Like Gene always says, don't trust AutoEQ...

[Byron Flagg]

I recently upgraded my home theater with new LCR speakers, subwoofer, and receiver:

NHT C-3
Rythmik F-12
Onkyo TX-NR747

After setting everything up, including running the AutoEQ function, I started listening to music and movies and while I was very satisfied with the upgrade in terms of overall clarity and fidelity, something was wrong. The audio seemed to lack "punch" or "impact". Intense and climactic movie scenes just didn't seem to intense and sick EDM beat drops just didn't do it for me anymore. My first thought was the new subwoofer but I quickly realized that the energy was missing from frequencies higher than the 80Hz crossover I had used. Realizing that the AutoEQ created during the auto setup could be to blame, I grabbed my Smaart rig and decided to capture a before and after as I turned off the AutoEQ and just ran flat.

First I captured response and coherence curves with the system as-is with AutoEQ engaged. I took measurements at three points across my primary listening area and averaged the results. Then I turned off the AutoEQ and repeated the process to generate a new response curve.

I was running pink noise through the system at 90dB when I first turned off the AutoEQ and subjectively I felt it made a significant improvement in the mid-low frequency range 100-200Hz. Sure enough the response curves confirmed that the AutoEQ had made few changes, except to pull out a massive amount of energy from 80Hz to 225Hz. See the images below: Purple = Before. White = After

Clearly there are still some acoustic fluctuations, but the AutoEQ had responded by pulling all the energy out of this band which just sounded awful. As a secondary repercussion of the AutoEQ, the subwoofer level was set about 3dB too low as well. In the second screenshot below you can see the before and after response curves after I turned off AutoEQ and readjusted the subwoofer level. Purple = Before. White = After

As for what could have caused the AutoEQ to fail so badly, I can't be sure. The theater space is well treated acoustically with short (< 1sec) reverberation times from 20-20kHz. From the response curves you can see the clearly there are some acoustic effects, but that's the whole point of AutoEQ so that's no excuse. What you can't see, because the coherence curves don't capture in my old version of Smaart, is that there is a significant acoustic "issue" at 250Hz where the coherence drops out completely and you can see a corresponding drop in the response curve. I don't know what that is, but it will have to wait for another day.

So please learn from my recent experience that confirms what Gene says all the time. AutoEQ doesn't always work well and can be worse than just running flat. I'm using a mid range receiver in a well controlled acoustic environment and it STILL caused problems. Never again.

I'm also toying with the idea of dealing with the "presence peak" that starts building at 10kHz. That data confirms my subjective feeling that the sound is a bit bright and can be fatiguing during action movies or high volume EDM sessions. But I'm going to start by making some adjustments to my acoustic panels which are no longer positioned correctly to mitigate first reflections. I'm hesitant to use the coarse EQ available in the receiver as it may do more harm than good.

Tonight is movie night so I'll giving the system a good long subjective evaluation.

 

[PENG]

So please learn from my recent experience that confirms what Gene says all the time. AutoEQ doesn't always work well and can be worse than just running flat. I'm using a mid range receiver in a well controlled acoustic environment and it STILL caused problems. Never again.

I don't know about Onkyo's but Audyssey XT and XT32 does work well for me. As you said, ..doesn't always work well.... "always" is the key word.

 

[lovinthehd]

AccuEQ isn't all that well regarded, probably for a reason. I've had great results using Audyssey myself like Peng. I don't trust it, but I've yet to beat it with manual controls.

Otoh why is your sub where it is? Why using such small speakers? EQ can only do so much....

 

[PENG]

I posted this several times before but here it is again just for comparison. Red is with Audyssey off. Aside from a dip at 50 Hz that I could reduce by adjusting manually at the subs, I can't really complain.

Again, I am sure REQ, including Audyssey and others don't always work for everyone. I do believe those that have been around for a long time, such as Anthem ARC, Dirac, Audyssey etc., could make some improvements for most people who take the time to do it according to the provider's instructions.

 

[Byron Flagg]

AccuEQ isn't all that well regarded, probably for a reason. I've had great results using Audyssey myself like Peng. I don't trust it, but I've yet to beat it with manual controls.

Otoh why is your sub where it is? Why using such small speakers? EQ can only do so much....

The subwoofer is located as a nominal starting point on the centerline of the room which is also the centerline of the screen and speaker layout. I did this with a mind to add additional subwoofers on the wall midpoints and visually it makes for a nice symmetrical setup. My initial testing and measurements indicate that I'm getting good flat response from the subwoofer, where it is, that extends well below 20Hz, so I may not end up proceeding with any further bass optimization for this system.

I'm using such small speakers for a couple reasons. One, while it looks like this is a dedicated home theater space, it isn't. The space is also used for video based exercise and as a play space for my two young children. So I wanted to minimize the amount of stuff sitting on the floor to maximize space for exercise and to minimize the potential for damage. I have a pair of floor standing speakers that have been tipped over several times and are also now full of little plastic toys and dried up PB&J sandwiches that were forced into the bass ports by my curious children. Two, this setup is an experiment in a fully acoustically suspended system. All three front speakers are identical and sealed, as is the subwoofer. After reading so much about the superior fidelity of acoustically suspended speakers, I wanted to hear it for myself. For the sake of living life to its fullest, my next setup will probably be fully ported. The NHT C-3 response extends well below my 80Hz crossover. Testing and measurements confirm this.

Of course I'm curious whether the additional drivers possible in floorstanding speakers or larger LCR wall mount speakers would provide improved transient response. I'm interested to hear people's opinions (data is even better) regarding how larger floorstanding speakers would be superior to smaller bookshelf speakers when used in combination with subwoofers.

 

[lovinthehd]

The subwoofer is located as a nominal starting point on the centerline of the room which is also the centerline of the screen and speaker layout. I did this with a mind to add additional subwoofers on the wall midpoints and visually it makes for a nice symmetrical setup. My initial testing and measurements indicate that I'm getting good flat response from the subwoofer, where it is, that extends well below 20Hz, so I may not end up proceeding with any further bass optimization for this system.

I'm using such small speakers for a couple reasons. One, while it looks like this is a dedicated home theater space, it isn't. The space is also used for video based exercise and as a play space for my two young children. So I wanted to minimize the amount of stuff sitting on the floor to maximize space for exercise and to minimize the potential for damage. I have a pair of floor standing speakers that have been tipped over several times and are also now full of little plastic toys and dried up PB&J sandwiches that were forced into the bass ports by my curious children. Two, this setup is an experiment in a fully acoustically suspended system. All three front speakers are identical and sealed, as is the subwoofer. After reading so much about the superior fidelity of acoustically suspended speakers, I wanted to hear it for myself. For the sake of living life to its fullest, my next setup will probably be fully ported. The NHT C-3 response extends well below my 80Hz crossover. Testing and measurements confirm this.

Of course I'm curious whether the additional drivers possible in floorstanding speakers or larger LCR wall mount speakers would provide improved transient response. I'm interested to hear people's opinions (data is even better) regarding how larger floorstanding speakers would be superior to smaller bookshelf speakers when used in combination with subwoofers.

Makes more sense if you're adding subs along each midwall, thanks. Figured you'd measured but thought I'd ask anyways....

Just curious about the speaker sizes as to what the reasoning is, thanks. FWIW sealed and acoustically suspended aren't quite the same thing.

I was thinking you'd just like larger speakers than smaller for the music you mentioned. A bookshelf with a decent sized driver I don't think is at a disadvantage to a similarly sized driver in a tower speaker when using subs and the tower can't be as easy to remove from the kids....

 

[highfigh]

I recently upgraded my home theater with new LCR speakers, subwoofer, and receiver:

NHT C-3
Rythmik F-12
Onkyo TX-NR747

After setting everything up, including running the AutoEQ function, I started listening to music and movies and while I was very satisfied with the upgrade in terms of overall clarity and fidelity, something was wrong. The audio seemed to lack "punch" or "impact". Intense and climactic movie scenes just didn't seem to intense and sick EDM beat drops just didn't do it for me anymore. My first thought was the new subwoofer but I quickly realized that the energy was missing from frequencies higher than the 80Hz crossover I had used. Realizing that the AutoEQ created during the auto setup could be to blame, I grabbed my Smaart rig and decided to capture a before and after as I turned off the AutoEQ and just ran flat.

First I captured response and coherence curves with the system as-is with AutoEQ engaged. I took measurements at three points across my primary listening area and averaged the results. Then I turned off the AutoEQ and repeated the process to generate a new response curve.

I was running pink noise through the system at 90dB when I first turned off the AutoEQ and subjectively I felt it made a significant improvement in the mid-low frequency range 100-200Hz. Sure enough the response curves confirmed that the AutoEQ had made few changes, except to pull out a massive amount of energy from 80Hz to 225Hz. See the images below: Purple = Before. White = After

Clearly there are still some acoustic fluctuations, but the AutoEQ had responded by pulling all the energy out of this band which just sounded awful. As a secondary repercussion of the AutoEQ, the subwoofer level was set about 3dB too low as well. In the second screenshot below you can see the before and after response curves after I turned off AutoEQ and readjusted the subwoofer level. Purple = Before. White = After

As for what could have caused the AutoEQ to fail so badly, I can't be sure. The theater space is well treated acoustically with short (< 1sec) reverberation times from 20-20kHz. From the response curves you can see the clearly there are some acoustic effects, but that's the whole point of AutoEQ so that's no excuse. What you can't see, because the coherence curves don't capture in my old version of Smaart, is that there is a significant acoustic "issue" at 250Hz where the coherence drops out completely and you can see a corresponding drop in the response curve. I don't know what that is, but it will have to wait for another day.

So please learn from my recent experience that confirms what Gene says all the time. AutoEQ doesn't always work well and can be worse than just running flat. I'm using a mid range receiver in a well controlled acoustic environment and it STILL caused problems. Never again.

I'm also toying with the idea of dealing with the "presence peak" that starts building at 10kHz. That data confirms my subjective feeling that the sound is a bit bright and can be fatiguing during action movies or high volume EDM sessions. But I'm going to start by making some adjustments to my acoustic panels which are no longer positioned correctly to mitigate first reflections. I'm hesitant to use the coarse EQ available in the receiver as it may do more harm than good.

Tonight is movie night so I'll giving the system a good long subjective evaluation.

In acoustics, when stereo or multi-channel audio, symmetry is your friend and your room isn't symmetric. This may be where your issues originate. You have one front corner treated with the panels close to the left speaker and the right side has a space to the side that isn't treated. While this is probably OK for mid-higher frequencies, it doesn't help the low frequency energy and the difference between the two ranges is: higher audio frequencies and light radiate and reflect in a similar manner- the sound impacts a flat, smooth reflective surface and, as in optics, the angle of incidence = angle of reflection. A rough surface scatters the energy and an absorptive surface reduces the energy in the wavelengths where it's structure allows it to. Your panels would need to be much thicker than what's showing, in order to affect the low frequencies. It's also possible that placing the mic in the center of the seating area is the sole factor in the response anomalies shown in your graphics. You have panels immediately to the left of the seating and nothing to the right- the effects of the treatments may look good on paper, but the perceived effect will be different. The absorption at the left A) has more of an affect than equal treatment of the right wall and B) equal treatment of the right wall + the different distance means the right side response has been affected more than the left.

Center the mic and test it again and I think you'll see a much different response.

There- I fixed it.

 

[Byron Flagg]

Thank you to everyone who has provided input so far. As I mentioned in my original post, that evening I watched the new Tarzan movie and have been doing some additional subjective listening tests with music to further explore this issue.

While turning off the Accu-EQ did make a significant improvement in the mid-low frequency response and restored a significant amount of "warmth" to the sound, the system still lacks any tactile bass performance. During the Tarzan movie, there were several scenes that involved stampedes of large animals as well as cannon blasts. During these scenes I could hear the walls shaking sympathetically from the LFE content, yet I felt no "chest thump" or any direct tactile experience of the bass. Most pointedly, the cannon blast was reproduced with sufficient bass energy to rattle the windows, but I felt absolutely nothing at the listening position.

My Smaart measurements confirm strong system response well below 20Hz and although there is some fluctuations in the 50Hz to 200Hz range that are likely acoustic in nature, I'm not convinced that acoustics alone can explain why I feel no tactile bass while sitting 8 feet away from a 12" subwoofer driver while the house around me shakes its bones. As a reminder, the response curves shown above are actually an average of three points spanning my primary listening area so the data should be free of highly localized phenomena. I've also moved around the room extensively in search of a spot where the transients are stronger and can find none. Even kneeling on the ground directly in front of the 12" subwoofer I can definitely hear very low frequency bass content, but not feel any tactile transients in my body.

As further information for your consideration, the current setup is replacing two BOSE 701 floor standing speakers that I purchased in the mid 1990's and with which I used no subwoofers, just the 701's in full range mode. While the overall fidelity and high frequency clarity of the NHT bookshelves are vastly superior to the 701s and while the 701s response dropped off below 40Hz, the Rythmik sub is going strong well below 20Hz. Yet somewhere in the middle, the tactile transient response is lacking from the new system. Even without a dedicated subwoofer, the BOSE 701s provided much better chest thumps than the new system. When something exploded in a movie, I could feel the blast wave in my chest, arms, and legs. Now I can raise the listening volume to the threshold of pain and still not feel anything in my body even during the most LFE intensive action movie scenes.

I'm beginning to suspect that my selection of acoustic suspension speakers and subwoofer, while it did provide the fidelity I was looking for, is not capable producing the sharp, low frequency transients necessary to provide a tactile experience. What I don't know is whether those missing transients are above or below the 80Hz crossover, or in other words whether I'm looking at the possibility of changing out the subwoofer or the bookshelves.

Any other thoughts as to why the new system produces every subtlety of a gentle whisper, and at the same time can shake the walls and rattle the windows, yet it cannot generate any tactile sensations at the listening position or anywhere I've tried standing in the room?

 

[lovinthehd]

The range you're looking for is probably somewhere in the range of 50-150hz for that chest thump from various things I've read. Some resort to a mid bass module or maybe you can add something like a minidsp for more eq'd response in that range.

 

[rnatalli]

Anthem ARC never lets me down and I've had good luck with Dirac and Audyssey XT32 as these 3 seems to produce consistent results. Here's my most recent ARC subwoofer measurement.