The No.1 Website for Pro Audio
 All  This Thread  Reviews  Gear Database  Gear for sale     Latest  Trending
Big echoy room measurements
Old 4 weeks ago
  #1
Gear Nut
Big echoy room measurements

Hello,

I moved to a new space and measured this big rectangular room with my measurement mic.

The room is 6 meters long, 5 meters wide, and 4-5 meters to the ceiling (big space). Of course, the room is full of nice reverb and such. Doesn't help that the floor is hard wood. Looks pretty but everything is all over the place here.

I'm thinking about treatment just to improve it to a decent level. Frequency wise it's not so bad I believe, a little low on the low end, but that alone could be corrected with software (I'm using Sonarworks).

I'd need to do something with all these reverb tails and reflections and such. I can hear the transients are completely smeared with drums and such, so this wasnt news to me that something is wrong.

I attached some measurements. I set the "time" on the waterfall to something like 700-800ms so there are pretty long tails here. I have attached a 20Hz to 20kHz, one 20Hz to 500Hz, one 20Hz to 300Hz and one 20Hz to 200Hz picture just for clarity.

I also attached a floor plan drawing, maybe this could get me started with some ideas. This is a dedicated room (kinda, its my living room, but I live alone), and barely has any furniture. So basically a blank space, I'm thinking about some acoustic panels and a big floor rug to start with.

Thanks
Attached Thumbnails
Big echoy room measurements-frequency_response.jpg   Big echoy room measurements-20-20000.jpg   Big echoy room measurements-20-500.jpg   Big echoy room measurements-20-300.jpg   Big echoy room measurements-20-200.jpg  

Big echoy room measurements-floor_plan.png  
Old 4 weeks ago
  #2
Lives for gear
 

Quote:
The room is 6 meters long, 5 meters wide, and 4-5 meters to the ceiling (big space)
Nice! That's a very good size for a room. Lucky you!

Quote:
Doesn't help that the floor is hard wood.
Actually, Assuming there's no air gaps under that floor and its resting on concrete, not joists over a room below, that's the best possible floor you could have (well, second best after solid concrete, of course).

Quote:
I'm thinking about treatment just to improve it to a decent level.
What do you mean by "decent level"? What one person considers "decent", another might consider "terrible", while a third considers it "amazing".

Quote:
Frequency wise it's not so bad I believe,
That's impossible to say from the FR graph you posted! You have it smoothed to 1/12 octave, and the scale runs from -200 dB to +200 dB. In other words, from quieter than intergalactic space with not even a grain of dust around for hundreds of light years, to louder than the loudest sound ever experienced on planet Earth! If you set your scale to something more reasonable, such as 1/48 octave smoothing and dB from 40 dB to 100 dB, I think you'll notice that your frequency response is very far from good.

Quote:
a little low on the low end,
Plus some rather large modal problems, as well as SBIR, floor bounce, and early reflections. You will need a bit more than just "some treatment" to get that under control. It looks pretty much like what you'd expect from an empty room.

Quote:
but that alone could be corrected with software (I'm using Sonarworks).
Actually, it's a common misconception that it is possible to correct modal issues, reflections, and SBIR in an empty room using EQ. It is physically impossible. But they don't tell you that in the manual! They make it sound like you could set speakers inside a concrete sewer piped, fire up their software, and all of a sudden it will sound like Blackbird Studio C... The other thing they don't tell you in the manual is that you can only successfully use EQ once the room has been fully treated acoustically, to eliminate or greatly attenuate all of those issues: modal, SBIR and reflections. Once those are dealt with, THEN you can use EQ to fix the final details. If you try to do it before that then you can indeed flatten the FR curve a bit for one single location in the room, but at the expense of causing other issues, and messing up other locations in the room. Technically, EQ can only be used for issues that are "minimum phase", and there's not many of those in a small untreated room! Suitable acoustic treatment installed at the correct locations for each problem can improve the minimum phase region, and make it possible to use EQ.

Quote:
I'd need to do something with all these reverb tails and reflections and such. I can hear the transients are completely smeared with drums and such, so this wasnt news to me that something is wrong.
Definitely. It's hard to see the actual acoustic details based on the very large scales that you used for your graphs, but its still evident that there's some rather large issues in there that need fixing.

If you post the actual MDAT file, then we can analyze it for you properly, to highlight the issues, then you can decide on what treatment to do for each.

Quote:
I set the "time" on the waterfall to something like 700-800ms so there are pretty long tails here.
Setting the time scale so large hides a lot of the details. If you the "use defaults" button, you'll get a lot more useful information. Also, you can find the actual decay times on the IR and RT60 windows, both for the entire spectrum and for the individual frequency bands, measured in several different ways. That's a lo more accurate than trying to estimate visually from a waterfall plot or spectrogram.

Quote:
I also attached a floor plan drawing, maybe this could get me started with some ideas. This is a dedicated room (kinda, its my living room, but I live alone), and barely has any furniture. So basically a blank space, I'm thinking about some Acoustic Panels and a big floor rug to start with.
Forget the rug. Carpet does the exact opposite of what small rooms need: it sucks up the highs, messes with the mid range randomly, and does nothing at all for the lows. What small rooms need is massive absorption in the lows, some controlled treatment in the mids, and little to nothing in the highs. Also, laying a rug on your floor is going to trash that excellent acoustic surface you have down there! Just leave the floor bare for now, until all the other treatment is in place, then if you find that the highs are a little to bright still, maybe consider a very small rug under the mix position, as the last tweak.

Before you start with treatment, you will first need to set up your room geometry correctly.

To start with, symmetry is critical: the left half of your room should be a mirror image of the right half, as seen from the mix position. At the very least it must be like that for the front half of the room: it's not quite so important for the part behind your head, but from your ears forward to the front wall it is extremely important to have a perfectly symmetrical setup. Also, your speakers should be pointing down the longest axis of the room, not across the short axis. So, rotate your layout 90° to the left or right, so you are facing one of the short walls. Probably the best orientation in your room would be to face the windows, which leaves the entire opposite wall available for treatment. If you faced the other way, then you'd have to cover your windows with deep, thick acosutic treatment.

Then get your speakers off the desk and onto stands, up tight against the front wall with just a small gap between the speaker and the wall (about 4" / 10cm). The stands need to be heavy, massive, rigid, and sturdy. Some people just pile up concrete blocks or bricks and wrap that with nice looking fabric, or use stands made from hollow steel section then filled with sand. Etc. Set up your speakers around 20% to 30% of the room width on the stands, set up the mix position (your head) between about 33% to 43% of the room length, and rotate both speakers so they are aiming at a spot about 18" or so behind your head. That's the theoretical "good" layout, and then you can use REW to optimize that.

Then, for treatment, firstly you will need massive bass trapping. I would suggest "superchunks" in your room corners, plus deep absorption across the rest of the rear wall, 4" or even 6" absorption panels on the first reflection points, a hard-backed ceiling cloud, and treatment on the front wall around and between the speakers. That should get most of the issues subdued somewhat, making the room "decent" but not fantastic, then you can decide on what further steps you might want to take.


- Stuart -
Old 4 weeks ago
  #3
Gear Nut
Thanks Stuart, I think I understand most what you said! I also attached the mdat file I did. First time I used Room EQ so not sure about some things (though read about it a little bit). The desk can be moved, I guess facing the windows at 2/3rd of the room would be good?
Attached Files
File Type: mdat first.mdat (3.16 MB, 4 views)
Old 4 weeks ago
  #4
With a room this size, all of your problems will be very low in the frequency range. The good news is you have enough space for a lot of treatment.
Old 4 weeks ago
  #5
Lives for gear
 

Quote:
Originally Posted by breakline View Post
Thanks Stuart, I think I understand most what you said! I also attached the mdat file I did. First time I used Room EQ so not sure about some things (though read about it a little bit). The desk can be moved, I guess facing the windows at 2/3rd of the room would be good?
Great! I downloaded your REW file and did some basic initial analysis, and things are pretty much as I thought they would be... and rather far from where you thought they were. So take a seat and a deep breath before you read on!

I have posted images below, with graphs from REW done at more realistic scales and smoothing. The first one below is the frequency response across the full spectrum, from 12 Hz to 22 kHz, and you can see that it is not so flat at all! It's not extremely terrible, so there must be some treatment in the room already: Some type of absorption, it seems... maybe furniture? Drapes? Something else? But its still not very good.

However, FR is not really the most important graph to be looking at: it's the time-domain graphs that are much more important. Those show how the levels change over time, after the speakers stop playing. In other words, they basically show how the ROOM reacts if you play pink noise through your system at a constant level, then suddenly cut that off instantly. Or another way of thinking about it: what would happen if you were to make an extremely loud, extremely short "impulse" sound in there, such as a gun shot, or a balloon popping. Time-domain graphs show how the room Responds to that Impulse, so they are called "Impulse Response" graphs. There's one button on REW that shows the actual impulse response, but making sense of that is rather complicated and convoluted. It's better to look at the waterfall plot, the spectrogram, and the RT60 graph, which are still based on the IR data, but are more intuitive and easier to understand than the pure IR graph.

So, that's the next graph in line: the spectrogram. What that shows is how levels change over time, where time is plotted up the page, frequency is across the page from left to right, and levels are plotted in different colors. Each long spike going up high means that there is some type of resonance going on at that specific frequency. The length of the spike tells you how long that resonance lasts after the speaker cut off, and the colors tell you the intensity of that resonance. So a long red spike means a major resonance, while a short blue spike is minor and low level. I have also added a spectrogram graph from the final test we did in the room of one of my clients, showing what the spectrogram looks like for a very well treated and tuned room. As you can see, there are no spikes at all: just smooth, clean bands of color all at the same level, across all frequencies. That's what it should look like, for a professional studio control room. I have also added the frequency response curve for that same room, so you can see how that should look. Compare those to graphs form the "good" room against your two graphs, and you can see the differences.

Next up after that, is your waterfall plot. That's sort of similar to the spectrogram, but in 3D: frequency still runs across the page, but now intensity runs up the page, and time is shown as running "towards" you, coming out of the screen. It's the same data, just displayed in a different way. You can clearly see the modal ringing (and other stuff) in that one.

Finally, the last two images are "RT60" graphs, which show the decay time for each specific frequency band. First comes yours, then comes one from a properly tuned room. And this shows your REAL problem in your room, very clearly! The decay time are all over the places, anc very, very long! They range from 500 ms in the high end (indicating that you have some treatment in there, perhaps drapes or furniture), to "off the scale" in the low end: longer than 900 ms. What you SHOULD have in there, is something similar to the last graph: basically, a flat line across the entire spectrum, showing identical decay times at all frequencies, and at around 280 ms. That's the target decay time for that specific room: your room might be different... it would need to be calculated. But somewhere in the range 200 to 300 ms would be about right, probably. Certainly not where you have it now, which is around 800 ms!

So, the final conclusion is that you need some pretty major treatment in your room! Proper layout, heavy bass trapping, first reflection points, ceiling cloud, then additional as required after further testing. The good news is that it's a good sized room, so it has great potential: it could be a very nice room, if treated properly and tuned well.

- Stuart -
Attached Thumbnails
Big echoy room measurements-breakline-rew-fr-12-22k.png   Big echoy room measurements-breakline-rew-sp-12-500.png   Big echoy room measurements-crfkus-rew-sp-20..500-1..48-final.png   Big echoy room measurements-crfkus-rew-fr-20..20k-1..24-final.png   Big echoy room measurements-breakline-rew-wf-12-500.png  

Big echoy room measurements-breakline-rew-rt-40-12k.png   Big echoy room measurements-rdmous-rew-rt60-final-40-12k.png  
Old 4 weeks ago
  #6
Gear Nut
Quote:
Originally Posted by Soundman2020 View Post
Great! I downloaded your REW file and did some basic initial analysis, and things are pretty much as I thought they would be... and rather far from where you thought they were. So take a seat and a deep breath before you read on!

I have posted images below, with graphs from REW done at more realistic scales and smoothing. The first one below is the frequency response across the full spectrum, from 12 Hz to 22 kHz, and you can see that it is not so flat at all! It's not extremely terrible, so there must be some treatment in the room already: Some type of absorption, it seems... maybe furniture? Drapes? Something else? But its still not very good.

However, FR is not really the most important graph to be looking at: it's the time-domain graphs that are much more important. Those show how the levels change over time, after the speakers stop playing. In other words, they basically show how the ROOM reacts if you play pink noise through your system at a constant level, then suddenly cut that off instantly. Or another way of thinking about it: what would happen if you were to make an extremely loud, extremely short "impulse" sound in there, such as a gun shot, or a balloon popping. Time-domain graphs show how the room Responds to that Impulse, so they are called "Impulse Response" graphs. There's one button on REW that shows the actual impulse response, but making sense of that is rather complicated and convoluted. It's better to look at the waterfall plot, the spectrogram, and the RT60 graph, which are still based on the IR data, but are more intuitive and easier to understand than the pure IR graph.

So, that's the next graph in line: the spectrogram. What that shows is how levels change over time, where time is plotted up the page, frequency is across the page from left to right, and levels are plotted in different colors. Each long spike going up high means that there is some type of resonance going on at that specific frequency. The length of the spike tells you how long that resonance lasts after the speaker cut off, and the colors tell you the intensity of that resonance. So a long red spike means a major resonance, while a short blue spike is minor and low level. I have also added a spectrogram graph from the final test we did in the room of one of my clients, showing what the spectrogram looks like for a very well treated and tuned room. As you can see, there are no spikes at all: just smooth, clean bands of color all at the same level, across all frequencies. That's what it should look like, for a professional studio control room. I have also added the frequency response curve for that same room, so you can see how that should look. Compare those to graphs form the "good" room against your two graphs, and you can see the differences.

Next up after that, is your waterfall plot. That's sort of similar to the spectrogram, but in 3D: frequency still runs across the page, but now intensity runs up the page, and time is shown as running "towards" you, coming out of the screen. It's the same data, just displayed in a different way. You can clearly see the modal ringing (and other stuff) in that one.

Finally, the last two images are "RT60" graphs, which show the decay time for each specific frequency band. First comes yours, then comes one from a properly tuned room. And this shows your REAL problem in your room, very clearly! The decay time are all over the places, anc very, very long! They range from 500 ms in the high end (indicating that you have some treatment in there, perhaps drapes or furniture), to "off the scale" in the low end: longer than 900 ms. What you SHOULD have in there, is something similar to the last graph: basically, a flat line across the entire spectrum, showing identical decay times at all frequencies, and at around 280 ms. That's the target decay time for that specific room: your room might be different... it would need to be calculated. But somewhere in the range 200 to 300 ms would be about right, probably. Certainly not where you have it now, which is around 800 ms!

So, the final conclusion is that you need some pretty major treatment in your room! Proper layout, heavy bass trapping, first reflection points, ceiling cloud, then additional as required after further testing. The good news is that it's a good sized room, so it has great potential: it could be a very nice room, if treated properly and tuned well.

- Stuart -
Thanks Stuart, very good info! So would you suggest trying to move things a bit, buying some basic absorption panels, and try measuring it again, or should I just get an acoustic engineer?
Old 4 weeks ago
  #7
Lives for gear
 

Quote:
Originally Posted by breakline View Post
Thanks Stuart, very good info! So would you suggest trying to move things a bit, buying some basic absorption panels, and try measuring it again, or should I just get an acoustic engineer?
That depends on how good you want it to be. I guess that sounds a bit trite, but it's true. What is your final goal for this room? That's the key. If you plan to just use it as a hobby studio to have fun with your own mixes, or maybe for a band of your buddies making tracks for friends and family, then you could just get by with some basic treatment. But if you are planning to use the room professionally, to produce good mixes that other people will pay you for, that's a different story: in that case, it would be worth spending more time and money to get it pretty good. And if you are planning to master the debut albums for the Next Greatest Biggest Artists in there, then you'd probably be wise to take it to the limit: get the acoustic response as good as it possibly can be.

Basically, that's your call! As a studio designer, I'm always eager to see EVERY room made as good as it possibly can be, and that's what I always recommend to everyone... but its the studio owner that has to decide if that's necessary for his studio. If it's just going to be a fun place to play around with personal, private mixes that will never see any public play, it's might not be justifiable to go for world-class acoustics, for that specific person. But there's still an awful lot of home-studio builders that want their place to be fantastic, even if the purpose is not high-end professional mixing! Just like many people want their houses to look great, even though they never plan to have them featured in an architectural digest, or they want the meals they prepare for themselves to taste fantastic, even though they are not running a Cordon Bleu restaurant in their home.... while others are happy to live in a hovel and eat McDonald's every day....

Realistically: re-orienting your room to get good geometric layout and symmetry, then installing some thick bass traps and treatment at your first reflection points, is going to get you to that first stage: fine for entertainment and impressing friends and family. Anything beyond that, for more professional work, is going to need more effort (and money).

Sorry I can't give you a more definitive answer!

- Stuart -
Old 4 weeks ago
  #8
Lives for gear
What's your goal with this room?

What's your budget?

We assume no Sound-Proofing is recquiered, right?
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump