Audyssey's Next Generation of Room Correction (MultEQ-X)

[Chromatischism]

Rip off is subjective, but in terms of added value DLBC simply adds more than this appears to right now. So the comparison really makes no sense.

Because features like multi-sub are missing from Dirac whereas they've been part of Audyssey XT32 for 10 years. That's why Bass Control seems to add a lot.

Also I have to disagree, the new software for Audyssey adds a ton more new features that Bass Control does not over its standard version. So the $500 they want is a huge ask IMO. It's arguably more tempting to buy a MiniDSP 2x4 HD and learn MultiSub Optimizer.

 

[Chromatischism]

Technically my poll vote is:

"I'm a current AVR owner. $200 price is acceptable, and though I don't like the restrictive terms, there are no better options, so I will buy it."

 

[Chromatischism]

Will this update make dual sub with different settings possible on a X4500H?

If you're asking if you can set them up in stereo, no.

Audyssey at its core time aligns and equalizes two subs together per best practice. The software interface gives us more control but it does not alter the core functionality (at this moment in time).

 

[jhaider]

Because features like multi-sub are missing from Dirac whereas they've been part of Audyssey XT32 for 10 years. That's why Bass Control seems to add a lot.

Sorry, but matching levels and delays of two subs isn’t “multi-sub.” That’s just a crapshoot.

 

[Chromatischism]

Sorry, but matching levels and delays of two subs isn’t “multi-sub.” That’s just a crapshoot.

I'm pretty sure any time two or more subs are integrated and equalized it's called multi-sub. At least on this planet.

 

[Sancus]

I'm pretty sure any time two or more subs are integrated and equalized it's called multi-sub. At least on this planet.

You cannot independently equalize two subs with any Audyssey AVR, they're treated as 1 channel in terms of EQ. So to compare with DLBC you actually need to buy a miniDSP 2x4 HD just to get close to the same capability. This app doesn't change that limitation, as was established earlier in the thread.

Also I have to disagree, the new software for Audyssey adds a ton more new features that Bass Control does not over its standard version. So the $500 they want is a huge ask IMO. It's arguably more tempting to buy a MiniDSP 2x4 HD and learn MultiSub Optimizer.

E: Oh, and I love MSO, it produces fantastic results especially for free software. But it would be misleading to say it's better than Bass Control(or this app makes Audyssey better). MSO can only optimize crossover with L/R or C(your choice). It doesn't optimize every crossover in a multi-channel system. I do plan to do a direct comparison whenever I buy a DLBC AVR though and post detailed measurements, because I'm not 100% sure how material the difference is.

 

[Haint]

Agreed.

AVS Forum, and unfortunately to too large a degree, this forum, are populated by what can be termed elementary school arithmetic audiophiles.

These folks have little to contribute and essentially understand nothing about the underlying technology, but they are quick to note that one option costs more than another The forum covering the editor application is prime example.

In almost any comparison of (1) two or more options there will be geniuses who manage to determine, likely after extensive contemplation, that one option costs more than another. This determination of which option costs more is typically their only contribution. Normally other geniuses will pile-on since these forums often have a cancel-culture on steroids. Evidently these individuals believe they are providing key insights and analysis that others will miss, having failed elementary school arithmetic.

Unfortunately the poll for this thread seems to fit with the thought process that is described above, with a total concentration on price.


(1) This isn't about $10k DAC's or ridiculous cables, the subject here is two or more realistic options.

Thank you for the valuable insight, what a thorough evaluation of Dirac and Audyssey. You really changed a lot of hearts and minds with such compelling and well supported arguments. Glad we have an Audyssey and Dirac engineer like yourself on hand to school all us elementary school arithmetic audiophiles.

You cannot independently equalize two subs with any Audyssey AVR, they're treated as 1 channel in terms of EQ. So to compare with DLBC you actually need to buy a miniDSP 2x4 HD just to get close to the same capability. This app doesn't change that limitation, as was established earlier in the thread.

It also does a poor job with the level and delay matching IME, you're much better off doing it manually and Y-cabling the subs to a single input before running Audyssey. XT32 is generally very good at frequency response correction, but terrible at everything else (sub alignment, crossover, etc...)

 

[Trell]

You cannot independently equalize two subs with any Audyssey AVR, they're treated as 1 channel in terms of EQ.

It was not always so and was changed later to todays functionality as they found it worked better.

Here is an article that is written in 2010 (checking the date on the forum posts) while the date on the article is from 2015 (probably republished).

Home Theater Multiple Subwoofer Set-Up & Calibration Guide

This article discusses methods of proper placement, and setup of multiple subwoofers to achieve the best measurable and audible results for your entire listening area not just a single sweet spot.

www.audioholics.com

>>>
As you can see Audyssey created a very ragged response, much worse than having no EQ employed at all. I even tried setting up my AVP-A1HDCI via 3 sub outs but had the same unfortunate results.

When I first setup the Audioholics Showcase Home Theater room, Audyssey's Chief Tech Officer Chris Kyriakakis paid me a visit. We spent countless hours calibrating measuring and calibrating my system. It was at that time we realized the best way to apply EQ would be to apply a simultaneous single correction curve to all of the subs. Chris agreed but unfortunately no Audyssey enabled A/V receivers or processors featuring multiple subwoofer outputs support this. Audyssey did however employ this in their new subwoofer calibration standalone processors: the Audyssey Sub Equalizer and the SVS AS-EQ1. Audyssey took it one step further by measuring each sub individually, adjusting phase, distance and level and then applying a single correction curve to both subwoofers. We pleaded with Denon to offer this for the AVP-A1HDCI Pre/Pro and they claim this upgrade is in the works. We keep our fingers crossed for that day but for now I pain through the process of using a single subwoofer output and making the best of a single delay setting for all four subwoofers not theoretically perfectly positioned in my odd shaped room.

Over time I became smarter and gave up the notion of independent subs. Thus until the above mentioned firmware update becomes available, we highly recommend you utilize a single subwoofer output for all of your subs when using Audyssey. This applies to most of the other room correction systems that I've tested. Below are the results of such a setup which you can see were quite favorable.
<<<

 

[Chromatischism]

You cannot independently equalize two subs with any Audyssey AVR, they're treated as 1 channel in terms of EQ.

Maybe you replied to the wrong post, but I said the same thing above:

If you're asking if you can set them up in stereo, no.

Audyssey at its core time aligns and equalizes two subs together per best practice. The software interface gives us more control but it does not alter the core functionality (at this moment in time).

 

[jhaider]

Audyssey at its core time aligns and equalizes two subs together per best practice.

That’s not, and never was, “best practice.”

The competent approach has always been to use level and phase/delay as degrees of freedom to achieve the smoothest response.

 

[Chromatischism]

That’s not, and never was, “best practice.”

The competent approach has always been to use level and phase/delay as degrees of freedom to achieve the smoothest response.

Of course you also use phase/delay and level. Then, you EQ as one. That is best practice.

 

[Kal Rubinson]

At one point they mentioned you cannot use an external interface, because volume would be difficult to calibrate, and latency would throw off timing.

1. Sensitivity should be part of the calibration file data.
2. Latency issues should not apply to analog microphones.

 

[Sancus]

It was not always so and was changed later to todays functionality as they found it worked better.

I didn't know that. That's an interesting bit of history. I have to assume it worked better to treat them as 1 because they just didn't have a very good algorithm for handling the sub EQ. Though in that example, multiple subs were grouped(with splitters I assume) to each of the two outputs, which is a pretty bad idea when you can't independently adjust the delay and output.

I think people often confuse doing separate PEQs for each sub, based on measurements of just that sub, with the multivariable optimization calculations that MSO and DLBC do. The first strategy produces very poor results, it's true.

 

[jhaider]

Of course you also use phase/delay and level. Then, you EQ as one. That is best practice.

I’m not sure who you’re arguing with here. That’s exactly what DLBC does, and why it adds considerable value.

Rote matching of level and delay, which is what Audyssey does, is almost never going to be the optimum solution for multiple subs.

 

[Sancus]

Of course you also use phase/delay and level. Then, you EQ as one. That is best practice.

Best practice is to treat phase, delay, level, and EQ filters for each sub as independent variables in a multivariable optimization function with the goal to achieve lowest possible average seat to seat variation. This isn't something that can be done manually at all, and it's basically what Geddes suggests in his papers although he skips the actual information you need to achieve it lol.

 

[jhaider]

Sound United posted their video on Youtube 3hrs ago:

It's 1.5hr long.

Thanks for that. I had a chance to watch it today.

To summarize, I no longer think MultEQ X a big step back from the iOS app as was my impression from the @joentell video, though on balance I still don’t see huge added value compared to the iOS App and obviously it’s an irrelevant software anyway unless they upgrade it to run on macOS. (Even the windows laptop I bought for VituixCAD and London/Audio Architect can't use it, because it has some "LTSC" version of windows 10 that does not support microsoft's own software store.) The Audyssey iOS app is such a huge multiplier for Audyssey that it transforms the system from distant also-ran to worthy competitor in an absolute sense, while significantly lowering the price of entry to decent room correction by allowing advanced adjustments to be made on relatively inexpensive hardware for a very nominal one-time fee. This is…not that.

The biggest advance over the iOS software I saw discussed was a sort of backdoor version of Anthem ARC’s “Quick Measure” tool, which should have been copied by everyone else long ago but unfortunately has not been. Oddly, this super useful system setup feature came up as an aside in a discussion of microphone extension cables!

I prefer to directly adjust a target curve over mediating such adjustments through pretend PEQ filters, but in fairness RoomPerfect doesn't tell you anything about how your "voicing" filters change the corrected power response either.
The PEQ section isn’t as bad as I thought. Using PEQ doesn’t seem to delete the automated target curve, as was my previous impression. (Or does it?) It’s also kind of odd to have the curtains there instead of on the target curve screen but I guess that doesn’t matter as long as this key functionality is there. There's no "preconditioning" functionality, i.e. the ability to knock down a big room mode peak present in all channels manually and let room correction handle more subtle stuff.

Unfortunately, MultEQ X does not even start to address the Audyssey equipped D+M units’ biggest acoustic design flaw: one hand (room correction) does not know what the other hand (bass management) is doing. A room correction system should not present a user with separate curves for “speaker x” and “subwoofers.” Rather, the target curve should consist of “aggregate channel x” with the speaker and subwoofers integrated. ARC and RoomPerfect have long approached this properly. With DLBC, Dirac is on the right page too. D+M and Audyssey need to work together to fix this foundational problem, and integrate bass management into Audyssey. I hope they do.

 

[Sal1950]

It will be interesting to see the way things develop over the next couple years.
On one hand they have promised this to be an ongoing project, constantly evolving, improving and becoming more powerful
The other side, how soon the package will improve to the point that the on-board processing power of
older units like my AV7703 will no longer support it.
I spent a bit of time playing with X and it's demo package only to find that at this point the PC package doesn't offer me any real improved config options over what I currently get with the android app. running on PC emulation.
Looks like I'll be sitting on my wallet and watching things from the sidelines for now.. I'm completely happy with my pre-pro and see no pressing reason to update it. Marantz has been great with support suppling regular firmware updates as late as this week.

 

[KMO]

Unfortunately, MultEQ X does not even start to address the Audyssey equipped D+M units’ biggest acoustic design flaw: one hand (room correction) does not know what the other hand (bass management) is doing.

The EQ filters have always known what the crossover is, and I was always under the impression that regardless of what the "after" display shows, they're actually setting up the EQ filters to get the crossover shape right.

That's part of the reason it's generally said that Audyssey doesn't filter below the crossover frequency (so don't lower it). The Audyssey Pro manual says:

For every crossover frequency option, MultEQ Pro calculates a satellite filter that will best blend it with the subwoofer. For this reason, it is extremely important that the bass management frequency entered in the preamplifier be the same as the crossover frequency selected on this screen.

And I believe the distinction is that Pro lets you choose the crossover frequency to do this work at, whereas non-Pro didn't. I would imagine MultEQ X does what Pro did.

Certainly it's not as integrated as it could be, but they're not totally separate.

 

[Chromatischism]

Best practice is to treat phase, delay, level, and EQ filters for each sub as independent variables in a multivariable optimization function with the goal to achieve lowest possible average seat to seat variation. This isn't something that can be done manually at all, and it's basically what Geddes suggests in his papers although he skips the actual information you need to achieve it lol.

If that is your goal. I do not agree that it is everyone's goal to have good bass in multiple locations. Some people want a perfect sweet spot, to hell with other spots

 

[Sancus]

If that is your goal. I do not agree that it is everyone's goal to have good bass in multiple locations. Some people want a perfect sweet spot, to hell with other spots

That doesn't really change things though, the method produces even better results if you use fewer spots or more tightly grouped ones. And in any event, it optimizes the MLP most. Unless you only sit in exactly the same way at all times, it's probably best to measure a few spots anyways. For example there was substantial bass variation on my one(1) middle sofa seat when sitting up straight and reclining before optimization. Now there's not too much, but reclining is pretty dead flat since that's what I set as my MLP. And I only have 2 subs and tried to optimize 3 seats, so it wouldn't be hard to do even better.