Audyssey MultEQ Room Correction Interview With Chris Kyriakakis

[AcuDefTechGuy]

I highly doubt any classical listener would want his system equalized to flat at the listening position.

As Gene mentioned and as the Audyssey Reference Curve showed, there is a roll-off after 4kHz. So Audyssey does NOT EQ to FLAT after 4kHz.

The exception is when people use Audyssey FLAT + Dynamic EQ.

So if you use just Audyssey Reference curve, there is a roll-off in the treble.

 

[TLS Guy]

As Gene mentioned and as the Audyssey Reference Curve showed, there is a roll-off after 4kHz. So Audyssey does NOT EQ to FLAT after 4kHz.

The exception is when people use Audyssey FLAT + Dynamic EQ.

So if you use just Audyssey Reference curve, there is a roll-off in the treble.

In any event for me raising the HF is a problem. Part of this may be the source being listened to. However it is a one size fits all in the domestic versions. The one size fits all is a problem, because of the varying dispersion patterns of speakers. If the speakers have nice wide optimal dispersion patterns, the reflected sound proportion is increased and the microphone will detect greater HF losses. However these speakers still sound balanced as long as their near field response is flat. A speaker that beams, like a Klipsch, will have greater HF at the testing position.

I don't know what speakers were used in listening sessions, but it likely had an impact.

I suspect this is also part of the reason for the over correction in my system.

My in room balance less Audyssey is excellent. All Audyssey curves downgrade the performance. That goes for two of my systems. It is the HF correction I object to and it needs looking at, or there needs to be a way for customer adjustment on all versions.

My hunch is that speakers with good well balanced sound, that measure well, do not need this type of program, except in the bass in some rooms.

 

[Irvrobinson]

You mean this curve?

This curve is used to equalize studio speakers to the listening position. That is pretty much the curve you will get at the listening position from speakers that measure well. That is my point, Audyssey corrects it, and I maintain it shouldn't. But if what you like is flat at the listening position go ahead.

That curve is almost exactly what I shoot for at my listening seat. I arrived at it by using recordings I made in my listening room, and trying to make them sound as realistic as possible. And here I thought I was running my bass octaves just a little hot; it looks like I'm pretty much on target.

 

[TLS Guy]

The graph I posted shows that Audyssey curve is down about -6dB @ 10kHz (~69dB, Reference is 75dB). At 20kHz, it is down -10dB. I assume it varies with speakers/rooms?

Green is the Audyssey Curve, Reference level is 75dB:


But yeah, I think all the Audyssey graphs I've seen showed a roll-off in the treble, unless you use Audyssey Flat + Dynamic EQ.

Here is one example, where 70dB is the reference level. Purple graph is FLAT + DEQ, Green graph is FLAT only.

Thanks for posting those curves. Uncorrected HF is down 10db from a reference level at 500 HZ. That in my view is where it should be. Audyssey should not have lifted the HF. Those are the db corrections to all speakers except the rear backs in my studio. They are not an improvement. I pretty certain the rear backs were not corrected due to proximity to three of the testing positions. In addition the speaker lobing tilt of those speakers is right down to the microphone.

This raises another issue of listening distance from the speakers. If you are further away, then Audyssey will provide more lift, unless it does compensate for distance. Again for natural sound there is no data that the HF loss because of the greater proportion of reflected sound should be equalized. I haven't checked to see if there is data on this, but I do not know of any. I doubt it would be the right thing to do, as we are conditioned to expect what takes place. This is almost certainly a factor in the preference for wide dispersion speakers in tests.

 

[ira]

This raises another issue of listening distance from the speakers. If you are further away, then Audyssey will provide more lift, unless it does compensate for distance.

I'm lost: can you explain in detail what you mean by this quote? I'm familiar that SPLs drop with distance in a frequency-independent manner, but I'm unfamiliar with any frequency-dependent variability over distance in air's ability to transmit sound that needs DRC to accommodate.

 

[TLS Guy]

I'm lost: can you explain in detail what you mean by this quote? I'm familiar that SPLs drop with distance in a frequency-independent manner, but I'm unfamiliar with any frequency-dependent variability over distance in air's ability to transmit sound that needs DRC to accommodate.

As you move from the source, the balance of the received sound shifts in favor of reflected sound versus direct sound. In the reflected envelope HF is preferentially attenuated.

This accounts for progressive HF loss as you move away from the source. This is not something that should be compensated for, and if you do it will sound unnatural. If you want it in plain language it sounds awful.

 

[AcuDefTechGuy]

This is not something that should be compensated for, and if you do it will sound unnatural. If you want it in plain language it sounds awful.

Again, this is not a matter of FACT. It is purely a preference regardless of which music you listen to.

I listen to both classical music and rock/pop/alternative music. I prefer the HF & LF increased a little at the listening position.

I think the best room correction software should offer a variety of PREFERENCES, not just ONE.

As Gene stated, Audyssey starts the roll-off after 4kHz. This is the Reference or default curve. Perhaps most people prefer this curve.

But in addition, Audyssey also offers a FLAT curve which does not roll off the treble. Some people prefer this curve.

People can choose whichever they like. It's not carved in stone and black-and-white. No point debating which sounds more natural or best.

 

[cpp]

I like this quote from James Tanner of Bryston but that doesn't mean I agree

I can tell you now though that I will never build in Room E.Q. into any of our products. After learning as much as I have in developing the Model T Bryston speakers I am more convinced than ever that it is simply the wrong approach.

james

The problem with EQ is when you listen to a speaker you are listening to the POWER RESPONSE of the speaker not just the ON AXIS response. The power response is all the direct and reflected energy in the room mixed together and when you change the on-axis response using EQ you totally disrupt the balance between the power response and the direct on axis response of the speaker that the engineers spent countless hours making as wide and even as possible.

james

 

[TLS Guy]

I like this quote from James Tanner of Bryston

I could not agree more. However, I would not discount the fact, that Room Eq might improve some poorer speakers. You can't hide from the fact that most speakers are not very good.

This approach, in my view, can only make good speakers sound bad. I don't discount the fact it might improve a few weaker ones by chance.

 

[cpp]

I could not agree more. However, I would not discount the fact, that Room Eq might improve some poorer speakers. You can't hide from the fact that most speakers are not very good.

This approach, in my view, can only make good speakers sound bad. I don't discount the fact it might improve a few weaker ones by chance.

+1, some poorly designed speaker could benefit from some EQ. But even with some good speakers, the room can kill a good design.

 

[AcuDefTechGuy]

It's really really basic and simple. It doesn't even require impressive knowledge of audio. Try the Room EQ. If it sounds better than Pure Direct, then use the EQ. If they sound worse than Pure Direct, don't use the EQ.

 

[AcuDefTechGuy]

I like this quote from James Tanner of Bryston but that doesn't mean I agree:


I can tell you now though that I will never build in Room E.Q. into any of our products. After learning as much as I have in developing the Model T Bryston speakers I am more convinced than ever that it is simply the wrong approach.

The problem with EQ is when you listen to a speaker you are listening to the POWER RESPONSE of the speaker not just the ON AXIS response. The power response is all the direct and reflected energy in the room mixed together and when you change the on-axis response using EQ you totally disrupt the balance between the power response and the direct on axis response of the speaker that the engineers spent countless hours making as wide and even as possible.

Translation: We don't want to pay anymore money because then we would have to charge $14K for our pre-pro, instead of just $10K. It doesn't matter how many people have used room corrections and love it.

 

[cpp]

Translation: We don't want to pay anymore money because then we would have to charge $14K for our pre-pro, instead of just $10K. It doesn't matter how many people have used room corrections and love it.

That should read, we don't want to design EQ into our products because I ( james) would have to eat my words about eq along with charging $$ for the software in our Sp3 pre-pro

 

[AcuDefTechGuy]

That should read, we don't want to design EQ into our products because I ( james) would have to eat my words about eq along with charging $$ for the software in our Sp3 pre-pro

Design?

Like the way they designed their speakers. Not.

If Bryston ever implement RC, the only thing they would do is add one of the existing RC (Audyssey, ARC, Trinnov, Dirac, Lyngdorf, ect.).

Bottom line is, everyone has a different opinion on this subject. Some people find that certain RC improve the sound. Some people find that RC worsen the sound. It doesn't matter if you are Bryston, ATC, Harman, or anyone else.

You use whatever makes the sound of your system better regardless of what the so-called experts think.

 

[cpp]

Design?

Like the way they designed their speakers. Not.

If Bryston ever implement RC, the only thing they would do is add one of the existing RC (Audyssey, ARC, Trinnov, Dirac, Lyngdorf, ect.).

Bottom line is, everyone has a different opinion on this subject. Some people find that certain RC improve the sound. Some people find that RC worsen the sound. It doesn't matter if you are Bryston, ATC, Harman, or anyone else.

You use whatever makes the sound of your system better regardless of what the so-called experts think.

Now we don't want to get into the speakers which few have heard any where around where I live like Fl or SC.

True, when I listen to music via my Denon, it's pure direct. Movies I might kick on Audyssey depending on the movie.

 

[AcuDefTechGuy]

True, when I listen to music via my Denon, it's pure direct. Movies I might kick on Audyssey depending on the movie.

I was singing the room-correction-is-bad song for about 15 years and adamantly refused to listen to anything other than Pure Direct for both music and movies until one day I turned on Dynamic EQ. And in an instant, I changed my tune.

 

[jliedeka]

I really don't use pure direct. It takes away the subwoofer and bass management. Only some types of music sound good that way so it's not worth it to me. I don't think quieting half the circuitry in my AVR makes any real world difference to sound quality. But then, my speaker wires aren't danceable so what do I know.

I have an older Audyssey EQ and I think, on balance, it helps. I'd have to look to see which profile I use. My ears are used to whatever it is.

Jim

 

[TLS Guy]

I was singing the room-correction-is-bad song for about 15 years and adamantly refused to listen to anything other than Pure Direct for both music and movies until one day I turned on Dynamic EQ. And in an instant, I changed my tune.

What you really discovered were the old Fletcher/Munson curves of the old loudness button.

That was a bad idea also.

 

[AcuDefTechGuy]

What you really discovered were the old Fletcher/Munson curves of the old loudness button.

That was a bad idea also.

I think you're confusing Dynamic Volume with Dynamic EQ. You should read about the differences over at the Audyssey forum by Chris Kyriakakis.

It's not about increasing the levels. For example, I could increase the subwoofer channel levels to be 12dB hot and the bass would still not sound as punchy, dynamic, and musical as when I turn on Dynamic EQ.

Dynamic Volume basically just increases all the low levels like dynamic compression. I don't like that. But some people have use for Dynamic Volume with certain situations so I won't presume to tell people that what they are doing is a bad idea.

 

[TLS Guy]

What you really discovered were the old Fletcher/Munson curves of the old loudness button.

That was a bad idea also.

They are similar as they both raise both ends of the frequency spectrum. Dynamic volume and the old loudness control just increase it as the volume was reduced. Neither have anything to recommend them.