A speaker placement tip that speaker manuals get wrong (2017)
sonicscoop.com
sonicscoop.com
One thing I would mention is that rigid fiberglass corner traps seemed to do a lot in my studio.
Also if you are going to talk about room modes, you are also going to want to think about the permeability of your walls and whether they are going to even reflect the frequency in question and set up a mode.
And also we may want to make a distinction between making a listening environment that sounds amazing, and one that has utility. It can be easy for an intermediate or beginner at mixing to get really tricked by a great listening environment in to doing things that won’t translate.
On the other hand, life is short, and listening to a beautiful set up is just a better quality of life if nothing else.
If I understand things correctly, this was missing a bit from the whole (otherwise very good) explanation in the article: it constantly mentions 'walls' and each time I can't help but think that the difference between a solid concrete 30cm thick wall and a 10cm plasterboard plus some thermal insulation makes a difference in how it reflects lower frequencies. To the point that in the latter case there might be not much reflection at all and you actually have to start to measure from the wall (or whatever) in the next room?
So my advice is to get a UMIK-1 measurement mic and a 31-band graphic EQ, then download roomeqwizard.com software and learn to use it. It involves walking around your room with the mic while playing constant test tones and watching a sound spectrogram graph on your computer screen. Even though the test tone is constant, when you walk around your room with the mic you can see and hear how the tone changes due to the standing wave phenomenon. It will open your eyes (and ears) to how the room you are in affects the sound coming out of speakers. And rather than reading articles on the web you can actually experience for yourself the scientific reality of how it works, just like in high school chemistry class you see what happens with your own eyes when you mix vinegar with baking soda, rather than just reading theory in a book. The actual physical experimental confirmation is crucial.
Adjusting anything over 250-500Hz is a waste of time because if you shift your head (or measurement mic) just a foot or two you get a completely different frequency response. When you test this yourself you won't need to take my word for it or read web articles shilling snake oil audio products. Your eyes and ears and calibrated measurement mic will confirm the scientific reality for you.
As to other equipment... at the affordable lower end you want a USB or Thunderbolt interface with balanced XLR connectors connected either directly to studio monitors or to an equipment rack with the amps, crossovers, equalizers, etc needed to power big passive subwoofers. The Focusrite Scarlett 2i2 (~$160 on Amazon) is a best seller USB audio interface and crystal clear. I'm listening through one right now and I love it. Be aware that if using a Mac with a T2 security chip (2016 and newer Macs) you need to use an external thunderbolt USB hub because T2 occasionally cuts out the internal USB; scroll down this article if curious:
https://tidbits.com/2019/04/05/what-does-the-t2-chip-mean-fo...
Here is a screenshot from the article of the T2 hiccup actually being recorded: you can see the audio waveform briefly cut out:
https://tidbits.com/uploads/2019/04/T2-hiccup.jpg
None of Apple's "fixes" fixed the problem. This was the last straw for me and many other audio guys so now we use both Ubuntu and Windows 10 with WSL instead of the Mac. Most photos of $1M+ recording studios I see nowaways have Windows 10 desktops. If you don't want Win10 telemetry spying on you you need to set up a pi-hole or external firewall box, but M1 Macs and all iOS devices report your location and other telemetry to Apple now so you also need an external firewall for Apple now too, but I'm digressing from audio....
Back to audio, the UMIK-1 USB measurement mic is usually ~$99 new on Amazon/eBay/wherever. The roomeqwizard.com is free (as in gratis, not open source) Java software that can use the mic. For studio monitors, Yamaha and Adam Audio are both good and affordable. Many like the JBLs as long as you aren't too picky about hiss. Genelec is the very best and provably so by independent scientific measurement (see audiosciencereview.com measurements). But Genelec is stupid expensive, over $1k for a pair of their budget monitors. Anyway, this post is long enough and turning into an article. To summarize, get a Scarlett USB interface (~$160), the UMIK-1 ($99), a good pair of 5" or 7" studio monitors (Yamaha, Adam Audio, or maybe JBL, $300-$600 for the pair), and a graphic EQ. The bargain-basement Rockville REQ231 for ~$130 is essentially a cheap but good clone of a more expensive EQ. You can also spend $250-$400+ for Behringer or dbx/JBL which looks prettier in the rack but honestly doesn't sound any different.
- Place your speakers such that they form an equilateral triangle, with the tweeters pointing directly at your head while you are seated in the control position. Near-field monitors are made for focusing on a 'sweet spot' like this. Using purpose-built speaker stands is the easiest route here.
- Treat the first reflection points with foam, or anything that absorbs/scatters sound (e.g. I hung a huge shag rug on the back wall)
- Break up smooth surfaces (walls/floors/windows) with other absorptive/non-smooth decorations and materials. Rug on floor, couch at back of room, curtains, even a bookcase; all will help absorb and scatter sound.
- A good rule of thumb is to spend as much money on 'treating' the room as the speakers themselves (even great speakers will sound like trash in a room with no treatment).
- When using the speakers for mixing/mastering work, listen at a conversational volume. The lower the volume, the less the room matters (and the room matters a lot).
- You will get far better sound from decent headphones (I have HD650s) than from all but the most pristinely designed/tuned monitor setups. Probably would cost well over $10k and a lot of knowhow to beat the headphones.
Mix— headphones
Master— monitors
Test— Toyota Corolla (Petty, said something like "just about everybody's gonna be listening in a Toyota"—they'd literally print a tape and run it out to someone's car)
It's definitely sage advice!
† edit: Looks like I fell victim to hearsay. It's just something I'd heard so many times for so many years I assumed it was true, and it may be partly, but here's a great write up on the subject—tangential to the main thread: https://www.soundonsound.com/reviews/yamaha-ns10-story
I have a few rock CDs from the 90s (Dizzy up the Girl is one) that ar completely unlistenable in many cars; too bad Petty wasn't involved in mastering those.
Yes, treat your first-reflection points with material that absorbs as much as energy as possible. However, scattering (aka diffusion) should be a waaay lower priority at your first-reflection points. In fact, if you're absorbing with enough mass at these points there shouldn't be much high or mid-range energy left to scatter. And for the low-mid and low freq energy that isn't being completely absorbed you'd need a hilariously-large diffuser to scatter wavelengths of those size. So don't even worry about diffusion paneling of any kind on your front walls. Just absorb, absorb, absorb at your early-reflection points.
I personally use several 5.25" GIK Acoustic panels with another 5" air gap between the back of the panel and the wall. At my listening position I have them placed on the wall to my left, my right, and have several panels hanging above my head with as much of an air-gap between them and the ceiling as I can allow to absorb the ceiling-reflections. There's more nuance to how I determine the exact positioning, but that's the general layout.
--------
I'd also like to reinforce the idea that for anyone who needs to make critical audio decisions you should absolutely be investing the time, money and effort to get your room treated/tuned to it's full potential. Otherwise it will undermine every decision you make until you do. You simply cannot make good decisions without A) excellent monitoring (speakers) and B) A properly-tuned room. To ignore either of these is to deceive yourself. Consider it your highest priority before you even consider spending money on additional hi-fi audio gear beyond your speakers. If you can't hear your speakers honestly and with minimal interference from the room then there's little point in focusing on the finer details.
---------
If you really want to hear how your room is affecting your audio then play a low and steady note (say a 60 Hz sine wave) and then walk around your room while it's playing. Listen for how dramatically the volume of this single note will fluctuate as you change position. It will likely vary from overbearingly loud to near-total silence. Now consider that every other note that you can possibly hear - from the lowest sub to the highest buzz) will also be subject to these kinds of boosts and dips in volume. Then you'll have some sort of idea of what you're up against.
The most simple and imperfect route: get a global EQ for your OS and play around with a sine generator and notch filters. Fix the 1-2 frequencies where it's just way too loud.
The fiddly and free route: https://www.roomeqwizard.com/ (search for tutorials on that) and than for windows there is "Equalizer APO" to load it. Use the best mic you have, in a really bad room even an sm57 will do wonders.
The more expensive and easy route: https://www.sonarworks.com/reference with a measuring mic (they are not that expensive).
I think the worse your room and speakers are the more this helps (I lived an a square tube, now in a better shaped room the effect is much less noticeable). For me it really cleans up the otherwise muddy lows. I can recommend boosting them afterwards for a more hi-fi sound, really neutral doesn't have the punch you might be used to :).
Disclaimer: this will be worse than the perfect physical setup. You can fix some things in software, but there will be losses and it and only fix one point in your room really well. It's also a hassle to change EQ settings when you change to headphones.
IE, if you want a pleasing sound, a loudspeaker that's playing 90dB at 100Hz should be playing 83dB at 16,800Hz. And the response curve should be consistent from point A to point B.
This recommendation can be impacted by the power response of the speaker, but that's another subject...
There are two very different goals that are easy to conflate and that the article isn't very clear about:
1. If you are a music listener or music producer focusing on making music, you just want a room that sounds great to you. If you're making music, you want a sound that inspires you and gets the creative juices flowing.
2. If you are a mixing or mastering engineer or a music producer focusing on polishing and mixing your work, you want a room that gives you insight on how the audio will sound in a wide variety of listening environments. Your goal is to get the song sounding as good on as many people's speakers, headphones, and rooms as possible and your own personal setup is a proxy for that.
For example, if you only have a couple of tiny speakers that don't go below 50 Hz, your song may have subbass that you literally cannot hear. But when someone with a subwoofer plays the song, all of that comes out. If it's too loud or out of tune, the song sounds like trash, but you have no way of knowing.
So the goal is a monitor setup that is relatively neutral so that the song doesn't sound better to you than it will to others. You want a fairly flat response so that you aren't inadvertently EQ-ing the song to cancel out peaks and valleys in your room because then that will imbalance the sound in other rooms. And you want to cover the full frequency range so you can hear everything others will be able to hear. If the songs sounds great in your room, but it sounds like garbage for everyone not in your room, it definitely is not great.
Most articles about monitor placement tend to assume the goal is 2 more than 1. These days, the line is blurry because many musicians and producers are also doing a lot of the mixing.
There are HiFi speakers that are designed to provide the correct frequency response when placed near a wall and sound wrong otherwise. "Klipschorn" would be an extreme example of this category.
There are other speakers (actually the majority) that are not designed to be placed against a wall, but, if you use digital room correction features in your HiFi receiver, they still have to be placed against the wall, so that there is less work for the DSP to do.
There are also so-called "electrostatic speakers" (e.g. "Quad ESL 2912") which absolutely must not be placed near the wall, because the rear wave is in counter-phase (unlike the traditional speakers), and will cancel the front wave exactly.
You could try boosting at the frequency that the valley is at, but that makes both the signal _and the reflection_ equally as loud. Therefore, they cancel out in almost the same manner for basically no net gain.
https://en.wikipedia.org/wiki/Room_modes?wprov=sfla1
Their influence can be be diminshed with absorbers, construction of a room without parallel walls or so big that they can't be heard (eg Concert halls)
Another example may be the Ohm Walsh line of loudspeakers. I've only ever heard positive things about how close-to-life it sounds, and that's accomplished with the unique upward-facing driver and using walls and corners to an advantage.
Even with measurements like that, it's pretty nontrivial to compensate for it. It's technically a proven problem[1] (solvable for some convex shapes) but its a little more complex than just adjusting the volume or phase of a given frequency. A room -and by extension all the little nooks and crannies of a room- acts as a resonator, and will gather and sustain notes. You need to counter that sustain ahead of time.
1: https://en.wikipedia.org/wiki/Hearing_the_shape_of_a_drum
It should take a (cheap) microphone with a known response curve and a sine wave generator to calibrate any number of speakers in a room.
Amazon Alexa does the same thing, except with an array of microphones.
The problem I have is with the bass. Standing in the normal position to interact with my PC, the bass strength is low. Now, the moment I move back about 1 meter, the bass strength becomes great (the one I'd expect from this pair of speakers).
I really don't know what to do :/
You'll need to set it on both speakers.
Low E has a wavelength of a couple meters, no problems hearing that close up.
The issue is that, because of reflections, waves whose wavelengths are on the same order of magnitude as the distance from the speakers to the wall are getting cancelled out, depending on where you listen from. It is a well-known phenomenon.
(Higher frequencies which are multiples of the missing bass frequencies will also be attenuated, but this is less immediately noticeable.)
For instance, if OP has his speakers located 57cm / 22 inches from the back wall, there is going to be a very deep null at 150Hz, due to the reflection from the back wall. (150Hz is 227cm long.)
If OP is listening in the nearfield, that dip will be obnoxious.
On the other hand, if OP is listening at a distance of two meters, the dip will be LESS obnoxious because there will be dozens of dips and peaks in the response, contributing constructive and destructive interference, simultaneously.
This is one of the reasons that loudspeakers are generally measured under two conditions:
1) very very close. For instance, a woofer can be measured with the microphone less than a centimeter from the cone
2) But the preferred method of getting a full range measurement is to measure the speaker outside, far away from any reflective surfaces.
That's exactly what I said.
I read the grandparent to imply that distance from the source was the only factor, that one needs to be a certain distance from a source to hear bass frequencies at all, which I'm sure you'll agree is incorrect, as we both seem to have a correct understanding of the physics involved.
Taken to the extreme is the "Geddes Approach" which use multiple subwoofers placed around the room and lots of EQ provide a smooth low frequency response at the listening position.
Each served it's own purpose and all the mixing guys I saw wound swap between them fairly regularly. In that situation it was about emulating different scenarios, from the 4" block speakers to get a feel for what someone would hear on an old car stereo to mid field Dynaudios that were more flat than hyped to get a feel for what the home audiophile might experience.
That's not a practical approach for a small studio (either in size or budget) but having been out of that world for a long time it makes me wonder if someone would be better served with two or three pairs of reasonably priced pair of speakers rather than trying to perfect the sound of one pair that aren't going to reflect (no pun intended) other playback environments that aren't acoustically optimized.
Aesop Rock is one that comes to mind where you can literally hear the hearing damage in the resulting mix. Late Zappa the same thing happened.
As always, Hi-Fi and Studio set-ups/advice are not entirely inter-operable, and a lot of online chat can be both camps talking simultaneously which is difficult for a reader to parse.
It took a good few days of moving the monitors & measuring the response, but I believe I was able to find a position that is usable & makes up for some of the deficiencies of my room.
I haven't even got to the room that I use as an office/studio, which is a small spare bedroom. The room works well as an office now that I've redecorated it and put in new floor and shelves, but I need to finish off power routing and suchlike before I start on speaker placement.
[0] Many serious audiophiles would argue that you should do just this for the best results but I don't have the space and I want a room that's usable for things other than just listening to music. I often think this is one of those situations where the perfect really can end up as the enemy of the good.
If you're listening alone, and don't care for a small sweet spot: In our study room I'm VERY happy using a DIY FAST/WAW system with a large full range speaker. The 4" FR (Tangband W4-2142) has some heavy beaming at higher frequencies. The sweet spot is rather small (I put it in front of my main display) - but reflections are a non-issue. The result is a very good sound stage. I implemented the cross over into a DSP (Beocreate Amp) and have one channel for each of the two drivers per side. But the only "real" correction I apply is at the low end (less room modes & extension to 40Hz at expense of ~9dB less SPL_max. Pretty nice for 4l encased volume and two 4" drivers).
> Among audio engineers, the term monitor implies that the speaker is designed to produce relatively flat (linear) phase and frequency responses. In other words, it exhibits minimal emphasis or de-emphasis of particular frequencies, the loudspeaker gives an accurate reproduction of the tonal qualities of the source audio ("uncolored" or "transparent" are synonyms), and there will be no relative phase shift of particular frequencies—meaning no distortion in sound-stage perspective for stereo recordings.
Generally, you want your speakers to be a half wave length away from the wall for the first order standing wave in your space. This is so the wave output by the speaker will cancel with the wave resonating between your walls, which will help the issue quite a bit (though it's not perfect either).
With something like the Klipschorn you're basically giving up on solving this issue with speaker placement and will have to resort to room modifications to ameliorate it. That said, you get a lot of benefits from the khorn also : )
It's pretty remarkable to play a sine wave at the frequency of your room's standing wave and then just walk the length and hear it go from completely silent to very loud and back again. Makes you realize how much distortion this is actually causing
I'd also recommend The British Audiophile's take on speaker placement and distancing: https://www.youtube.com/watch?v=VHOYXjVJKKY
https://www.youtube.com/watch?v=AfMg5P0_vJk
Listen to him wax sarcastically about blind tests. Does he not get it? Or does he make his living from not getting it?
Most modern studio monitors put the tweeter in a waveguide. By doing that, it accomplishes two things:
* The loudspeaker frequency response is consistent, no matter whether you're standing directly in front of the speaker, or to the sides. IE, the placement of the loudspeaker is less important than with older designs.
* Most importantly, a waveguide takes the energy radiated by the driver and it focuses it into a narrower beam than a conventional tweeter. By doing this, reflections off of the sidewalls, ceiling and floor are less of an issue.
To make a long story short, modern studio monitors with a waveguide are less fussy about placement.
Reflections off of the sidewalls, ceilings and floor can still cause issues with lower midrange and midbass response, but these issues can be fixed with EQ to a large extent.
The reason that EQ is less effective at high frequencies is because the wavelengths are so short. For instance, 5khz is 6.8cm long. Because it's so short, equalizing that frequency for one point in the room can screw it up at ANOTHER point in the room.
Low frequencies are much longer, and because they're longer, EQ is more effective. For instance, 500Hz is 68cm long. Because of this long length, if you EQ a speaker to flatten out a peak at 500Hz, that EQ will be effective over a broad range of positions in the room.
Wait, isn't the conclusion of the article to not overly worry about placement and just use your ears as a guide?
The build-up is just background info, which I actually appreciated.
I don't really believe this. Ever put speakers in an empty, stone-walled, wood-floored room? No EQ in the world is going to fix that. You need acoustic treatment in some shape or form, unless your walls are made out of cotton.
A typical room with drywall, carpet (or rug) and upholstered furniture is going to be EQ'd pretty well. I suppose that could be considered "acoustic treatment" but most people just call that "normal"
I can understand the other reasons speakers are nice; being able to hear other things besides the audio playback, sharing the sound with other people at the same time, etc.
As for precision: it's easier to hear the panning / stage in good headphones just because it's so augmented. The low end - no matter how good the headphones are, you can't rely on them to mix it properly, you need good monitors. If you mix the low end on headphones, there's always a high probability that it will only sound good on this pair of headphones and not translate to other setups.
Is there a name for this effect? I'd like to hear an example!
[0] Here's a simulation of where the modes of such a room might be: https://amcoustics.com/tools/amroc?l=300&w=300&h=300&r60=0.6
Obviously, any information directed at people with no knowledge in area is simplified and possibly also intentionally incorrect.
The simple articles on the internet and brochures assume you are stupid and you need safe way to set up your speakers. So that's why they ask you to place them far from wall. It would be stupid to try to explain physics of wave propagation and interference to people who need to read a leaflet to be able to set up their speakers.
But if you have bit more knowledge it is not difficult to control your situation while also having speakers closer to the wall.
Sound is distorted when you receive multiple waves that are out of phase and especially, as it usually happens, when different wavelengths are received at different phases (some in phase, some completely out of phase).
While it is not possible to remove all reflections in a closed room (and some reflection is actually welcome because we want to feel being in a room rather than void) it is possible to make sure that the ratio of the wave from the speaker to the one from reflections is high enough that we do not perceive them as distortions.
When a speaker is placed close to a wall, any energy radiated from the back of the speaker is reflected. There is also some energy from the front of the speaker, but because of the speaker itself only little of it reaches the wall behind and by the time it is reflected in the direction of the listener there is just not enough energy to be perceived as distortion.
It is important that a lot of this is due to relative distances of sound sources (or reflections). If you sit 1m from the speaker that is 50cm from the wall, the sound from the speaker travels 1m but the reflection travels 2m to reach you. This means the reflection is attenuated by 6dB by distance alone (assuming omnidirectional emission from the speaker and that it is perfectly reflected from the wall).
When a speaker is emitting very close to the wall and uses back port to emit bass, that back port is very close to the wall and not only gets distorted (due to being closer to the wall than wave length) but also reflected with a power very close to the original wave but out of phase, depending on frequency. That's what gives horrible result.
Back ported speakers sound better because they elliminate sound of moving air, but they have disadvantage in placement close to walls. Just use front ported speakers when placing close to wall (see bookshelf speakers) which are designed exactly for that use case.
More tips:
- don't place speakers close to the wall if you are going to sit far from them, because this causes relative ratios to be close
- it is ok to place front ported speakers close to the wall if you are going to sit close to them (for on your desk)
- don't just focus on what's in front of you, if there is a wall behind you it is going to cause reflections that can cause powerful distortions if you are relatively close to it compared to the distance of speakers.
- you can put something soft and permeable behind your speaker and behind your head to reduce high frequency distortions
- you can play with some types of shapes around speakers and behind your head to reduce low frequency reflections distorting your sound. By having different parts of the reflection arrive at different times you reduce perceived effect of the distortion.
This is why acoustic panels only work for mid-range frequencies and higher. Alternatively, you can also use bass traps with a membrane tuned to a specific frequency, which can attenuate said frequency without taking up ridiculous amounts of space.
It is better to have less deep bass (you can add more of it if you want) than have deep bass that is very distorted.
Also, as I have added later after you posted, you can play with objects around your speakers to have the reflection broken and arrive at different time, effectively reducing its impact on the main wave. But as you pointed out that will only work for waves up to some length depending how large objects we are talking about.
Easier to place one subwoofer than to tune five speakers.
The are other reasons:
- producing low end requires large speaker, and large == costly
- large speaker == lots of space required on desk, restricting your placement. Having small speakers on desk is a functional compromise. People working professionally with sound will buy studio monitors which take way more space but they can take it because sound quality is important because it is their job.
- multiple speakers producing low end == timing and interference issues. When you have multiple speakers producing low end, you will hear interference as you move your head around as different lengths of waves cancel or reinforce at difference places.
Since you don't need stereo for low end (your brain can't discern direction of very low end sound anyway) it is better to avoid that issue by having single speaker.
Now, single speaker avoids those issues and allows you to be basically anywhere in the room and have acceptable sound, but adds discontinuity between your low and mid range. Audio professionals will prefer studio monitors because they care for for the sound to come undisturbed so much, they can place their speakers symetrically and keep their head at exactly right spot when listening.
I see a lot of people putting up ugly "sound absorbing panels." What they don't realize is that it takes a LOT of material to absorb sound. 100Hz is over three METERS long.
There are ways to reduce the impact of reflections, but sound absorbing panels are probably one of the least practical ways to do it.
I also don't think placing objects around the speakers is going to help much with the main issue which is the low frequency is it?
For example, when you have two bare, parallel walls.
Since you can't move the walls to not be parallel (although some people actually did it) what you can do is to look at other parts of requirements:
-- ensure energy can be absorbed or dissipated before it can bounce multiple times
-- ensure there are other objects that cause distances as seen by the wave to vary, this means there will be no single dominant wavelength that is going to be reinforced and if there is any standing wave it is going to be weak (will start with less energy and have much less of wall surface -- much smaller angle in which it can initially gather the energy).
Absorbing bass is very difficult because, as mentioned by somebody else in another comment, it would require very thick material. And so the other way is to cause the energy to be reflected in various different direction so that it is not perceived as single sound.
And so, putting objects in your room helps a lot as any person that has removed all their furniture from their room can attest.
When you remove furniture the room acoustics suddenly becomes very noticeable due to powerful reflections and standing waves from bare, parallel, hard wall surfaces.
(admittedly I read both quickly, with distractions, so I almost certainly missed something...)
Well, it is not necessarily wrong to omit a lot of complexity when explaining something to someone who is new (ie. speaker buyer who had to look at the leaflet). Even if the advice is not strictly correct, it is not wrong.
We do this constantly, especially with children. We feed them with not strictly correct information because the alternatives are to not give them any, to give too much or to give information that is not useful.
You would not tell your kids their uncle committed suicide by blowing his head off because he was unsatisfied with his miserable life because he made a bunch of lousy choices and because of this aunt left him for a younger guy which caused severe depression. That is way too much information for a child to handle.
People who know better will know why the advice is structured this way and will know to ignore it.
There is also incentive for producers to give information like that, because it is easier to require you to put your speakers far from any wall than to delve into particulars of your speaker placement. "We guarantee our product offers excellent quality if you put it far from walls, but should you decide to ignore that advice you are on your own"
"(And what does “behave” even mean? It’s not like we are talking about children here.)"
Are they serious? :-D