What's a typical room? You might have lucked out with your room as is but it would be a very different story for say a condo dweller surrounded by poured concrete, CHB and glass. Then there's the issue of symmetry.
Psychoacoustics of Room Reflections
- Thread starter amirm
- Start date
I am a playback guy, not a recording guyHello Amir,
Would you have any comment on acoustic parameters to be considered for an optimal tracking room, optimal mixing room, and optimal mastering room, i.e., three separate rooms/spaces with three different functions
Thanks
. So I can't give that advice. From what I have read, for mixing and mastering, folks prefer to have the front of the room and especially the sides fully absorptive. The so called reflection free zone or LEDE came from that school of thought. For tracking, that is just something I have not studied. Likely reflections are good there but again, I am not an expert at all in that field. Perhaps Bruce can opine there or Ethan.Well, it depends on whether it is furnished well or not. Dr. Toole goes through a great scenario of taking an empty room and progressively adding the right furnishings to it to bring down the mid to late reflections. Here is a nice graph from that:
If you have the typical modern living room with many hard surfaces for example, the fact that it is a living space does not make up for the fact that it will be too "live."
As Wendell nicely said, some of the issues we face these days is easy availability of tools to measure our rooms has resulted in too many people acting on data that is not what it seems when we consider how a human with two ears and a brain hears things (vs. a single microphone and a graph). How good your music would sound if we used a single mono mic and recorded it raw? Likely not that good yet that is what we do when we measure with microphones.
I am a playback guy, not a recording guy
Live end/dead end mixing rooms are an idea that doesn't work very well in the view of many people. FWIW, the best control rooms/ mixing rooms I've sat in were not heavily treated. They don't tend to have a lot of soft furniture in them, so they are treated, but with a mix of soft and hard surfaces that sounds much like a normal listening space.
There is varying mileage on that, though, and there are people out there who love to mix in pretty dead air.
Tim
Exactly. The measurement metric for it is an RT60. This measurement is usually reserved for large listening spaces (our home is NOT that). However, in this narrow context it can be used to measure how "reverbrant" the room is. Per previous graph, the target is around 0.4 seconds in the 500 Hz region. If you hit that, then you are within the "typical" range. If you are higher, then more absorption is necessary. if it is below, highly unlikely, it means the room is too dead.
There is another test for that is to have a person stand where the speakers are and speak in normal voice. If the voice is easy and comfortable to hear, then the room is probably good in this regard. The reason this test works is that the human voice has similar directivity to a typical speaker and so it simulates the real thing for us.
I think this statement is simply untrue. Until about 1960 or so (the advent of home stereo with some pretentions to actually reproducing the sound of live recordings), we didn't listen to reproduced music in rooms. In fact most live performances spaces were and are "treated" specifically to improve the sound. So we went from roughly 1960 to roughly 1985 (when companies began larger scale marketing of "home" acoustic treatments) listening to "untreated" rooms, and I suspect (but this I don't know for sure) that developing and marketing products for home listening rooms actually occurred because of a perceived problem in the sound of home listening, not just someone out to make a buck.
I think what he is saying is that for countless years we have lived in closed spaces where reflections are always there. The brain has adapter to use them to better its recognition of what is there. Who do you know that enjoys singing outside in the open air? I think most prefer a shower. Ooops. I said I did not know anything about recording space.
Well, it depends on whether it is furnished well or not. Dr. Toole goes through a great scenario of taking an empty room and progressively adding the right furnishings to it to bring down the mid to late reflections. Here is a nice graph from that:
If you have the typical modern living room with many hard surfaces for example, the fact that it is a living space does not make up for the fact that it will be too "live."
As Wendell nicely said, some of the issues we face these days is easy availability of tools to measure our rooms has resulted in too many people acting on data that is not what it seems when we consider how a human with two ears and a brain hears things (vs. a single microphone and a graph). How good your music would sound if we used a single mono mic and recorded it raw? Likely not that good yet that is what we do when we measure with microphones.
Amir,
I think that fig16.1b is rather alarming. If we see the error interval we can see that one third of the typical rooms are outside the band and are not acceptable for music reproduction. Also considering that since the time it was built there was a systematic change in furniture style of living rooms, making the rooms much more "live" as you say and we see are relying on statistics with reduced scope. In this aspect I think audiophiles should be selfish - statistics are nice to establish general trends and writing scientific papers, but each audiophile is an individual case. Assuming as a general rule that we do not need audio treatments above the transition zone does not seem acceptable. Perhaps we do not really need them, but IMHO it should be checked with measuring instruments. At less I would sleep easier after I do it or have someone doing it ...
Last edited:
I think what he is saying is that for countless years we have lived in closed spaces where reflections are always there. The brain has adapter to use them to better its recognition of what is there. Who do you know that enjoys singing outside in the open air? I think most prefer a shower. Ooops. I said I did not know anything about recording space.
I prefer singing in the car, personally. I get the first reflection directly off of the windshield back to my ears, then the seats, headliner and carpeting absorb, the curves and angles of the glass diffuse, and I get great sound. Closest thing to hearing a recording of myself played back in headphones.
Beyond that light comment, I think we have entered a discussion zone that may be far to pragmatic for an audiophile board. I agree with most of it, though. I have a good friend who is a good musician and a better audio engineer. The guy simply has great ears and knows what happens to sound once it exits the speakers. I recently saw him walk into a room with a playback system to tweak, and get it sounding great just by moving speakers and furnishings. But it was more work than you might imagine. He has a thing about shelves full of books and nicknacks. He used them to great advantage, and annoyed the bejeezus out of the lady of the house in the process.
But a pretty modest system was cooking when he left.
Tim
Just a few minor quibbles (perhaps) here.
In that 'other thread'
I mentioned we would be doing a room to room comparison and videoing the results, the comparison being the same system in two identical rooms, one 'treated' and the other untreated. Btw, the differences kinda did come across even on you tube but as suspected it was no where near as clear and dramatic as in person. That's by the by.To 'many' it was clear proof of the need for room treatment, but I am trying to reel that enthusiasm back a bit by pointing out that my treated room is NOT really treated in the manner most conceive of when room treatment as a term is used. Rather, due to the size of these rooms and their construction all I did was (and only on hindsight did I come to realise this) reduce the decay time of the room down to an acceptable level.
In the 'untreated' room we found the decay time to be over 0.8 sec, it came down in my room to about 0.3 sec. BIG difference, especially in the hearing.
As I tried to point out in my reeling back, this was NOT an exercise in treating FR points or such (my room and layout kinda don't support that approach in any case) nor any other associated approaches normally used in 'treating rooms'.(eg, my chair is at least two metres from any side or rear wall for example, so I am not constrained by factors most other people meet...tho it can and does tend to place me in bass nulls at times which needed a lot of gradual re-arranging to mostly over come)Anyway, the slight quibble here is maybe that 'anything under .4 is 'too dead'. When I FIRST achieved that level I wondered if it were too dead or just such a dramatic difference to what I was accustomed. I happen to love it, it is calm and quiet rather than too dead to me, it is actually a pleasure to just sit in the room (compared to the other) even if there is no music on. In any case, whether that is my own personal taste rather than the general has kinda been answered in that no-one has at any stage ever mentioned they thought it 'too dead', rather it is just right to all that have listened. Not that I am out to prove this, but it might be worth noting that 'these people' I speak of are many, spose around a hundred by now all up...tho a few repeats!...and they are audio guys as well.
Maybe all it is is a 'warning' to not take these numbers as gospel? You know, 0.41 is heaven, 0.39 is hell. I am sure amir (or toole) did not mean it that way either, but a gentle reminder of that.
I am a playback guy, not a recording guy
Cool, Amir...thanks
I was in Vintage King's L.A. location, aka VKLA, Saturday night, and had a brief opportunity to listen in the "Critical Listening Room" featured in this video at ~ 6:45...http://www.youtube.com/watch?v=uqm1gqwz-Ys
Fascinating...
Thanks for the report Terry. I should have been more clear. The 0.4 metric is not a precise number to pick. 0.3 seconds is certainly in the comfortable and recommendation range. As you go below that, you may still like what you hearing but the space will become less comfortable to interact in. 0.2 seconds is probably the limit in that regard.
One other guideline I have seen from Tony Grimani is to have it range based on volume of the room. Smaller the room, the smaller the RT60. At the 2000 cubic feet, he suggests 0.25, going up to 0.31 for 4000.
One other guideline I have seen from Tony Grimani is to have it range based on volume of the room. Smaller the room, the smaller the RT60. At the 2000 cubic feet, he suggests 0.25, going up to 0.31 for 4000.
Amir,
Very nice article! Keep up the momentum and write more! What I find interesting is that for every Martin Logan or Soundlab owner that likes a speaker with a curved panel to beam the sound, or an MBL owner, you will always have a "science minded" person like Sanders who likes a flat, non-beaming panel - and will have a "science reason" to prove it works best. What would you say to him?
As to your conclusion, "... It is not every day that we get to like some distortion and save money not trying to eliminate it! ... " Do you really mean this?? One man's distortion is one man's realism. Vinyl and tubes may be distorted to "scientists", but sound like real music to many of us... Don't see seriuos audiophiles saving money by moving away from these technologies! And in fact, they are growing!
Very nice article! Keep up the momentum and write more! What I find interesting is that for every Martin Logan or Soundlab owner that likes a speaker with a curved panel to beam the sound, or an MBL owner, you will always have a "science minded" person like Sanders who likes a flat, non-beaming panel - and will have a "science reason" to prove it works best. What would you say to him?
As to your conclusion, "... It is not every day that we get to like some distortion and save money not trying to eliminate it! ... " Do you really mean this?? One man's distortion is one man's realism. Vinyl and tubes may be distorted to "scientists", but sound like real music to many of us... Don't see seriuos audiophiles saving money by moving away from these technologies! And in fact, they are growing!
no worries amir, thought my last sentence 'salvaged' it! I did not really think you meant it as iron clad. Still, the important point (?) to grasp is that..well the lesson I have learned is that I did not 'treat my room' per se, or if I used that term as I did in the past then it could be misleading. I remember when you first raised this over 'there' I protested somewhat until after thought I realised that a) my room is NOT your typical 'average household listening space' (whatever that term means) and b) due to that the problems I encountered, and treated, are not ones any 'normal' person would come across. Ie, most would not listen in the home equivalent of an underground parking station!
Just tried to do a quick calc of my volume, and came up with something like 270 cu m, which loosely is maybe 7000 cu f. Going on those above figures, I'm heading in the 'wrong' direction and back to the original figures for the untreated room! (maybe it would not reach that high tho).
Think I'd rather stay where I am haha.
The usual way IS to determine RT60 targets. In my case I wanted .25 at 20KHz rising to .4 at the midrange remaining flat and rising again from 70Hz to .7 at 20Hz. To achieve this one has to take account how the room is built and everything that will go in it. While many obsess about the absorption coefficient of Corning fiberglasses, absorption coefficient charts do include drapery, carpet, furnishing and even people. It then becomes about percentage of area coverage.
At some point clapping your hands just doesn't cut it. The whisper test is a much better one. How far can you be from another person in the room and understand what he or she is whispering without reading lips? Consonants are high frequency while vowel sounds are mid and low frequency. Intelligibility is the first priority after all. If you have a hard time telling a violin from a viola, where they are placed in the stage is less important in my mind. As is expected RT60 is also representative of the noise floor and the frequency response curve.
I always say that speaker anechoic measurements are a very good thing. They are the best we can hope for precisely because there is no such thing as a "typical room", we need an anchor. I also always say that the in-room response in one's own home is much more important for us the users. I don't use this only in the sweet spot either. Since I entertain a lot we shot for even coverage, that's where the diffusive elements came into the picture.
Like it or not we have to accept that the room will contribute to the final outcome, heck, how many people are in the room will too. Killing first reflections ain't it, going back on topic. Killin' the ringin' is much more important and heavily treating the first reflection point is one way of doing it and depending on the cases usually not the wisest way unless you are in a very small room where you have no other choice. It has become a mantra unfortunately, so I am glad Amir has written this article and started this thread. Very glad.
At some point clapping your hands just doesn't cut it. The whisper test is a much better one. How far can you be from another person in the room and understand what he or she is whispering without reading lips? Consonants are high frequency while vowel sounds are mid and low frequency. Intelligibility is the first priority after all. If you have a hard time telling a violin from a viola, where they are placed in the stage is less important in my mind. As is expected RT60 is also representative of the noise floor and the frequency response curve.
I always say that speaker anechoic measurements are a very good thing. They are the best we can hope for precisely because there is no such thing as a "typical room", we need an anchor. I also always say that the in-room response in one's own home is much more important for us the users. I don't use this only in the sweet spot either. Since I entertain a lot we shot for even coverage, that's where the diffusive elements came into the picture.
Like it or not we have to accept that the room will contribute to the final outcome, heck, how many people are in the room will too. Killing first reflections ain't it, going back on topic. Killin' the ringin' is much more important and heavily treating the first reflection point is one way of doing it and depending on the cases usually not the wisest way unless you are in a very small room where you have no other choice. It has become a mantra unfortunately, so I am glad Amir has written this article and started this thread. Very glad.
Jack,
I think you are missing one point. What is the correct way of determining the RT60 targets? As far as I understand we are seeing that for small rooms the RT60 measurements can be used as diagnostic tool, but not as a target.
How did you determine the RT60s for your room?
Hi my friend,
Look at it this way. The anechoic output characteristics of your system are constant. Using noise which by nature is random but also constant over time, the immediate implication on FR is the time it takes for that pulse to drop in amplitude at each frequency within the same given time. There are many published and accepted specifications. Some here advocate the FR of B&K, I'm a bit more formal and follow DIN specs for specific applications. My chosen FR is a modification of DIN 4 (because I do prefer a bit more bass overhang than is accurate, I like my highs a bit sweeter too). DIN 4 is for spaces for live performances in a range of volume where my listening room lies. I could have chosen RT60 that was mastering quality however these have practical consequences. Since I was building for enjoyment and not work, I chose to design according to my preferred FR curve which is intertwined with the reverberation time. Unlike work where I would need to hear what's only on the recording sans room added reverberations, for enjoyment, I simply didn't need to go that length. I'm perfectly comfortable with a facsimile of "a" concert rather than "the" actual event captured or guessing the intent of a construction.
When I met with my acoustician, even if I am a certified acoustician myself (He's a lot better than me), I told him I wanted the acoustic profile of a large recording room and not a control room plus the modifications I wanted for my personal tastes. As I said, I wanted a flat midband. The reason for this is that my loudspeakers have adjustability in the treble and bass regions. I wanted a downward slope up top because the factory setting was attenuated (if lacking in presence I would not be adding boost if ever I felt things were dulled) and I wanted some bass lift because I wanted to save on amplifier power between 100hz AND 50HZ, the area before the active subs kicked in.
I've spent hundreds of hours in control rooms and while I see the need for those specifications, it just isn't what I'd want when listening for pleasure. If I had pursued a career in recording and production, you would have seen me tow the objectivist line of "fidelity": like medicine, first do no harm. I did not. My thinking is that for pleasure, nobody is going to die if I follow what I like. I'm buyin' with honestly made money, I feel I have the right to do as I please! I really don't care if others find my system too neutral or too oozing with license. Believe me, I've had both reactions! LOL. Bottom line is, I'm the one that lives with it. Besides, I've heard more of what's in a recording at home than I ever had in the studios.
There are other acousticians on our forum, that have categorically stated that RT60 is old news. Until I experience otherwise, I'll just have to disagree. Like I said, it is intrinsically intertwined with FR. FR alone is like a snapshot. FR and RT60 is like moving pictures.
Look at it this way. The anechoic output characteristics of your system are constant. Using noise which by nature is random but also constant over time, the immediate implication on FR is the time it takes for that pulse to drop in amplitude at each frequency within the same given time. There are many published and accepted specifications. Some here advocate the FR of B&K, I'm a bit more formal and follow DIN specs for specific applications. My chosen FR is a modification of DIN 4 (because I do prefer a bit more bass overhang than is accurate, I like my highs a bit sweeter too). DIN 4 is for spaces for live performances in a range of volume where my listening room lies. I could have chosen RT60 that was mastering quality however these have practical consequences. Since I was building for enjoyment and not work, I chose to design according to my preferred FR curve which is intertwined with the reverberation time. Unlike work where I would need to hear what's only on the recording sans room added reverberations, for enjoyment, I simply didn't need to go that length. I'm perfectly comfortable with a facsimile of "a" concert rather than "the" actual event captured or guessing the intent of a construction.
When I met with my acoustician, even if I am a certified acoustician myself (He's a lot better than me), I told him I wanted the acoustic profile of a large recording room and not a control room plus the modifications I wanted for my personal tastes. As I said, I wanted a flat midband. The reason for this is that my loudspeakers have adjustability in the treble and bass regions. I wanted a downward slope up top because the factory setting was attenuated (if lacking in presence I would not be adding boost if ever I felt things were dulled) and I wanted some bass lift because I wanted to save on amplifier power between 100hz AND 50HZ, the area before the active subs kicked in.
I've spent hundreds of hours in control rooms and while I see the need for those specifications, it just isn't what I'd want when listening for pleasure. If I had pursued a career in recording and production, you would have seen me tow the objectivist line of "fidelity": like medicine, first do no harm. I did not. My thinking is that for pleasure, nobody is going to die if I follow what I like. I'm buyin' with honestly made money, I feel I have the right to do as I please!
I really don't care if others find my system too neutral or too oozing with license. Believe me, I've had both reactions! LOL. Bottom line is, I'm the one that lives with it. Besides, I've heard more of what's in a recording at home than I ever had in the studios. There are other acousticians on our forum, that have categorically stated that RT60 is old news. Until I experience otherwise, I'll just have to disagree. Like I said, it is intrinsically intertwined with FR. FR alone is like a snapshot. FR and RT60 is like moving pictures.
Thanks, Jack. Your long post was excellent and honest.
I am really enjoying these (and other) debates about room acoustics and trying to to create an enlighted opinion. It is not easy for someone with almost none previous experience in acoustics as the diversity of the opinions can be sometimes extreme. It is why only understanding the "behinds" can be of any use. Otherwise it is just one more tube versus SS debate!
I am really enjoying these (and other) debates about room acoustics and trying to to create an enlighted opinion. It is not easy for someone with almost none previous experience in acoustics as the diversity of the opinions can be sometimes extreme. It is why only understanding the "behinds" can be of any use. Otherwise it is just one more tube versus SS debate!
Thanks Amir and all for this great conversation. I'll chime in with my current leading thoughts that kinda cover everything from first reflection control, later reflections and even RT60.
A lot of the confusion is from the reality that understanding of small room acoustics is a moving target. Why is this a moving target? My opinions (please note), first is that many acousticians had brought thinking from large room acoustics or hearing testing acoustics to small room environments (meaning rooms in less than 7K cu ft...a generalized threshold). This has created early standards that have been hard to break.
Second and I think most important is the positive trend of speaker design: improved directivity control and off-axis response. It wasn't that long ago that the on-axis pressure amplitude response was the driving factor in speaker design. In this condition, absorption of early/first reflection and more absorption in general can be a positive thing.
In a conversation I had with Dr. Toole, knowing that I'm an acoustics guy, he said "don't $%^& with my speaker". Shaking in my New Balance's, this one sentence has taken a long time to integrate into practice. I am of the personal opinion that Dr. Toole has an agenda and it worked. He wanted to get the attention of the "industry" to design better speakers and set the direction for all to see. Speakers should have great off-axis response and controlled directivity is a great feature.
Rephrasing, above the modal threshold the speaker is the dominate factor, so don't let the room mess it up.
Bridging that to Amir, I suggest you have to bring the specific speaker into the equation of set of factors to evaluate first reflection choice. Speakers as just described can easily be left alone at first reflections if....and here's where the silly qualifiers come back into the equation...if the room allows. If your room is narrow, you'll likely have to address these acoustically to not "%^&* with the speaker".
The one thing about RT60 and ETC is that they are direction blind without context. Acoustics can be approached in a directional manner so that the environment the speakers sees and the environment the human enjoys for speech can be optimized for each. And this would be the loftiest of goals and approaches.
As for RT60 (and removing the sub-text that reverberation does not exist in a small room), there are ranges of comfort and preference...I kinda like the .3 range personally. But I've done rooms with lesser T60's that sound and feel VERY comfortable due to directional use of acoustical fittings.
The more domestic the room the easier it is to get balance from general household furnishings and acoustical tweaks and a T60 in the .4 range general fits very nice. The more dedicated the room, the more you should think of all surfaces as part of a system. And finally, I consider myself an equal opportunity offender in that I like absorption, reflection & diffusion equally...just that the right widget has to be in the right place.
An example of how diffusion can play tricks in aT60 study...2-D diffusors spread the energy ~equally in all directions and there is an inherent absorption loss with this process. To the "eye" of the RT60 calculator this looks drier than the ear hears. This can be affirmed by doing octave-band filtered studies within the energy/time data. You could have room A and room B with a near identical looking T60 and room A could have carpet/drapes and upholstery while room B could have more reflective furnishings and 2-D diffusors at specific locations. The ear would in fact hear that room B has more mi/hi energy than room A.
What is "correct" is a case by case approach. However if you have quality speaker test data and a working knowledge of acoustical options/tools, it is very possible to get a GREAT sounding room.
Above the modal threshold, respect the speaker first and make sure the designed response reaches the listening position.
A lot of the confusion is from the reality that understanding of small room acoustics is a moving target. Why is this a moving target? My opinions (please note), first is that many acousticians had brought thinking from large room acoustics or hearing testing acoustics to small room environments (meaning rooms in less than 7K cu ft...a generalized threshold). This has created early standards that have been hard to break.
Second and I think most important is the positive trend of speaker design: improved directivity control and off-axis response. It wasn't that long ago that the on-axis pressure amplitude response was the driving factor in speaker design. In this condition, absorption of early/first reflection and more absorption in general can be a positive thing.
In a conversation I had with Dr. Toole, knowing that I'm an acoustics guy, he said "don't $%^& with my speaker". Shaking in my New Balance's, this one sentence has taken a long time to integrate into practice. I am of the personal opinion that Dr. Toole has an agenda and it worked. He wanted to get the attention of the "industry" to design better speakers and set the direction for all to see. Speakers should have great off-axis response and controlled directivity is a great feature.
Rephrasing, above the modal threshold the speaker is the dominate factor, so don't let the room mess it up.
Bridging that to Amir, I suggest you have to bring the specific speaker into the equation of set of factors to evaluate first reflection choice. Speakers as just described can easily be left alone at first reflections if....and here's where the silly qualifiers come back into the equation...if the room allows. If your room is narrow, you'll likely have to address these acoustically to not "%^&* with the speaker".
The one thing about RT60 and ETC is that they are direction blind without context. Acoustics can be approached in a directional manner so that the environment the speakers sees and the environment the human enjoys for speech can be optimized for each. And this would be the loftiest of goals and approaches.
As for RT60 (and removing the sub-text that reverberation does not exist in a small room), there are ranges of comfort and preference...I kinda like the .3 range personally. But I've done rooms with lesser T60's that sound and feel VERY comfortable due to directional use of acoustical fittings.
The more domestic the room the easier it is to get balance from general household furnishings and acoustical tweaks and a T60 in the .4 range general fits very nice. The more dedicated the room, the more you should think of all surfaces as part of a system. And finally, I consider myself an equal opportunity offender in that I like absorption, reflection & diffusion equally...just that the right widget has to be in the right place.
An example of how diffusion can play tricks in aT60 study...2-D diffusors spread the energy ~equally in all directions and there is an inherent absorption loss with this process. To the "eye" of the RT60 calculator this looks drier than the ear hears. This can be affirmed by doing octave-band filtered studies within the energy/time data. You could have room A and room B with a near identical looking T60 and room A could have carpet/drapes and upholstery while room B could have more reflective furnishings and 2-D diffusors at specific locations. The ear would in fact hear that room B has more mi/hi energy than room A.
What is "correct" is a case by case approach. However if you have quality speaker test data and a working knowledge of acoustical options/tools, it is very possible to get a GREAT sounding room.
Above the modal threshold, respect the speaker first and make sure the designed response reaches the listening position.
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |