well...I didn't mention it only because I don't have it yet. Also, I haven't seen it yet. I've been waiting for it to come out on DVD but didn't want to see it until I got the new subs into the system. On your recommendation though, It's now on the list.

Thanks

Javry
 
#60 ·
I have YET to hit max on my B4+...I've hit uncorrected 120+ with 3-4 db hot at 75 calibration on my Lex, I am endangering myself structurally going too much beyond that...I know damn well it'll do more...amp wasn't even soft clipping (seating about 10 feet now)...one day I'm gonna let it rip..if I break something, I'll send the bill to ya Vod...:) tom...This isn't a boomy, ****** bass..its a damn tight bass...spl's are good, but only when everything sounds great...so spl shootouts can get pretty ugly, and distorted.
 
#61 ·
Quote:
Originally posted by Threecard
I have YET to hit max on my B4+...I've hit uncorrected 120+ with 3-4 db hot at 75 calibration on my Lex,
on my CBII, I'm typically around a gain of 35. Sometimes I crank it up to 40 for the heavy bass scenes but I hardly ever go beyond that. When my B4s arrive, I'll probably let her rip for a little bit. But I'm really looking forward to what they'll do at ambient.

Javry
 
#62 ·
Hi all, saw something posted and thought I might be able to clarify a few concerns:

Quote:


I wish he would have done more testing on it, but his methods and my methods are so different, no one should expect our data to absolutely parallel anyway. Different PCs, different mics, different measuring distances, different input signals(for FR for example) and probably a lot of different setting in the Soundtechnology software itself.
If a test system is calibrated, then from a frequency response standpoint the type of mic, type of PC/soundcard/digitizer, software settings, and even test signal should not make a difference. I've done extensive sets of measurements in the past for FDA-approval of new Ultrasound systems, and have been "through the wringer" in terms of measurement types.


MLS (MLSSA, LAUD, many others), FM Slide/CHIRP (predominantly TEF), stepped sine (LMS), pink/white noise (RTAs), all will produce the same frequency response given the following:


- Data is properly truncated/windowed to eliminate effects of reflections

- System is properly calibrated.


I've spoken at length with Dr. Johnson in the past, and believe he is properly windowing the data, to eliminate reflections. A good examination of the impulse response shown at his site will show that the data is windowed correctly; there isn't any extra reflections in the data. As is shown in his review of the B4+, it's clean to 60 msec, which would indicate good data acquisition down to 16 Hz.


If there is a discrepancy between measurements, then both measurements should be called into question, until one or the other can be verified as accurate. Given that the only measurements posted have been from Dr. Johnson, and personally knowing how he captures and analyzes his data, I'd say his results are most likely quite accurate.


As far as MLS accuracy showing 12-13 dB of drop at 20 Hz, this is not an artifact of the measurements; rather, this is showing the true anechoic measurement of the system. In-room will of course have room gain, flattening the curve. Properly windowed, MLS is as accurate as any other test signal, from a frequency response standpoint. The difference is the environment where the measurements and the claimed performance are taken.


On another note:

Quote:


Did anyone ever figure out what was up with the frequency response curve which is posted? The maximum output numbers cited are also more in line with a 90-92dB/1W/1m sensitivity. Were you able to duplicate these measurements? I am somewhat curious if there were any proximity issues in measuring at 1m off the side.
My guess is that the environment being discussed is different. I suspect the B4+ may have the claimed sensitivity when in-room, where 8-12 dB of gain is expected down below 50 Hz. I know Mark is typically working in "largish" rooms (like 4,000 seat theaters), so that free-air sensitivity is truly anechoic. An anechoic efficiency of 90 dB @ 1W, 1m would fall in line with Dr. Johnson's findings of 122 dB SPL @ 1500W (~32 dB of power gain, meaning a 1W output of ~90 dB SPL).


I personally would be VERY surprised to see an efficiency of 100 dB SPL @ 1W, 1m anechoic for the B4 at any frequency, anechoic. If it does perform that way, then SVS has truly shattered the limitations of Hoffman's Iron Law, and should immediately patent when they are doing. An efficiency of 100 dB @ 1W, 1m is 6%, which is extremely high, especially in the bass range. Even large bass horns have a difficult time reaching that kind of sensitivity down in the sub 50 Hz range, free-air. For example, the LAB Subwoofer, a large (32 cubic foot - nearly 3X the size of the B4+) horn subwoofer drops below 100 dB efficiency at 100 Hz. At 30 Hz, still within the cutoff of the horn (and thus benefitting from the loading and directivity gains of the horn), the LAB Subwoofer is down to 95 dB @ 1W, 1m anechoic.


My guess is the difference in claimed efficiency is in-room versus anechoic. I know we've seen upwards of 95 dB SPL @ 1W, 1m efficiency for the Sadhara in our home theater room (2300 cubic feet), but anechoic measurements in our smaller 77,000 cubic foot warehouse (a bit more than 2100 cubic meters) show a true efficiency around 84 dB @ 1W, 1m.


Dan Wiggins

Adire Audio
 
#64 ·
Tom,


Then your problem is with Dr. Johnson's measurements, not MLS. MLS is most famously used in the MLSSA measurement system, which is used in a huge variety of applications by some of the biggest names in the industry. Properly used, MLS is as accurate as gated sine wave approaches, and in fact can be more accurate, since it is much more noise-immune than gates sine waves (as is FM Slide/CHIRP systems).


Personally, having done thousands of measurements, including FDA-life-critical measurements (for example, we designed and validated the entire audio system in the Siemens Antares ultrasound system - speakers, amps, signal processing, mics, etc), I've found wonderful correlation between MLS, gated sine, and CHIRP when properly used. I tend to use CHIRP because of the speed or the signal, but in high noise environments (particularly outside where wind can skew the bottom end) MLS with it's pseudorandom noise pattern is superior. Higher noise rejection. Personally I rarely use gated sine since it is very susceptible to noise.


Gating in a room can be stretched out to 1 msec per foot from the nearest boundary or significant object; in a larger room, it's possible to get 100 Hz. In a huge room/warehouse, or outside, you can get much further. Note that gated sinewaves can have the same reflection issue. if the reflection happens within the measurement window, then the measured data will be corrupt. Regardless of measurement technique, proper setting of the gate window is critical, and will skew the measurements accordingly.


The issue, therefore, is operation of the system, and not the measurement system itself. How was the gate set? If the gate is 60 msec out, as shown in the B4+ measurements, then the data should be clean down to 16 Hz. If the gate was shorter, then the lower end of the frequency spectrum would increase.


So I guess the question is what could be the issue with Dr. Johnson's measurements? Perhaps we should ask Dr. Johnson to join us and explain how he takes his measurements. If Soundtechnology is as difficult to set up and configure as you state, then the issue is with the particular tool used (not the measurement technique), and would mean that results found by Dr. Johnson should be discounted, for all tests so measured. Even "comparative" samples within the same test set should be circumspect, because the potential errors can't at this time be determined to be precision or accuracy.


As far as the sensitivity of the B4+ goes, the SPL measurements of Dr. Johnson (the only published review I can find for the B4+) shows 90-92 dB, and 100 dB in room. Just wanted to clarify. Also, be aware that the RS meter is inherently inaccurate above 116 dB SPL, and is limited to 126. I've seen it swing wildly low AND high at the upper ends of things. Personally I would not trust it at all for SPL levels above 110, but then again I think you and I have access to gear beyond the usual HT enthusiast.


Dan Wiggins

Adire Audio
 
#65 ·
Dan,


Let me try to make this as simple as possible...


I don't have a *problem* with the way John Johnson measures the FR. I just don't think it is going to parallel a 2m GP measurement. I personally feel 2m GP is the most accurate way of getting quasi-anechoic data. You may disagree...that's fine...I don't really know what all of your measuring preferences are. And to be honest, I really have no interest in them.


I have spent the better part of the last 5-6 summers(May thru Oct) doing nothing but measuring speakers and subwoofers outside...sometimes making hundreds of measurements a day in the process. You have the methods you are comfortable with...I have methods that I am comfortable with. You have equipment you are comfortable with...I have equipment that I am comfortable with. That I am put in a position to defend my measuring preferences by a competitor in a thread solely dedicated to a recent review of a SVS product is a little surprising. Maybe I am just envious of the free time others have...



I have already listed multiple examples of MLS not matching 2m GP data at all in DOCUMENTED cases (Keele) above. And I would like to see the look on Bruce Halls face if someone told him his 18" servo just measured 15dBs down at 20hz quasi-anechoic..:) The fact that these MLS measurements show the smallest PCi and the largest Velodyne having *about the same* FR should be a good indicator that they aren't the most accurate of data.


Tom V.

SVS
 
#66 ·
Quote:
Originally posted by TV
I have already listed multiple examples of MLS not matching 2m GP data at all in DOCUMENTED cases (Keele) above.
Tom,


What bothers me is that when properly calibrated with known conditions and suitable resolution, measurements should be highly repeatable. Outdoors on the ground is one condition that most anyone can easily replicate. While in-room measurements are very difficult to compare with those taken in different rooms, with the exception of temperature and wind, the great outdoors is pretty constant in acoustic loading of low frequency devices.


Point being, that outdoor ground plane measurements of suitable resolution with a calibrated system should be easily comparable. This is nothing new to the professional audio world, or even to designers in the residential/consumer world who are measurement savvy. I hope that reviewers follow the lead of Keith Yates and John Johnson in taking measurements which are in many ways easily compared and repeatable. If measurements are not accurate and repeatable, there's no point in taking them.

Quote:
Originally posted by TV
And I would like to see the look on Bruce Halls face if someone told him his 18" servo just measured 15dBs down at 20hz quasi-anechoic..:) The fact that these MLS measurements show the smallest PCi and the largest Velodyne having *about the same* FR should be a good indicator that they aren't the most accurate of data.
Two points need to be considered here. First, John Johnson does not list the level at which his frequency response measurement was taken. It is only a realtive level plot. I'm sure you have seen some of the measurements our friend Deon has taken of the Velodyne and Sunfire subwoofers, and the measurements Keith Yates published also show how frequency response is rarely constant with level. With the sophisticated limiting executed in the HGS subwoofers, I would be that at some output level, the frequency response IS 15dB down at 20Hz quasi anechoic. This would also explain the response of the SPL-8. This is not much different that what is seen with a Bag End sub where it is flat to 8-10Hz... at ~75dB.


Finally, there is nothing in John Johnson's measurements which indicate an included reflection or lack of resolution at the frequencies of interest. This is why I had inquired as to how this response compares to your measurements. Citing that users have reported very high levels of bass on program material measured with a Radio Shack SPL meter on C weighting only tells us the levels are high, not if they are high at low frequencies. The measurements being discussed are not "computer theory/estimation" but rather well understood measurement methods. Nothing in this is hypothetical modeling, but real world measurements. Measurements which somewhat contradict the "Anechoic" frequency limits quoted in the B4-Plus manual. As the measurements differed, I inquired how these measurements correllated to yours.
 
#67 ·
>>>Tom,


What bothers me is that when properly calibrated with known conditions and suitable resolution, measurements should be highly repeatable.Two points need to be considered here. First, John Johnson does not list the level at which his frequency response measurement was taken. It is only a realtive level plot. I'm sure you have seen some of the measurements our friend Deon has taken of the Velodyne and Sunfire subwoofers, and the measurements Keith Yates published also show how frequency response is rarely constant with level. With the sophisticated limiting executed in the HGS subwoofers, I would be that at some output level, the frequency response IS 15dB down at 20Hz quasi anechoic.
 
#68 ·
I could give a rats ass about the measurements..my ear is the one that counts...If I'm happy and with the way SVS backs their products, I'll be svs for life. Johnson must have thought the B4+ was worth testing....was it a mistake...I doubt it...get in line.
 
#69 ·
Quote:
Originally posted by javry
Hey Brian,

Please do let me know how this goes. A red subwoofer huh?

Javry
Didn't have a chance to do much this weekend other than subjectively compare scenes from The Matrix Reloaded with and without one port plug (20hz). Having one plug definitely enhanced the bass you could "feel" in the freeway sequence and others and did not require too much upward adjustment in level with Avia. I will know more after I do my freq. sweeps in preparation for BFD programming.


BTW, what's wrong with a red subwoofer? I wonder if Tom could chime in and let us know what color (roughly) SVS ships the majority of B4's with?
 
#71 ·
Quote:
Originally posted by Vetinari
BTW, what's wrong with a red subwoofer? I wonder if Tom could chime in and let us know what color (roughly) SVS ships the majority of B4's with?
Now Brian.....don't go gettin your ti.......'s in a wringer! A red subwoofer is a very nice thing....sure to be loved and admired by all. Why, just the other day, I was strolling through the park....and whaddya know.....a cloud came floating by in the shape of a red subwoofer....just kiddin, of course. Actually, It sounded kind've cool. TV just sent me some samples to pick from and guess what......mercury red is on my list of potentials. Now, aren't you ashamed?.............don't answer that.


Now on to more serious matters. I noticed that 3 has his ports pointed in several different directions re: his photos. Does this matter? In other words, does the relative position of the ports [outward vs facing the wall, et al] impact sonic performance? How about the drivers? Maybe TV can weigh in here, as well.

Javry
 
#72 ·
Quote:
Originally posted by TV
Mark, I'm sorry you are *bothered*. In my experiences when you introduce a single variable into the experiment, all bets are off. In this case, we have different computers, different sound cards, different mics (I believe), different distances between the mic and the subwoofer, different orientation between the subwoofer and the mic, different input signals, different generators of the input signal, probably different weather conditions (and if you don't think that will make a difference...try measuring the same subwoofer in 90 degrees in the middle of the afternoon and 50 degrees later in the evening) and almost assuredly a minimum of a half dozen differences in the settings of soundtechnology itself. The ONLY constant is both methods had the subwoofer *outside*...and we really don't even know how close each outside environment was. And thru all this...someone is *bothered* that the two methods aren't showing identical results? Please.
Then someone's measurements AREN'T CORRECT. I think that's what Mark is getting at. If they are different, then one (or both) are incorrect.

Quote:
MLS is a computer estimation of how a speaker MIGHT measure in a true anechoic chamber. Because it is *well understood* doesn't mean it is more accurate than any other method.
I'd suggest a quick read of http://www.purebits.com/mlsteo.html to understand that MLS is NOT a computer estimation. It allows you work in closed environment because it has the ability to gate echoes out (which are much more difficult to do with CW signals).


To summarize: MLS uses a pseudorandom noise pattern that has the wonderful property that when it is cross-correlated with itself shifted by one sample, you get back a perfect impulse. In other words, it's a way of creating a VERY high S/N impulse.


Now, take an impulse, and run it through a Fourier Transform. You get the frequency response over the time window input. Provide a 1 second impulse response, you get 1 Hz out as the lower limit.


And, because you cross-correlate (a true mathematical function, in that you get a unique output for any unique input, thus there is no "estimation" involved), you get time domain data to work with. Which makes gating out echoes a LOT easier.


There's no "estimation" of MLS at all; it's a wonderful, highly-noise-tolerant test signal. If it's an estimation, so is gated sine. So is a CHIRP. I can assure you that all 3 have their uses, but MLS and CHIRP are used in most high-end measurement systems (such as TEF, MLSSA, Audio Precision One and Two, etc). Gated sine as done by Linear X/LMS is rapidly being dropped because of it's susceptibility to noise, and inability to effectively deal with echo-filled returns.


There's several other signals and systems books I can direct you to, if you'd like to learn more about use of MLS and CHIRP in measurement systems. Both approaches are used heavily in some of the most advanced acoustics measurement systems on earth - ultrasound and SONAR. High resolution of extremely small signals in high noise environments are why these types of signals - and not CW (gated sine wave) pulses - are used.


MLS is at least as valid as CW, if not more so, for the audio professional. The freedom from noise, and the ability to easily identify echoes (readily apparent in the time domain) prior to frequency domain transforms is very beneficial in ANY situation (including true anechoic chambers, such as the one at Siemens Medical in Issaquah, WA). CHIRP has the same benefits over CW as MLS, but because of the shorter pulse length is slightly more susceptible to noise (but still quite a bit better than CW).


I'll ask Dr. Johnson to pop on by, and see if he can clarify his measurement technique.


Dan Wiggins

Adire Audio
 
#73 ·
at least as I understand MLS according to JJ's reveiw of the Ultra but I think it would be similar to the B4 or other non-front ported subs.

Below is a quote from the Ultra review and shows the problem as I see it.


"Since we don't have an anechoic chamber, we used MLS filters, which simulate anechoic response. I placed the microphone 8" from the subwoofer, and used white noise with MLS filters, and a 65 ms window, which means that everything which the microphone receives 65 ms after it starts receiving sound is eliminated from the measurement."


When he was performing other 8" microphone measurements he said

"The red line at the top is with the calibrated microphone placed 8" from the subwoofer output region (bottom of the subwoofer)." There's no reason for me to think the microphone was anywhere else for the MLS test than located at the subwoofer output region.


Herein lies the problem. The microphone is over 3 feet from the ports.


Perfoming the 8" test with my Ultra and RS meter (C-weighted slow), a 90 db reading at the sub output area yields a a 99- 100 db reading at the port area.


As you go higher in frequency, you will get a blended response between the ports and speaker. As you get to 50 Hz it is all speaker.


While the B4 is tuned naturally higher than the Ultra, I think the same logic applies. Assuming of course what I am saying is logically.


In a nutshell, a single microphone may not be enough when measuring output that close to a non-front ported speaker. I think you would need one 8 inches directly in front of the ports and another 8 inches by the speaker. With a sealed or front ported speaker I don't see an apparent problem with using a single microphone.


As you get further away, such as with ground plane measurements, the ports and speakers blend better, giving a more accurate indication of what the frequecny response is. With my Ultra at 3' and 20 hz, the output is still greater above the ports but it is only ~ 2 dbs instead of 9-10 dbs. At four feet therer is not much of a difference.
 
#74 ·
>>>Now on to more serious matters. I noticed that 3 has his ports pointed in several different directions re: his photos. Does this matter? In other words, does the relative position of the ports [outward vs facing the wall, et al] impact sonic performance? How about the drivers? Maybe TV can weigh in here, as well.

Javry
 
#75 ·
TRC,


I think you have found one of MANY potential details that have been ignored during this debate. As much as I like the little $549 25-31PCi...the thought that is going to have about the same anechoic FR as a Velodyne 18" servo is a stretch to the extreme. I covered some of these issues in general when I said *mic distances*. There is also the effect that driver orientation can have on the measurement. When I test subwoofers that have a driver/port/PR on different baffles...I orient the subwoofer so that the port/driver will be approx the same distance to my mic (which I keep at 2m). Just rotating the subwoofer 45 degrees can make a big difference in the measurement.


I have seen quasi anechoic data from at least a half dozen sources now...Don Keele, All the audio ideas guide reviews, my own, two close friends of mine I can't name(they work for large companies that probably wouldn't like them sharing data :), and a nice cross section of FR measurements in AUDIO magazine that weren't performed by Don Keele. There is also a ton of quasi-anechoic stuff in all the old SB magazines...and some data in the misc audio critic.


NEVER, not once has any of sources shown identical measurements when measuring the same subwoofer. I have gotten close to Keele's measurements...but never identical. And i have seen big differences between MLS and 2m GP in the past...this isn't anything new.


The myopic view of this would be to simply say everything must be "wrong" then. But that isn't the case. The data I measure is correct...it is correct in that it is how I CHOOSE to perform my measurements. This doesn't make everyone elses data inaccurate. Their data is just as accurate to THEM...as they are obviously not going to copy every piece of gear, and every measurement method someone else uses. All of Don Keele's data isn't suddenly "wrong" because someone measured something different using a COMPLETLY DIFFERENT METHOD.


Tom V.

SVS
 
Status
Not open for further replies.
You have insufficient privileges to reply here.
Top