You are quite correct...
RT60 as a spec is normally only quoted in large venues...
Where you worry about music sounding like it's in a big cave...
And even the intelligibility of voice or audio presentations...
HOWEVER, in the context of normal sized rooms, it more relates to how absorptive various surfaces are at different frequencies.
Basically, in an open space, or an anechoic chamber, all you hear is the sound reaching you directly from the speaker.
But, in a room, you are obviously going to hear a mix of direct and reflected sound.
Another factor is that our ears and brains are quite good at differentiating direct and reflected sounds under certain conditions but less so under other conditions.
We can easily recognize sounds that arrive a long time after the original as "echoes"...
Yet we hear sounds that arrive only a few milliseconds after the initial sound as "part of the original sound"...
So, let's say you're in a large room, with very hard walls, that are especially reflective at high frequencies.
In that situation you will hear sounds followed by a distinctive echo - and the echo itself will sound rather bright.
But, if you were to hear the same sound in a small room, with equally hard walls, equally reflective at higher frequencies, those echoes will tend to blend with the original sound.
And, as a result, rather than hearing echoes, what you will hear is that the original sound seems excessively bright (because, in your brain, the bright reflections are summed with the original sound).
And, in our small very reflective room, you're going to have difficulty adjusting things to sound the way you would like.
You can adjust the system so that the direct sounds (short window) have a flat frequency response.
But, if you do that, the longer term "power response" measurement will show a rising high frequency response.
Neither measurement is wrong...
Over the course of a few seconds, the low frequencies will die down more quickly than high frequencies, leaving an excess of high frequencies on average.
The result will sound good with short transients but, with pink noise, or with complex musical passages, or cymbals, over a few seconds the high frequencies will build up and sound harsh.
Alternately you can adjust the system for a flat long term power response.
And, if you do that, the room will generally sound balanced, and you'll avoid that nasty "glare".
HOWEVER, because you have used EQ to reduce the high frequency content in direct sounds, sharp sounds like cymbals may sound muted and dull.
The problem is that, since music contains a combination of both sorts of sounds, you will be unable to find a compromise that sounds correct for both.
You are quite right... and the RT60 itself is meaningless here...
However an RT60 that varies widely over the audible frequency range tells us that the absorbency of the room probably varies widely over that frequency range.
And, if that's the case, we're probably going to find a wide disparity between the direct response and the power response.
And that may suggest that we're going to have a problem finding EQ settings that work well for both.
In that situation... if I had a small and relatively bright listening room... and had to minimize room treatments...
I would try to select speakers with a somewhat narrow dispersion pattern at high frequencies (so I hear a higher proportion of direct sound)...
Then, if it was an option, I would try to put some sort of padding on the rear wall (which would absorb high frequencies, and so minimize the amount that end up reaching, and being reflected from, the other walls)...
By minimizing reflections I would help balance the amount of direct and reflected sounds reaching the listening position...
This would allow me to calibrate the sound system to deliver a frequency response that would be roughly similar between direct and reflected sound...
In real life, you wouldn't need to measure RT60... just clap your hands and listen to the echo.
My point is that, without some sort of room treatment, short window and power response measurements would be very different....
And it would be VERY difficult to find a compromise that would make both "good"...
To answer your question.... pretty much yes.
You can play a single impulse, measure the response, and basically close the window before any reflections from the nearest surface can reach the microphone.
Likewise, you can play a series of separate tones, and do the same for each.
You can also play a sweep tone... and basically have a filter that follows the tone.
You are basically saying that "from the instant the 1000 Hz tone is played, we will only listen to THAT FREQUENCY for a short amount of time, and stop before reflections have time to arrive".
You essentially have "a moving tone and a moving window that follows it".
This is part of the idea behind chirp tones.
There are only two basic limitations with most of these theories...
Everything we've talked about is actually "one dimensional".
We haven't accounted for the fact that, in our room with shiny side walls, and a carpeted rear wall, the DIRECTION the sound is coming from matters.
Sound that hits that rear wall first will be absorbed by the carpet while sound that hits the side walls will ricochet back and forth.
With a single measurement point we can measure a sort of overall average but we cannot account for more complex interactions like that.
(Arguably, with only direct sound, we only have one dimension, but things get very complicated when you're discussing the acoustics of the room.)
The other limitation is that our time windows themselves introduce artifacts.
For example, if you were to take a 400 Hz test tone, and turn it on for 20 milliseconds, you have also MODULATED that 400 Hz tone (it has been modulated by the frequency that corresponds to how long it was on).
Therefore, when you analyze your measurements, you'll find "50 Hz stuff that wasn't there before".
It is impossible to make a 400 Hz test tone that lasts for 20 milliseconds but is still a PURE 400 Hz tone.
And, if you turn it on and off too quickly, those artifacts will fall within the range of frequencies you're measuring.
Similarly, you cannot technically "open a filter for some number of milliseconds" without the results being affected by "the length of time the window was open".
(Luckily you can calculate exactly what these effects are, and account for them, but you cannot avoid them entirely.)
You can demonstrate this easily by taking your favorite audio editor and generating a 400 Hz tone of any reasonably short length.
When you look at the result in spectrum view you will see other frequencies around the beginning and end of the tone.
These are "modulation products" between the tone itself and the time interval during which you "switched it on".
(In simplest terms "a 400 Hz tone for 20 milliseconds" is not and cannot also be "a pure 400 Hz tone"; in order to be a perfectly pure tone it must remain on forever.)
RT60 is a measure I see frequently applied to large venues.....Concert halls and large clubs, perhaps.
In a smaller, reverberent space, the RT60 measure might be fairly long. And be helped by room treatments....maybe some diffusion and absorption.
You DO NOT want an RT60 of ZERO. That would be a Anachoic space and is an AWFUL place to listen to anything but your heartbeat.
Certain types gett all wrapped up with this measurement thing. I'd rather do a good basic setup and at some point trust my ears.
That being said, I'll admit to being PRONE to go all measurement geeky. In my profession I did Metrology as part of my job, including stuff the home user
would NEVER EVER consider, like real NIST calibrations of measuring equiopment and daily consistency checks based on internal standards. Careful record keeping
was a MUST while each year I did a Gage R+R study to determine a given tools fitness to measure our process...
All sorts of OTHER questions occur to me. Just ONE? Could you do a VERY time limited impulse measure at a series of frequencies.....and 'string' them together to form an idea of
what the anachoic response would look like?
asq.org/quality-resources/gage-repeatabilityYou're welcome to drive yourself NUTS with the above link.....More than you ever would WANT or NEED to know about measurment systems.
