Graph Interpretation Graph

[fuzz092888]

Can someone help me interpret these two graphs and what exactly they mean in the area of loudspeaker design. I did some googling, but am still a little confused.

 

[Steve81]

The Schroeder Plot is decay in dB vs time.

Reverberation Time computation, Schroeder Plot

The reverse time integration transforms the normal Room Impulse Response (fig. 1) into a decay plot (fig. 2) where it is possible to evaluate the decay time and consequently the reverberation time (60 dB decay).

EDT is a calculation of the reverberation time by frequency band.

Reverberation Time Measurement - Noise and Vibration Basics

Reverberation Times may be labelled EDT, T20 and T30 respectively for those three evaluation ranges. EDT is used in room acoustics only, while T20 and T30 may be used in all applications. Reverberation Time is measured in 1/1- or 1/3-octave frequency bands, some of which may be averaged to provide a single-number result for the most significant bands.

Reverberation Time may range from 0.1 seconds (or less) in anechoic chambers, to 10 or more seconds in large public spaces.

 

[fuzz092888]

So these charts would be the decay of sound in the bass region? As in "the attack and decay of the bass notes was excellent" ?

 

[Steve81]

As far as I can tell, the charts are the decay & reverberation of a full band impulse, in room.

The Schroeder plot doesn't have frequency as a variable at all. The EDT shows the average reverb time in milliseconds for each octave. For example, in the first graph, at 1000Hz, in room reverberation time was 963ms vs 158ms in the second plot. At 63Hz, the first image shows a reverb time of 691ms versus 5345ms for the second image.

 

[fuzz092888]

The Schroeder plot doesn't have frequency as a variable at all. The EDT shows the average reverb time in milliseconds for each octave. For example, in the first graph, at 1000Hz, in room reverberation time was 963ms vs 158ms in the second plot. At 63Hz, the first image shows a reverb time of 691ms versus 5345ms for the second image.

I can follow what you're saying here

As far as I can tell, the charts are the decay & reverberation of a full band impulse, in room.

But what does this bit mean? Explain as if you were explaining it to a 3 year old Unfortunately that's where I feel like I sometimes whenever I read about some of this stuff

 

[Steve81]

But what does this bit mean? Explain as if you were explaining it to a 3 year old Unfortunately that's where I feel like I sometimes whenever I read about some of this stuff

I guess the best way to describe it would be like an Audyssey chirp versus say, running a 63Hz sine wave, then a 125Hz sine wave, and so on. As a real world example, when Dennis measured the FR of my speakers I took to compare to the Phils, it wasn't with individual sine waves (which would take a while), but with a single such Audyssey-like chirp that lasted a fraction of a second.

 

[fuzz092888]

So the graphs were comparing two different speakers. Each had whatever frequency tone played at the specified intervals. Since one had a much greater reverberation time, what does that mean when comparing the two speakers?

 

[Steve81]

Since one had a much greater reverberation time, what does that mean when comparing the two speakers?

Not much as far as I can tell, at least not without a lot more information, ie testing methodologies. On the face of it, with the times recorded, it doesn't even look like the same room. I mean, I don't see how you can go from ~700ms of reverb time at 63Hz to over 5 seconds without some pretty big variables changing.

 

[fuzz092888]

This was on a product page to testing methodologies weren't available (on the page).

Here are the two captions:

"Early decay time of a conventional loudspeaker
system in an average room. Notice elevated midband bars that degrade midrange resolution."

"Early decay time of the 8.3 in an average room shows dramatic improvement through the midband area. This results in improved midrange intelligibility and tonality"

Sounds to me like it is design difference that is claimed.

 

[Steve81]

Sounds to me like it is design difference that is claimed.

Sounds like a load of bull to me, but I've been wrong before I guess I just don't see how a speaker is going to cut in room reverb time like that. Room treatments, sure, but the speaker? I mean, once the speaker's energy is decayed, which doesn't necessarily take much time depending on frequency, what's left?

Nothing special there, an Emo X-ref XRT 5.2, but you see the time scales involved vs the time scales of the claimed in room decay/reverb. Not sure how a speaker can address that gap further.

 

[fuzz092888]

Sounds like a load of bull to me, but I've been wrong before I guess I just don't see how a speaker is going to cut in room reverb time. Room treatments, sure, but the speaker?

How equipment would you need to test something like this out? An oscilloscope?

 

[Steve81]

How equipment would you need to test something like this out? An oscilloscope?

I'd expect an REW setup (computer with a decent sound card, software, microphone) would do the trick as it can give you a waterfall graph.

 

[fuzz092888]

I'd expect an REW setup (computer with a decent sound card, software, microphone) would do the trick as it can give you a waterfall graph.

Perfect, I've got all that. I also shot an e-mail to the designer and asked him some questions. He seems like a good guy and a friend of mine knows him so I think I'll get honest answers.

Thanks Steve I appreciate the help.

 

[Steve81]

Glad to be of assistance. I am curious to hear what the designer says. Like I said, I've been wrong before, and I don't mind a little humble pie for dessert.

 

[3db]

Sounds like a load of bull to me, but I've been wrong before I guess I just don't see how a speaker is going to cut in room reverb time like that. Room treatments, sure, but the speaker? I mean, once the speaker's energy is decayed, which doesn't necessarily take much time depending on frequency, what's left?

Nothing special there, an Emo X-ref XRT 5.2, but you see the time scales involved vs the time scales of the claimed in room decay/reverb. Not sure how a speaker can address that gap further.

I concur. Reverb is a function of a room's acoustic. Vibration is a function of a speakers cabinet resonance. Cabinet resonance is and should be far less in duration than a rooms reverb.

 

[fuzz092888]

So here is what I was told: (For the purposes of the following the first graph is "speaker A" and the second graph is "speaker B")

The plots show that speaker B has a faster early decay time at frequencies above 500Hz relative to the floor reflection as defined by the measurement window.

Both speakers were measured in the same room and from the same position.

Thoughts? Any follow up questions I should ask? Anything else that would help you guys understand this so that you can then explain it to me?

 

[Steve81]

Thoughts? Any follow up questions I should ask? Anything else that would help you guys understand this so that you can then explain it to me?

Just "how". Only thing that pops to my mind to potentially reduce in room reverb/decay time is very narrow directivity.

 

[fuzz092888]

Just "how". Only thing that pops to mind is perhaps very narrow directivity.

Does this help? From the product page.

Model 8.3 Focused Array™ is a wide range controlled dispersion line source loudspeaker system. It features new Quiet Cone™ technology coupled with an advanced two stage inclined front cabinet baffle and 3 unit line source tweeter array. This highly advanced approach minimizes room interactions that affect wide band intelligibility. The 3 unit tweeter array has a precise dispersion pattern that minimizes diffraction effects while minimizing distortion. A customized/proprietary 8"midbass-midrange wool-carbon composite driver is used in concert with a highly sophisticated crossover and very powerful 12" sub bass driver to produce extended bass to 30 Hz, negating the need for separate powered subwoofers. Every attention to detail and sonic performance has been implemented to achieve a speaker system that is truly amazing and full range. Marginal rooms now have the ability to sound like well balanced concert halls. The full weight of an orchestra is integrated with depth, clarity and pinpoint placement.

 

[Steve81]

Does this help?

There are a few buzzwords in there that would lead me to say that my narrow directivity guess might be going in the right direction. However, I'd leave it to a real expert to dissect the claims and design details.

 

[fuzz092888]

There are a few buzzwords in there that would lead me to say that my narrow directivity guess might be going in the right direction. However, I'd leave it to a real expert to dissect the claims and design details.

Where's TLSguy or Granteed when you need them