To EQ or not to EQ, Is auto room EQ a good thing?

[Tony V.]

Last year I tried playing with turning audyssey off and then manually adjusting the EQ on my two Yamaha YDP2006 EQs that I have on my front 3 channels and sub.

With all the automated room EQ flavors available now (Audyssey, MMAC, YAPO, Dirac) one begins to think how much is too much. Ive been wondering if in my case I could accomplish the same or even better results myself using outboard EQ on the front sound-stage. Onkyo got me thinking as they are no longer applying EQ to the front channels with their proprietary room eq that is in their receivers now.


One of the things Ive learnt long ago is that applying EQ filters on frequencies above 400Hz that are not identical on both channels will cause your imaging to collapse across the front.
To prove this I set up a preset with no filters applied. I played some well recorded music that I use for demos and that I know. I played the track and then took a filter with average Q and added it only to the left channel at 4K (so it was a nice boost from about 2K up to 6K) and then another at 2K on the right channel and I was actually astonished how much just that caused the image to move into the center of the room or what seemed to pull it from where it should be.
I then restored the flat setting and tried several different frequencies again different on the two channels cutting and or boosting and each time the difference was noticeable (last night I asked my wife to sit in the room while I did this and even she commented at the difference and how it pulled the sound into the center of the room).
I then did the same test but this time I locked the left and right channels so any adjustment would be done evenly on each channel, this did not effect the image at all other than boosting/cutting the frequency range of that part of the music.

I know that with auto room correction it is quite common to have filters automatically added that are different on each channel as not many rooms are perfectly symmetrical and speaker placement is never ideal so using auto room EQ would be concerning to me as that is not the right method and should be avoided. I wonder what others thoughts are on this is and if I am correct thinking Onkyo may be on the right path by eliminating EQing the mains and only focusing on the sub and surrounds and also that room acoustics should be addressed and take a high priority over EQ.

NOTE: I should also clarify that EQ on the subwoofer channel is a very good thing and should always be done.

 

[Leonard Caillouet]

I have used it on my system and depending on how I locate the speakers in the room I have found that it works from adequately to very well. Sure, you can almost always do better manually, but for most people, I think it works pretty well. It has to do with what you expect from a system.

 

[Sonnie Parker]

One of the things Ive learnt long ago is that applying EQ filters on frequencies above 400Hz that are not identical on both channels will cause your imaging to collapse across the front.

Think about this... if your two front speakers are NOT frequency matched, the imaging will have less chance of being accurate. Mismatched frequencies in the midrange will cause a voice that the recording engineer intended to be center to be skewed left or right of center. We've witnessed this on several occasions. EQ fixes it. And it very logical reasoning why.

When the recording engineer records the source, his studio is treated to extremes and/or his system is equalized to death...possibly a combination of both... and his speakers and/or headphones are frequency matched precisely. So unless your speakers are symmetrically placed in a completely symmetrical room and everything in the room is symmetrical (basically an an-echoic chamber, you will probably need EQ. There is no way I can manually do what Audyssey or Dirac Live can do... they take phase/delay into account for each filter... and up to 512 filters on Audyssey. Manually I don't have anywhere close to the adjustments it has... nor can I perform the algorithms that it can perform. Most "home" systems are mixed with a great room or rec room... anything but symmetrical. I have yet to hear one that sounded good without EQ. Even the shows we go to I am scratching my head on the setups of tens of thousands of dollars worth of equipment that needs better imaging. If they can get the speakers far enough out into the room where holes in the wall and reflections make less of a difference, it can sound pretty good in many cases... but it's hard to do in some of those small rooms.

This is why every time we have setup different speakers in my room, the imaging improves with equalization. Although my room is nearly symmetrical in all aspects, the left and right speakers are not always frequency matched. We've actually had some that were off pretty bad. EQ fixed it and in every case improved the imaging, sound stage and depth acuity. There is absolutely no way I could live without my Audyssey and/or Dirac Live. Gotta have it!

 

[Leonard Caillouet]

Great summary, Sonnie. And what you describe is not just your experience but is confirmed by others that have done countless hours of testing and listening in that room. And it has been consistent over dozens of speakers that have been carefully placed and tested, and many changes in room conditions such as acoustic treatments large sub and speaker moves and having multiple speakers in the room at any time.

 

[1_sufferin_mind]

One of the things Ive learnt long ago is that applying EQ filters on frequencies above 400Hz that are not identical on both channels will cause your imaging to collapse across the front.

No argument there, but curious: Why 400Hz? Because that's roughly around the Schroeder frequency?

I know that with auto room correction it is quite common to have filters automatically added that are different on each channel as not many rooms are perfectly symmetrical and speaker placement is never ideal so using auto room EQ would be concerning to me as that is not the right method and should be avoided. I wonder what others thoughts are on this is and if I am correct thinking Onkyo may be on the right path by eliminating EQing the mains and only focusing on the sub and surrounds and also that room acoustics should be addressed and take a high priority over EQ.

NOTE: I should also clarify that EQ on the subwoofer channel is a very good thing and should always be done.

At least one other device I know of leaves upper frequencies unprocessed. That's the DSPeaker Anti-Mode 2.0 Dual Core, and it's (configurable) upper limit is 500Hz. The full version of Dirac Live LE which runs on the Emotiva XMC-1 operates to the frequency extremes, but has user-adjustable "curtains" that define the upper/lower limits for processing. And while measured frequency response can vary widely between left and right channels, I have yet to see Dirac specify anything but identical target curves for each (at least in the upper frequency region). I'm unfamiliar with Audyssey's flavor of auto-EQ, except to say that it's soundstage and imaging performance has been shown to be significantly below that of Dirac Live's (Ref: Post #51 Here).

I'm no seasoned vet when it comes to acoustics and room correction algorithms, but my experience has me agreeing with Sonnie. I, too, think that Dirac Live can not be bested by manual calibration. There are too many interrelated variables to account for in both frequency and time domains. Add in the finer points of specifying the correct type of filter needed (FIR or IIR), then optimizing it for a minimum phase system, and you have a formula for human error. The way I understand it, DSP processing power only recently (within the last few years) became sufficiently robust to handle the task. So while in the past it may have been true that manual calibration could equal or beat auto-EQ, I believe that's no longer the case.

 

[Sonnie Parker]

I currently have the DDRC88-BM with Dirac Live processing my two front channels... with Audyssey handling the remaining channels for movies. I set Audyssey to bypass front L/R channels so Dirac can handle them. It works well, as you will see in the upcoming MartinLogan Expression 13A review.

 

[tesseract]

My system sounds and measures pretty good without EQ. Dirac put the icing on the cake and made it sound great.

The best way to overcome the room is measurements, treatments, EQ and more measurements. All rooms/speakers can benefit from this.

 

[JBrax]

I've been plodding along with this hobby for a long time and never taken any measurements. I know I should and I also know I could most likely improve upon what I think as is sounds great. Audyssey does a good job in my room and definitely sounds better engaged than not. Sonnie, since you've used both Dirac and XT32 do you feel Dirac is that much of an improvement?

 

[Sonnie Parker]

The best way to overcome the room is measurements, treatments, EQ and more measurements. All rooms/speakers can benefit from this.

And then there are those who also are beginning to suggest more and more that we don't need acoustic treatments... reflections (even first) can be a good thing. And ... AND... we now have manufacturers making AMPS that can help eliminate room modes. Brilliant perhaps.

Sonnie, since you've used both Dirac and XT32 do you feel Dirac is that much of an improvement?

It's close as far as how it sounds in my room. The measurements look better with Dirac, but it's not easy to tell a difference between the two.

 

[Tony V.]

No argument there, but curious: Why 400Hz? Because that's roughly around the Schroeder frequency?

Yes, that is my thought.

I agree as I said that room treatment should be addressed, in my opinion if you start there your auto eq will have less to do in those delicate upper frequencies.

I still have Audessey engaged on my Onkyo and feel it is doing a good job but if my tests above show anything is that there can be too much of a good thing in rooms that are not symmetrical for acoustics.

 

[tesseract]

And then there are those who also are beginning to suggest more and more that we don't need acoustic treatments... reflections (even first) can be a good thing. And ... AND... we now have manufacturers making AMPS that can help eliminate room modes. Brilliant perhaps.

Not so much that we don't need treatments, but that we should use the first reflection instead of absorbing it, especially in small or medium-sized rooms.. Secondary, tertiary, etc. reflections and modal build up in corners are still good to treat.

 

[Todd Anderson]

Think about this... if your two front speakers are NOT frequency matched, the imaging will have less chance of being accurate. Mismatched frequencies in the midrange will cause a voice that the recording engineer intended to be center to be skewed left or right of center. We've witnessed this on several occasions. EQ fixes it. And it very logical reasoning why.

When the recording engineer records the source, his studio is treated to extremes and/or his system is equalized to death...possibly a combination of both... and his speakers and/or headphones are frequency matched precisely. So unless your speakers are symmetrically placed in a completely symmetrical room and everything in the room is symmetrical (basically an an-echoic chamber, you will probably need EQ. There is no way I can manually do what Audyssey or Dirac Live can do... they take phase/delay into account for each filter... and up to 512 filters on Audyssey. Manually I don't have anywhere close to the adjustments it has... nor can I perform the algorithms that it can perform. Most "home" systems are mixed with a great room or rec room... anything but symmetrical. I have yet to hear one that sounded good without EQ. Even the shows we go to I am scratching my head on the setups of tens of thousands of dollars worth of equipment that needs better imaging. If they can get the speakers far enough out into the room where holes in the wall and reflections make less of a difference, it can sound pretty good in many cases... but it's hard to do in some of those small rooms.

This is why every time we have setup different speakers in my room, the imaging improves with equalization. Although my room is nearly symmetrical in all aspects, the left and right speakers are not always frequency matched. We've actually had some that were off pretty bad. EQ fixed it and in every case improved the imaging, sound stage and depth acuity. There is absolutely no way I could live without my Audyssey and/or Dirac Live. Gotta have it!

Sonnie: Have you heard a performance difference between Audyssey and Dirac Live? I have no experience with the latter... curious to hear what you have to say!

 

[Leonard Caillouet]

The best way to overcome the room is measurements, treatments, EQ and more measurements. All rooms/speakers can benefit from this.

I would say that ideally one should start with placement, then measure, treatments, and EQ. Use what the system gives your for free (system including the room) and do the least amount of treatment and EQ necessary. I agree with Sonnie regarding the matching between the speakers.

 

[Sonnie Parker]

Sonnie: Have you heard a performance difference between Audyssey and Dirac Live? I have no experience with the latter... curious to hear what you have to say!

As stated previously... I can measure a difference, but it is not so easy to hear the difference. The measurements are only slightly different from about 2K on up (IIRC) ... Dirac being smoother in that area. In the lower and mid frequencies they measure close to the same. Listening... man it's hard to tell. There is not an easy way to switch back and forth... it's takes 30 minutes to swap them out and run the setup for each. They both sound good, but I would still prefer the Dirac for two-channel listening.

 

[Leonard Caillouet]

Now that is a post you won't see on many forums. Can measure the difference but can't hear it. Not picking on you Sonnie. It is just so indicative of what we are about. We are more interested in understanding what is really going on than going on endlessly about what we believe we hear and see. And the bottom line is are you happy with the listening and viewing. So much of this industry wraps around hype and blather. The bottom line is enjoying the music, and for hobbyists like us, also getting to better understand the technology.

 

[Sonnie Parker]

Good point Leonard... and my mind tells me to that I want to use Dirac over Audyssey for the front two channels, even though I may not be able to notice any significant difference. It's psychological for sure (for me)... just seeing that smoother response makes me believe it likely sounds a little more accurate.

 

[tesseract]

Good point Leonard... and my mind tells me to that I want to use Dirac over Audyssey for the front two channels, even though I may not be able to notice any significant difference. It's psychological for sure (for me)... just seeing that smoother response makes me believe it likely sounds a little more accurate.

I think the key to Dirac's success lies less with its treatment of frequency response and more with the time domain, which Dirac seems to manage better than just about anyone else.

 

[Leonard Caillouet]

For overall clarity, maybe, and for me that is the first thing that I notice about a system. But remember, we have learned that small variations between channels in balance at different frequencies can affect the image presentation.

 

[Sonnie Parker]

But remember, we have learned that small variations between channels in balance at different frequencies can affect the image presentation.

Super key point here that I think more miss than not, as evidenced by what we heard at Axpona and what we've heard at other shows in the past. Dirac or Audyssey on many of those systems would have helped, regardless of how much those setting them up might have cringed at the thought.

 

[AustinJerry]

I currently have the DDRC88-BM with Dirac Live processing my two front channels... with Audyssey handling the remaining channels for movies. I set Audyssey to bypass front L/R channels so Dirac can handle them. It works well, as you will see in the upcoming MartinLogan Expression 13A review.

Sonnie, just curious, why would you mix Audyssey and Dirac Live? Is it because you are running more than 7.1? Users I am familiar with either use a second 88A for Atmos systems, or leave the ceiling speakers without correction (after setting trims and delays). While I have never tried it, and don't want to judge, mixing room correction technologies seems like an unusual approach. I switched to Dirac after being a long-time Audyssey user, and feel it is a significantly better technology.

 

[Sonnie Parker]

Agree... I believe Dirac to be superior as well, and you just gave me an idea that I really did not think about, although I am really not mixing the two systems to a great extent since the L/R being bypassed is not equalized by Audyssey... only by Dirac. However, I am running Dirac in Stereo mode, so there is some dual eq going on with the subwoofer. It seems to work fine, although certainly not the preferred method.

I think that is a great idea to not equalize the Atmos speakers... not that big of a deal anyway really. At least that will suffice until I get the Emotiva XMC-1 when it's ready for Atmos and DHCP 2.2.

Btw... welcome to AV NIRVANA!

 

[AustinJerry]

Agree... I believe Dirac to be superior as well, and you just gave me an idea that I really did not think about, although I am really not mixing the two systems to a great extent since the L/R being bypassed is not equalized by Audyssey... only by Dirac. However, I am running Dirac in Stereo mode, so there is some dual eq going on with the subwoofer. It seems to work fine, although certainly not the preferred method.

I think that is a great idea to not equalize the Atmos speakers... not that big of a deal anyway really. At least that will suffice until I get the Emotiva XMC-1 when it's ready for Atmos and DHCP 2.2.

Btw... welcome to AV NIRVANA!

Thanks for the welcome.

A number of people report that EQ on the ceiling speakers doesn't add much. Just run Audyssey to determine the proper trims and delays, then turn Audyssey off leaving the trims and delays configured for the ceilings. Zero out trims and delays for the ground-level speakers, and then utilize the full 7.1 capabilities of Dirac Live. As noted, this avoids conflicts on the sub channel, and utilizes Dirac on the center channel. Dirac on my center channel dies something magical--you really should try it. Good luck!

 

[Todd Anderson]

As stated previously... I can measure a difference, but it is not so easy to hear the difference. The measurements are only slightly different from about 2K on up (IIRC) ... Dirac being smoother in that area. In the lower and mid frequencies they measure close to the same. Listening... man it's hard to tell. There is not an easy way to switch back and forth... it's takes 30 minutes to swap them out and run the setup for each. They both sound good, but I would still prefer the Dirac for two-channel listening.

It's funny you say that... doing an A/B comparison is never an easy task. I always find myself keying-in on something that (in real time) sounds unforgettable. But the second there's a pause, small delay, or switch, it's very hard to discern a difference.

 

[AustinJerry]

It's funny you say that... doing an A/B comparison is never an easy task. I always find myself keying-in on something that (in real time) sounds unforgettable. But the second there's a pause, small delay, or switch, it's very hard to discern a difference.

I agree that A/B comparisons are extremely difficult. However, I would like to point out that the MiniDSP 88A has four memory presets that can be used to store four different Dirac calibrations. Switching back and forth between presets can be accomplished using a remote control, and occurs in ~2 seconds, close enough to compare the two preset effectively. Add to that the powerful custom curve editor in Dirac Live, and the configurability of the bass management plug-in on the 88A, you end up with a room correction system much more versatile than Audyssey. I admit I have not evaluated the new Audyssey mobile app that allows new configurability options (it is not supported on my Marantz 8802A), which may narrow the configurability gap.

 

[AudiocRaver]

YES! In rooms that are absolutely awful, full of extra-live reflections, then auto eq can make things worse, but rooms that are that bad should be shunned as listening areas anyway.

I agree with Sonnie, especially where steps have been taken for best setup possible before EQ, then the EQ helps polish and match the two main speakers. Imaging is invariably improved.

My system sounds and measures pretty good without EQ. Dirac put the icing on the cake and made it sound great.

Dennis's room is a great example. It sounds very good without Dirac, but when he kicks it on, it sounds even better.

And then there are those who also are beginning to suggest more and more that we don't need acoustic treatments... reflections (even first) can be a good thing. And ... AND... we now have manufacturers making AMPS that can help eliminate room modes. Brilliant perhaps.

It's close as far as how it sounds in my room. The measurements look better with Dirac, but it's not easy to tell a difference between the two.

Early reflections CAN be used to advantage, if understood and handled carefully. Fine detail and depth acuity are definitely better with Dirac.

A number of people report that EQ on the ceiling speakers doesn't add much. Just run Audyssey to determine the proper trims and delays, then turn Audyssey off leaving the trims and delays configured for the ceilings. Zero out trims and delays for the ground-level speakers, and then utilize the full 7.1 capabilities of Dirac Live. As noted, this avoids conflicts on the sub channel, and utilizes Dirac on the center channel. Dirac on my center channel dies something magical--you really should try it. Good luck!

Thanks for joining the conversation, Jerry, and welcome.

Great tip on the ceiling speakers not needing EQ so much. That makes the DDRC-88BM into potentially the ideal device for even bigger (11.2) ATMOS systems.