Room Acoustics - Large Room and Small Room

[PrestonTom]

WHOOPS!!!!

I finished reading the rest of the Large-Room Acoustic chapter.

Guess what, I just answered many of my own questions.

Sorry, I did not mean to get things off track,

-Tom

[mas]

Thanks Tom...

Thanks for taking the time to ask, as if you are having trouble understanding something - as least at first reading, I suspect others are too! And I fear many will not ask questions for fear of looking - well, 'less then intimate' with all of the details! And the important thing is that folks understand - at least in concept what is going on!

So I hope that all realize that the only 'dumb' question is the one that someone fails to ask and get an answer to! And I hope no one is scared to ask!

Besides, with the silence, I have no way of knowing where everyone is - whether they are just not looking at the material, or if they haven't had time, or if they understand it all! And folks, you are ALWAYS welcome to PM me &/or to have me call you on skype if you don't want your question posted in public - & I can always take it anonymously and address the topic in general - whatever makes you most comfortable!

Tom, with your questions, I will try to provide a bit more info on those issues for all as well! All of the questions are 'most excellent'! (ie Q 1-3, the 1.6 sec RT60 etc and the frequency dependence and issues related to critical distance, etc....And I will provide some additional info for 4 too!)

I will post as soon as I have a chance !!!!!

Mark

[DrWho]

It's been test week for me so I haven't had time to read anything - so the soonest I can comment will be late next week.

[Arky]

mas

I intend to join your class this weekend. My knowledge is zilch so anything I learn will be a benefit. I have a foam by mail shipment due next week, and no plan, so this is timely.

I'm curious and may have missed it, what is your background?

[mikebse2a3]

Hey Mark

Just wanted to give you a litle feedback and let you know that I think your doing a great job with this.

I am following along some but unfortunately I've got alot going on for the next 3 weeks and I hope to get more involved with this thread when time allows it.

Anyway I appreciate all your Time and Effort!

mike tn[]

[gaspr]

The new edition of "Sound System Engineering" is now available from

amazon.com for $79.95 USD. From amazon.ca it is priced at

$59.09 CAN.

[mas]

If you'd like to save a few bucks, you can order the 3rd Edition of Sound System Engineering, ISBN 1-240-808304, for $71.96 from NSCA:

http://www.nsca.org/Home/Store/tabid/100/List/1/ProductID/112/Default.aspx

I think we need to work something out with our Canadian friends to buy it and ship it media rate to us!

Darn, I guess that means we need to be nice to them![][][]

[mas]

Just checking in...

Well, I don't know how many have taken the time to read and to attempt to understand the material, but I want to make a few comments regarding the acoustics of large acoustical spaces (LAS).<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

You should recognize that that a room must be of certain size to support a reverberant field. And this size is also frequency dependent! the lower the frequency, the larger the room!

Common to all spaces are the direct signal Ld, the early specular reflections Lre, and the later reflections Lr. In a sufficiently large space, the establishment of a 'late-time' statistically reverberant field is possible. What is unique about the reverberant space is that it is 'well-mixed' and uniform in level. You cannot identify any particular reflective paths, as every path is equally probable, and the gain of this reverberant field is uniform. Thus the reverberant field is an extremely diffuse field. And as few here have need to deal with LASs, we will refrain form suggesting how this should be 'treated', as we will hold off until after we have a better understanding of the various acoustical components as we begin to focus on the small acoustic space (SAS). And it is also important to note that a reverberant field does not manifest itself as a perceived echo, as it is often the case!!

With the establishment of the reverberant field, we have several concepts that are very important in dealing with LASs - among them being the Critical Distance Dc. RT60 measurements are useful; the Q (directivity) of acoustical sources is very important;

And lest anyone have gotten too hung up with the sample calculations, I sure hope that no one has spent too much time with them! They are by NO MEANS complete! They were presented (in a 'cheap and dirty' manner) simply to provide a very basic example of the origins of the mathematical calculations and the simple models employed. They also introduced a few simple modifications to illustrate both the potential and the need to address several environmental variables. And even more significantly, they presented the calculations algebraically rather than in a multi-variant differential form. In this way these calculations are alike those for LF room modes - there is a mathematical aspect to them, but in all but the most basic ideal cases, the calculations required are far too complex to reasonably mess with. So, aside from understanding their origin and the variables that enter into the equations, rather than dealing with complex abstract calculations whose simplified assumptions render results that deviate sufficiently from the more complex reality which they approximate, we instead cut to the chase and defer to modern measurement techniques that are much more accurate in that they measure 'what is' to many orders of magnitude greater than we can hope to calculate.

Oh, and there are employed in the chapter several values based upon empirical measurements of a great many acoustical spaces in order to simplify the examples. For example, the use of 1.6 sec for the RT60 of a moderately mixed reverberant field. They are valid, but (as I quickly discovered as I tried to think of a way to simply present the determination of the value!) they are far beyond the necessary understanding required to understand the basic examples offered in other words, you are welcome to delve into their origin after you have mastered the fundamentals! But at this juncture you are encouraged to accept them on faith as their derivation exceeds the basic concepts being addressed at this point! A Catch-22! (Plus, they are not critical values!)

Let me offer what might provide a more concrete example of the complexity. One is free of course to calculate the volume of ever single part used to build a car. Every single whatchmacallit, do-jiggy and widget, from each wire to the exhaust pipe to the firewall. And with the complexity of shapes, you should be kept busy for quite awhile. You can then attempt to provide an acurate density for each component, including those that are composites as well as those comprised of multiple materials. After calculating the mass of each component, you can then add them together and ideally you will have the total mass of the vehicle. Of course you will have to ascertain just how much gas and wiper fluid, and any other number of time variant variables that are presently in the car at a given time... A task that will take you far longer that you probably have patience, and still after you are done there remains a significant source of error. Of course with the advent of sufficiently large, high quality calibrated scales, the alternative is to simply measure the mass of the vehicle with a high degree of precision. You are welcome take your pick of the most expedient method that you prefer. Personally, I'll weigh the sucker and benefit from a more precise measure with a substantial saving in time and effort. Just as I prefer measure the response of a room rather than trying to calculate the respective characterisitics.

Hence, the point is that you should NOT try to calculate reverberant fields based upon calculations. (That is one of the more sadistic graduate school slave labor tests for witches deleted from the MP&THG movie left on the cutting room floor as it was simply too gruesome for all but the most stalwart of audiences [] )**. But you now have a very simple intro and appreciation into the exciting(sic) foundation of classical acoustics. ;-)

I am quickly getting off track here, but perhaps this provides a small illustration of how the tool of math has progressed as well. While most see differential equations (Calculus) as more complex than algebra, and LaPlace transforms as more complex than basic Calculus, each progression was in fact a simplification for the method previously employed. As the previous method became sufficiently complex and 'less well-behaved', new methods were derived to overcome limitations and obstacles inherent in the older methods. So, rather than the new methods being 'harder', they were actually shortcuts enabling more powerful methods that were easier to deal with. Thus this same situation has been addressed in practical acoustics with new and more powerful tools, which have the benefit of being both easier to use as well as being more accurate.

With Schroeder, we begin to move out of the old models and enter the more contemporary model. And we now have methods that can measure the characteristics of a real space more accurately than our simplistic calculations are able to accomplish. And lest one think that they are capable of calculating such issues (as with the low frequency modal characteristics), except in extremely simple rooms, few are ready to tackle issues such as the coupling of spaces, irregular volumes, reflective, diffractive, refractive qualities of environments and an entire host of variables that render even the most thorough attempt to model an acoustic space mathematically as a grand masters or doctoral thesis. And that is for only one space! We haven't even begun to examine how the two or more spaces couple and beheve in a manner different than either single space... And we are not limited to 2 coupled spaces! And then the primary focus of the research would appropriately be to explain the differences and limitations of the calculations versus the response of the real room! If one wants to delve into this arena, you are welcome to explore the realm of such tools as EASE and CATT-A after we have finished this topic. And there is quite healthy debate into just how real world response may vary from the calculated models. You see what fun awaits you if you choose to enter into such an exciting(?) group discussion?

So...I hope I have presented a sufficiently complex morass and provided you with sufficient reason to want to escape from it (and this discussion!) as quickly as possible.

In short - forget trying to calculate the reverberant characteristics. Focus on appreciating the basic approach and some of the variables taken into consideration in calculating such a complex multi-variant characteristic. The inordinate complexity of calculation combined with the advent of a few new tools renders this a character building exercise. Rather, focus on understanding the necessary conditions and the meaning of the component and summed characteristics. Instead, what you should come away with is an appreciation and a good understanding for just what a true reverberant space is...and what it is not!

So you should understand what constirutes a large acoustical space (LAS) as well as the fundmental and necessary criteria for the esistence of a reverberant field. This is important as we are going to quickly move to the subject of the small acoustical space(SAS) . And the criteria for a true LAS reverberent field are not met. Thus you must understand the limitations encountered in the SAS, and why the existence of a true reverberant field is not valid. As we get there, this understanding is going to become critical in identifying the actual situation at hand, and it will help you to understand the variables upon which you will need to focus.

And a thorough understanding of what a reverberant field is not brings us face to face with the small acoustical space (SAS) - an understanding of which will form the focus of what we will be dealing with in home theater and home audio listening rooms.

Please review pages 210 and 211! And note page 220 for what will become more useful in a SAS.

For all intents and purposes, we can forget reverberant fields. Instead, we are going to begin focusing on the precursor to such fields, and to attempt to understand the characteristics of the component variables and the nature of the reflections within a SAS which lacks the establishment of a significant well-mixed reverberant field.

As such, you can forget your RT60 measurements, the important Critical Distance, and many other fundamentals of LASs! I would think many would cheer this! But while we are able to dismiss the existence of such relatively well-behaved characteristics, it becomes very important that we understand the component variables! And it becomes necessary to identify and focus on addressing each of these component variables! Thus, while we will not be dealing with a reverberant field, the differences exhibited by a SAS and an understanding of these variables and components are critical. The realm of the SAS is more complex and much less well-behaved then the LAS. Our understanding of reverberant fields focuses on the ideal, which the focus of the SAS is upon a very 'non-ideal' space.

So, in a sense, Life has become more difficult, as these variables are not well behaved. They are 'situation' dependent, and they do not constitute a static model for which easily standardized recommendations can be made. Let me qualify that statement A general standardized model can indeed be suggested. But the behavior of the variables in a SAS cannot be easily predicted in a manner sufficient to state a standard prescription for how the unique combination of variables in each space must be specifically treated! In other words, I can tell you what should be done to treat the various variables in a SAS. But I cannot tell you exactly what should be done where, as each room will present unique characteristics that influence these specifics. And without measurements, we are rendered able only to make general abstract suggestions and procedures about what 'should' occur. But with measurements, these suggestions and processes can be honed with razor precision, and the effect of each specific treatment can be verified.

At this point I will digress and mention the subjective quality of psychoacoustics. Much is known regarding how the various phenomena manifest themselves and how we perceive them. There is even a large body of generally agreed upon preferences and general guidelines. Issues such as intelligibility can be measured and quantified. The science of pyschoacoustics is objective. Various measurements quantifying and qualifying this have been developed and are generally well agreed upon (%ALCONS, etc.) But ultimately, you must decide what is optimal for you! You decide what effect you like. You may prefer a particular psycho acoustical anomaly or shading. Rather than tell you what you like, we will attempt to stay within the bounds of what are generally accepted current best practices. What I state will attempt to reflect this. But I have no desire to debate individual deviations from this except in a separate thread. My desire is to address the variables and the combuinations that enable you to make and refine the ice cream. You will have ample opportunity after these concepts are understood to experiment and discover what flavor of ice cream you like!

I hope that this does not sound harsh or seem a dismissal of differing tastes. But as it does not fundamentally alter the understanding of the component acoustical variables but simply suggests alternative manipulations of them, the discussion is superfluous to our understanding of the basic behavior of the component variables at hand. Does this make sense? Whether we add a little more sugar or a little less sugar to taste, it does not alter our fundamental need to understand the behavior and effect of sugar!

May I suggest that AFTER you thoroughly understand the fundamentals, THEN we may intelligently and objectively debate these various subjective preferences, and how the various objective variables may be employed to achieve them, and why they are preferred, at our leisure. But this is a separate discussion for a later date. We have our hands more than a little full right now with the basics!!

So, if we can defer to the subjective issues and focus on the physics of the component variables, we can hopefully begin to understand the nature of the variables faced in the SAS, and begin to focus on how to objectively identify them, and then finally, what we can do to modify them and then how to objectively verify that what we think we have done has indeed been accomplished! All of this can now be objectively accomplished with great precision. And with a thorough understanding of these principles, you can then become the master artist, with the necessary tools and understanding to effectively create your own masterpiece.

So, for those interesteddo you have questions regarding what a reverberant space is? If you read the first several pages of the Small Acoustical Space chapter, you should be introduced to the limitation that differentiates a small acoustical space from a LAS by virtue of the SAS not satisfying the necessary criterion for the establishment of a reverberant field. Do you understand this? For those who don't and won't ask questions, maybe I should threaten to repeat the last circular statement! [] OK, enough scary threats! But if you don't, PLEASE ask, as an understanding of this characteristic distinction is critical to go forward!

If you can, ask focused questions referring to specific concepts or definitions and avoid trying to leap to applications or uses at this point as we lack sufficient foundation to go there yet. After we get through this there will be ample opportunity to explore all of your questions! And trust me, nothing would please me more than for a substantial group of people on this site to be able to intelligently explore the pros and cons of a particular approach or preference! In the meantime, it would be nice to try to keep everyone close to the same page without folks falling behind (even if some of you are a little bored and impatient to move ahead)! We will get there quickly enough! But I do not want those a bit confused to be discouraged or lost or to give up!

And as I have mentioned beforeIt is much easier to address questions directly and in a focused manner with real-time interaction rather than from a 60,000 foot vantage point! So please, feel free to PM me and we can chat via IM or by voice on Skype (www.skype.com) and I will try to specifically address ANY question or issue you might have A few have done this, and I think they will agree that it is much easier to zero in on any issue quickly and easily in an interactive manner and to spend as much time as necessary to explore and clarify any issue! BTW, Skype computer to computer calls are always free and computer to phone calls in the US are free - at least through the end of the year. So Please, just ask for assistance. Unlike this diatribe [], it is rather easy one-on-one to cut to the chase and address any misunderstanding or confusion.

**You might want to talk to the folks at RPG regarding Room Sizer and Optimizer. These two programs which attempt to calculate room modes and to provide dimensional analysis of PERFECTLY RECTANGULAR rooms ONLY (without open entries, alcoves, stairways - perfect closed rectangles with uniform surface characterisitics) strain even the latest multi-CPU PCs! If you inquire about them and pursue the topic of a new updated release that will approximate more realistic room topologies, depending upon with whom you speak you may hear them refer to Stephan's work for the next version of EASE for a more comprehensive model. And EASE costs in excess of $5K! ...and while good, for these characterisitics - LF room modes and reverberant fields (in large rooms), we still measure the room !!! The required algorithms are simply too complex to adequately represent all of the variables of a real space. And that is after the program completes execution on a quad CPU Opteron workstation - which can easily take several days of realtime run-time!

Mark

[DrWho]

hey mas, I haven't had a chance to read through it all yet. I'm like 2 weeks behind on correspondance and school continues to pile on the workload. Just thought I'd chime in so you didn't feel like everyone was ignoring you. To be honest, I haven't even read this last post of yours all the way through either []

I would be interested in calling you sometime though - perhaps after 9pm when it's free (provided it's not too late).

In fact, I have a new application at work that I'd like to do some time-based measurements. I finally have my "calibrated" microphone (good enough for what I'll use it for) and hopefully I'll get ETF or something similar. If I'm feeling really nerdy I might try writing one of these programs for a school project, but maybe that's a bit overkill and just reinventing the wheel []

[mas]

Mike, you have a PM. (that just doesn't sound quite right!)[][]

[gaspr]

Fascinating thread so far. I think I have read everything at

least twice, some of it three and even four times. I know

that some may have not had time to "do their reading" but I, for one am

anxious to learn more.

So when do we start classes again Professor mas?[][][]

[DrWho]

Ok, I've heard this term thrown around quite a bit, but I didn't see it defined very well:

The Haas Kicker:

Why the heck would you want to intentionally introduce the Haas effect in the control room? It makes sense that you want the ISD to be greater in the control room than in the studio, but I totally don't get the Haas thing. I did notice that it was mentioned that the kicker should be adjustable - is this more of a referencing the mix type of exercise?

To bring it back home a bit - what do we want to do in terms of a Haas kicker in the home environment? And what about multichannel? The geometries of the rooms discussed are geared towards 2-channel playback - I don't think it outrageous to claim these rooms wouldn't be ideal for the surround channels.

Another concept I wanted to throw out there is the ability to "ignore noise". I've never tried to share this approach before, but one thing I try to do when tracking and mixing down in the studio is to ignore the sound of the room and focus on the initial original sound coming out of the monitors. Essentially ignoring the "reverb" of the room. I almost want to say this is a frequency dependant thing too - being much easier with the higher frequencies and much harder with the lower frequencies. So when dialing in the low end I'll reference on a pair of dry headphones (AKG K240) and a pair of wet headphones (Sony MDR). It's hard to do imaging and depth on headphones which makes mid/high frequency adjustment impossible with headphones (though they are a great check to make sure the stereo spread isn't too wide).

Anyways, is it possible that what I'm experiencing is the ISD gap? And would it be too extreme to argue that a recording engineer should only be mixing for that small section of the sound and rely on the end user to optimize the "reverberant" structure of their own room to suit their tastes? I'm sure those familiar with the nature of "referencing the mix" will appreciate how loaded this question is.

Referencing the mix is a topic that has really pissed me off because any good engineer should be able to make anything sound good on the system to which he's listening. Maybe that's overflow from styles of mixing live sound (which is often less than ideal) where the deficiencies of the equipment and room become part of the music. I've always felt that a good recording engineer should be able to optimize the sound for a specific room if he so choosed, which would imply we should all be listening with the same equipment in the same room. Afterall, isn't the goal of our playback systems to "hear what the artist intended"? And then I sit and think about the dry boring nature of the studio that optimizes the ability of the engineer to hear the flaws and remove them - often resulting in a mix that is everywhere more enjoyable than the studio. I feel understanding this disparity is the most important aspect of getting the most out of our systems because the "fidelity" of the source material is the only thing we don't have any control over.

And then in the back of my mind I'm always thinking about how one can still enjoy a good song, overdriving the crappy stock speakers in a car with the windows down at 60mph. Sometimes I wonder if all this "ideal" crap is really worth the trouble - does it really enhance the enjoyment? Seems "better" sound systems always result in revealing more flaws - effectively limiting the breadth of music that can be listened to. Maybe Bose should be getting more credit than they're given... []

Ok, that's enough for now - hopefully it'll spur a little more conversation. []

Thanks for the 'light' read []

[WMcD]

I must catch up on reading too.

The kicker is something of interest.

My first thought was that it serves as some sort of a psyhcoacoustic marker. As if our ear brain combination needs it to mark the upturn to the reverberant portion and the end of silence of the ISD.

My second thought turns to what I understand is being done with data reduction in audio recording. One effect is that we do not perceive sonic events just before a loud noise. It is as if the pre loud event gets ignored so that system resources can turn to figureing out the big loud one. The data reduction systems exploit this and don't bother to preserve the pre loud event data.

Maybe that is the reason the kicker is important. It makes our brain ignore anything in the ISD period. The ISD should be silent. If it not, the kicker turns off our brain from hearing it.

Gil

[DrWho]

Hey Mas, just wanted to say thanks for the conversation the other night. I can't believe we talked for 6 hours! []

A few things clicked and I'm totally wanting to dig into this a bit

more...I was even able to explain a few of the concepts to my fellow

AES members at a meeting tonight. Perhaps I'm running with scissors,

but it really doesn't seem too complicated once you really understand

what the measurements are showing. Of course I'm sure picking the

treatments and the implementation is a whole different world, but at

least I can identify the crap []

You out there Damon? I would love to work with you on your new room and

collaborate on the treatment process. I don't have the means to

experiment with this stuff on my own, but I'd love to help any forum

member to the best of my ability. It would be educational for everyone

involved. I believe Mas is willing to help out quite a bit too - even

mentioning some DIY room treatments that should work quite well.

I also believe we have an article coming up shortly that's pretty much

an introduction to how one would go about the room treatment process -

which should prove to be a most practical learning tool. It would also

be fun if someone on the forum wanted to document the process of

treating their room. Maybe we can stumble across some aesthetically

pleasing ideas (which seems to be the major deterent in typical rooms).

[gaspr]

Just a bump...hoping there is more to come in this thread.

Garth

[mas]

Funny, I was wondering the same thing!!! [] My free time has gotten crazier as well as many others here! But not to worry...

Actually, I was waiting to reply to Doc's and Gil's comments (as they have jumped ahead to a very interesting topic (ok, ok, so I don't have so much of a life to find this interesting! [] )) and I was rather waiting to know when others had had a chance to at least look at the material so that we could resume reviewing and adding more to it - as there is allot more.

It is much easier to comment on that which others have at least looked at rather than trying to qualify a concept where no one knows to what you are referring![]

What I wanted to do was to review a bit about the various parts of the room and the measurements of which we will need to be aware, and then start addressing their relationships and how we go about tailoring the relationships.

Then I have some more info that will take us through some additional methodology for treating a room - after which we can talk about the various ways of approaching a space and the various treatment options.

But one thing at a time! As I know that many will want to jump to the end. But going through the steps will actually make many of the treatment questions easy, as many will then have the tools to determine the answer to their own questions - and many more. Or, at least that is my goal! []

{I am also fighting with ways to be able to present graphics here - as this site is not very multi-media friendly - and I need to experiment with embedding graphics - do I have to upload the graphic to a folder here?}

__________________________

So here is the proposed itinerary:

First: Just read the material - do NOT get caught up trying to do any calculations! Especially reverberation! Hint - the calcs are like room modes - existing real world spaces are not calculated. They are measured. Secondly, in a small room, you won't be using reverberation or RT60s anyway. (Hint: The entire chapter on Large Acoustical Spaces was meant simply to introduce you to the concept of reverberation so that we could ween you off the notion that you will be using it in Small Acoustical Spaces!! See how easy that is! So, while you need to know what reverbation is in order that you can tell what it is NOT, reverberation will not be on the test![] You can revisit reverberant sound fields when they call you to tune the Superdome.)

Be aware of the ETC diagram structure and the 3D waterall plots - as we will be referring to them constantly - especially the ETC.

Direct signals Ld (hopefully we won't have to spend too much time on this concept [] )

Early reflections Lre - oblique and perpendicular behavior

The ITD - a relation between the direct, Ld, and the early reflections, Lre, plus an intro into the psychoacoustics of which one must be aware - and how we determine the ITD and why it is significant... And just what is a 'kicker' anyway?

Semi-reverberant fields and what we do in a Small Acoustical space when you have lots of 'early' reflections, but the space does not support a reverberant field.

General conceptual room tuning article...

Once we know what the various fields and pieces and parts are we can start to examine 'how to manipulate them'. Which leads to...

Methods for tuning and adjusting these regions so as to create a comfortable balance and relationship between the various 'components'.

And then we can address lots of neat tricks and methods - especially as most rooms have weird anomalies making such a bag of tricks pretty useful.

[Deang]

I feel stupid and my head hurts.

[colterphoto1]

I'd like to read the assignments, but can't wade through the fuzzy text. Is there any better way to get this material other than the $71 to buy the text? I've gotten through all the posts and understand what you're saying, but understand that I need to get the basics down before we go further.

Michael

[Deang]

I'm going to buy the book and use a black highlighter on the math.

[gaspr]

The way I read this is that for the most part, the math is there

as a sort of "proof" for the general concepts. So, if you

accept the general concepts, no need to do the math.[]

Math....that's what computers are for.[]