Room Treatment

[Skol303]

And here are the results...

I’ve included some notes to help make sense of the different graphs for anyone new to acoustics…

Frequency response
The first graph shows the full range frequency response from 20Hz - 20kHz with 1/24th octave smoothing (pretty much “warts and all”). The second graph is the low end from 20-200Hz with no smoothing at all.

Notice that the dip around 60-70Hz has been greatly reduced. The frequency response below 200Hz is now within +/- 5dB right down to 30Hz… that’s quite an achievement for a room this size, if I do say so myself! (+/- 10dB is generally regarded as “good”; +/- 5dB as “excellent” and +/- 3dB as being highest professional spec… and very rare, even in the best studios).

In fact the response is now within +/- 5 or 6dB across much of the frequency range, not counting the sharp dip of around -10dB between 600-700Hz caused by a desk reflection (nothing I can do about that without changing the desk). Thankfully, steep nulls of this type - and at that frequency - aren't too problematic, as one's brain mostly 'fills in the gap' when listening. Clever brain!

Edited May 5, 2018 by Skol303

[Skol303]

Low frequency decay (waterfall graph)
This shows how long it takes sounds to decay (reduce in energy) across the low frequency range from 20-200Hz. Notice the longer decay times below 50Hz, partly caused by a room mode (peak) at 32Hz and general background noise. You can spot some other modal frequencies between 150-200Hz: they're the spiky bits sticking out of the graph, showing frequencies that are 'ringing out' for longer than others.

It's difficult to control the very low frequency decay times (below 50Hz) in such a small space without using excessive amounts of fibre and/or pressure-based bass traps, which I just don’t have sufficient space for. If your own room has long decay times in the more audible range above 50Hz, then you might want to think about trying to fix them… otherwise they can lead to frequency masking and ‘one note bass’ problems.

[Skol303]

Reverberation time

Reverberation is measured by the time taken for sound to decay by 60dB from its initial level (known as RT60). In a professional control room, the reverberation time should be between 0.2 - 0.4s from 200Hz - 4kHz in order to meet industry standards. Based on the size of my room, the recommended reverb times are much quicker as follows:

• Recommended average of 0.1 - 0.2s from 200Hz - 4kHz. My result: 0.11s
• Recommended <0.46s at 63Hz. My result: 0.32s

(NB: you can calculate the recommended RT60 time for your own room here).

So my room has a much quicker reverberation time than the industry standard, but it's within the recommendations for a room of its size.

It's interesting to note that the recommendations for small rooms basically result in what's called a "dead room" (i.e. very short reverberation). This makes sense for two reasons: 1) Small rooms need a lot of acoustic treatment to control low frequencies, and the more treatment you add the shorter the reverb times will be; 2) Small room reverberation nearly always sounds terrible! So don't be put off by the term "dead room". Unless you have a large space with a nice, natural reverberation, dead is exactly what you want... resulting in what some call “Big Headphones” - i.e. a room with minimal reverberation.

I'd argue this applies to recording rooms as much as it does mix rooms. Better to record dry and add high quality reverb using software (or re-amping in a nice big room), rather than printing crappy, small room reverb all over your tracks.

[Skol303]

So how good are these results?

My room now has a well-controlled frequency response that meets "pro" standards. Its decay/ reverberation measurements are also in good shape, but they don't meet professional standards - and they never will. It's just not possible in a room of this size.

Perhaps most crucially, the room sounds good to my ears. My reference material has a new lease of life: clear, punchy lows and smooth highs that don’t have anywhere near the harsh/brittleness that I was experiencing before. The soundstage appears wider and I can hear a lot more detail. Everything just sounds better. And my old mixes sound quite flat and lifeless by comparison… so I’m looking forward to putting this newfound clarity to good use.

Ok, but how does it compare to a professional studio?

Very few pro studios publish their acoustic measurements for obvious reasons - i.e. they’re typically not as pretty as they’d like! So it's difficult to find any data to compare against.

But as an example, here’s a measurement from Sunny Side Studios, a professional recording, mixing and mastering facility in Belgium (taken from the article in the link above). This is the low end frequency response from 20-200Hz, which is typically the most difficult section of the frequency response to get right. The measurement from Sunny Side Studios shows a variation of around +/- 5dB (10dB from highest peak to lowest dip).

And here’s the same frequency plot at the same scale from my room… a converted garage in sunny Manchester… which also shows a variation of just +/- 5dB across the same range (well, at least down to 30Hz). Not too shabby at all.

Of course this doesn't mean that my room acoustics in any way measure up to those of a high-end studio like Sunny Side. They are leagues apart. But it hopefully shows that with the right amount of stubborn, methodical persistence, you can get very usable acoustics in a small room.

Job done. Time to make some music...  

Edited May 5, 2018 by Skol303

[lurksalot]

A brilliant adventure in sound there Paul  and a superb explanation .

i can't help visualising the padded cell you referred to early on , especially as that is where I would be after tackling a project like that 

[Skol303]

^ Thanks bud! Acoustics is a big subject and I didn’t really know where to start or what to aim for when I set out... so hopefully some of the detail above might be useful to other noobs.

But yeah, it took quite a lot of faffing around - probably more than the industry standard recommendation for faffing around!  - but it was also strangely enjoyable.

...I clearly need to get out more.

[Skol303]

In case anyone finds this useful... slim chance, but you never know!... here's a link to a whole bunch of acoustic measurements from different rooms, which I compiled over on the Gearslutz forum:

https://www.dropbox.com/s/bt7xvsok7w3r4u6/Gearslutz.mdat?dl=0

It allows you to view and compare what the acoustics of different people's rooms 'look like'. Each measurement contains notes on the size and volume of the room, plus a link to the relevant thread on gearslutz.com where you can read up on the methods of acoustic treatment used.

You'll need a copy of Room EQ Wizard software to open the file, which is free to download from here:

https://www.roomeqwizard.com

Maybe useful to anyone who's planning to improve the acoustics of their room and wants to get an idea of what the end result might look (or more importantly sound!) like.

 

Edited August 2, 2018 by Skol303

[Mornats]

Homework for the geeks amongst us. I like it

[funkle]

Any comment on room EQ software like Sonarworks?

I’ve had a good experience with it.

I have a small room that I have absolutely zero treatments on. It’s not politically possible, as it were, to make that happen (yet). Sonarworks did an amazing job in that setting of making the monitoring usable. At least IME. 

EQ of course doesn’t sort out early reflections, reverb, etc in a room. And the effect is really only good for a small area between my near field monitors. But it has made an appreciable difference here...

[Skol303]

1 hour ago, funkle said:

Any comment on room EQ software like Sonarworks?

Yep, I would certainly vouch for Sonarworks (having used it myself) and also Dirac Live (which I use now). The latter gets a better grip on the low end in my experience; at least that's what I observed from acoustic measurements taken in my own room.

Room EQ software does nothing to fix boundary reflection nulls and slow decay time problems, as you mention. But it's great for reducing peaks and allowing you to control the overall shape or 'house curve' of your room's frequency response, in ways that are difficult to achieve with acoustic treatment alone. Some pro studios also use Trinnov, but you need around £4K to burn for that.

I'd normally recommend EQ software as the 'icing on the cake' - something to use once your room acoustic treatment is otherwise finished. But if circumstances dictate that it's the only option available then sure, go for it! 👍

[funkle]

Here's my completely untreated, carpeted (eeeek!), weirdly shaped attic room which I am forced by necessity/politics to use as my little music mixing room. Or at least what Sonarworks makes of it....

The bass null around 80Hz and mids null around 300Hz are pretty significant, as is the huge chunk of treble lost around 9-14kHz. You can see my monitors don't really put out anything significant below 50Hz (Equator D5s). And some interesting resonant peaks at 150Hz and 7-8kHz.

(My headphones do have more low low bass, oddly enough - Denon AH2000. Graph below.)

Sonarworks doesn't do FFT analysis so I can't tell you about reverb time. Given that the entire floor is covered in carpet, I assume it's not going to be huge...idk though...

 

 

Given that I can't uncarpet the room, knock down walls, or turn it into my own personal studio (the space is shared with the good lady), all I can say is that Sonarworks has seemed to take what was a very weird sounding space and at least made it usable.

The more I have been reading about acoustics, the more I am convinced that I'm going to need to design a room for this in the fullness of time. I'm just going to do the best I can with what I have for now.

Quick question for the experts: how bad is this room as it is? (Sonarworks obviously does a lot of corrective EQ on the various nulls and peaks, which seems to work in the very small area that is my listening zone...)

Thanks. 

 

Pete

 

 

[Mornats]

I'd be happy with a curve like that in my room  The caveat is that I've got a massive dip in the lows. I've not measured it but it must be -15db or more. As a result I can't mix anything with bass on my monitors and have to rely on headphones. I'm sure Sonarworks would sort your dips and peaks out. If I can catch the upgrade from headphone edition (that I have) to the studio with mic in a sale I'll share my measurements on here.

[funkle]

Thanks Mornats. 

I’m going to fire up Room EQ Wizard and see if I can get an idea of what the reverb decay is like in the room as well. The ‘clap test’ sounds fine, if somewhat unscientific. 

Sonarworks has a setting (with long latency) which minimised phase issues with the EQ correction. Gives high latency, but it sounds good....

[Skol303]

14 hours ago, funkle said:

Here's my completely untreated, carpeted (eeeek!), weirdly shaped attic room which I am forced by necessity/politics to use as my little music mixing room... how bad is this room as it is?

Hi Pete,

Firstly, it's "not too bad"! It looks like a typical untreated room and I've seen a lot worse.

Sonarworks is a great tool for room EQ; but a poor one for actually measuring room acoustics. I assume you must have a measurement microphone? If so, download a copy of Room EQ Wizard (for free) and use that instead - it's a different league and you'll get a much clearer picture of how your room is behaving.

That said, from the Sonarworks image you've posted I can suggest that:

• The low frequency dip around 80hz is very common, to the extent that literally every small room will show a similar dip (usually somewhere between 60-100Hz). It's caused by a standing wave (aka a 'room mode') and/or what's known as SBIR (surface boundary interference response). Likely a combination of both. Try pushing your speakers right up against the front wall - literally within a few millimeters, which should help to lessen it a bit. Beyond that you need install acoustic treatment, especially on the rear wall. 60cm depth of fluffy insulation (same as what's in your loft) is ideal.
• The peak at around 160Hz is likely caused my another room mode, this time resulting in a peak rather than a cancellation (here's a handy tool for working out the modes that are present in your room). Sonarworks should be able to get that under control to a large extent.
• What looks like another cancellation dip around 300Hz (i.e. caused by sound reflecting off walls/ceiling).
• Things then get a bit weird in the high end! The peak at 8Khz, followed by the dip around 12kHz, are also likely to be boundary reflection/cancellation problems (aka comb filtering), but it's unusual to see such a large dip in the higher frequencies. Possibly something to do with the shape of the room itself.
• Carpet will do very little to reduce reverb time (other than in the high frequencies). What you're not seeing here are the decay times, which are arguably just as - if not more - important than the frequency response, especially in the low end (where long decay times are notorious for causing 'frequency masking' - i.e. making it difficult to discern between different bass notes). Get REW hooked up, as you mention, and give it a shot with that instead. Unless you're able to clap at 50Hz...  

[Skol303]

13 hours ago, Mornats said:

I'm sure Sonarworks would sort your dips and peaks out.

Peaks most likely yes; dips most likely no 

[funkle]

Thanks Skol. I’ve downloaded REW and will get going with it over the next few days. Thanks for your very detailed input!

[Skol303]

You're welcome!

This is one of the best introductions to REW and acoustic measurements that I've found.

Follow the advice there and you won't go far wrong  

[funkle]

Ok, I'm still wrestling with REW and how to get it to show what I want. The measuring side seemed straightforward enough - Sonarworks calibrated omni mic and mic cal file loaded into REW, getting the software set up etc.

It's also complicated by the fact that I got a second screen and had to shift my studio monitors around a little to accommodate it. So the Sonarworks measurements will likely not be directly comparable, as I've moved the monitors forward a good 6 inches.

Nonetheless I have some stuff I can post.

SPL/freq chart smoothed to 1/24. This is averaged from a few responses in the sweet spot and moving it around by an inch or two: 

I'm guessing -55 dB is the baseline here.

 

Lots of ongoing ringing out in the low end. And the mids....lol

I'm going to need a little help interpreting this one. 

My take on it is it looks like I could use some fairly major bass trapping. That's an understatement...

@Skol303 what do you reckon mate?

Thanks kindly once again. 

 

NB. The shape of the space is pretty odd. It's in an attic room; 3 of the 4 walls are basically shaped/angled to match the roof (!); the other wall is flat but has virtually no space to put corner bass trapping in as a radiator is wedged in there. I'll take pics of it, but basically if I'm going to do bass trapping I'll probably have to do it on the ceiling, on the single flat wall and where the flat wall meets the ceiling....

 

 

[Skol303]

REW results! Excellent. Now we can get to work...

Email the REW .mdat file to me here and I’ll take a proper look at it for you: [email protected]

You ideally need to measure left speaker on its own; right speaker on its own; and then both speakers together (but not a big problem if you haven’t done so).

At first glance your frequency response looks very typical of a small room; no concerns there other than the big dip in the high end which is unusual.

I’d need to adjust the view on your waterfall graph as it’s currently just showing a load of ‘noise floor’ (background hum in the room) which is masking a more clear interpretation of the results.

I can also check the impulse response for any significant reflections (again hard to tell from the view you have above).

Email over the file and I’ll post some thoughts here, along with some new graphs 👍

[Skol303]

PS: photos or a rough sketch of the room with measurements would be useful.

[funkle]

Ah mate. You’re a gem. I really appreciate this. 

I’ll redo them as L on own, R on own, and together. I haven’t figured out how to do both together yet but will do. I’ll try and do a sketch of the room as well. It will shed some light I suspect...

[funkle]

Email coming @Skol303. Very much appreciated.

[charic]

This thread is awesome!  I'm hoping to be planning out a garage conversion in the near future so I'm sure I'll be referring to it plenty!

Although I'm thinking of going slightly more simplistic and aiming for a "practice room" type environment.  Keep noise in and neighbours happy type thing.  Although, still not going to want horrible reflections mushing up the sound either.

Just more taming of a room than "treatment" is what I'm expecting!

[Jus Lukin]

-

Edited February 27, 2022 by Jus Lukin

[WinterMute]

Been absent from this thread for a while, been building studios for other people, just popped in to say it is entirely possible to increase isolation in any room, but it's likely to be reasonably expensive and not especially easy to do unless you've a good builder or a VERY handy DIY'er.

I did mine with the help of a very decent local builder, but I had to lay out exactly what I wanted in terms of the construction and materials, it cost about £8K to fully proof the garage to the point where I can work at proper levels into the early hours without disturbing the good lady wife or the boys.

As pointed out above, isolation is not acoustic treatment, and you'll need to do both if you want the room to be useful with other people in the vicinity.

 

Skol: your room looks grand mate.