Spot treating a room at mirror points theoretically seems useless
I am setting up a room for accurate monitoring. So far, all sources I have found regarding where to place mineral wool or acoustic panels indicate that you only need to place these panels on the side walls at mirror points, on the back wall directly behind the listener, and on the front wall right behind the speakers.
Looking at the picture below, this only appears to absorb the 1st reflections only, and does not take into consideration the sound that bounces off of two or more walls. The thick orange lines are just a couple examples of how sound can end up bouncing off the walls and reaching the listener by completely avoiding acoustic panels that were strategically placed. There has to be a million different ways sound can reach the listener by avoiding strategically placed panels, especially when taking into consideration a 3-dimensional space.
My question is, does spot treating a room like I have mentioned really make a huge difference in monitoring accuracy, even though reflections can reach the listener by avoiding these carefully placed acoustic panels?
.............
My question is, does spot treating a room like I have mentioned really make a huge difference in monitoring accuracy, even though reflections can reach the listener by avoiding these carefully placed acoustic panels?
"Huge difference" is relatively easy to reach, with any treatment or with changing positions of listener and loudspeakers, but "better response" is not that easy to achieve.
Worse room problem is non-articulated bass response. Treating only points of first reflections (with thin 4" panels) virtually doesn't solve this problem at all. Later, we need this reflections back and diffused, for better mix translation, btw.
EDIT: And yes, whole sound doesn't propagate like light from flashlight... only high frequencies has that thin sweet spot. So LF sound waves can't even "see" this panels.
Worse room problem is non-articulated bass response. Treating only points of first reflections (with thin 4" panels) virtually doesn't solve this problem at all. Later, we need this reflections back and diffused, for better mix translation, btw.
EDIT: And yes, whole sound doesn't propagate like light from flashlight... only high frequencies has that thin sweet spot. So LF sound waves can't even "see" this panels.
I know side panels do not take care of the low end. I'll be using 12" thick side wall panels made from R19, so that should absorb sound fairly well down to about 150hz. I'll of course put up bass traps to handle the rest of the low end. I'm wondering if spot treating the walls like I said with also adding bass traps will leave me with an accurate/dry sound in the mid to high end part of the frequency spectrum.
........I'm wondering if spot treating the walls like I said with also adding bass traps will leave me with an accurate/dry sound in the mid to high end part of the frequency spectrum.
This way you will obtain a tiresome room with not that well mix translation. Strong diffuse sound field (from about 1kHz and up) is a key for good mix translation and path to acceptable room where people will like to work.
EDIT: I mean, please, don't (only) kill first reflections, you need it. Or... combine absorption and diffusion at first reflection points
OP - the point is, the first reflection points are by far the loudest and closest in time to your source. Therefore they have the worst effect on your direct sound. Doesn't mean that there arent other issues - but these are your primary and worst offenders.
OP - the point is, the first reflection points are by far the loudest and closest in time to your source. Therefore they have the worst effect on your direct sound. Doesn't mean that there arent other issues - but these are your primary and worst offenders.
I know these are the worst offenders. I just need to know if the "other issues" that are still present should be something I need to worry about or not.
This way you will obtain a tiresome room with not that well mix translation. Strong diffuse sound field (from about 1kHz and up) is a key for good mix translation and path to acceptable room where people will like to work.
EDIT: I mean, please, don't (only) kill first reflections, you need it. Or... combine absorption and diffusion at first reflection points
Your saying its not good to have a completely sonically dead environment for monitoring then? I actually need some early reflections in my monitoring environment to achieve a good mix translation? This is a new idea to me.
Your saying its not good to have a completely sonically dead environment for monitoring then? I actually need some early reflections in my monitoring environment to achieve a good mix translation? This is a new idea to me.
Yes.
At first reflection points you don't have biggest problems with high frequencies but with non resonant interferences, called here SBIR (Speaker Boundary Interference Response), which is usually below 200Hz, if they exist, so you need absorption at first reflection points but NOT for high frequencies. You need high frequencies reflected back into room, to be aware that you are in the room, because you are born in the room and you live in the room like all your ancestors (it is about genetics), and most of your clients will listen your products in the room. To enable your brain to easily differentiate some recorded first reflections in your mix from first reflections in your listening room, you need diffusion over existing LF absorption, at first reflection points.
In short, we must keep first reflections of high frequencies in the room to enable people to work easily in it. The way how we disable negative influence of them, is diffusion combined with absorption... literally.
Have you ever been in an anechoic chamber? Uncomfortable as hell, terrible place to try to mix.
you don't need to restrict yourself to an "uncomfortable" anechoic chamber to achieve an anechoic (effectively anechoic) speaker-listener response whereby no indirect specular energy is incident from the room's boundaries.
Your saying its not good to have a completely sonically dead environment for monitoring then? I actually need some early reflections in my monitoring environment to achieve a good mix translation? This is a new idea to me.
all models strive for good LF response and decay times, so there is little room for debate there.
in the specular region where wavelength is small in relation to boundary size, energy behaves and can be modeled as a ray (angle of incidence = angle of reflection, thus geometric "reflection points"). however, as you stated, there are 2nd order reflections that could still be detrimental and the "mirror" will not detail these issues.
some models (NE) aim for an effectively anechoic speaker-listener path. thus the direct signal is the only signal processed and no indirect signals based on room impede the listening position. the room itself is not an anechoic chamber, however - so not necessarily an unconformable environment to work in. LEDE (and RFZ - a way of achieving LEDE specular response) are other models that do impose indirect energy to the listening position, but with specific criteria in time, gain, arrival, and type of energy (eg, diffused). so there are many ways to achieve "accuracy".
in your room, instead of relying on blind application of treatment based on mere "possible reflection points" - you can utilize the Envelope Time Curve (ETC) response to identify the boundaries incident of the ACTUAL high-gain, early arriving indirect specular reflections that are destructive to intelligibility, localization, and imaging, and attenuate as required. the ETC displays how ALL of the specular energy impedes the listening position: from early-early energy (if there are coupling issues, as vibrations will travel faster in solid than that of air, arriving before the direct signal), to the direct signal (shortest path between speaker and mic), to the early reflections (edge diffraction from cabinet or objects placed close to speaker/receiver - or that from sidewalls, ceiling, floor, desk, work surface), to the later arriving reflections/decay, until the last of the energy has been damped and is below ambient noise floor.
it will display how specular energy impedes the listening position - gain with respect to time. and since speed of sound is a constant within your room, you can work backwards once a high-gain signal has been identified to determine the total flight path (thus, total distance traveled) and the boundary or object incident of the indirect energy.
simply placing broadband absorption at any and all possible reflection points is a quick way to create a highly damped room. whether this is beneficial is up to your design requirements. using the ETC to identify only the actual high-gain/early arriving signals and their actual incident boundary allows one to surgically place absorption at only those areas that require it - minimizing the amount of absorption within the room. bear in mind there are other ways to attenuate an indirect signal while maintaining as much energy within the room via the use of a large (with respect to wavelength) flat reflector panels to redirect the geometric reflection away from the listening position and (for example) towards the rear wall / rear side-walls where it can be diffused and presented back to the listening position. redirection vs absorption of these early, high-gain reflections maintains more energy within the room which can be managed and reintroduced to the listening position more appropriately (if that is the specular response you choose). LEDE/RFZ, for example, creates an effectively anechoic time period (referred to as ISD-gap) of which no high-gain energy impedes the listening position such that the direct signal is all that is processed and localization and imaging are not skewed by the high gain signals arriving within the haas period. the longer this gap (the longer it takes for indirect energy to impede the listening position), the larger the perceived acoustical size of the space. since in this model's example we want energy to eventually be reintroduced, this is done via the use of reflection phase grating diffusers (QRD/PRDs) - such that the termination of the ISD-gap arrives as a lateral, exponentially decaying (semi) diffused sound-field. the gain of the first significant energy to impede the listening position signifies "liveliness", and the diffused decay contributes to "spaciousness" of the room. so this is one example of a way to have accurate direct signal representation, while still maintaining a sense of liveliness and spaciousness to your room. in NE room, the ISD-gap is essentially infinite, as there is no indirect energy ever presented to the listening position; the direct signal is all that is processed.
there are of course other methods, but LEDE (RFZ) and NER are two that can be fairly easily emulated within your room. regardless of the specular response, LF/modal region issues are still relevant and must be addressed (but this is somewhat independent of the specular response - even though treatments may still need to take into consideration each aspect).
there are many ways to achieve suitable "accuracy" within your bounded acoustical space to work in - your own personal constraints, tastes, etc can dictate this.
your primary question here was with respect to "reflection points" - and to truly understand the actual behavior of the space, measurements will be required. it is quite certain that you will still be attenuating the early reflections from the sidewalls - but utilizing the ETC will detail you all of the ACTUAL high-gain early arriving reflection points, as the mirror does not detail you gain, time arrival, or any other "non-obvious" sources of indirect energy. and as always, appropriately broadband treatments are required: in the case of velocity-based porous absorption, thickness and gas-flow-resistivity are key values, but also physical size - as the treatment must be 'large with respect to wavelength'.
I am setting up a room for accurate monitoring. So far, all sources I have found regarding where to place mineral wool or acoustic panels indicate that you only need to place these panels on the side walls at mirror points, on the back wall directly behind the listener, and on the front wall right behind the speakers.
Looking at the picture below, this only appears to absorb the 1st reflections only, and does not take into consideration the sound that bounces off of two or more walls. The thick orange lines are just a couple examples of how sound can end up bouncing off the walls and reaching the listener by completely avoiding acoustic panels that were strategically placed. There has to be a million different ways sound can reach the listener by avoiding strategically placed panels, especially when taking into consideration a 3-dimensional space.
The complete and proper direction for reflection control is "all surfaces that reflect sound arriving at the listening position within 20 ms of the initial sound from the speakers."
Nothing to do with just side walls or ceilings.
Andre
__________________ Good studio building is 90% design and 10% construction.
This is what I do. You can also find the second reflection points by making a second mirror of the room. Still trying to figure out how to get reflections for multiple surfaces with this method, if it's even possible
The complete and proper direction for reflection control is "all surfaces that reflect sound arriving at the listening position within 20 ms of the initial sound from the speakers."
Nothing to do with just side walls or ceilings.
Andre
Or even more accurate; ... arriving at the listening position within the ISD-gap (whatever it might be; anything from about 12 ms to about 25-30 ms depending on the requirements of the control room and the possible recording room connected to it).
The complete and proper direction for reflection control is "all surfaces that reflect sound arriving at the listening position within 20 ms of the initial sound from the speakers."
If you are unable to understand how to track these down then treat the early reflections left/right, ceiling and the back wall if less then 8 feet from the mix spot.
You could have tens of reflections inbetween the speakers and listening position. The good part is they are usually close enough to one another to treat with a single panel.
This is what I do. You can also find the second reflection points by making a second mirror of the room. Still trying to figure out how to get reflections for multiple surfaces with this method, if it's even possible
why not utilize ETC and blocking +/or string method?
why not utilize ETC and blocking +/or string method?
I would use ETC too. My interest in this initally was in the design of a new room. Most of the echo mangling I do is just by experience, a panel here and there, to find out what color makes the music more whole. But the most important thing in finding reflection points is finding the largest groups of them in a small area. This involves analyzing many orders of reflection, which the schematic mirroring method makes quite simple. The hardest part is keeping the schematic organized, because the constant mirroring usually ends up in superimposed layouts and quite an ugly mess. If you use the AutoCAD or graph paper method, the easiest room shape by far is a square or rectangle. I would not recommend it to an amateur in geometry or mathematics for other room shapes. The reason I say this is because, with a square or rectangle, the higher order reflections are easily observable by using a simple rectangular array of the footprint. Any other shape requires very odd mirroring patterns, superimposing the surfaces over each other, and a large decrease in productivity.
I would not be suprised if the specialized acoustics programs had functions to find reflection points of a certian order, but those are really intended for business people with their large prices.
The thick orange lines are just a couple examples of how sound can end up bouncing off the walls and reaching the listener by completely avoiding acoustic panels that were strategically placed. There has to be a million different ways sound can reach the listener by avoiding strategically placed panels, especially when taking into consideration a 3-dimensional space.
yes, but notice how much longer of a flight path the thick orange lines are in the example diagram you provided - the longer the distance traveled, the longer (in time) it takes for that energy (indirect reflection) to impede the listening position. as stated above, time is also a factor in how such energy is addressed and managed.
I would use ETC too. My interest in this initally was in the design of a new room. Most of the echo mangling I do is just by experience, a panel here and there, to find out what color makes the music more whole. But the most important thing in finding reflection points is finding the largest groups of them in a small area. This involves analyzing many orders of reflection, which the schematic mirroring method makes quite simple. The hardest part is keeping the schematic organized, because the constant mirroring usually ends up in superimposed layouts and quite an ugly mess. If you use the AutoCAD or graph paper method, the easiest room shape by far is a square or rectangle. I would not recommend it to an amateur in geometry or mathematics for other room shapes.
im confused by this statement - what do you mean by finding the "larger groups of reflections in a small area"? why must you analyze many orders of reflection? what is the goal in such a scenario?
im confused by this statement - what do you mean by finding the "larger groups of reflections in a small area"? why must you analyze many orders of reflection? what is the goal in such a scenario?
Some groupings have a direct correlation with flutter echo. Such as the area in between the monitors/speakers and the listener, assuming parallel side walls. Not only does the first reflection happen in that area, but second, third, all the way up until the material damping takes effect. Yes, in time those higher order reflections are far delayed, yet they are all focused in one small region. Resonance might occur in areas like this as a result of the closely correlated time delay gaps. This doesn't only apply to specular reflections you see, but the path of travel of the sound energy to the position of interest. It is in these areas that room treatments have the biggest impact.
Some groupings have a direct correlation with flutter echo. Such as the area in between the monitors/speakers and the listener, assuming parallel side walls. Not only does the first reflection happen in that area, but second, third, all the way up until the material damping takes effect. Yes, in time those higher order reflections are far delayed, yet they are all focused in one small region. Resonance might occur in areas like this as a result of the closely correlated time delay gaps. This doesn't only apply to specular reflections you see, but the path of travel of the sound energy to the position of interest. It is in these areas that room treatments have the biggest impact.
forgive me but i'm still confused on the specific issue and why you say this "doesnt only apply to specular reflections" - and how it is these areas that room treatments have "the biggest impact". what does "biggest impact mean"? a drawing/diagram may help, but im still confused on the issue you're trying to raise. apologies -
It's simple, really. Because each reflection contributes to reverb, at large groupings the least surface area is required to make the larger changes in ETC. Why are you assuming I am talking some kind of sound trick? It is what it is.
jd, advice on websites and fora tends to be simplified. Like Everest and others' work it's primary goal is to entice newbies into the joys of acoustic treatment. Much of todays music is produced in at best Prosumer level rooms. So any treatment is encouraged, but we rarely see the job full well done. So look out for context here.
The first reflection points are likely to be much more significant than the later paths due to level loss. In a highly damped small room, the level falls off very quickly. Even in a lightly treated room they are both way louder and way earlier. Surely a priority.
Many find highly damped rooms tiresome in one way or another.
Personally I don't. I value the accuracy over creature comfort. I mean, we are pretty comfortable compared to coal miners.....
I have noticed over and over that speakers sound too bright in such rooms. Bright and clear, just like vocals sound in highly damped booths.
Strangely all that fluff promotes sparkling clarity.
Turning down the tweeters makes life a lot less stressful and in my experience (and many others) enhances mix translation.
It's simple, really. Because each reflection contributes to reverb, at large groupings the least surface area is required to make the larger changes in ETC. Why are you assuming I am talking some kind of sound trick? It is what it is.
statistical application of absorption is relevant in reverberant (Large) acoustical spaces, but im not sure i follow your logic here. in the rooms of context, the primary concern of the specular region is control of early, high-gain reflections while opting either to permit or restrict a later arriving sound-field - and the ETC details all of this. the individual reflections can be resolved in terms of their gain, time-arrival, and vector - unlike energy within a reverberant sound-field which in these rooms is generally below the ambient noise floor and above our hearing range (in terms of frequency).
when you make the statement: "This doesn't only apply to specular reflections you see, but the path of travel of the sound energy to the position of interest. It is in these areas that room treatments have the biggest impact."
i'm just trying to understand what you mean by making the "biggest impact" in this scenario?
While it is true that with distance and time the reflections don't "color" as much, these surfaces are still the number one contributors to reverberation at one location. Through the use of diffusion, the ETC can be fixed far faster and easily than any other method.
i'm afraid i still do not see how this is relevant. is this a suggestion that the primary concern regarding small acoustical spaces is NOT the early-arriving, high-gain indirect specular signals that arrive within the haas period and are destructive to accurate localization and imaging of the direct signal - but instead is to utilize a minimal amount of treatment in order to attenuate later arriving specular decay and flutter echo?? if so, then i disagree.
the first statement of the thread gives the context: "I am setting up a room for accurate monitoring.".
Quote:
Originally Posted by OpusOfTrolls
While it is true that with distance and time the reflections don't "color" as much, these surfaces are still the number one contributors to reverberation at one location.
there is no such thing as "reverberation at one location" - that defies the very definition of "reverberation/reverberant sound-field" ... eg, statistically random-incidence energy flows equal/probable in any/all directions/locations. and then you are also making some significant assumptions about the room if you are somehow able to tell us what boundaries will be incident of the most energy that makes up the locally variable later arriving decay trail. 0_0
can you tell expand more on how the later arriving decay trail colors the signal, and why this should be the primary concern regarding treatment placement (eg, absorption) vs the early arriving higher-gain signals that are detrimental to accurate localization and imaging?
Quote:
Originally Posted by OpusOfTrolls
Through the use of diffusion, the ETC can be fixed far faster and easily than any other method.
the ETC is a measurement tool; hence, it doesn't require "fixing". what exactly should the ETC look like in your opinion? is it mandatory?
Since I don't like to do post breakdowns (more efficient this way), I will try to sum it up.
In between parallel surfaces, exists constant time delay per reflection for the normal incident angle. This is actually impossible in practice, because of speaker cabinent obstruction. There cannot be an exact constant delay between parallel reflections that are specular. Actually, there can be, if speakers are soffit mounted with parallel front/rear wall. But this is the case of bad acoustics that we solve immediately.
Following this idea, concept, that the time delay changes per reflection, there are many reflections in the case of parallel sidewalls between the speakers and listening position. As the order increases (the number of reflections until arrival at the listening location), the angle of incidence becomes more narrow, and reflection nodes more cramped into one area.
Now, applying absorption to the said area will just kill flutter. Because of the minimal surface area absorption in use, will also maintain the reverberant field. Statistically, less absorption equals more echo. But we just got rid of most colorful form of it.
I would think the trained Acoustician can see from here where the diffusion actually is. It is the result of late arrival from every other surface, eventually coming into the listening position. It is not guaranteed that there will not still be resonance, because resonance can happen with more than two surfaces. Tangential and Oblique modes come to mind.
The grouping of reflections into one small area, combined with angles that are close to one another, is the flutter zone. This is exactly what we want to get rid of for critical listening.
There are other examples that don't exactly create flutter, but they would require severe deviations from parallel in the room shape. In almost all circumstances, it is best to remove parallel surface reflections from the echo. This is why I recommend the treatment of small protruding and inset areas with absorption.
As for the use of diffusers in the spaces without absorption, the best configuration is a matter of debate, as well as dimensions. From diffusion, reflection becomes less ray like and more conical. Some of the direct lower order reflection points will benefit from diffusion, in regards to the echo delay and sound scattering. There is no one size fits all room treatment.
I do recommend side wall AND ceiling treatment for flutter echo. Remember, to break the second order reflection, only one parallel surface needs absorption. The first reflection from the floor can be treated with the proper device, which varies on application.
New Reply 
Facebook
Twitter
LinkedIn
