Mix Room Acoustic Treatment - Let the fun (again) begin.. Update #12

grappagreen

Hi All,

So I thought I'd start a post on this just in case anyone is interested in understanding a more budget orientated approach to creating a decent sounding environment. This isn't going to be an @Octatonic style build(!) but the results will be worth the effort. The rationale for this thread is that anyone who fancies improving their space can hopefully follow some of this, understand the what/why and tailor to suit.

I've been working in this room for a while but predominately on cans due to the challenging acoustics; In honesty I hate headphones and whilst it's entirely possible with some strategies to mitigate the challenges, I just find it tiring and somewhat less enjoyable. Even though stuff like Waves NX and the like give more of an out-of-the-head experience for me it's just less fun and much more fatiguing. 

So.. the criteria is;

• Effective as far down the spectrum as is physically possible within the space constraints
• Easy self-build
• Look good
• Room should be equally tight across the spectrum (not dead though) and retain a sense of decent acoustic size
• As little change to the physical structure as possible
• Mobile so that I can take it with me if/when I move 
• Not offensive to the wife (failed already!)

The room itself is a reasonable size at 5m x 4.5m x 2.4m which is good, but anyone in the know will immediately spot the first challenge from a room modal build-up perspective. More on this as we progress through the thread over time.

As a teaser here's a couple of the prototype panels that I put together to validate the design and work out the production creases.

Wall Panel Construction - Dims: 1500 x 600 x 425



Walls Panels Complete + Corner/Ceiling Traps Dims: 1200 x 600 x 225



Finally I want to give you a loose analogy and maybe you'll see why this stuff is important if you actually want to hear music at decent resolution on speakers. It will also highlight some of the madness of human beings. Imagine you purchased an SLR camera, lens and a filter that fits to the front of the lens. Lenses come at certain quality levels and so do filters. Let's say 1 = poor and 10 = amazing. If you bought a 1 lens and a 3 filter then moving to a more expensive 2 or 3 lens would make sense. If you went and purchased a 10 lens and popped the 3 filter in front of it then you wouldn't notice any improvement and, whilst you'd have a warm and fuzzy feeling about your new purchase and have bragging rights to all your mates (and everyone on FB blah, blah), you'd have wasted all that money. You'd tell yourself you notice the difference but you'd just be fooling yourself...
In our work the lens is your speakers and the filter your acoustic space. An untreated room is usually about a 1 on the quality scale and yet people spend fortunes upgrading speakers, buying better AD/DA converters, buying expensive cables, adding speaker isolation stands and leaving the room as-is... it just makes me die.. the world is a mad, mad place.. Of course I do understand that filling up a space with acoustic treatment is not always possible but even a very modest amount of treatment can make a massive difference and it's cheap in comparison to that new pair of KH310's you're drooling over and you won't actually be able to hear!

Throughout the process I'll highlight some of the theory in a simple and understandable way, I'll discuss what is actually going on and what is important and will also talk about and show some measurement data.

Catch you soon (or not if you're completely uninterested!),

Si

Musicwolf

Have you considered employing a limp-mass membrane approach in order to extend the lower end effective range of the traps?

https://www.soundonsound.com/techniques/studio-sos-making-small-room-sound-good 

https://www.soundonsound.com/techniques/studio-sos-building-diy-vocal-booth 

I'm planning to do this on my next major re-fit

grappagreen

Musicwolf said:

Have you considered employing a limp-mass membrane approach in order to extend the lower end effective range of the traps?

https://www.soundonsound.com/techniques/studio-sos-making-small-room-sound-good 

https://www.soundonsound.com/techniques/studio-sos-building-diy-vocal-booth 

I'm planning to do this on my next major re-fit

Thanks for the post (one more than I expected!)

I already have 6 MLV pressure traps... more on this later

Si

Musicwolf

grappagreen said:

I already have 6 MLV pressure traps... more on this later

Si

I am experiencing panel envy already.  We need details / pictures ASAP

roberty

I treated my living room for listening. It's basically a 3x3x3m cube, the worst possible size. Put floor to ceiling traps in three corners, first reflection filters on the side and back walls, then added room correction DSP. It is incredibly clean and accurate in there now

I'm getting my office room ready for recording and am thinking about treatment in there, too. I might just bring the panels upstairs from my living room when I get into it though

I'll watch this thread with interest, thanks

Supportact

I'll be following this! I've already built some wall panels for our live room but I'm hoping to upgrade some other bits later this year. It's probably going to involve making or buying a ceiling cloud and some corner traps and/or diffuser panels, so I'm on the lookout for ideas. 

grappagreen

Update #1

Let's ask ourselves a question. What is sound? A sensible answer might be 'stuff we hear'. Nothing wrong with that (albeit we do actually feel some it when it comes to the very low end). Another answer might be 'it's a response to phenomena that is captured and interpreted by our auditory system'. No matter, we all know what it is. 
Ultimately its vibrational energy that is propagated through a medium (air, water etc.) and is captured and then perceived by our brains.

Why the heck am I talking about this in a post about treating an acoustic space on a  guitar-centric forum you ask? Well I think it's important to have some idea what we are dealing with if we want to understand and potentially influence it. You may not and that's cool.

Your ability to hear is absolutely amazing (I'm assuming your body is working normally when I state this!). You are equipped with a binaural hearing system i.e. two ears and when stimulated they allow you to perceive and map a 3 dimensional image in your head of what is going in your environment, the type of environment you are in and much more. It's incredible really. Why do we have such a system? I don't definitively know but a strong contender is that it is tied to our desire to survive. It's useful to know where a sabre-tooth tiger is, how far it is away, whether it's coming or going etc. Whilst sight is your primary sense, sound is secondary and they are tightly integrated. Eyes have a narrow field of vision but ears allow you a much wider conceptualisation of the environment you are in and the dangers that lurk therein.

Our perception of sound is the subject of study in psychoacoustics and whether you know it or not, whenever you are listening to a decent mix on a stereo system (set up properly), the folks who put the music together are using their knowledge of psychoacoustics to fool you.

I implore you to go and listen to some music, close your eyes and see with your ears. On quality program material you should be able to literally 'see' a image in front of you. It will have height, depth and width. Voices and instruments will be portayed with clarity across the stereo image. You will be able to hear everything in exquisite detail, It will be literally like the stuff is happening right in front of you. It isn't though. It's a ghost, a phantom if you like and it's all happening inside your head.

The challenge is that most people have never experienced the above really. If we want to experience it we need a 'good' environment.

Next time we'll talk about what 'good' means in this context. What we actually need is some sort of standard. Luckily, we're in luck!

Si

grappagreen

Update #2

Don't worry, we'll get to designing and implementing our treatment plan soon! Notice I said designing.. that's because we actually want to achieve a specific goal(s) and in order to do so we need to a) understand the goals b) implement a solution that is going to meet the goals!

So what goals are we setting out to achieve. That's where standards are useful. The great thing about standards is there are so many choose from Well in our case there aren't but no matter we're going to pick an old(ish) but useful one and see what it sets out.

With the snappy title of 'Listening Conditions for the Assessment of Sound Programme Material: monophonic and two-channel stereophonic' (EBU Tech. 3276 - 1998) we have a set of requirements that set out an acceptable quality target for critical listening. It's fairly brief at 19 pages and not overly heavy in terms of content. In brief it sets out some definitions and recommendations for configuration and then the following among other things;

• Early Reflection criteria
• Reverberation Field criteria
• Room Response criteria
• Background Noise criteria
• Speaker performance criteria

Notice the focus here: time domain stuff first and then frequency response. It's time that is the most critical despite what people think. We're not going to spend time on noise and speakers as that's not were focusing on here.

Early Reflection Criteria
These represent the sound that bounces of surfaces i.e. not directly from the speaker to our ears. Obviously the time of this reflection is dependent on the distance the sound travels and hence the size and shape of your room determines what you experience. The specification states that ER's arriving within 15ms of the direct signal should be at least 10db below the direct signal.

Reverberation Field Criteria
You know what reverb is. Basically the standard sets out the acceptable reverb time and quality of its decay. Long and short is that reverb time needs to be controlled and it should be within certain tolerances in it's tonality i.e. uniform across its range.

Room Response Criteria
This represents the frequency response. It sets out an acceptable frequency response based on a test signal (3rd octave filtered pink noise). Basically the test looks for a +/- 3db across this test between 50hz and 2000hz and then slopes off above 2k at 1db/octave i.e. a shallow slope down on the -db side.

We'll talk about some considerations and design options and then pick a design that will work for us in our challenging spaces next time. It will be brief as basically our options are very limited but it's fun to understand the different approaches. Well fun for me anyway!

Regards,

Si

grappagreen

Update #3

Now we've talked about standards and what we'd like to aim for let's briefly talk potential approaches and then swimming pools! What?!?

As you can imagine the problems we face have all been faced many times before and a number of typical solutions have been developed and applied. We'll go from the sublime to the ridiculous. We'll also see what is an option for us and our challenges in meeting the self-imposed criteria in post #1;

#1 - LEDE - Live End/Dead End. Basically absorption in the front half of the space and diffusion at the back. Generally requires lots of absorption to be effective up front and physical space for effective diffusion at the back. Smaller rooms are a challenge here.. not going to be my approach due to physical size and, the space I would need for diffusion, would limit my ability to achieve more important things (there's only so much real-estate). Correct application of diffusion also requires a certain distance that I don't have.
#2 - NE - Non-Environment. Basically a room that is acoustically invisible! As you can imagine this is a heavy lift as we need to absorb all of the sound energy. Move along, nothing to see here..
#3 - Diffusion. Now we're getting into crazy territory. Imagine if we could build a room where the energy was simply(!!!) deflected in all different directions (and timings) across the spectrum (I'm simplifying massively here). You would end up with a space where the only energy was the stuff coming out of your speakers and the 'ambient energy' enveloping you. If you fancy having a crack at this approach Google 'Blackbird Studio C' to get some ideas and get your order in for a warehouse to house it in and some MDF blocks.. Holy shit..
#4 - RFZ - Reflection Free Zone. A design where the listening position is 'free' from early reflections. In the 'proper' world this is done using geometry. The shape of the room is designed such that, in the listening position, no early reflections hit the listening spot. Given we're not changing room geometry this is probably not an approach for us. But.. if we accept a compromise i.e. that we cannot eliminate all ER's and we refer to the standard (ER's at <10db to direct signal - which doesn't say eliminate - just attenuate), then if we were to somehow absorb the energy at the points where reflections occur we could be on to something. This is what we will do to start with as it sounds like the only viable option.

I want to give you another somewhat lose analogy to what's going on in your room when a sound comes out of your speakers. Imagine a hypothetical swimming pool the exact dimensions of your room. It's full of water and the water is perfectly at rest. No energy is being added or removed. We drop two rocks into the pool at the exact positions of your speakers. As you know from experience with water in pools, chaos ensues. The kinetic energy from the rocks causes waves in the water as the energy is transferred to the water molecules. Through compression/rarefaction, energy in waves move, hit boundaries, mix together and over time it's a 'wavey' mess. Through frictional losses and transference to boundaries, the energy dissipates until at some point the water returns to a static state. This is what happens in your room with sound in air (albeit much slower), except that we have an additional boundary (the ceiling) to make matters worse. As you can imagine trying to hear what's coming out of your speakers in this melee is quite the challenge! Rooms (and swimming pools) actually have another problem; they resonate. The energy in your room will, at certain frequencies, cause standing waves to be generated that relate to its physical dimensions. These waves not only cause havoc to the frequency response (ever walked around your room and noticed that you hear the bass response change as you walk across?) but the resonant energy takes time to dissipate which cause some frequencies to 'hang about' after others have decayed away. Not good if want to hear that kick drum stop as it does in reality. We're referring here to room 'modes' and if you think you're going to achieve a nice flat sub/bass frequency response in your space after you've popped a couple of 75mm acoustic panels on a wall then think again. Imagine an elephant running at full speed towards you, trampling everything in its way. You want to stop it and all you have is the ability to clap your hands. The elephant is the low-end. Good luck!

Given the above you would think that trying to solve these problems is hard. It is and we can't actually solve most of them. The purpose of acoustic treatment isn't to 'solve'; it's to make things better to the point of being acceptable (and living with the stuff that is still unacceptable after you've tried but at least being aware of it). Let's not be downhearted however. The first change we will make will result in a change in your ability to hear sound that is so profound that, even if you do nothing else, it will have all been worth it. Seriously..

Si

grappagreen

Update #4

Ok.. so let's stop talking all this nonsense and treat the room. Actually, no. Before we do anything we want to make sure that ourselves and our speakers are in the optimum position. It's absolutely essential this is right as we build everything off this in our space. It's also free and, as you will see, correct positioning can make a significant difference to the performance of our system.

So what's the best place? Depends on your room of course! My room is 4.95L x 4.5W x 2.4H. Generally we look to run the speakers down the longer dimension but in may case I'm going to rotate through 90 degrees along the shorter dimension. Why? Because my room is symmetrical when I orient this way. Symmetry is important; we ideally want everything as symmetrical in a room as possible because it will help with the stereo imaging. There is also the practical aspect that I can treat the room in this orientation more symmetrically and also access in/out and room use is easier this way. 

We know that we will be sitting half way between the side walls (remember the symmetry bit?) but along the length we have to pick a spot. As a random choice we're going to sit 38% from the front wall. This isn't actually a random choice at all and relates to the best theoretical room mode frequency response position. Don't want to jabber on but @ 38% you 'should' be at the point where the response is flattest for the first mode. This is theoretical however and in the real world we're going to experiment and find the best spot. 

How do we do this? I'm going to use REW and do a number of tests.

Test 1
I'm going to point a speaker down into a corner with the mic at the 38% listening position (pointing at the ceiling) and then sweep a sine. I'm then going to move the mic forward and backward in 5cm increments and do the same. I will then compare and find the best listening position for MY room.

Test 2
I'm going to set up my monitors in an equilateral triangle (based on the best listening position I found in test 1) and then move the speakers forward and backwards along the sides of the triangle and sweep a sine at each position. I'll do this for each speaker individually and then for both. I can then find the best position for my speakers in relation to the best listening position identified in Test 1.

Most of the analysis here is smoothed at 1/48th octave. This is what we should be using to understand what's going on. Not trying to make ourselves feel good (and kidding ourselves) with 1/3rd octave stuff although this is useful.

Results

Test 1

Listening Position - One speaker pointing down into a corner



Notice the magnitude of change between the position highlighted and all the others!

Test(s) 2

Speaker Position - Low End View (both speakers)



Again note the magnitude of change here... all I'm doing is moving things! It ain't pretty but's certainly much better!

Speaker Position - Low End Time Domain (both speakers) in worst position



Speaker Position - Low End Time Domain (both speakers) in best position



Notice how the modal ringing at 34/38 is reduced. I have a very interesting ring at 107hz!!

Speaker Position - Overall Frequency Response @ Best Position. Right Speaker



Speaker Position - Overall Frequency Response @ Best Position. Left Speaker



Speaker Position - Overall Frequency Response @ 1/3rd smoothing. Right Speaker



Speaker Position - Overall Frequency Response @ 1/3rd smoothing. Left Speaker




So what's the takeaway here. Listening & speaker positions can make a significant difference and would should put in the effort to get them optimised. 

Actually, it might be worth highlighting some things that jump out of this initial analysis even though we haven't yet treated anything. 

• Note that a lot of the peaks and troughs in the low end line up with room mode lines (at the bottom of the charts)
• Note the peak around 160 hz that coincides with all of the modes that build up in this area
• Note the low modes that ring out most are the axial modes for the length and width of the room. Ignore the 100hz ish one at this point
• Note the frequency response charts for L&R are pretty similar
• Note the drop-off in low end below 100hz. My speakers are sealed box AE22's and have a 12db/octave roll off from 100..

The above, along with other data we will collect, will guide what we do as we start to address the issues.

If anyone is actually reading this rubbish and has any questions just let me know.

Si

grappagreen

Update #5

Maybe it's time I actually did something rather than talk about all this nonsense! 

So, we've got some targets to go for based on the EBU spec and we've optimised the listening position based on some data.

What's the approach going to be then?

Basically in small rooms we have two tools we can use; broadband based velocity absorption and narrowband pressure based absorption. Without wanting to get too deep, velocity based treatment presents friction to air particle movement and should be placed in positions where the molecules are in motion. Pressure based treatment presents resistance via a membrane, again converts energy into heat through and should be placed in positions where molecule velocity is lowest and therefore pressure highest i.e. against boundaries. A big difference between the two is velocity based covers a wide frequency spectrum, it's a bit like a shotgun whereas pressure based covers a very narrow frequency range and is more like a sniper rifle. Velocity based absorption is very low tech and is generally in the form of acoustic panels filled with a material like mineral wool. It's easy to build, install, works and is low cost. The downside of it is that it is limited in its ability to deal with low frequency energy. Designed and implemented properly it can be effective down to 40hz however this requires deep panels (or less deep with air gaps behind) and takes up lots of space. Pressure based panels are much more effective but only over a very narrow frequency range. They are much more difficult to make and the level of quality in construction is critical as, incorrectly built, they just won't work. They are however the only viable treatment if we want to solve some very low frequency issues (unless you want to look at active devices such as PSI AVAA - not really an option at £2.5k a pop and we'd need a few!).

Given we have limited space and a whole host of issues to solve across the whole spectrum we're going to velocity-based absorption to tackle as many issues as we can and then turn to pressure-based to focus on anything we can't improve. Velocity-based panels will allow us to manage early reflection objectives, decay time objectives and assist in removing energy that will improve our frequency response.

As a general principle I will be using only two types of panels. I'm doing this to just keep things as standardised as possible, Just makes constructing everything way easier and quicker.

Panel #1 Design - Early Reflection management - side and back walls.
The goal here is to design a panel that will absorb as much of the spectrum as possible at the reflection points on the side and back walls. Given we want as much absorption as possible these panels are 1500 high x 600 wide x 425 deep. I could have build panels at 200 deep and placed them 200 mm off a wall and have achieved a similar performance (slightly worse low frequency) however practically these would have to mounted to walls with a bracketing system. I don't want to 'fix' stuff and mess with the room wherever possible and hence these 'coffins' can just be floor standing and be easily moved. They are deep as I want to absorb as much energy as I can. We come to an interesting point here. Most commercial panels are way shallower than this (even those called things like mega bass trap etc.). The key thing here is what do we call effective? If a panel absorbs 1% of the energy is it effective? I would argue that it isn't and my threshold for this is that for a panel to be called 'effective' at a frequency it needs be at least 50% effective. A 100mm panel mounted to a wall is about 10% effective at 40hz! It rises to about 20% if mounted 100mm from the wall. My panel is 60% effective at the same frequency. My wife was right after all, size does matter.. I can't comment on 'marketing hype' but I think the reason you don't see commercial manufacturers making very deep panels is simply down to practical and commercial reasons. How many idiots are going to buy and install a 425 deep panel? Not many..

I'll be placing one of these on each side wall and two on the back wall at the reflection points. These points are best found by getting someone to hold a mirror on the wall. If you can see the tweaters you need a panel there! Also remember that ER's are reflections hitting you within broadly 20ms of the direct sound. This is one of many reasons why bigger rooms present less challenges as the walls are father away. Strictly speaking my back walls do not fall into the ER category as the reflection takes about 26ms to hit me but I need to treat it for other reasons so..

Panel #2 Design - Early Reflection management - ceiling.
Given these these need to be ceiling mounted we need something more manageable so these will be 1200 x 600 x 225 panels mounted 100 mm from the ceiling. These will be about 55% effective @ 40hz.

I'll be mounting 3 of these slightly forward of the listening position in between me and the speakers.

Panel # 3 Design - General Decay Time, Rear Wall Bounce and General Frequency Response Smoothing.
These will be the an additional 'coffin' and the standard 1200 x 600 x 225 panels.

I'll be situating the coffin on the back wall to supplement the two ER panels and positioning 6 standard panels straddled across the three corners of the room from floor to ceiling. I can only treat 3 as one corner has a door in the way! Grr.. We mount them in corners as low frequency energy builds up in corners (go and listen) and they are straddled so that they 'catch' both room modes. Basically we want to get most bang for our buck!

Panels are all built and I just have the ceiling panels to install next weekend. Will post some pics and some data to see where we are and how things are looking/sounding.

Si

Supportact

Finding this very interesting so thanks for going so in depth with it. I wanted to ask, for the room tests, can you post a link to the software you used if possible? Also did you need to use a special microphone to make the tests?

Looking forward to seeing pictures of what you've done so far as well!

grappagreen

Supportact said:

Finding this very interesting so thanks for going so in depth with it. I wanted to ask, for the room tests, can you post a link to the software you used if possible? Also did you need to use a special microphone to make the tests?

Looking forward to seeing pictures of what you've done so far as well!

Hi,

It's good to know that someone is actually reading some of this stuff

The software I use is REW (Room EQ Wizard) which is free to download and use. In terms of measurement microphone you basically need something omni pattern with as flat a frequency response as possible. I use a Behringer ECM8000 which is cheap and plenty good enough. People obsess about these things (like everything on the Internet) but when you're dealing with a room that could have swings of +/- 15 db across the spectrum an expensive mic is a complete waste of money! Does it really matter that the microphone could have a 0.1db variation

If you want to have a go, look on Youtube for the tutorials by Music City Acoustics on setting up and taking measurements using REW which are excellent.

Have a go and see how you get on! It's dead easy and enlightening/depressing..

Si

Supportact

Thanks Si that's good to know, I will look into it. 

robinbowes

This is all really interesting stuff. I have a degree in Acoustics, and I know enough to know that it's a difficult field to master. Your approach is sensible - you have set goals and are working through them methodically.

I'll certainly bookmark this thread for later, and investigate REW (I have an ECM8000, which I use with my DEQ2496).

Thanks!

R.

roberty

Been enjoying your posts too @grappagreen. I used REW to set up room correction in my living room and it made a world of difference

octatonic

Question- why 1/48th smoothing?

grappagreen

octatonic said:

Question- why 1/48th smoothing?

With 1/48 you can actually see what the room is doing where it matters and is most troublesome i.e. the sub/low and, to a certain extent, mid range. It's good if you want to identify issues/challenges and 'attempt' to deal with specifically with them. In reality however the human auditory system is very clever and as such stuff that looks terrible objectively isn't an issue subjectively because of the way our brains process the data. I'll talk a little about the psycho angle in the results/analysis to follow in a sec! Understanding a little about this is actually pretty important as nobody should waste time and effort on tackling problems that don't matter.

Folks sometimes lose sight of the fact that a swept sine and a microphone is nothing like music and your ears/brain! It's easy to go down rabbit holes if we're not a bit careful..

grappagreen

Update #6a

Sorry it's taken a while to get to this. A strange, orange, orb shaped object appeared in the sky and as such the wife decided 'we' had better things to do.. she was right of course

So.. recapping I've applied some treatment to try and improve 3 specific things and I'm using EBU 3276 to assess how I'm getting on.

This post is going to be a bit image heavy so bear with me..

First, a couple of pictures of the room and treatment to date. Don't have anything to hand that's wide angle so took some photos looking from the corners. Just a note that the rack and guitar gear was out of the room when I measured the environment! The spare treatment stacked above the rear panels is stuff that I have on hand to help with finding and isolating issues.. more on this later.









First area we're focused on is early reflection management. REW has a the ability to derive data to give a view on this from the impulse response that is generated as part of the sampling process. A slight theoretical side track but it's worth noting here so that we can understand more. A room is basically two rooms when it comes to acoustics. At low and sub frequencies, sound can best be understood as behaving like waves. There is a theoretical point, as first described by a clever guy called Schroeder, at which point the behaviour of sound starts to be better understood via statistical analysis. In a nutshell, sound above this 'Schroeder Frequency' can be understood as being more specular i.e. it behaves more like rays and bounces off boundaries like a snooker ball off a cushion. With early reflection management we're basically trying to limit the impact of this ray-like energy hitting us in our listening position and impacting on the frequency response and the clarity of the sound coming out of the speakers. Given the snooker ball analogy we can see why, with early reflections, we position the absorbers in very specific places i.e. the point where the ball would hit the cushion and bounce directly at our ears.

Let's look at some data. I'm only going to show the majority of this for the left speaker as the post would be very long..

Impulse Response dBFS



Some things to note here. Overall looks pretty good. We have some specular reflections early (before 2.5ms) that are being caused by my desk and ceiling however these are way better than the empty room performance. I quickly knocked up some desk angle plates that really helped with the desk reflection. I do have some reflections around 12.5 and 17 ms that I may look into further. 

A better way to see this is via energy %



This is the same data over a longer time period (and different scaling) and allows us to more easily see where reflections occur. In an ideal world (which we don't live in!) we would like to see the black 'Schroeder Integral' line be a smooth sweep downwards. We see steps in the line where we have a peak of energy.

The standard is actually pretty specific about the requirement. It states that ER's should not be above -10db for frequencies between 1 to 8k.

Luckily REW allows is to filter by frequency bands. Let's take a look. We can ignore the first bump.. These are all 1 octave band filters.

@ 1k



@ 2k



@ 4k



@ 8k



The good thing here is that for the most part is close to -10dBFS. We have a couple of peaks slightly over but not by much. We have to pragmatic here and given that it's the desk/ceiling causing these (and I've pretty much done all that can be reasonably done) we'll live with it. To improve this I'd need to consider removing the desk (or getting smaller one) and sticking a fabric ceiling in with absorption behind it. Bigger fish to fry..

We'll focus on decay tomorrow..

Si

octatonic

grappagreen said:

octatonic said:

Question- why 1/48th smoothing?

With 1/48 you can actually see what the room is doing where it matters and is most troublesome i.e. the sub/low and, to a certain extent, mid range. It's good if you want to identify issues/challenges and 'attempt' to deal with specifically with them. In reality however the human auditory system is very clever and as such stuff that looks terrible objectively isn't an issue subjectively because of the way our brains process the data. I'll talk a little about the psycho angle in the results/analysis to follow in a sec! Understanding a little about this is actually pretty important as nobody should waste time and effort on tackling problems that don't matter.

Folks sometimes lose sight of the fact that a swept sine and a microphone is nothing like music and your ears/brain! It's easy to go down rabbit holes if we're not a bit careful..

Yes, I know why we use smoothing.
Why 1/48, rather than another value such as 1/24 or 1/6?
1/48 is pretty detailed and is the treatment you are doing going to be so targeted to specific frequencies with that much granularity, or more broadband?

grappagreen

octatonic said:

grappagreen said:

octatonic said:

Question- why 1/48th smoothing?

With 1/48 you can actually see what the room is doing where it matters and is most troublesome i.e. the sub/low and, to a certain extent, mid range. It's good if you want to identify issues/challenges and 'attempt' to deal with specifically with them. In reality however the human auditory system is very clever and as such stuff that looks terrible objectively isn't an issue subjectively because of the way our brains process the data. I'll talk a little about the psycho angle in the results/analysis to follow in a sec! Understanding a little about this is actually pretty important as nobody should waste time and effort on tackling problems that don't matter.

Folks sometimes lose sight of the fact that a swept sine and a microphone is nothing like music and your ears/brain! It's easy to go down rabbit holes if we're not a bit careful..

Yes, I know why we use smoothing.
Why 1/48, rather than another value such as 1/24 or 1/6?
1/48 is pretty detailed and is the treatment you are doing going to be so targeted to specific frequencies with that much granularity, or more broadband?

I know you know James 1/48 is very detailed. I will be using a combination of velocity and pressure based specific treatment (and some diffusion/scattering) based on the challenges I find. The posts on decay time and freq response over the next couple of days will bring it all together and make more sense..