Audyssey Room EQ Review

[Dj7675]

I believe the flat lifeless default curve audyssey uses is intentional, as to force people into using Dynamic EQ to get a listenable system

It had been years since I had turn on the flat curve. My oh my... my ears did not agree with it. Bright and light on bass.

 

[rxp]

DynEQ also messes with the rear/surround speaker levels.

Only this last week did I manage to get my bass back to how I wanted it - I used to just increase the subwoofer level, which obviously only boosted low bass. Now I get using Ratbussy to get a house curve in there.

Since disabling DynEQ I'm noticing all sorts of good surround tracks in TV shows and the height channels are way more noticeable on upmixing and native content.

 

[Sal1950]

I believe the flat lifeless default curve audyssey uses is intentional, as to force people into using Dynamic EQ to get a listenable system

That's just silly, what would they gain, you don't pay to use DEQ

DynEQ also messes with the rear/surround speaker levels.

That's true and I have no idea why. When properly setup DEQ would have the capability to be a valuable feature if it wasn't for this.
You have to know what your doing to get the software to really work well. A tool that's still in development, though I believe that's true for all the other DRC programs. I believe in another 10-15 years these programs will really come into their own. Unfortunately I'll be dead by then. LOL

 

[rxp]

On Ask Audssey the creator said they had some research to show that listeners had a hard time localizing sound from the rear at lower volumes. It sounds like a good idea in principle but they've not updated it with the addition of height channels. So now the rears are out of whack with the heights.

I find DynEQ is a really simple idea. My dad's stereo amp from the 90s had loudness compensation! All we need is a simple loudness compensation algorithm. Same with all this room correction stuff. Humans don't hear any of these dips and tiny peaks in the bandwidth without significant training. All we need is tone controls like Toole is always saying.

Heh: what do you know, since I disabled DynEQ tone controls is now available in both my SR6011 and SR6012. The manual actually says they're available: http://manuals.marantz.com/SR6011/EU/EN/SEHFSYrsdmuabm.php . No need for damn house curves on the app anymore!!

 

[Matt S]

I believe the flat lifeless default curve audyssey uses is intentional, as to force people into using Dynamic EQ to get a listenable system

Tested and tested .. far too many times with various settings. I prefer flat to reference. Voices are clearer. I guess it depends on your room and other factors. I definitely do NOT have too little bass and its not particularly bright either. Dynamic EQ gives me too much bass so I have thst switched off.

My brother has lent me his miniDSP to test later. I cant wait to hear the difference between Audyssey and Dirac.

 

[Kal Rubinson]

I believe the flat lifeless default curve audyssey uses is intentional, as to force people into using Dynamic EQ to get a listenable system

Since that curve came before DEQ, you might say that it was invented to compensate for "the flat lifeless default curve."

 

[davidc]

I'd like to share my experience with Audyssey MultEQ XT32. I started using MultEQ XT32 back in 2013 when I pudchased a Marantz AV8801 processor. My small listening room, in effect my living room, measures 13 x 13 x 8 feet (4 x 4 x 2,4 m). Despite the 13 channels available on the AV8801, because of the limited space I have and the right side of the room flanked by a pair of windows and a door leading to the balcony, I only really have space for a 5.1 multi-channel setup although I'd like to run a dual subwoofer configuration sometime in the future.

The goal for the acquisition of the AV8801 at the time was to have an up to date AV processor that sounded great with my growing collection of multi-channel DVD-Audio, SACD, and later on Blu-ray audio along with support for the new Blu-ray losless audio codecs (Dolby True-HD & DTS Master-Audio) and the addition of Audyssey's MultEQ XT32 digital room correction to help reduce room modes.

Before the use of Audyssey, I had horrible room modes where the bass resonated way too much for my taste. You could really tell that it muffled much of the sound from the midrange. Audyssey really did a great job of taming the room modes and opening up the sound letting through more midrange detail. I mostly favored the Audyssey flat curve.

However, through the years and with repeated critical listening tests with Audyssey active and off on the same tracks, it became clear to me that there was a side effect that I was hearing and that I had indeed heard before but that I had pushed to the back of my subconscious mind since the results in terms of reducing the room modes were so successful. This side effect that had resurfaced from the inner depths of my mind and that I could no longer ignore is a compression of the soundstage. It's a clear reduction in the width of the stereo image and also of the depth of the soundstage. Many musical instruments also seem to be somewhat slightly elevated vertically. Later on after some searching online, I found a couple of Audyssey users on other forums that reported observing a similar compression of the soundstage.

When a good audiophile friend of mine would come over to listen to music, he was never too convinced with my demos of Audyssey althoughh he couldn't quite put is finger on what didn't sit well with him in spite of the clear reduction of room modes. He would always ask me to turn off XT32. A few years later on one occasion of critical listening to some new recordings, I pointed out to him what I had come to realize or rather accept. We listened to a few tracks with Audyssey both on and off and he agreed with my observations.

I haven't given up on finding a DRC solution that will again tame those foom modes without affecting the soundstage. I want to try Dirac Live on a future HTPC build project as it appears to have become a much preferred solution in the last few years. Some audio companies have like NAD and Wisdom Audio have actually ditched Audyssey in favor of Dirac over the last few years. One characteristic that seems to be superior with Dirac versus Audyssey is the improved impulse response allegedly made possible by their particular use of mixed FIR and IIR filters or so they claim. Measurements by a few users on some forums seem to lend some credence to such claims like the one posted by AustinJerry's measurements posted on AVS Forum. According to the Dirac white paper that has been pointed out by a couple of previous posters on this thread, "ON ROOM CORRECTION AND EQUALIZATION OF SOUND SYSTEMS" , Mathias Johansson of Dirac Research makes the following assertion:



I can't say for sure that the phenomenon related in the excerpt above is the exact explanation of what I'm observing with respect to the soundstage compression but it may be a possible cause.

In the last few years I use Audyssey on movies but when I sit down to listen to music, I turn it off and live with the room modes. I actually prefer the sound coming out of my Oppo UDP-205 straight into my Parasound A21 amp and their little brothers. The Oppo's DACs and analog outputs walk all over the AV8801's TI PCM1795 DACs and much overhyped Marantz HDAM analog section. Furthermore, in light of Amir's review of Marantz's latest but lackluster AV8805 measurements sporting the much newer and arguably better AKM DACs with improved HDAM modules versus my now aging AV8801, it follows naturally from the Oppo UDP-205's excellent measurements that it sounds better.

Damn! this little hobby becomes expensive over the years.

It's interesting how people can get varied results with Audyssey. I'm assuming you have tried running Audyssey setup again? Have you used REW to perhaps make small changes in speaker positioning to improve response as much as possible and then run Audyssey again?

 

[davidc]

My experience with soundstage has been the opposite. We all know that the variables are infinite.

Last night I literally heard a creaking noise coming from my lap...had to pause the music to make sure. The sound was that three dimensional.

I didn’t A/B that...yet, but I’ve never heard my music sound better, and Audyssey EQ is a big part of the equation for ME.

What was in your lap?

 

[Kal Rubinson]

It's interesting how people can get varied results with Audyssey. I'm assuming you have tried running Audyssey setup again? Have you used REW to perhaps make small changes in speaker positioning to improve response as much as possible and then run Audyssey again?

This phenomenon is not limited to Audyssey. Here is a statement from one of its competitors but which applies to all of its ilk when not supplemented by other operations:
"................., those poor results are most probably related to each subwoofer and the main speakers being measured and corrected individually. As a result, their respective interactions/phase cancellations are unpredictable and can produce the unpleasant result that you mention."

 

[davidc]

I found DEQ to produce a bit more bass than I liked. The curve I created was more natural. Taste definitely plays a role here so there is no right answer to a precise point.

In the old days, all manufactures offered a "Loudness" button that was supposed to follow the Fletcher-Munson curve. But...who's old enough and has a good enough memory to remember than Marantz also offered a continuous gradual control to decrease the effect of the Loudness button also

 

[thewas]

Many current Yamaha stereo amps still have a nice variable loudness functionality where you are supposed to set the a loud listening level with the volume knob and then only use the loudness knob to adjust to your desired listening level.

 

[flaviowolff]

That's just silly, what would they gain, you don't pay to use DEQ

But they want you to keep using the ecosystem, and that means making it impactful to the average consumer.

I believe DEQ is a big selling point for audyssey. “The ability to hear clear bass at low volume”. That sells. Engage DEQ, and the bass becomes very nice (best at 10db offset for most content IMO). DEQ is also engaged by default after setting audyssey up, at least on the two avrs I tested.

And DEQ would not work very well if the base curve weren’t flattish at the bass. The boosts would be too overpowering if the default curve were like Dirac’s, unless they changed it to compensate, ofc, but I believe they keep the base curves (both reference and flat) bass light for it to work seamlessly with dynamic eq, and get people disappointed when they turn DEQ off and think to themselves “I need DEQ, it really improves sound. I won’t use a receiver without it never again”.

or do you think they are just dumb and believe everyone loves flat bass?

 

[rxp]

Review/Comparison of level settings for my Marantz SR6012/XT32 using a built in Mic vs UMIK-1 with calibration file loaded and REW's SPL meter (slow, C weighting).

After a recent re-calibration my wife mentioned that recently some movies we'd been watching had poor dialogue intelligibility in loud scenes. I initially thought it was the Nolam movies - a director notorious for crazy sound mixes that even lead some to clip. Then I remembered that on first setup, 3 years ago, I had funky levels too. I asked Marantz for a replacement mic in fact. So now and back then I corrected my levels using my UMIK-1 with 90 deg calibration file loaded.

Here are my results, the 3 older measurements are from my old theater with an SR6011 and it's own mic. The bottom one is from yesterday on an SR6012 with a replacement mic.

As you can see, totally different. Some channels have 4db difference. If we can't get a better mic to feed Audssey, something that will have the biggest audible difference - levels- is something it can't seem to set properly. Unless you get lucky with your mic.

Next experiment will be to point the UMIK 90 deg away from the ceiling speakers to see if there's any difference in the SPL levels REW reads.

Edit: there is a difference, it's only 0.5db. The heights run 0.5db lower when you point the mic 90 degree away from them

 

[Sal1950]

In the old days, all manufactures offered a "Loudness" button that was supposed to follow the Fletcher-Munson curve. But...who's old enough and has a good enough memory to remember than Marantz also offered a continuous gradual control to decrease the effect of the Loudness button also

The big problem with all the older "Loudness" buttons is that the Fletcher-Munson curve is based on knowing the volume level. The Marantz DEQ has a chance of being right in that it "know's" how loud the music is at X volume setting from the measured Audyssey 0 level

 

[davidc]

The big problem with all the older "Loudness" buttons is that the Fletcher-Munson curve is based on knowing the volume level. The Marantz DEQ has a chance of being right in that it "know's" how loud the music is at X volume setting from the measured Audyssey 0 level

You are so right.

I sort of left the hobby for a while and then came back. I remember when I came back in and learned that AVR's had effective room correction, and we could actually measure SPL, and we could calibrate the volume control, and we could match the actually loudness that the director intended...well, this was like the holy grail from the 60's to 00's!

 

[flaviowolff]

The big problem with all the older "Loudness" buttons is that the Fletcher-Munson curve is based on knowing the volume level. The Marantz DEQ has a chance of being right in that it "know's" how loud the music is at X volume setting from the measured Audyssey 0 level

it could be even smarter and detect the average level of the track as to adjust the DEQ offset, since music is usually louder than movies. I use streaming services with volume leveler via pc or smartphone (Spotify on normal sounded about as loud as some movies I tested), but an internal solution would be much better.

 

[John Galt]

In the old days, all manufactures offered a "Loudness" button that was supposed to follow the Fletcher-Munson curve. But...who's old enough and has a good enough memory to remember than Marantz also offered a continuous gradual control to decrease the effect of the Loudness button also

Reminds me of this support article I read at https://audyssey.zendesk.com/hc/en-us/articles/212347383-Dynamic-EQ-and-Reference-Level?page=5

”The old-style "loudness" controls on receivers simply did not work. The reasons:

1. They were based on the wrong curves. Fletcher-Munson curves were developed for telephone headsets and they are not right for speakers in a room.
2. The control was a simple bass boost tied to the volume control. A very inaccurate match to the curves (which were wrong to begin with).
3. There was no way to calibrate the input or the output so the system had no idea what reference level was.
4. The old loudness adjustment was not dependent on the content. Only the volume setting.”

 

[rxp]

Here is some more info from Audssey's creator on how Dynamic EQ works

In our research on loudness we found that human perception of loudness falls off faster behind us than it does in front. So, Dynamic EQ compensates for that using a human loudness spatial model that looks at the master volume setting and makes the appropriate adjustment to the level of the surrounds. This is designed for content where the overall surround impression should remain the same at all volume levels: i.e. surround movies and music.

Games, are a different story... The notion of "front" and "rear" changes constantly, particularly in POV games. Unfortunately, there is no way that Dynamic EQ can deal with that without a direct connection to the game engine that allows us to predict the current perspective.

The technology is working just fine and doesn't depend on the speaker type you are using. However, if your current surrounds are not pointing to the listening area it could be that MultEQ applied too much trim correction.

You can't really simulate Dynamic EQ manually. The amount of bass boost and surround compensation depends not only on the playback volume, but also changes dynamically with the content.

And another

We are aware of the ISO work, but have also done our own experiments to develop Dynamic EQ. There are some additional considerations in loudness estimation that account for the acoustics of small rooms. Furthermore, Dynamic EQ also monitors the content variations and not just the setting of the volume control. Proper loudness compensation is not a simple implementation of the ISO curves.

No wonder it's so unruly especially with height channels and side surrounds. Ruins the height channels for me. Nice idea in principle if only they'd stop the rears getting impacted.

 

[KEW]

Many current Yamaha stereo amps still have a nice variable loudness functionality where you are supposed to set the a loud listening level with the volume knob and then only use the loudness knob to adjust to your desired listening level.

... and they come with a remote that has no control of the "Loudness", only the "Volume"!
WHY?????

 

[thewas]

... and they come with a remote that has no control of the "Loudness", only the "Volume"!
WHY?????

Exactly, I guess because most upmost people can not, don't want or don't understand to use it correctly. As we know loudness was long time frowned upon and guess still is from audiophi̵l̵e̵s̵ools.