Listening Room Intelligibility Test

[Art Noxon]

Hello Mark,

Great input. I don’t know if I have met Dr Steeneken. Certainly know of his work. I understand he was involved with B&K’s RASTI test. The RASTI , TEF S/N 80ms, STI and %AlCONS have all been sync’d together.

I wrote two papers about MATT. In 88 I presented Articulation and the Small Room and in 89 I presented Articulation-Prerequsite to Performance. These can be downloaded from the AES section of published articles located at http://www.acousticsciences.com/articles.htm

The first paper was about MATT. The second paper was a generalized perspective. I described the 3 axis of MTF testing: Spectrum (20 to20k), Modulation rate (0 to 20 Hz) and Modulation Amplitude (0 to 20 dB). Inside this 3 axis cube I defined a few sub volumes, one for speech intelligibility and another sub volume for audio.

I have tested many hifi and recording studios over the years. People can deal with 5 dB modulation, like 10 dB modulation and love 15 dB modulation. The difference between 15 and 20 dB modulation is very slight. High performing recording studios register 15 to 17 dB modulation between 28 and 780 Hz and are also have a flat “EQ” without any equalization.

An audio room test subculture has grown up around this early work. Go to http://www.ramsete.com/Public/Papers/153-AES110.PDF and also google ACT Acoustic Quality Test, which is what MATT has evolved into, at lease in Europe. Here you find downloads, Utube and all kinds of good work.

Clearly, In the Netherlands is where the real work in evolving the MATT test system has been going on, much more than anything I could do alone. Art Noxon

 

[Mark Seaton]

Hello Mark,

Great input. I don’t know if I have met Dr Steeneken. Certainly know of his work. I understand he was involved with B&K’s RASTI test. The RASTI , TEF S/N 80ms, STI and %AlCONS have all been sync’d together.

I wrote two papers about MATT. In 88 I presented Articulation and the Small Room and in 89 I presented Articulation-Prerequsite to Performance. These can be downloaded from the AES section of published articles located at http://www.acousticsciences.com/articles.htm

The first paper was about MATT. The second paper was a generalized perspective. I described the 3 axis of MTF testing: Spectrum (20 to20k), Modulation rate (0 to 20 Hz) and Modulation Amplitude (0 to 20 dB). Inside this 3 axis cube I defined a few sub volumes, one for speech intelligibility and another sub volume for audio.

I have tested many hifi and recording studios over the years. People can deal with 5 dB modulation, like 10 dB modulation and love 15 dB modulation. The difference between 15 and 20 dB modulation is very slight. High performing recording studios register 15 to 17 dB modulation between 28 and 780 Hz and are also have a flat “EQ” without any equalization.

An audio room test subculture has grown up around this early work. Go to http://www.ramsete.com/Public/Papers/153-AES110.PDF and also google ACT Acoustic Quality Test, which is what MATT has evolved into, at lease in Europe. Here you find downloads, Utube and all kinds of good work.

Clearly, In the Netherlands is where the real work in evolving the MATT test system has been going on, much more than anything I could do alone.
Art Noxon

Thank you for the background info and links Art. I'll have to look at the papers later, but from what you describe it might suggest that a different scalar readout from testing such as what I've tinkered with previously in my TEF25 could allow some correlation and quantification to subjective changes in quality that aren't easily depicted in response curves or decay times. If we can isolate bandwidth in measurement, that would make it easier to confirm what sort of effort/fix is required, and verify if the efforts are fruitful.

 

[dougsmith]

I tried the test in my room. The only significant difference I heard between the speakers and headphones was what sounded like a slight desynchronization of the cadence between clicks and tones at certain frequencies of the tones in the room (as though the clicks speeded up ever so slightly relative to the tone bursts). But I'm not sure if my mind was playing tricks on me. My room is fairly large (15 x29 with cathedral ceiling), the speakers are Geddes Abbeys + subs at ~12' spacing with an equilateral triangle to the listening position. The room is basically just typical a furnished living room. My wife was not too enamored with the sound. I'll try some more experiments when she's not around.

 

[KlausR.]

Hi Art,

For now, report into the forum about what you heard.

Room (living) is 8.6 x.4.6 x 2.5 m, brick ‘n mortar, windows double-glazed, floor tiled, ceiling stretched synthetic fabric with 26 cm air above, loosely filled with rockwool, no other acoustic treatment. Speakers are on long wall, 2.8 m apart, front baffle 85 cm from front wall, listening distance is 3.5 m. Burnt the track to CD.

The clicks seem to be somewhat softer (as if onset duration was longer) in the low frequencies, and get sharper when frequency rises, both over headphones and speakers. At some points the sine pulse gets louder and almost overwhelmes the clicks, must be hitting an eigenfrequency. At some points the sines also get less loud as if one sat in the dip of a standing wave. Wavelength at 780 Hz is 44 cm, so I can move the head from peak to dip.

In any case, all clicks are well separated, clear and distinct, no click is lost on the journey, no smearing, garbling or yodeling.

I wonder how much of what one hears or doesn’t hear is due to the room, how much to the loudspeakers themselves, and how much to HRTF.

Klaus

 

[Art Noxon]

Guys,

1) If anyone wants to have an experience, just make a recording (no AGC recordings) of what you hear when the MATT test plays and send it to me, I'll process it and send results back to you. If we want we could post the curve on this site, if anyone can stand the heckling, or just send it by email to you. I'll process it at no charge (normally $35) and get a copy to you with discussion.

What's interesting to me is that for the most part, the people responding seem pretty satisfied with the performance of their system/room, by that I mean the MATT isn’t telling them much news. I’m not used to having that response. On the other hand, I usually only get to work on rooms that have become a real pain.

A direct wave that experiences sound canceling makes for a very interesting study. The direct is usually clearly heard when it first arrives at the listener. Then other reflections occur which return to the listening position. Here the two waves “add” and the listener gets to listen to the sum. Here the sum of the ongoing direct tone and it reflection is close to zero, because they are close to equal and opposite.

Now when the direct stops, all that is left is the loud reflection and the sound level jumps up and stays loud until the reflection is finished. And so when you experience a cancel type tone burst you actually experience a complicated set of sound level change: silent, loud, quiet, loud, decay to silent. Imagine Bow-Wow and you’ve got it. Who let the dog out? Woof … woof …

Now there are three things that can interfere with the direct. A) a simple reflection
B) standing wave reflection and C) a resonant mode.

A) A simple reflection might be a side wall bounce.

B) The standing wave is when the wave goes past the listener, hits the rear wall and reflects straight back to where it came from. A standing wave seems like a mode, because it is filled with phase adds and phase cancels (hot spots and suck outs, as they say in audiophilese), however it is not stored energy. A standing wave is created instantly as soon as the wave folds back on itself and it stops as soon as the end of the direct hit the rear wall.

C) The room mode. The room mode is like a 3 dimensional standing wave and some. It occurs when the reflected sound off the back wall hits the front wall and also folds back on itself, in time with the direct wave. Once the reflection pattern couples up with the speaker the speaker can keep pumping energy into the strength of the standing wave, now a mode, and it gets louder and louder. It only stops getting louder when the sound absorbing power equals the sound delivery power. That’s all for now…
Art Noxon
Inventor of the TubeTrap and President of Acoustic Sciences Corp

www.acousticsciences.com

 

[RBFC]

Art,

That's a very generous offer. As soon as I can get a hold of some recording equipment, I'll do this.

Thanks,

Lee

 

[RUR]

Art,

That's a very generous offer. As soon as I can get a hold of some recording equipment, I'll do this.

Thanks,

Lee

Shoot, Lee, I'm so keen to do this I may have to actually figure out how to use the gear I already have!

 

[garylkoh]

Thanks you, Art. That is a very generous offer.

In the interest of education for everyone, I'm willing to have the results of my room posted on the forum and try to develop the thick skin to withstand the heckling. I think, that many times, the heckling can even be instructive because it does illuminate the heckler's perspective.

 

[Bruce B]

Well I just shot my room. Let's hear the bad news. I'll PM you...


http://www.pugetsoundstudios.com/hirez/tones/ASC MATT test final.wav

 

[garylkoh]

FWIW, here's mine:

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

I'm putting my heckle-proof armor on. I duplicated the channels so that I could have a listen on headphones.

 

[dougsmith]

I'll post mine too, but it won't be until tomorrow evening.

 

[Bruce B]

Guess Gary and I are the only ones so far brave enough!

 

[RBFC]

Guess Gary and I are the only ones so far brave enough!

You're the only ones with instant access to recording equipment. However, your bravery is commendable.

Lee

 

[dougsmith]

I posted mine HERE... shoot!

- Doug

 

[garylkoh]

This is fascinating - how the same thing can sound so different on three different systems. Now, I wonder if we all have recording equipment - what a specific piece of music would sound like when played back through each room - especially if we all had the same mic

 

[dougsmith]

Yes, I wonder also. I used an Audio-Technica condenser mic because I wasn't getting enough sensitivity out of the ECM8000 with my recording utility for some reason (although the ECM was working fine with Holm Impulse). The AT is has a cardioid sensitivity pattern - more omnidirectional than the Berhinger.

(USB Mobile Pre as soundcard on my laptop)

 

[Bruce B]

This is fascinating - how the same thing can sound so different on three different systems. Now, I wonder if we all have recording equipment - what a specific piece of music would sound like when played back through each room - especially if we all had the same mic

Well my chain was: Earthworks M50 => DAD AX24 => Pyramix

 

[garylkoh]

I didn't want to use my loudspeaker measurement rig, so I bought a new Dayton mic, an E-mu 0404 and downloaded Audacity. Total cost to be able to make recordings - $225.

 

[Bruce B]

I didn't want to use my loudspeaker measurement rig, so I bought a new Dayton mic, an E-mu 0404 and downloaded Audacity. Total cost to be able to make recordings - $225.

You forgot your time at $300/hr. !!

 

[garylkoh]

You forgot your time at $300/hr. !!

With you and with Art contributing your time as well, this forum has got to be one of the most valuable places for audiophiles to hang out!!