The astounding new AS-EQ1 SubEQ goes on sale!

[Ed Mullen]

Gamelover360 wrote:

Just to be clear, because I have searched and I am not entirely sure about the answer to this question:

If you use 2 subwoofers located up with each of your mains, what benefit would the as-eq1 provide over Audyssey MultEQ XT......IF you only care about 2-3 listening positions (the seats on a couch)? I have heard that Audyssey will do a good job of EQ'ing the combined response of dual subs by seeing them as one. Isn't that the point anyway is to EQ the total subwoofer reponse at your listening position anyway? Thanks.

The least advantage for the AS-EQ1 over an AVR would be colocated dual subs (which the consumer version of MultEQ XT will simply view as a single subwoofer). In that scenario, the advantage of the AS-EQ1 would strictly be limited to its superior filter resolution and processing power to provide a more optimal EQ file than the AVR can generate.

If you split the subs up on either side of the front stage, the AS-EQ1 (even in dual combined mode) will look at the distance and FR of each subwoofer independently, and then will also look at the combined response and build the EQ file accordingly. If one subwoofer is showing an anomaly the other isn't, it will be corrected earlier in the signal chain, instead of applying a blanket correction to both subwoofer. So you'll get this additional benefit over the first scenario above (with colocated subs).

If you have independently verified that MultEQ XT has done a fine job with your corrected FR and you are satified with the sound quality, then by all means stand pat - we never recommend buying gear when there is no obvious improvement to be had. But enough enthusists have found their AVR solutions to be either sorely lacking or (in some cases) practically non-existent, and for them the AS-EQ1 is a great addition to their arsenal in the quest for great bass.

[Gamelover360]

Ed Mullen wrote:

The least advantage for the AS-EQ1 over an AVR would be colocated dual subs (which the consumer version of MultEQ XT will simply view as a single subwoofer). In that scenario, the advantage of the AS-EQ1 would strictly be limited to its superior filter resolution and processing power to provide a more optimal EQ file than the AVR can generate.

If you split the subs up on either side of the front stage, the AS-EQ1 (even in dual combined mode) will look at the distance and FR of each subwoofer independently, and then will also look at the combined response and build the EQ file accordingly. If one subwoofer is showing an anomaly the other isn't, it will be corrected earlier in the signal chain, instead of applying a blanket correction to both subwoofer. So you'll get this additional benefit over the first scenario above (with colocated subs).

If you have independently verified that MultEQ XT has done a fine job with your corrected FR and you are satified with the sound quality, then by all means stand pat - we never recommend buying gear when there is no obvious improvement to be had. But enough enthusists have found their AVR solutions to be either sorely lacking or (in some cases) practically non-existent, and for them the AS-EQ1 is a great addition to their arsenal in the quest for great bass.

Thank you very much Ed.

[pj121391]

Are they Shipping this EQ yet?

[CharlieU]

pj121391 wrote:

Are they Shipping this EQ yet?

I believe the last I heard, it will be the middle of next month.

[BigPines]

Ed Mullen wrote:

...and then I remeasure my own chair at the end of the session to weight/skew the response a bit for the sweet spot.

That is hilarious because I do the same thing. ...and I thought I was being deviously clever.

I just figure nobody else in the room will appreciate it as much as I do so my listening position should get twice the weight. Of course, I don't mention this little trick to anybody in the family...

Mike

[Ed Mullen]

BigPines wrote:

That is hilarious because I do the same thing. ...and I thought I was being deviously clever.

I just figure nobody else in the room will appreciate it as much as I do so my listening position should get twice the weight. Of course, I don't mention this little trick to anybody in the family...

Mike

Right on, brother. It's good to be the king (of at least one thing anyway).

[stevefish69]

My EQ1 arrived today (In London )

Ed - I wonder if you could help out to see if i've set up my EQ1 correctly

Initial impressions are pretty good. I have no idea how this unit has done it but, but the complete soundstage has improved 10 fold. The fronts and rears seem to have moved out of the room big time, and have an amazing sense of depth that i've never managed to get on my equipment before. The pans, and swings from front to rear are now seemless and i still cannot see how just be EQ'ing and time aligning the Sub this can be done

I will probably try another set of readings when i have more time, with maybe a few less readings. My only concern is that I've lost a lot of the chest impact that used to be had, but seeing how high my Sub channel was actually set in comparision to my mains was a interesting to say the least. IIRC when th emain channel was producing 75db the Sub channel was pumping out about 90db

My PCU used to be set to 12 o'clock gain and the Audiolab to +4db on the sub trim. This has now dropped to 9 o'clock and -1.5db trim to give the ruler flat response.

Now if i knock the Sub channel up a notch again and run it hot (although not 15db :grin i'm wondering whether this immense change to the other speakers will dissapear as the mains will be running lower than the Sub. The PCU still dug nice and low when needed, but just in a more lifelike manner than i'm used to.

The one thing thats confusing for me is that when doing the channel level matching it says to set the master volume to 0db and then adjust the individual trim level down until you reach 75db. The Poxy Audiolab AP does not allow you to alter the trim of the left channel. You can only move the other levels up / down to suit. On the Audiolab the master volume on the fixed left input needed to be at -12db to get 75db. I then adjusted the gain on the back of the sub down to 9 o'clock to match the 75db. At the end of the calibration, it aksed me to reduce the sub trim by a further -1.75 on the AP. DOes this mean that reference level on the AP is actually -12db

[Sonnie]

That is no fair...

However, I suppose congratulations are in order

[CharlieU]

Steve,

Are you running a 15Hz tune on your PC13-Ultra? From the first graph, it looks like you put some work into getting the sub right before getting the EQ1. In any case, the second graph is beautiful. Thanks for taking the time to post your results.

[stevefish69]

CharlieU wrote:

Steve,

Are you running a 15Hz tune on your PC13-Ultra? From the first graph, it looks like you put some work into getting the sub right before getting the EQ1. In any case, the second graph is beautiful. Thanks for taking the time to post your results.

Hi ya,

Yes the Sub is being run in 15hz tune. It's in the front corner of the room and is the third and best position i could find for it using REW.

I will eventually (when i get time to fart) get some REW sweeps done. I'd like to be able to show the following.

Old results with BFD
Sub and mains together with no EQ
Sub and mains together with EQ1 recommended settings and gain set to 9 o'clock
Sub and mains together with EQ1 setup but sub gain back to 12 o'clock gain

Regards

Steve

[markus76]

Hi,

although the AS-EQ1 sounds promising it has a fundamental flaw—to have the subwoofers calibrated, the mains need be part of the calibration procedure. Otherwise the whole frequency range where subs and mains cross over will be uncorrected. I've searched the AS-EQ1 manual but couldn't find any way to make the mains part of the calibration. A simpel level match of mains and subs won't do the trick.

Best, Markus

[Ed Mullen]

markus76 wrote:

Hi,

although the AS-EQ1 sounds promising it has a fundamental flaw—to have the subwoofers calibrated, the mains need be part of the calibration procedure. Otherwise the whole frequency range where subs and mains cross over will be uncorrected. I've searched the AS-EQ1 manual but couldn't find any way to make the mains part of the calibration. A simpel level match of mains and subs won't do the trick.

Best, Markus

I'm not sure I understand your statement, but the product is not "fundamentally flawed". That's a pretty strong statement. I will help you work through any setup issues and answer any questions you may have about the unit.

[markus76]

Hello Ed,

in the modal region we won't be able to make assumptions on how mains and subs will cross over. I'd like to direct everybody that is interested in the topic to the scientific work of Todd Welti:
http://www.aes.org/e-lib/download.cf...80&name=harman

Best, Markus

[Sonnie]

Fundamentally flawed...





I'd rather believe that someone has fundamentally flawed eyes.

[Ed Mullen]

stevefish69 wrote:

My EQ1 arrived today (In London )

Ed - I wonder if you could help out to see if i've set up my EQ1 correctly

Initial impressions are pretty good. I have no idea how this unit has done it but, but the complete soundstage has improved 10 fold. The fronts and rears seem to have moved out of the room big time, and have an amazing sense of depth that i've never managed to get on my equipment before. The pans, and swings from front to rear are now seemless and i still cannot see how just be EQ'ing and time aligning the Sub this can be done

I will probably try another set of readings when i have more time, with maybe a few less readings. My only concern is that I've lost a lot of the chest impact that used to be had, but seeing how high my Sub channel was actually set in comparision to my mains was a interesting to say the least. IIRC when th emain channel was producing 75db the Sub channel was pumping out about 90db

My PCU used to be set to 12 o'clock gain and the Audiolab to +4db on the sub trim. This has now dropped to 9 o'clock and -1.5db trim to give the ruler flat response.

Now if i knock the Sub channel up a notch again and run it hot (although not 15db :grin i'm wondering whether this immense change to the other speakers will dissapear as the mains will be running lower than the Sub. The PCU still dug nice and low when needed, but just in a more lifelike manner than i'm used to.

The one thing thats confusing for me is that when doing the channel level matching it says to set the master volume to 0db and then adjust the individual trim level down until you reach 75db. The Poxy Audiolab AP does not allow you to alter the trim of the left channel. You can only move the other levels up / down to suit. On the Audiolab the master volume on the fixed left input needed to be at -12db to get 75db. I then adjusted the gain on the back of the sub down to 9 o'clock to match the 75db. At the end of the calibration, it aksed me to reduce the sub trim by a further -1.75 on the AP. DOes this mean that reference level on the AP is actually -12db

Thanks for the positive feedback on the AS-EQ1. The 10-fold improvement in sound stage resulted from 2 things - running the subwoofer calibration level flat/even with the speakers, and 2) the perfectly flat FR you have now achieved.

You lost that chest impact because it trimmed about 5 dB off a 5 dB peak centered at ~35 Hz. The sound is more accurate now (as you noted), but will take an adjustment period. You can run the subwoofer calibration level a little hotter, but eventually (as you fear) the sound stage will collapse and the subwoofer will start to dominate and become localized. A flat/even calibration level with the mains is the best way to keep the subwoofer blending seamless and non-localized.

The level matching feature is important for the exact reasons you cited - you were running the subwoofer about 15 dB hot. Simply correcting for FR is not enough - it also need to be level-matched with the front stage speakers.

To answer your question a -12 dB master volume setting does not necessarily mean that is now Dolby Reference Level (although it could be). It was simply the master volume required to get your left channel (which has no trim control) to 75 dB with the voltage input from the AS-EQ1. Then you adjusted all your other speakers and the subwoofer to that reference point.

The fact that DVDs have mastering/encoding levels which vary considerably makes moot the whole concept of achieving Dolby Reference Level at any given master volume anyway. Play Star Wars I and Star Wars II at the same master volume setting and you'll immediately notice SW II is much louder. Ditto for Sea Biscuit and Underworld (the latter being extremely loud at -13 and the former being comfortable at -5). Throw in DTS tracks which lack dial norm and which are typically mastered 4 dB louder anyway, and you have an 8 dB disparity in bass peaks between Master & Commander cannon blasts on DD vs. DTS at the same master volume setting.

So in the end, don't worry about achieving Dolby Reference Level at any given master volume setting, because the varying encoding levels on DVDs pretty much throws that right out the window. Instead, just set the master volume at a comfortable level for any given DVD and sound format and be happy.

[markus76]

Sonnie,

what do we see in those diagrams? Is it an average of multiple measurements? What smoothing? What window is applied?

Best, Markus

P.S. Please don't quote me wrong. I didn't say that the subwoofer calibration is fundamentally flawed. I said it's a fundamental flaw to NOT make the mains part of the calibration.

[BigPines]

markus76 wrote:

Hello Ed,

in the modal region we won't be able to make assumptions on how mains and subs will cross over. I'd like to direct everybody that is interested in the topic to the scientific work of Todd Welti:
http://www.aes.org/e-lib/download.cf...80&name=harman

Best, Markus

From a sub EQ perspective, we don't need to make any assumptions on what the crossover will be. We just need the FR to be as flat as possible. The AVR has the duty to integrate the subs and the mains not the sub EQ. If the AS-EQ1 calibration is run first (as recommended right?) and then the AVR calibration is run, the AVR has the sole responsibility of integrating the sub and the mains. I don't understand the problem.

Mike

[markus76]

Ed,

I have another question: the device uses FIR filters that introduce a delay in the signal path—how big is the latency? Is it fixed or variable?

Best, Markus

[Sonnie]

I see exactly what you see.

If you want graphs to get more technical beyond where we can really hear a difference, there will be plenty of before and after REW measurements right here at the Shack.

I am confident SVSound did not release a flawed product knowing it will be tested over and over again by numerous people, only to their own product demise should it not live up to its claims. A little common sense tells us this much.

[BigPines]

markus76 wrote:

Ed,

I have another question: the device uses FIR filters that introduce a delay in the signal path—how big is the latency? Is it fixed or variable?

Best, Markus

Where is all this coming from? Do you have an interest in a competing product by any chance?

Why don't you measure the unit and find out?

Mike

[Ed Mullen]

markus76 wrote:

Hello Ed,

in the modal region we won't be able to make assumptions on how mains and subs will cross over. I'd like to direct everybody that is interested in the topic to the scientific work of Todd Welti:
http://www.aes.org/e-lib/download.cf...80&name=harman

Best, Markus

The only assumption on how the mains and subwoofer will cross over is the application of the high pass filter on the speaker and the low pass filter on the subwoofer at the selected crossover frequency.

The entire sordid affair of applying a 2nd order high pass and a 4th order low pass is firmly rooted in THX, and will only work correctly if the speakers are sealed with a natural 80 Hz acoustic F3 and a 12 dB/octave roll-off profile. Virtually none of the speakers on the market today meet that criteria, most of them being bass reflex with varying F3 levels (both higher and lower than 80 Hz). This results in asymmetrical roll-off slopes and attendent phase and FR distortions.

So the crossover between the speakers and the subwoofer is almost always compromised by using non-THX certified speakers and a 12/24 high/low pass filter profile. Even if the FR of the speakers was also EQd to be flat (which does occur with an increasing number of auto-EQ AVRs), that alone does not ensure a correct speaker/subwoofer crossover transition unless the speakers are THX certified, and that is rarely the case.

And if you think an "all in one" auto-EQ set-up like an AVR with MultEQ XT will somehow correct for the inherently flawed 12/24 crossover network when using non-THX speakers, you're dead wrong. No AVR can alter the acoustic F3 of a speaker, nor can it alter the acoustic roll-off profile (all bass reflex speakers roll-off 4th order). Furthermore, I've never seen any consumer AVR select an optimal crossover frequency for any speaker, even ones that are THX (sealed 80 Hz F3). Typically the AVR sets the crossover far deeper than optimal, and often sets even small and bass deficient speakers to Large (full-range)! Overriding the AVR speaker size selection is mandatory in order to protect the speakers and also allow all deep bass to be played by the subwoofer. In addition, no consumer AVR ever adjusts the FR of the subwoofer to compensate for perceived or actual FR/phase deficiencies of the speakers - if they adjust the FR of the subwoofer at all (and some don't even do that), it's always for a flat resonse to 200+ Hz, so the sub will track the low pass filter optimally. The AS-EQ1 is no different in that regard.

None of the above problems with the speaker/subwoofer transition is the fault of the AS-EQ1, nor is it caused by the AS-EQ1, and it does not represent in any shape or manner a "fundamental flaw" of the product. A subwoofer with a flat FR will always sound better, and will also track any predicted low pass filter slope perfectly, thereby getting you "that much closer" to an ideal response.

[Ed Mullen]

markus76 wrote:

Ed,

I have another question: the device uses FIR filters that introduce a delay in the signal path—how big is the latency? Is it fixed or variable?

Best, Markus

A fixed 8.5 ms latency which is included in the subwoofer distance recommendation by the AS-EQ1. For 2 channel users with pre/pros which lack a distance control, this latency must be physically compensated for by moving the subwoofer closer to the listener (often impractical), or by digitally delaying the input to the mains using an external DSP crossover like those available from Behringer. We're quite clear on this point in the owner's manual and with 2 channel customers who inquire about the unit in Sales and Tech Support.

[Ed Mullen]

markus76 wrote:

Sonnie,

what do we see in those diagrams? Is it an average of multiple measurements? What smoothing? What window is applied?

Best, Markus

P.S. Please don't quote me wrong. I didn't say that the subwoofer calibration is fundamentally flawed. I said it's a fundamental flaw to NOT make the mains part of the calibration.

And my response (again) is that "all in one" AVRs will always target a flat FR for the speakers, and always target (if they EQ the subwoofer at all) a flat FR for the subwoofer. The phase/FR flaws inherent in the asymmetrical 12/24 crossover network are always present whether the subwoofer is EQ'd internally by the AVR or externally by the AS-EQ1. The AS-EQ1 just does a better job of EQing the subwoofer than the AVR can.

And if we think the above state-of-the art of auto-EQ AVRs represents a compromise (and who could argue otherwise), consider all the owners of legacy AVRs out there with no auto-EQ of the speakers at all. They are the ones who suffer the worst with just level matching, picking a global crossover frequency, and setting the distances. Again, the AS-EQ1 will also get legacy AVR owners "that much closer" to an ideal/optimal response by at least providing a flat subwoofer FR which perfectly tracks the predicted low pass filter slope, but it can't fix the speaker/subwoofer transition problems attendent with a 12/24 crossover, nor can even modern auto-EQ AVRs.

[Ed Mullen]

BigPines wrote:

From a sub EQ perspective, we don't need to make any assumptions on what the crossover will be. We just need the FR to be as flat as possible. The AVR has the duty to integrate the subs and the mains not the sub EQ. If the AS-EQ1 calibration is run first (as recommended right?) and then the AVR calibration is run, the AVR has the sole responsibility of integrating the sub and the mains. I don't understand the problem.

Mike

If the AVR has auto-EQ, it should be run first. It will target a flat FR for the speakers down to their respective F3 points and then allow them to roll-off at either 2nd order or 4th order (depending on whether the are sealed or bass reflex).

The AVR will typically make the error of setting any speaker to Large (full-range) which has usable output at/near 40 Hz (don't ask me who dreamed up this arbitrary cut-off). This must be manually overidden by the user of course. And if a given speaker is set to Small, the crossover frequency selected by the AVR has nothing to do with properly integrating with the subwoofer - rather it is always based on the acoustic F3 of the speaker. That is why you will see varying crossover frequencies for each channel rather than the global crossovers in legacy AVRs. And the subwoofer FR is always EQd flat.

And (unfortunately) every speaker channel/subwoofer crossover transition is fraught with the potential for FR/phase problems, none of which are the fault of the AS-EQ1 and are simply something we all live with given the compromises of the 7 speaker/1 subwoofer, 12/24 crossover network. Yes - it's a compromise vs. 7 full-range speakers - we all get it. And frankly, these compromises loom far more ominously on paper and in forum threads than they do in reality. In reality, the speaker/subwoofer crossover transition is fairly benign - even with the inherent limitations discussed above - and doesn't cause untoward problems for the listener. What we typically here when the speaker FR is flat, the subwoofer FR is flat, the levels are all matched, and the distances are set correctly - is a smooth and seamless transition from the speaker channels to the subwoofer.

[markus76]

Ed,

I don't think looking at low frequency modal problems like that will yield optimal results. Even perfectly matched filter slopes for subwoofer and mains won't give you a flat frequency response in a real room due to a rooms unique modal behavior.
When you are trying to optimize for a single listening location you might get good results with EQs but when optimizing for a larger area, then EQs will fail. Only multiple low frequency sources (and/or absorption) can smooth spectral variance, but ALL low frequency sources have to be part of the calibration, i.e. (multiple) subwoofers AND mains.

Best, Markus