The astounding new AS-EQ1 SubEQ goes on sale!

[Sonnie]

And as I just mentioned in another thread... electronic equalization devices are generally going to be the best bang for the buck when it comes to frequencies under 80Hz. Multiple subs around the room will get you a long way towards better overall bass response in multiple listening positions. Equalizing it electronically will get you even closer to that perfection you are looking for. No doubt SVS has really got something special here... dealing with many of the drawbacks to previous electronic device solutions.

REW is like gold... I love it... I love tinkering with it because it can show you soooo much about your room AND your equipment. It is amazing how many products have poor response... and REW will tell you all about it.

SVS EQ, multiple subs, treatments and REW = priceless!

[spearmint]

cavchameleon wrote:

Richard,

GREAT write-up (and incredible room!!!). Thanks for the link to your review, I did not see that before.
~SNIP~

Thank you Ray,

I appreciate the kind words.

[jpk]

see next post, pics and text!
JPK

[jpk]

Sometimes it better to leave well enough alone? Now I opened a learning curve for REW. Actually I've been wanting to try it for a while.
SVS AS-EQ1 before doing anything;



SVS corrected;


Thought I'd join the crowd and try REW, although with RS digital meter for mic?


Now I'm not so happy? Would the CM-140 be better? ( I believe I need an adapter for the signal?) Should I buy a Berringer mic and pre?
Where do I go from here? REW forum?
I thought I'd add some graphs to this, I didn't expect what I consider a poor response displayed from REW.

Have I made a sophomore error somewhere?

Ideas, comments?
Thanks,
Joe

Next I'll learn how to get larger pics in the post!

[cavchameleon]

JPK,

REW does take a learning curve. The RS mic IMO is a terrible mic to use, it's only good for relative level matching. Also, if comparing REW to what the AS-EQ1 is doing with the freq, you need to do an average of the same mic positions for setting up Audyssey. Waterfall plots are also better indicative of what is happening in the room, not just a freq plot. Yes, the Behringer ECM8000 is a decent mic. There is a cal file here at the Shack, generic, but pretty good for home use. If you want absolutely accurate readings, you'll need a calibrated mic and also to calibrate you sound card. The ECM8000 does need phantom power an there are may choices (do a search here on the Shack as this question has come up a lot and there are some pretty good experts here to give you help). I use the Tascam US-144, but as I said, there are may others to choose from. Thanks for the work so far, once you pass the learning curve REW is an awesome tool!!!

The most important thing, how does you system sound with the AS-EQ1 on and off?

Ray

Ray

[jpk]

Ray,
I took a break and remembered to try moving the mic around, top pic RS meter straight up, middle pic, mic at 45 degree, AS-EQ1 on for both of those. Last pic, mic at 45, As-EQ1 is off. Impressive on minimizing the suck-outs. It does sound awesome , I'm just trying to reassure myself it's a wise investment. The SVS before curves could be like a "house curve"?
Joe
I do feel better now.

[brucek]

Now I'm not so happy? Would the CM-140 be better?

The RS meter is fine for home use as long as you stick to measuring below ~3KHz and use the cal file provided on our download site.

top pic RS meter straight up, middle pic, mic at 45 degree

The recommendation is oriented vertically with a forward angle of about 10 - 20 degrees.

The SVS before curves could be like a "house curve"?

Personally, I preferred your before curve with no EQ, although it's always hard to tell when the plots are smoothed.
As suggested already, it's more revealing to use waterfall plots combined with your response plots.

brucek

[jpk]

Thanks Bruce,
I’ve re-read the manual for REW and decided to do some more listening before any more tweaks and measurements. I also did some reading on the REW forum. I had a few minutes last night and my daughter wanted to hear some Beatles, so I played the Love songs DVD-A. There was definitely more bass with the EQ on. Sometimes almost too much bass? Maybe the recording over emphasizes it? (Subs cancelling each other w/o EQ?) I’ll try to get a with and w/o waterfall in the next day or two.
Joe

[cavchameleon]

Joe,

Glad to see you're on track with REW. It'll be good to see your waterfall plots. A LOT of older recording were equalized when recorded (emphasized bass and upper treble) due to the short comings of much of the equipment at the time, especially in the speakers area. Most recording studios track and mix down on high quality speakers, but also check the recording on 'cheap' speakers with limited bandwidth since the assumption is that most of the audience will have cheaper equipment (i.e. car audio, etc.). Not to say that all newer recordings are all that great either - way too much compression used IMO on a lot of them (it's amazing that we have >100db capable dynamic range on a red-book cd and a lot of studios compress the recordings to only have 5db-10db dynamic range).

Ray

[jpk]

Brucek , Ray, everyone,
Here are two waterfalls, the first with SVS off and the second w/SVS on. I’m still surprised how slight movements of the mic radically change the response. I’ll refrain from saying anything else until you see them. Of course, both of these were taken w/ mic in same position.
Joe

[brucek]

Interesting that the EQ didn't appear to do much for the decay problem at ~20Hz.

Even after a half second the signal is still in the audible range.

brucek

[Sonnie]

brucek wrote:

Interesting that the EQ didn't appear to do much for the decay problem at ~20Hz.

Even after a half second the signal is still in the audible range.

brucek

I thought that was what we called "rumble"... Oh wait... as we use to call it in our cars... that sub can sure "hold a note".

[jpk]

2ea x 15" sealed subs at 125 lbs ea holding a note? I know you were kidding Sonnie.

I was hoping someone would have looked at the waterfalls and said “Your problem is a,b and c, you need to do x,y&z to fix it!

I think a couple of things are going on here:

REW could be using a longer duration signal / sweep than SVS's (relative) ping?

I have a resonance problem in the Man Cave , foot thick concrete walls better not be resonating! There is a partition wall with cheap paneling that may contribute to the problem, but I’m thinking the ductwork and primarily the air return sheet metal is at least the major cause of things.
Has anyone experienced this or know where I can find information to verify and remedy it? Tapping on the air return seems to have a resonance.

Man Cave = 22 x 26 x 7 ft. w/drop ceiling, 0.5 inch rigid fiberglass ceiling panels. Padded carpeted floor, one side wall partitioned and paneled.

Have a great 4th of July weekend!

Joe

[brucek]

I have a resonance problem in the Man Cave ,foot thick concrete walls better not be resonating!

The modal resonance (at ~21Hz) is a function of the room size. The concrete won't stop that from happening. Your 26x21x7 dimensioned room has length and width primary axial resonance modes at 21.7Hz and 25.7Hz. That looks about right for the long decay you're experiencing. Your duct work and air return sheet metal have little to nothing to do with the problem.

Parametric EQ filters operate both in the frequency and the time domain (just as a modal resonance does). So applying filters, whose gain and bandwidth are carefully matched to a resonant peak, both the excess amplitude and the excess decay time will be nullified.

brucek

[Kal Rubinson]

brucek wrote:

The modal resonance (at ~21Hz) is a function of the room size. The concrete won't stop that from happening.

Right. In fact, they will support this node better (for worse consequences) than more flimsy, lossy room boundaries. I know this from theory and practice.

Kal

[jpk]

brucek wrote:

Parametric EQ filters operate both in the frequency and the time domain (just as a modal resonance does). So applying filters, whose gain and bandwidth are carefully matched to a resonant peak, both the excess amplitude and the excess decay time will be nullified.
brucek

I’m wondering if I did something wrong setting up the SVS? (Although SVS's graphs looked good.) Am I expecting too much from the unit or if my unit has a problem?

Although it’s advertised:
“Harness the power and refinement of your system, and take the room out of the equation, finally.”

Ideas, Comments?

Thanks,
Joe

[Sonnie]

It might be good to have Ed chime in, but I would not blame them for taking the weekend off for the 4th either.

[Wayne A. Pflughaupt]

brucek wrote:

Parametric EQ filters operate both in the frequency and the time domain (just as a modal resonance does).

brucek wrote:

Interesting that the EQ didn't appear to do much for the decay problem at ~20Hz.

Perhaps this is why (emphasis added)?

Ron Stimpson wrote:

The SubEQ results are just jaw-dropping. The worse your room is, the better this box works to make it right. It corrects for acoustical room anomalies not with the brute-force manipulation of crude IIR filters found in conventional PEQs, but instead with the finesse and sophistication of FIR-based filters which don’t cause phase/time-domain anomalies and distortions which corrupt your sound.

But then, that does seem to contradict this (emphasis added)...

Ron Stimpson wrote:

Features

• Driven by an advanced Digital Signal Processor (DSP) using sophisticated Audyssey customized room-correction algorithms
• Utilizes Adaptive Low Frequency Correction Technology (ALFC) from Audyssey Lab the most advanced bass calibration method available to consumers today
• Performs in both frequency and time domain, for superior artifact-free bass adapted for your room
• Dual subwoofer processing for better blending and integration with main channels
• FIR filters to avoid time/phase distortion inherent with IIR filters for superior bass clarity
• Corrects bass for an incredible 32 positions in horizontal and vertical space to cover any listening area
• Heavy-duty steel chassis with SVS's magnetically-retained CNC'd aluminum face-plate
• Twin sub in, twin out, with multiple configuration modes. Mains level calibration signal output jack.
• Easy-to-use laptop PC graphical user interface (GUI) and CD operating software and users' guide
• Dedicated Audyssey microphone and AV receiver external calibration pass thru cables included
• High quality, isolated 12V. power supply, and USB cable included
• On-line SVS tech support page for easy software updates, upgrades and news
• Power and unit status LEDs
• Front-panel easy access for calibration microphone and USB connection
• Final test and assembly in SVS's Ohio headquarters
• Limited production, world-wide distribution only by exclusive SVS resellers

Wonder which it is?

Regards,
Wayne

[Timoxx4]

I think it maybe means that the FIR filters don't induce any unwanted time/phase distortion when it dose its thing. It only adjust for time/phase if needed and is not a by product of its normal frequency adjustments.

Where as the IIR filters found on your run of the mill PEQ or GEQ will actually cause some time/phase anomalies of its own when applying its more basic filters.So IIR can add time/phase problems when in use by accident, where as the FIR method only corrects for time/phase if needed and will not add anything that it shouldn't.

If my understanding is correct and makes sense ?

I think that’s what the above means anyway.

[Ed Mullen]

Timoxx4 wrote:

I think it maybe means that the FIR filters don't induce any unwanted time/phase distortion when it dose its thing. It only adjust for time/phase if needed and is not a by product of its normal frequency adjustments.

Where as the IIR filters found on your run of the mill PEQ or GEQ will actually cause some time/phase anomalies of its own when applying its more basic filters.So IIR can add time/phase problems when in use by accident, where as the FIR method only corrects for time/phase if needed and will not add anything that it shouldn't.

If my understanding is correct and makes sense ?

I think that’s what the above means anyway.

This is essentially correct; the minimum phase FIR filters do not introduce phase anomalies unlike IIR filters can, particularly with a high Q factor.

The AS-EQ1 generates an impulse response and does indeed look at the time domain component (ringing profile) of the room when it builds the minimum phase FIR filter solution, as opposed to just looking at the magnitude response.

With that said, it naturally gives precedence to the magnitude response. Completely suppressing that 21 Hz room mode would probably require a significant depression of the FR at that frequency. If you have a conventional PEQ, you can try to confirm this by cascading the PEQ over the AS-EQ1 solution and notching out the response at 20-22 Hz until the decay profile over that bandwidth falls into line with the remainder of the pass band.

As an aside, it would be interesting to see the magnitude response of an EQ solution which gives precedence to the time domain response. I suspect it would be anything but flat.

Probably the best way to summarize would be the AS-EQ1 gives precendence to the amplitude response, it does address the time domain response to the extent doing so won't significantly compromise the amplitude response, and the minimum phase FIR filters do not introduce phase anomalies.

[brucek]

FIR filters do not introduce phase anomalies

Ed, it's been my rudimentary understanding that while FIR filters are considered superior due to their linear phase, IIR filters (BFD, et.al) are quite well suited when used at frequencies below ~100Hz, as their characteristics of a 2nd order biquad match that of room modes. Room modes behave as 2nd order biquads and can be nullified (at a given listening position) if the IIR filters gain, bandwidth and center frequency carefully match the mode.

In experiments where detractors claimed the BFD didn't operate in the time domain, experimentation (here) showed how both magnitude and ringing tail can be completely nullified (at the listening position of course).

What was the reasoning for going the FIR route?

brucek

[Wayne A. Pflughaupt]

On top of that, why is “not introducing phase anomalies” such a big deal? There are “phase anomalies” all over the room because sound is bouncing around all over the place, especially low frequencies which are omnidirectional.

The following is not from Rane’s Exposing Equalizer Mythology, "#6: An ideal equalizer would add no phase shift when boosting or cutting:" (emphasis added)

Phase shift is not a bad word. It is the glue at the heart of what we do, holding everything together. That it has become a maligned term is most unfortunate. This belief stands in the way of people really understanding the requirements for room equalization.

Associated with each change in amplitude is a corresponding change in phase response. Describing them as unbelievably jagged is being conservative. Every time the amplitude changes so does the phase shift. In fact, it can be argued that phase shift is the stuff that causes amplitude changes. Amplitude, phase and time are all inextricably mixed by the physics of sound. One does not exist without the others.

Regards,
Wayne

[Ed Mullen]

Let's not confuse phase change as it relates to amplitude change, and phase anomalies caused by the filter itself. Of course phase will change if FR changes; the two are inextricably tied by Fourier transform. And when the amplitude of a room peak is reduced, the room ring time at that frequency is also reduced.

Audyssey feels that IIR filters with high Q (narrow band) factors have phase problems which are indeed audible. And attempting to avoid high Q filter values (by using wider filters with less resolution) simply causes them to overlap neighboring bands, thus reducing the efficacy of the EQ solution. Furthermore, Audyssey applies a proprietary pyschoacoustic weighting to its filter coefficients in order to apply filter correction appropriately. The combination of eliminating audible IIR phase problems by employing minumum phase FIR filters, and the use of pyschoacoustic weighting on the filter coefficients, results in a superior sounding EQ solution. This has been proven out by enthusiasts comparing their previous EQ solution (e.g., SMS-1, Behringer BFD, etc.) to the AS-EQ1 - the overwhelming consensus is the AS-EQ1 simply sounds better. And filter theory aside, isn't that what really matters the most to the end-user?

Enthusiats are certainly free to disagree with Audyssey's position on filter theory, and should they choose to, I recommend taking-up the debate directly with Audyssey. While SVS had a major hand in developing the AS-EQ1, we obviously did not design MultEQ XT Pro EQ correction technology - we chose Audyssey for that, because we feel they are industry leaders in the field of EQ correction technology. I can help anyone with AS-EQ1 set-up and operation, but if you want to debate filter theory and what you like or dislike about MultEQ XT Pro, take it up with Chris Kyriakakis.

[jpk]

Ed Mullen wrote:

If you have a conventional PEQ, you can try to confirm this by cascading the PEQ over the AS-EQ1 solution and notching out the response at 20-22 Hz until the decay profile over that bandwidth falls into line with the remainder of the pass band.

As an aside, it would be interesting to see the magnitude response of an EQ solution which gives precedence to the time domain response. I suspect it would be anything but flat.

Ed,
I don't have any other EQ devices to try.
For good, bad or otherwise I believe I'm the only one that posted waterfall graphs / plots, it would be interesting to see others!
The resonance at about 21 Hz looks pretty bad to me. Where do I go from here?
Thanks,
Joe

[Wayne A. Pflughaupt]

...if you want to debate filter theory and what you like or dislike about MultEQ XT Pro, take it up with Chris Kyriakakis.

Heh heh – very diplomatic response, Ed! I’m not smart enough to debate filter designs, just the oft-touted nonsense that phase introduced by equalizers is inherently a bad thing. But on the other hand, I can understand that with FIR filters in the game, that stance may need to be re-evaluated...

Audyssey feels that IIR filters with high Q (narrow band) factors have phase problems which are indeed audible. And attempting to avoid high Q filter values (by using wider filters with less resolution) simply causes them to overlap neighboring bands, thus reducing the efficacy of the EQ solution. Furthermore, Audyssey applies a proprietary pyschoacoustic weighting to its filter coefficients in order to apply filter correction appropriately. The combination of eliminating audible IIR phase problems by employing minumum phase FIR filters, and the use of pyschoacoustic weighting on the filter coefficients, results in a superior sounding EQ solution.

Cutting through Audyssey’s techno-babble, what you’re saying is that FIR filters allow the AS-EQ1 to use a multitude of high Q filters to achieve flat response by addressing every little ripple?

This has been proven out by enthusiasts comparing their previous EQ solution (e.g., SMS-1, Behringer BFD, etc.) to the AS-EQ1 - the overwhelming consensus is the AS-EQ1 simply sounds better. And filter theory aside, isn't that what really matters the most to the end-user?

No argument there. I’ve always said that an improvement in sound quality is the goal of equalization.

Regards,
Wayne