Pat,
> The basis of effective room tuning, as taught by ... Now try Googling "modal resonances" and see who's name comes up. <
In logic courses this is called "argument from authority" because it relies on invoking the names of experts. I like to think my answers and explanations stand on their own merit.
> Specifically we would request any SPL vs. frequency graph conform to our format shown here: <
I'll be happy to oblige when it's not too much effort. I'm not looking to rock the boat here, but the graphs I posted are as they came out of the ETF software. With ETF you can display decay versus frequency over only a few fixed ranges, but not the specific range you mentioned. Even if I were to display a wider range in ETF and then use Paintshop Pro to chop off the stuff past 500 Hz, that would show less resolution below 200 Hz because this is how ETF works. That is, to see the most detail below 200 Hz you have to set the upper limit to 200 Hz.
> We would also request that if a z-axis is to be included that it be labeled clearly for easy readability. <
The label says 400 milliseconds. Again, I'm glad to oblige, but how would you have me make that clearer?
> We have had loudspeaker manufacturers intimate that their products might look better on a 90dB or 100dB scale which would of course make the measured response look flatter. <
Yeah, no kidding. I always laugh out loud when I see graphs like that in the audiophile magazines, because it
completely hides the speaker's true response.
> The bandwidth about which we're speaking is 20Hz - 100Hz. Please keep this bandwidth for any submissions on this thread. <
In my experience the bass range extends up to about 300 Hz. This is where the worst of the modal
and positional peaks and nulls occur. Ignoring frequencies above 100 Hz is to ignore the most important range where bass instruments "speak."
Many people do not realize this, but most bass instruments have relatively little energy at fundamental frequencies. For example, with an electric bass the second harmonic is typically 8 to 12 dB higher than the fundamental when the string is plucked at the usual place over the pickup. Brass instruments like tubas and trombones have an even larger disparity between the fundamental and its harmonics. So when someone plays that big fat low A note at 55 Hz, what you're really hearing is mainly 110 Hz.
> There is no labeling on this chart designating where in the room this reading was taken. <
Here's a layout of that room showing the microphone position:
> Referring to the three peaks within the 20Hz -100Hz bandwidth I can truthfully say that I've never measured,
at the listening position, such a severe, triple-peak case in any room I've ever encountered. <
Most people do not realize how severe the peaks and nulls in all small rooms really are. When you measure at 1/3 or even 1/6 octave what is shown is an
average of all the frequencies in that band. However, the true response is always far worse, which is why I use the ETF software. It resolves to 0.7 Hz, and that's the
only way to know what a room is really doing at those very low frequencies. This graph shows the same data expressed as 1/3 octave and 1/12 octave at the same time:
Sorry if the above graph does not show the frequency range you prefer, but it's what I have handy. As you can see, the
exact same data displayed at 1/12 octave shows the true extent of the peaks and nulls far better than the 1/3 octave version.
> That task, overall is to obtain a listening space with an RT60 of between .2 and .4 seconds through the full frequency bandwidth ... below 200Hz the .4 reverberation time is allowed to rise <
In rooms the size you find in most homes there
is no true reverberation below a few hundred Hz. Rather, ringing at specific modal frequencies dominates, as shown in the ETF waterfall graphs in my earlier post above.
> parametric equalization, either manual or automatic is the most direct, cost effective and acoustically correct answer. <
Again, EQ can help, but it does nothing for modal ringing and its benefit is highly position dependant. I've measured a change of 15 dB across a physical span of only four inches - equal to turning your head - and 22 dB when moving 18 inches - the person one seat over on the couch. Here's a more complete description of that test, from my post about a year ago in one of the pro audio newsgroups:
I did this test in my fairly large (18 x 34 feet) home recording studio. This room has 17 wood panel bass traps and 11 broadband absorbers based on 703 and 705 rigid fiberglass. In a typical room without acoustic treatment the results will likely be even more dramatic.
I played a 100 Hz sine wave at a medium-loud level through my large JBL 4430 speakers. Then I walked toward and away from the walls in different parts of the room, listening for a place where the tone was softest. Once I found a quiet spot I placed an audiotechnica 4033 microphone where my ear had been, aimed toward the loudspeaker, then moved it by small amounts while watching the level meter to find the deepest null.
The minimum level I could find read -37 dB. I then moved the mike exactly four inches to the side and the level jumped to -22 dB, which is a 15 dB difference. Then I moved the mike to 18 inches away from the original null location, and there the level was -15, which is a 22 dB difference.
> the supposed importance of
model bandwidth or decay time in other areas of the room are no longer of import to us enjoying our music and film
at the listening position.
Yikes. If you don't consider excessive ringing important, I'm afraid there's nothing else I can to convince you other than my previous explanation, "Without traps, some bass notes ring out for as long as 1/3 of a second, so they muddy other subsequent bass notes."
I understand that many people want very much to believe that room acoustic problems can be solved completely using a simple and unobtrusive solution like EQ. If only that were so. Even Mr. Grimani you referred to is in the business of selling bass traps, so that must tell you something, no?
Thanks.
--Ethan
