Listening Room Intelligibility Test

[garylkoh]

Art, I think you mean Doug's MTF. Mine is just a mess below 40Hz - that might be the truck outside my window. The speakers have a lower frequency response of 52Hz.

 

[dougsmith]

For fun, go to Gary's MTF, that's Modulation Transfer Function. Start on the far right. We are going to count tone burst peaks. The first peak in fron the right is under the 2 of 20. That's 1. Now 1 thru 6 are clean peaks. 7, 8 and 9 are scrambled peaks, not spikes as before. And then we see 10, a double peak. And if we continue we count over to 15 and there is another double peak. We talked about this before, I think. Anyway, here is the perfect example of a sound cancel reflection effect. Remember, time here moves from left to right. The left hand peaks of 10 and 15 are the rise in the loudness of the direct sound. All of a sudden a reflection hits the mic and instead of the peak continuing to grow, it gets partially cancelled. The pressure of the reflection is in time with but out of phase with the direct signal, and quieter. Still, the out of phase collision of the two wave trains gives us a dip in the sound level. And as long as the tone is on, the cancel reflection effect stays there, canceling. Then the tone turns off and what do you hear? Not the speaker, you hear the tail end of the sound canceling reflection whipping by your head until it also runs out of sound. I just love that. Art

So, are those "remnants" the source of the scrambled peaks like the ones evident in my MTF at around70 Hz? What is the effect of equalization - a high-Q notch filter, for example (I am using a bunch of those across the 30-100 Hz region)?

 

[garylkoh]

Here's a picture of my room from the perspective of the listening position. Pardon the mess.

When I did the recording, the mic was placed where I would be listening from. I'll have to do a recording one evening when HVAC is off and the construction crew are not working.

The room is typical industrial space office construction. 13' 7" wide by 22" long. Ceiling 8' 11" acoustic tiles backed by 1' fibreglass backed by 1/2" plywood. HVAC unit sits on top

Mic was placed 15" 2" from the front of the loudspeakers. and 32" from the concrete floor. Behind the curtains you see are floor to ceiling glass windows.

 

[Art Noxon]

Gary, I double checked. I was looking at your printout. It is the last one on the previous page. I'm glad we had that truck, so we could demo what happens to data with the noise floor is loud. Intelligibility is reduced. Do another and I'll run it also. ........................ART

 

[dougsmith]

Here are some pics of the front and back of my room (it has a cathedral ceiling).

 

[fas42]

Mic was placed 15" 2" from the front of the loudspeakers. and 32" from the concrete floor. Behind the curtains you see are floor to ceiling glass windows

Gary, would you have an idea what the volume would have been at the mic position? Also, curious with Doug's room, how much an effect the large glass area behind the listening position would make ...

Frank

 

[Art Noxon]

So, are those "remnants" the source of the scrambled peaks like the ones evident in my MTF at around70 Hz? What is the effect of equalization - a high-Q notch filter, for example (I am using a bunch of those across the 30-100 Hz region)?

Yes, Doug. Starting with the peak under 10 on the right side of the low frequency range of your test, call it 1, and then count to the left to peak 13. That is a good tone burst. Peaks 14 through 17 are not so good, especially 15 and 16. Notice we have 4 crummy peaks 14 through 17. That is a bandwidth of 8 Hz, centered round 65 Hz. Take a pencil and connect the peaks with a line and connect the reverse peak with a line. this is the envelope of the dynamic response, The dB between the top and bottom line is the dB modulation. Notice that with Doug and Gary, at 65 to 70 Hz, they both have low amount of modulation but the peak levels barely change, just a couple dB lower than adjacent good modulation. That means we are not involved with a loud resonance problem. Because it is the lower part of the envelope that causes the modulation to get bad, we know we have too much local reverberation in this frequency range up in the front of the room.

When we see double peaks we usually see an overall lower level of tone burst loudness. Tall single peaks are turned into lower level double peaks when some sort of out of phase reflection is involved. Reverberation after the signal shuts off reduces how quiet the sound gets between bursts. So phase canceling reflections reduce modulation strength and excess reverberation also reduced the modulation strength. Together, they can reduce the modulation strength even more.

EQ is nothing but level adjustment. It can turn a loud section of poor modulation into a quiet section of equally poor modulation. You can imagine that if you payed around with your volume know during the test, the overall level of the test goe up or dow but the modulation remains the same. It take a while to become comfortable seperating the idea of volume level, broadband volume adjustments and EQing, narrow band volume adjustments from the world of high speed level changes due to rapid sound level change in the program material.

Art

 

[Art Noxon]

Here's a picture of my room from the perspective of the listening position. Pardon the mess.

When I did the recording, the mic was placed where I would be listening from. I'll have to do a recording one evening when HVAC is off and the construction crew are not working.

The room is typical industrial space office construction. 13' 7" wide by 22" long. Ceiling 8' 11" acoustic tiles backed by 1' fibreglass backed by 1/2" plywood. HVAC unit sits on top

Mic was placed 15" 2" from the front of the loudspeakers. and 32" from the concrete floor. Behind the curtains you see are floor to ceiling glass windows.

View attachment 1771

Gary, Acoustic ceiling tiles tend to produce weak bass. Dealer showroom always learn this lesson the hard way. You have added weight to your ceiling tiles with the plywood. fiberglass transfers the weight evenly to the ceiling tile and allows no rattle between the two hard sheets. this was a good thing to do. double up your fist and thump the walls of your room, using the heel of your fist to transfer the blow. Listen to the thunder. Frequently in the 60 to 70 Hz range, eextensive reverb can come from the vibrational resonance of the sheetrock and stud wall assembly. Art

 

[Art Noxon]

Doug. draw a side view of your room to scale. squint your eyes and you see a parabolic dish comprised of your rear wall and the back half of your steep ceiling...and this dish is directly focused on the lower front corner of your room, which is where your speaker are located. The room width is 15.25', but your parabolic dish surfce is on average 17' from the lower front corner. Your room goes into instantaneous resonance at 1128/34 = 33 Hz. Check modulation test. Your listening position has pretty low modulation at 33 Hz. Check your frequency sweep around 33 Hz and you see nothing alarming, pretty flat. What does that mean? It means: "EQ'd flat and still sounds flat", which means it isn't musical, even if it is EQ'd flat. There is more to musicality that a flat frequency response.

Open the windows on the front wall and let that focused pressure out of the room. Or, when it gets cold, pick up some burly bass traps on ebay and put them near the front corner. You want to put an acousic resistor load all along that parabolic dish resonance focal point. ............................ART

 

[dougsmith]

Wow Art, you sure nailed that one! Take a look at what my in-room response was before equalization (the pink curve).

http://www.dougsmith.lisalandy.com/myalbum/before and after.png

One of my subs is up front. I should probably move that to another position in the room that would not reinforce that frequency. A third sub might help as well. Another thought I had was to replace the wallboard in the area between the windows with 1/2" fiberglass panels. That would turn the front wall into a 4.5" thick fiberglass absorber.

You can also see the mode at 65 Hz.

Frank, The 7 ms reflection you can see in my impulse response (previous post) is due to that window. Putting some pillows behind my head improves the imaging slightly. I would like to add some thick (velour) curtains for that window someday.

- Doug

 

[garylkoh]

Gary, would you have an idea what the volume would have been at the mic position?

It was about listening level - 79dB.

 

[garylkoh]

Gary, Acoustic ceiling tiles tend to produce weak bass. Dealer showroom always learn this lesson the hard way. You have added weight to your ceiling tiles with the plywood. fiberglass transfers the weight evenly to the ceiling tile and allows no rattle between the two hard sheets. this was a good thing to do. double up your fist and thump the walls of your room, using the heel of your fist to transfer the blow. Listen to the thunder. Frequently in the 60 to 70 Hz range, eextensive reverb can come from the vibrational resonance of the sheetrock and stud wall assembly. Art

Thanks, Art. That's absolutely correct. The entire ceiling absorbs bass, and I have four Bass Traps in addition. But most of my speakers have a built-in active subwoofer, so I can just turn up the gain if necessary. The HVAC above the room puts even more weight on the ceiling.

The wall on the left is tilt-up concrete and the wall on the right is a weight-bearing structural wall. The only lightweight wall is the one 1ft behind the listening seat. I also have absorbtive Cinema Panels on the wall behind the listening position, and diffusion Cinema Panels flanking the Bass Traps on the left and right and also at the first reflection points on the left and right walls.

The room is also almost perfect 0.618 : 1 : 1.618 proportions. That was one of the reasons why I picked this particular room to put the listening room, we lucked out when we moved into this space because the office was already built.

Not really conventional acoustics, but all my speakers are dipoles and it seems to work for me.

 

[garylkoh]

"EQ'd flat and still sounds flat", which means it isn't musical, even if it is EQ'd flat. There is more to musicality that a flat frequency response.

+infinity!!

 

[Mark (Basspig) Weiss]

I just happened along this thread. Fascinating read. Kind of a corollary to my square wave test that I posted earlier last month.

I ran the MATT test signal and threw a mic in the front row of my theater. Not sure if I did it right, but here's the recorded WAV file:

http://www.sendspace.com/file/ml752m

There is a known dip at about 34Hz, which has been traced to the back wall reflection. It's severe in this room. For me, the tones were pretty distinct across the test. I think I could discern the individual blips better on speakers than on the headphones.

 

[garylkoh]

Hi Art, here's a second run through - with the HVAC off and the road outside fairly quiet..... unfortunately, Boeing landed one of their new 787's at about 40secs.

http://www.genesisloudspeakers.com/downloads/MATTest1.wav

 

[Art Noxon]

It was about listening level - 79dB.

Speaking of listening levels. Most hot rodders know how fast their car can go. And, most audiophiles know how loud they can play their room. It might be interesting to honestly compare listening level limits for different system/rooms among the participants. Who knows how loud they can play their room and still be listening to quality playback? Remember, this level is an overall average program sound level. It includes the distortion free playback of typically very loud and short lived dynamic transients in program material.

Why do rooms have level or loudness limits? Thought you'd never ask.

Houses, their structure and interior surfaces have a lot of mass in general, small amounts of elasticity or spring back ability and small amounts of internal friction. Speakers apply alternating pressure forces onto the surfaces of the structure. The mass of the structure provides an inertial resistance to being pushed around by sound pressure. But speakers are powerful, subwoofers are running at input power levels between 1 and 2 horsepower. When they overpower the inertia of the room the surfaces and structure of the room begin to move. As the room is deformed due to pressure, it is bent and like anything else, when bent it pushes back, like a spring does. And as it is moving into its bent position, the joint of the structure are stressed and some slipping occurs on the microscopic scale, causing friction which puts the brakes on every time the wall is moving.

So, natural rooms have natural weight (mass), strength (elasticity) and friction (damping). When alternating pressures are applied, the surfaces deform and a little energy is absorbed. When the speaker is turned louder, the structure if more strongly excited but still there is enough friction to absorb the energy invested in getting the structure to move and bend. In general this is the law of steady state, which is where we live. When we bump into things, they tend to settle back to where they were before they got bumped. An example of an unstable steady state condition is a little ball balanced on top of a large ball. It stays put (steady state) but as soon a it gets bumped, it falls off the ball, instead of settling back to its original position. A stick of dynamite is another example of a stable condition of an unstable system.

Anyway, eventually, as we turn the volume up toward 11 and keep listening to our favorite music, we discover that the room is beginning to sound bad. It’s as if the room is breaking up. We know about cones breaking up if over driven, well, rooms mechanically break up of over driven a well. Breaking up mean that the system is no longer elastic, it somehow slips or folds over or something like that. More power I being applied to the structure than it can absorb and so it begins to overshoot and shake….that’s breakup. Think of a diving board. A bird lands on the end and the diving board calmly rocks up and down, twangs at its natural frequency. When the bird pushes off and fly away, the end of the diving board again calmly rock up and down until its internal friction damps out the vibration. But if a 180 pound teenager lands on the end of the diving board and flys away, what happen? We all know the sound by heard. The board rattles. It vibrates so hard that it overhoots and the board flys off its support structure and then crashes back down. The diving board structure I no longer a simple elastic thing, it flys and crashes in addition. The diving board is “breaking up”, to use the language we use to talk about drivers.

So it is….with listening rooms, and just about anything else in the world. And so, audiophiles know how fast they can run out their room without it starting to go out of control. They know that in general, they can play music in their room with the volume set at or less than 6.7 or 82 dB,A (I’m making this up) but no more. Who knows what their playback level before breakup limit is? And then, what is your preferred average listening level?

Art Noxon
Invented TubeTrap, President of ASC
www.tubetrap.com

 

[RBFC]

Very interesting, Art. This explains why things don't work so well when you try to turn it up to impress folks, past your typical louder limits.

Question: When you design acoustic treatment for clients' rooms, does the typical listening level they employ affect the type and scope of room treatment you suggest? I feel that room treatment and spectral balance, etc. do not maintain a linear relationship with respect to playback volume.

Lee

 

[Bruce B]

Speaking of listening levels. Most hot rodders know how fast their car can go. And, most audiophiles know how loud they can play their room. It might be interesting to honestly compare listening level limits for different system/rooms among the participants. Who knows how loud they can play their room and still be listening to quality playback? Remember, this level is an overall average program sound level. It includes the distortion free playback of typically very loud and short lived dynamic transients in program material.

Why do rooms have level or loudness limits? Thought you'd never ask.

This is one thing that I noticed in my room. I keep wanting to turn it up louder and louder. It just kept sounding better the louder it got. I've never experienced that before! The bass remained articulate and nothing fell apart.

 

[garylkoh]

Why do rooms have level or loudness limits? Thought you'd never ask.

Thanks, Art. Not only do rooms have a level or loudness limit, loudspeakers do as well. Every driver, no matter how well designed, will have an area of lowest distortion or linearity. Drivers also have to couple to the air in front of the driver to drive the air. As air has a reactance and impedance, it is easy to over drive the air and the air itself distorts. (My analogy is that the driver punches through and the air gets out of the way.)

I was using a pair of monitors with 1" ring radiator ribbon tweeter and 5.5" mid/woofer. The loudness limit in that room is determined by the speaker (I would guess around 85dB at the listening position). With my larger speakers, the room is the limiting factor - bass frequencies more so than high-frequencies in my room. I'm using 1440W Class D amps - so the electronics pose no problems.

 

[terryj]

I agree (know!) that rooms can have their volume limits, but the offered explanation does not ring true to me.

It seems the reason given to the room hitting it's limits is the bass. IME it is not the bass (at least not in my case, but I'm having trouble imagining how it would manifest anyway) but far more the upper frequencies.

Sure, the subs/woofers may be getting one r two horsepower applied to them (which sounds impressive) but what are the actual figures??? It is very low I'm pretty sure, ie 100db efficient still only means a few percent of the applied power is converted to sound. I am sure the correct answer will be on it's way!!

If it were the bass, then it would imply a small two way for example could not be played to loudly for the room, which I doubt very strongly is the case.

I get the idea it is somehow connected to the uncontrolled/excess of reflections, or at least it is how it manifests in MY room. The unintellgibility suddenly goes thru the roof.

Funnily enough, the initial tendency to make it easier to understand is to turn it up, which only worsens the situation. Usually happens in the TV room (untreated), hard to understand so turn it up, makes it harder to understand.

Still, treating the room certainly helps the ringing (not bass in this example) which certainly allows a higher volume with intelligiblity.

I am late to the record the room party, but as it is only 35 bucks I'll send a payment to Art and have it dissected here.

So will havta start mucking about wioth audacity again soon.