Room measurements

[amirm]

Most people are referring we do need a large dynamic range for for audio measurements - however in order to properly measure RT60 we need a low noise floor. Although most of the time software interpolate RT60 from RT30, as we do not have absolute noise levels in the shown measurements, how can we be sure that these horrible spikes from the UMM-6 do not affect the decay measurements?
Yes, Amir I know you are not an RT60 dependent ...

RT30 is just fine and is actually part of international standard for such measurements. No one tries to measure true RT60. Programs like REW use sine sweeps that provide much gain over traditional impulse response anyway. Here is a sample using UMM-6:

This was at moderately loud levels and as you can see, it has nearly 60 dB of dynamic range anyway due to above technique. And what is left, is true noise and not the distortion spikes. For the rough measurement that RT60 is in small rooms, there is no limitation here.

 

[microstrip]

RT30 is just fine and is actually part of international standard for such measurements. No one tries to measure true RT60. Programs like REW use sine sweeps that provide much gain over traditional impulse response anyway. Here is a sample using UMM-6:

This was at moderately loud levels and as you can see, it has nearly 60 dB of dynamic range anyway due to above technique. And what is left, is true noise and not the distortion spikes. For the rough measurement that RT60 is in small rooms, there is no limitation here.

Thanks Amir - I feel better as I have ordered a calibrated UMM-6 yesterday from the site you referred - they ship internationally and accept paypal. As you say not having to carry all the extra hardware is a real convenience, although it helped to impress friends.

 

[amirm]

I think you will be happy.

One note of caution though: don't drop the mic. I have heard a couple of people drop them and something breaks lose in them. Also, the thing comes with a tiny tripod which I guess is intended to sit on your chair or something. I use a proper mic stand.

 

[DonH50]

Does RT60 really apply in the average home listening room?

* I am guessing I'll be sorry I asked that question. *

 

[amirm]

Here is all you ever wanted to know about RT60 and small rooms: http://www.whatsbestforum.com/showt...nt-thereof-to-quot-small-acoustic-spaces-quot

And even a deeper dive here: http://www.gearslutz.com/board/stud...ll-rooms-have-real-reverb-14.html#post8887203



Short answer is that it has some value in assessing how live or dead the space is (i.e. general amount of absorption).

 

[andromedaaudio]

here is a scheme of " standard " reveberation times in holland
between 0,3 -0,5 for homecinema
Soundstudios/recording room 0,2-0,4

Which are the smallest of them all , al other rooms/ buildings/óffices are higher
http://www.easy-akoestiek.nl/gratis-advies-op-maat/akoestiek-vragen/gewenste-nagalmtijd.html

 

[microstrip]

Here is all you ever wanted to know about RT60 and small rooms: http://www.whatsbestforum.com/showt...nt-thereof-to-quot-small-acoustic-spaces-quot

And even a deeper dive here: http://www.gearslutz.com/board/stud...ll-rooms-have-real-reverb-14.html#post8887203



Short answer is that it has some value in assessing how live or dead the space is (i.e. general amount of absorption).

Amir,

As perhaps it is not only me, so I post my doubts here, as the more I read about RT the more confused I get.
You wrote in another post

"If you search for my posts and countless peer reviewed literature in Journal of ASA and AES which use RT60 measurements in small spaces, the above is precisely the recommended practicethey do: the RT60 measurements are for "mid-frequencies". I use 500 to 1000 Hz; others use the same or go up to 3 KHz or so. RT60 is not provided as a measure to evaluate modal (e.g. subwoofer) region. "

In the same thread I could find that Ethan supports your posts:
"Quote Originally Posted by Ethan Winer : Nice post Amir. I suspect it's falling on deaf ears"

However the Realtraps site is full of graphs and analyzes of RT60 measurements taken between 30 and 300 Hz, as the one I attach. I have to say I took similar measurements in the past when treating my room. What is their real value?

Can you tell me what I and perhaps some others are missing? I think I am not alone after reading Don post.

 

[DonH50]

Thanks Amir, and Ethan.

I glitched and left unstated a key assumption, the one microstrip picked up on: Does RT60 matter in the average listening room at LF?

Anybody who has sung in a shower (or just dropped the shampoo bottle) knows the impact of reverb at high frequencies in small spaces. For subwoofers, the argument has always been the wavelengths are such that true reverb does really happen in small spaces (hidden by room modes and/or SBIR), or something like that...

My listening room is very dead by design. Any ambiance comes from the recording. If I had a bigger room I would have less treatment and a little liver room, but I have gone back and forth on how much I want the room to contribute and tends toward less live. That's another topic...

In any event, I do not want to sidetrack this discussion so will shut up and read. - Don

 

[amirm]

Amir,

As perhaps it is not only me, so I post my doubts here, as the more I read about RT the more confused I get.
You wrote in another post

"If you search for my posts and countless peer reviewed literature in Journal of ASA and AES which use RT60 measurements in small spaces, the above is precisely the recommended practicethey do: the RT60 measurements are for "mid-frequencies". I use 500 to 1000 Hz; others use the same or go up to 3 KHz or so. RT60 is not provided as a measure to evaluate modal (e.g. subwoofer) region. "

In the same thread I could find that Ethan supports your posts:
"Quote Originally Posted by Ethan Winer : Nice post Amir. I suspect it's falling on deaf ears"

However the Realtraps site is full of graphs and analyzes of RT60 measurements taken between 30 and 300 Hz, as the one I attach. I have to say I took similar measurements in the past when treating my room. What is their real value?

Can you tell me what I and perhaps some others are missing? I think I am not alone after reading Don post.

That is a waterfall display. It is not an RT60 computation. Waterfalls are created by taking time slices and performing FFT on them to produce the frequency response for that slice. Fourier transforms are only valid across infinite series. To use them on a slice, one has to use a window function which then smears the samples over a larger time slot than the slice. In English , you can't use waterfalls to compute RT60. The right analysis is as I showed which uses the impulse response.

Ethan's use of the waterfall is to indicate variations in frequency response and time domain ringing. So in that sense what you post from his site is fine and not related to this topic.

 

[andromedaaudio]

On the site linked it was described as follows .

The reveberation time is the time that is needed for the initial soundlevel ,usually delivered with a specially designed gun to fall back 60 db in level .
The reveberation time is the time needed for that soundenergy to be absorbed by the walls/floor/ ceiling and air .
Closed rooms with hard surfaces dont absorb soundenergy very well so they have long reveberation times

 

[Ethan Winer]

That is a waterfall display. It is not an RT60 computation ... Ethan's use of the waterfall is to indicate variations in frequency response and time domain ringing. So in that sense what you post from his site is fine and not related to this topic.

Exactly, thanks for saving me having to type the same answer.

--Ethan

 

[microstrip]

That is a waterfall display. It is not an RT60 computation. Waterfalls are created by taking time slices and performing FFT on them to produce the frequency response for that slice. Fourier transforms are only valid across infinite series. To use them on a slice, one has to use a window function which then smears the samples over a larger time slot than the slice. In English , you can't use waterfalls to compute RT60. The right analysis is as I showed which uses the impulse response.

Ethan's use of the waterfall is to indicate variations in frequency response and time domain ringing. So in that sense what you post from his site is fine and not related to this topic.

OK, thanks to both. I know there are limitations in waterfall diagrams at low frequencies due to limitations of the window function process. But ignoring them, the interpretation of a waterfall at a fixed frequency is just an amplitude decay versus time - my interpretation if that if you fit a line to this decay you could get an approximate RT60 number at this frequency. Isn't the information of a waterfall similar to what we get from the RT60's at many frequencies?

Ethan use of waterfalls seems to me to to be rather important for room bass behavior : See his text "These waterfall plots compare the before and after low frequency response, ringing, and modal bandwidth in a typical small room when empty versus with 12 pieces of rigid fiberglass placed in corners." Can I ask you what does "modal ringing mean"? How is it related to decay time and bass sound quality?

http://www.realtraps.com/art_testing.htm

http://www.realtraps.com/art_testing.htm

 

[Ethan Winer]

what does "modal ringing mean"? How is it related to decay time and bass sound quality?

Modal ringing is similar to reverb decay, but it's highly frequency-selective so it gets its own category. As for bass sound quality, when some bass notes linger for half a second before decaying to inaudibility, those notes will conflict with subsequent notes. If you've ever played chords on a piano, you know that closely spaced low notes sound muddy compared to closely spaced high notes. So of the bass plays a walking line (like a scale), two or three notes might sound all at once.

Another problem with modal ringing that's often overlooked is it can make bass notes sound out of tune. If a room has a string resonance at, say, 108 Hz, music in the key of A can sound flat. The mode at 108 Hz is excited via sympathetic resonance by bass notes or harmonics at 110 Hz, and both pitches sound at once.

--Ethan

 

[microstrip]

Modal ringing is similar to reverb decay, but it's highly frequency-selective so it gets its own category.

So I can analyze it as the decay time of modal frequencies? Or just because it "sounds" different we must use different tools?

 

[mep]

Modal ringing is similar to reverb decay, but it's highly frequency-selective so it gets its own category. As for bass sound quality, when some bass notes linger for half a second before decaying to inaudibility, those notes will conflict with subsequent notes. If you've ever played chords on a piano, you know that closely spaced low notes sound muddy compared to closely spaced high notes. So of the bass plays a walking line (like a scale), two or three notes might sound all at once.

Another problem with modal ringing that's often overlooked is it can make bass notes sound out of tune. If a room has a string resonance at, say, 108 Hz, music in the key of A can sound flat. The mode at 108 Hz is excited via sympathetic resonance by bass notes or harmonics at 110 Hz, and both pitches sound at once.

--Ethan

I don't understand if how if a bass player is playing a walking scale that two or three notes "might sound all at once." That would mean your room was ringing like crazy and all of the notes had really long decay times and bled into each other.

 

[Ethan Winer]

So I can analyze it as the decay time of modal frequencies? Or just because it "sounds" different we must use different tools?

If you look at a typical waterfall the frequency range is from 20 Hz up to around 200 or 300 Hz, with a resolution high enough to distinguish between 70 and 72 Hz. If you look at a typical RT60 graph it's full-range and shown in octave or (better) third-octave bands. Either way you're looking at decay times in different frequency ranges. So the basic concept is the same, but the application is different.

--Ethan

 

[Ethan Winer]

I don't understand if how if a bass player is playing a walking scale that two or three notes "might sound all at once."

I'm sorry you don't understand. Maybe this helps?

--Ethan

 

[mep]

Nope, no help. Their is no explanation to go along with the graph you posted in order for it to make any sense on its own. If you play a scale on a bass guitar and the notes all blur together, there is something seriously wrong with your room. I have heard bass guitar players playing bass in sub-optimum rooms and never heard this effect.

 

[Ethan Winer]

I have heard bass guitar players playing bass in sub-optimum rooms and never heard this effect.

I'll guess you're not a musician? You probably have heard this effect but just didn't realize it. When reverb / ringing makes bass notes sound muddy, that doesn't mean that all notes are completely indistinguishable. If you were able to A/B clean versus ringing you'd immediately hear it. At least most people with normal hearing would.

--Ethan

 

[mep]

My brother is a guitar player and he plays at least 3 times a week with a professional bass player that I have known for many, many years. My brother's room downstairs where they play has all hard reflective surfaces including a tile floor. This room rings like crazy. I have heard my brother and the bass player together many times in that room and I have never heard "2 or 3 notes sound at once." And no, I don't make any claims as far as being a musician. I have said before that I own three guitars (Strat, Tele, and Seagull acoustic) and I can't play any of them. But I do know what a scale sounds like and I'm pretty sure I would hear notes blurring together if played in a scale.