A Beginners Guide to Room EQ Wizard (REW)

if you enjoy content like this please take a moment to like and subscribe watch it and then join the conversation at AV nirvana calm all right so this is gonna be a basic room EQ wizard demonstration I'm calling it the best free tweak you will ever buy so when I did this at ex Bona they didn't know what rule was necessarily so they wouldn't know you guys obviously know what ruh is you will know that it's free but you also probably recognize that it's a pretty cool tool that we have free access to so I just want to quickly mention my affiliation so that you know where I'm coming from so I am from av Nirvana which is another forum like a BS from the home theater Shack guys and they also are hosting now roomie Q wizard and so that's kind of my role with that you go to our website you'll see our reviews I write some of them so please read them so I already mentioned that roomie q wizard is which we call ruh is basically being hosted on this site it's sort of the official support site av Nirvana John Mulcahy who's the developer of it provides free support there it's really the only site where he's routinely providing this level of support so it's great because he wrote the software you got a question he's got the right answer me on the other hand I don't know what I'm gonna tell you here it's free teachers well here is like a really small sampling of its features and I know you can't read that which is sort of the point this software is extremely capable so what I'm gonna show you today is really dealing mostly with how to take basic amplitude response measurements the frequency response but it's capable of quite a bit more so you can use it for doing like you guys have seen those cea 2010 subwoofer tests that like James Larson is here does and others so this software actually has like a built-in ability to do that it's got a full tone generator that you can use for testing all sorts of things it's got the ability to take impulse response measurements they can do RTA measurements it has different kinds of spectrograms including wavelets which is my new obsession so lots of cool things in there so tools of the trade things that you need to actually do this so we'll start with that you need to have a measurement microphone and I I think for home theater guys you guys have had the beat into your head that you need to use a measurement microphone I actually get a lot of people who come more from like the studio world and they'll be sticking their standard studio microphone in a stand and measuring and saying hey how come my response looks so bad so besides the fact that most speakers don't actually look very flattering in a frequency response measurement you do need to make sure you're using a microphone with a flat frequency response so the nice thing is that the wonderful people at mini dsb and at Dayton parts Express have produced very inexpensive under $100 microphones that are have a very flat response and when you use the correction files that they provide they are basically close enough to a reference microphone for our purposes and they plug in through USB which is really really easy then and then what you can do is you take your laptop or whatever computer you're using if it has an HDMI connector like this one does and you're doing like a home theater where you've got an HDMI input you run a nice long HDMI cable like I have over to the receiver you plug in your USB cable and you put it in a microphone microphone stand I actually intentionally brought my cheapest one I had this is an Amazon basics I think it was like $15 something like that it's really cheap so you don't have to spend a lot of money to do this and you don't need a lot of fancy cables rooo is then going to send test tones through your receiver through that HDMI connection the same way it would for music or if you were watching a movie so one thing that I think people sometimes don't fully understand is they think to themselves all I care about is how my system sounds I don't really care what it measures like not a scientist I'm not interested in the physics here so why is root why am i calling Roo a tweak so it provides useful repeatable and objective information that can be used to optimize the setup of the system so that's like the huge thing most of us even really experienced audio engineers let's say don't necessarily have ears that are attuned well enough that they could sit down listen to a system and say oh for sure there's a big boost of 50 Hertz there's a big cut at 80 Hertz and to be able to and then listen in different positions for instance and figure that all out so it's hard for us with our ears alone to really recognize everything that's going on and we'll hear it but we'll have trouble really quantifying what we're hearing and it takes with Roux you can actually do this very quickly and for those of you that don't have the experience of listening to lots of great systems it's actually even harder since you don't have a point of reference to know what you're listening for the microphone on the other hand is is measuring against flat so if it's not flat then there's something wrong and the trick for us is really figuring out a are we getting measurements that are in fact representative of what we hear which we'll talk about a little bit and then the second thing to think about is how flat is flat enough and in the reality as it does not actually have to be that smooth to sound pretty good our ears are really good filters so we've talked a lot about EQ business on the the queuing systems on the forums and you guys are familiar with this and I think a lot of you are what kind of wondering what this is all about so you're literally modifying the signal before it goes into the system URIs and essentially preconditioning it in order to compensate for the distortions that the room causes and to some extent the speakers themselves so your EQ it to make it sound better and some people get confused by this because they think to themselves why would I want to modify the signal somebody spent all this time engineering a perfectly good speaker and a perfectly good system and the problem is what people don't necessarily know how bad the room itself actually is even really well-designed engineers really well designed and engineered rooms I mean have problems and they caused huge distortions huge linear distortions in the system what that means is the frequency response gets to be really really wavy and and it sounds poor and so the EQ is one of the ways we can compensate for them and to do that well you really need to use measurements so the basic steps as I mentioned earlier are you need this microphone you need the laptop you plug it into your system that's your connection you plug your mic into the computer that's how it gets sound in into it and then you have to place the microphone at the first measurement position so that should probably ideally be your main listening position where you typically listen that's where I would recommend to start it's not really that critical you can measure in any order you want but make sure you label them so you place your mic in the listening position now I typically recommend that you place the microphone pointing up so here's the thing I'm going to say about if you have a microphone that came from cross spectrum where they've given you zero a forty five-degree in a 90 degree compensation pointing up with the 90 degree compensation is probably the best thing to do if you have one of these and you don't have that correction for low frequencies only it's probably fine but to be honest you should otherwise be pointing it out the speakers because the response won't be accurate pointing if you have one of the many DSPs they have a 90 degree correction it isn't actually a measured 90 Gert correction it was derived and so some people believe you shouldn't point it up you should point it at the speakers and use the zero Degree correction some people say pointed up it low frequencies it doesn't matter and that's really where you should be focused anyway so I'm just gonna say I think you should point it up since you can be measuring all the speakers it means that the microphones response shape is gonna be roughly the same for each speaker it's going to get the most information from the room and it's just the easiest way to do it even if you don't have the proper 90 degree curve so you place your mic in that position and you're going to take a measurement we'll go through that a little bit and you keep repeating these steps in each position so if this was my room and I was doing it this might be my main listening position where you're sitting I'd probably place my mic there I'd probably put the stand behind your head and I put the mic so that the ste neck she goes down over the seat with the mic pointing up and I would take a measurement there my next measurement position might be over where you're sitting my next measurement position might be over where you're sitting then I might go to the backseats now I'll say this Mike's room is set up relatively well so that none of the seats are overly close to a wall what I wouldn't do is let's say Mike actually had four seats here and five seats there and they actually pushed right up against the wall I wouldn't bother measuring those n seats against the wall because those measurements are likely gonna be fairly poor from the reflections off the wall and I don't want to compensate for that because whoever gets those crappy seats gets crappy seats and I don't want to make my seat worse because of their crappy seats right so you there's certain you have to look at your room if you've got like a nice space like this you measure basically in all the seats and I mentioned that I would do what would total probably seven measurements I actually would do more than that I would take more measurements around the main listening position I would just move it a little bit over to one side a little bit over the sign a little bit up a little bit down and the reason you do that is you just want to see how the response changes again we're going to talk about this that's part of the analyzing the data so we've talked about the connection process using the HDMI you can get the full 7.1 you can send discrete each year channels like that you can also actually do it through the analog sound I've actually on my computer got it through the headphone output before I've done it through a sound card a sound interface there's lots of ways this can be done and even if you have a surround system and you want to go to each speaker is if your surround processor has like a 5.1 input of some kind you can do it through that that's so that's something I've done before it works fine ideally you want to use the signal path that you're listening to just because sometimes the receiver or your source somehow modifies the signal and you want to make sure you're capturing that so as I mentioned place the mic at this primary listening position I said face it up you want to make sure you've loaded the correction file for it at low frequencies it's not a huge deal but one of the nice things is that the correction file does help to ensure that it's basically flat to 20 Hertz and if you have the cross spectrum one and actually make sure it's flat to below 20 Hertz as I said the mic Corrections are not all that big a deal below 500 Hertz these capsules are actually pretty flat they get a little bit rough by actually closer to like 5 kilohertz and that's restart to have Corrections so what I'm saying is in a room that I'm using this 500 Hertz actually means something it's not 500 Hertz like generically it's that 500 Hertz is above the rooms what's called the Schroder frequency the res the FS of the room but it's still within what we call that transition zone so that in there in a room there's sort of these acoustic zones if you will and at higher frequencies there's what's called a statistical zone and what that means is that statistically you can figure out where the sound is coming from but when you take a measurement you could retake that exact same measurement and get a slightly different revolt result in the high frequencies because the way in which the sound is reflecting off of the walls and coming from the speakers varies a little bit even over time at low frequencies that isn't is true and the room actually starts to play a bigger and bigger role in the sound and so you get into a what's called a modal zone and instead what you tend to see is that the response stays the same except that these modes caused the response to get big dips and Peaks and people you guys have heard about the modes that's what we're trying to actually typically fix there's then a transition zone between that modal zone and that stochastic or statistical zone that is an area where depending on what's going on and what it looks like it is okay to maybe apply a little EQ moving the subs around a little bit might actually have an effect up maybe not to 500 Hertz but it could have an effect even up to like 200 Hertz and so this is a zone where you're still playing around and so 500 Hertz and below is a place where we care about and it happens to also be a place where most microphones all measure about the same your as long as there's no filters on even your own smartphone's microphones can pick up sound pretty flat down to that so the other thing I mentioned here is you might want to take a picture when you do this your and I need to repeat the same measurement points when you apply EQ so that you can see if what you did worked the way you expected it to and so it's I'm not gonna say if you're off by an inch it's gonna make a huge difference but it does change small changes in the mics position can change the response and if for some odd reason that's why I was saying I usually take a bunch of measurements around my main listening position sometimes you'll get like a weird reflection off the ceiling or something with the mic or you'll just happen to hit some weird resonant zone where you get a big cancellation in the response and you move it a little bit it goes away so you want to make sure that when you find that good spot that you're able to repeat that same good spot and you don't accidentally do all this EQ and then measure in the bad spot and you're like I don't know what did I do when in fact all you did was put the mic in the wrong spot so what I sometimes do is I'll put the mic in the spot and I'll take a picture and I'll go to the site I'll take another picture and then that way I can use the picture to get you know close to note the other thing you could do is maybe put like a sticker or I know somebody who actually he puts a pushpin in the ceiling and hangs a string down I don't like to put holes in my ceiling but you know if you don't care you can do that so here we've got we're just going to go through the basic steps here so in the upper left corner that okay I apologize you cannot see but over here would be something that says measure you'd click on it and this would come open so when this comes up what you want to do is you're going to measure the you're going to set your frequency response you want to measure - so I'm in here I'm doing 20 Hertz I know a lot of you guys have hello Tony hurt you you would like to find out if you have bass below 20 Hertz you can do that keep in mind anything probably below 30 Hertz potentially could start doing damage to your system so if you really think you've got bass below 20 Hertz and you want to do it a I would make sure it's not up to loud until you have a sense of what your distortion curve looks like and B I would make sure you know your system can do that if you built your own subs and they're ported for instance and you ported them at 20 Hertz 10 Hertz to cause the woofer to exceed its xmax so significantly to damage it so I typically cuz for acoustic purposes below 20 Hertz serves no purpose so I typically set it at 20 Hertz I go up to 22,000 Hertz and I always do for them unless I'm in a hurry I always do full frequency response measurements and I'll describe why later but even if you don't need that data for things like a queuing having it helps because for instance the harmonics of distortion are at multiples of the of the fundamental right so a second harmonic of a hundred Hertz is 200 Hertz but what if we're looking at the ninth harmonic for instance which is it near its 900 Hertz so I like to do full measurements so that I don't get rid of any of that and it helps me to because sometimes I can look at other things like distortion to make sure that the system isn't distorting oddly in some way that I didn't expect okay so here the the 128 KN it says link that's the FFT length so the way that this system works is it uses a way of processing the signal and the law and it's called FFT or fast Fourier transform the longer that you make that the lower the noise floor and that's good for things like distortion or any of the decay measurements like waterfall plots or the wavelets for most of you 128 K is probably fine if you have a noisy room with like a lot of constant noise go to 256 K if you really want like the best measurement and you don't care about the added length of time it takes to take the measurement you can go to 1 1 Meg and that's fine in terms of output you'll see that I picked the left and I'm using an acoustic reference I'm also using the left for that that's fine when I you know I might do all the measurements I told you about like that the way I have my system set up that would actually cause the subwoofer to come on and the left speaker which is a good way to do it because you can see how they integrate while you're playing around the EQ and everything I might do it with the right speaker next I might then do it with the center channel if you use the ASIO drivers you can hit them each discretely as well that's fine it's not all that important but I would say this you do want to make sure that when you're measuring and tuning a system you don't rely on a measurement of just the subwoofer in just the mains because that doesn't actually tell you how they integrate so you do want to make sure that you do at least like I would actually do all of your EQ measurements based on like the left speaker and the subs playing so you hit start and it's gonna do its thing I have here Shh so the way the system works a little bit of talking in noise isn't gonna totally corrupt the measurements and I used to be really lacks about this but I've actually gotten some measurements from people where I saw really odd distortion results and some kind of weird dips in the response and I took the file the wav file out of Roo and I stuck it in a program called audacity which is just like a sound recording program and I listened and I could hear people talking loudly and it turned out that the weirdness and the response was background noise so it's pretty good dealing with that but you need to be quiet and you know my recommendation is try to find a quiet time to do it all right so as I mentioned it's pretty important to take multiple measurements and this actually came up on a forum recently where somebody had brought up weather microphones measure like our ears so I'm just gonna really briefly talk about this idea here measurements don't measure like a ears they don't hear what we hear and while in some ways like the resolution the frequency resolution that they can capture is better than our ears what matters ultimately is how it sounds to us so it doesn't matter that this for instance can measure in a more flat way or have greater resolution than our ears because it's what our ears hear that counts and there's a problem our ears don't have the same kind of polar pattern the way they pick up frequencies at different angles as one of these do so these are actually picking up sound in a room differently than we hear it so if all you do is stick the mic in that position and take one measurement you have not captured what you hear and I hear people sometimes say oh no but I don't move my head so why do I need to measure more than one position it doesn't matter that microphone didn't pick up what you hear even in that one position so you do got to take more and so that's what this idea of getting some spatial variances want to actually cover a few different spaces in where your head would be to get a sense and you're gonna find that for the most part in most rooms that low frequencies the response doesn't change much if you're only moving at 6 6 to 12 inches but you still got to take those measurements they're pretty important so the next thing you're gonna do is once you've collected all this information you've got to analyze the data so what I do is I take my measurements I label them pretty carefully as to where I took them so like I said I usually for the purposes of acoustics so I will check all the speakers because I want to see that they all look right and everything's ok but for acoustics purposes tuning the system I don't use all the different speakers I use usually just the left or the right speaker and the subs playing at the same time and that's it and I do all my analysis based on all the different space in the room from that and what I look for is how it varies so what I would expect is that it will vary if I have just a single subwoofer it will vary quite a bit in a room this size and all those different seats the modes are just going to cause that to happen but I would not expect it to vary all that much around my head and I want to see that and if it does I want to see how and why and I want to look for big peaks and big dips so if I take a measurement of this system and I see that there are some really big peaks and dips the first thing I'm going to do is part of my analysis process is try to figure out the cause of that so if I didn't move the speakers and I took the measurements like I mentioned to you then that gives me a sense of how the response changes from the room modes which is the sound the low frequencies hitting the walls and interfering with the direct sound that you're getting in your position but there's another thing called sbir which is speaker boundary interference response which is what happens when the speaker's sound radiates for instance behind the speaker and at low frequencies it does up down that way and back and interferes with the direct sound coming directly from the speaker so in one case it's an interference from where you're sitting the other one is an interference from where the speaker is so I need to figure out which of those it is because if it's a room mode there's not a lot I can do about that multiple subwoofers can help some treatment of the room to a point can help eeeek ewing some of the peaks can help but if it's speaker boundary interference moving the speaker's can actually make it so I want to know that so that I know what I need to treat I checked distortion I don't make a big deal about it like to me if you get into this sort of what I call like desktop drag racing where you're like oh my system has 0.01% purses you're 0.1% that's silly that's it's like totally meaningless all you're looking for is does the distortion curve remain relatively smooth and flat at the I mean you should be measuring it like 80 I didn't say this earlier but around 80 to 85 decibels is a good level to measure at so at that point you shouldn't see any big rise in distortion maybe at low frequencies as you get closer to the lower limit of your system you're gonna see the distortion rise and that's okay but over most of the trains you want to see that it remains low and and flat and as long as it's below 1% it's probably fine if it has a couple peaks that get up above to 3% that's probably fine to what I look for is sometimes you'll see like a sudden rise especially if the speaker has a damaged tweeter basically starting with the tweeter starts that whole area will rise up pretty high and it'll actually probably exceed 5 10% maybe higher yet if it's really damaged I also sometimes have found some speakers that just weren't that well designed and the crossover caused the tweeter to overload at 85 decibels and I saw that in the measurement where as soon as the tweeter came in there was a big rise in the distortion and then it went down again so I want to know that that tells me what the system is like and how it's working in that particular case you can't fix that but I still would want to know it Archy 60 refers to the way that sound reverberates and decays in the room so I look at it but I want to warn you guys I look at them I think most people look at them and most experts even though they look at them will tell you they're not very valid it's repeatable so from that standpoint I consider it valid but these rooms are acoustically way too small for the concepts that underlie our T 60 to be accurate so I tend to look at what's called EDT early decay time and there's well I figure what stands for but it's T opt those are the only two I really look at and I I would take it a little bit with a grain of salt your goal is you want to see that it's under half a second 500 milliseconds ideally if it's a home-theater it should actually be like under 300 milliseconds and you want it to be as flat and smooth as possible so this is not something you can EQ this is something you have to tree with acoustic treatments like this so if you see that like for instance it's really wavy and bumpy that tells you how to treat the room now at low frequencies it's not valid so below 100 Hertz that pretty quickly becomes invalid and you're not going to use them for that purpose but ideally if you've treated the room well enough you do actually see that they stay flat down you know potentially you know like 50 dat hurts I resist the waterfall so how many of you raise your hand have red professionals even tell you that when setting up a system you need to look at waterfall plots yeah lots you so here's the problem with that and here's why I say resist the waterfall these systems operate generally at what's called steady state a steady state system operates in a particular way that's really important whatever happens at one point in time is carried out over time so if there's a peak in the response at time zero there is a peak in response at time thirty and so that means that what they're telling you to do is look for ringing right they're telling you that there's ringing you need to get rid of it well there will only ever be ringing in a system if there's a peak in the steady-state response at time zero so you don't need to look at the waterfall you know what's happening over time based on what's happening at time zero that's why I say resist it because the problem is there it's not a very good resolution it's telling you nothing different than what the steady-state told you and it can give you misleading information you'll think you have ringing when in fact all you have is some noise in the background that it picked up or you'll think you don't have ringing when in fact you do that's in the steady state response but the resolution of the waterfall is too poor to pick it up so if you really want to look at what's happening over time a a trained eye and it doesn't take a lot of training can pick it up from the steady-state and be there's a better thing called a wavelet that you should use so check the wavelet instead and then 1/3 octave filtered impulse response so one of the things you can do this is somebody was asking about more advanced this is probably more advanced if you go into the impulse response there's a filtered interest fault and altered impulse response and you can set frequencies if you look at the different ones you'll see what you'll get a sense of like what it should look like which is basically a nice kind of round looking impulse as it gets to be higher frequency it's gonna go more from like a almost ball shape to something more pointy until it gets to be really thin and pointy again we're only focused on like 500 Hertz and below everything above that these measurements are not really a great way of assessing acoustics so what I do is I look at 500 Hertz or so and below and I look at the shape and what you're gonna find in a room is that where you should have like 250 Hertz is a really common one 250 Hertz happens to be the point we're ceiling in floor bounce that's the kind of sbir causes a certain cancellation and what you see then is this instead of seeing like one nice round impulse you actually see a doublet two of them together and they're miss shapen it's the cancellation so you can look at those and it's a really nice way of assessing the improvements you've made in your system as you move the speakers you add treatments things like that it's much better than using the waterfalls or the even the steady-state so I mentioned wavelet again I don't know James laughter he's around but I I put this in just to annoy him because I've been literally obsessed with this so this is a really good wavelet that's what a system looks like that has a really well integrated low frequencies in a room that measures fairly well these are not on the same scale I apologize so that one ends at one kilohertz this one ends at 20 kilohertz but you can compare so here's one kilohertz on this one and down there's one kilohertz down on that one you'll see this one doesn't come out past 10 milliseconds this system is full range speakers there's no subwoofer this one has a subwoofer in it look at that where the subwoofer comes in at 80 Hertz the time comes all the way out to 40 milliseconds and then it comes off up like that this system has too much delay coming in as a result of the subwoofer and it's causing poor integration and I picked it up in the wavelet so I like these because to me that's so easy to see there's nothing I just look at that I know that's wrong I gotta fix it all right so filtered in cross responses here's a filter impulse response from that speaker I mentioned that was full range 250 Hertz that looks really good this is actually what it should look like this is 250 Hertz taken on a speaker outside so there's no reflections affecting this one so to see what I was saying actually can't see a radish I guess instead of a ball but you get the point this is a little misshapen pretty good that's what I bet yours are gonna look like this I think this didn't come out of my room this came out of that one but my room doesn't look that much better than this this is what happens when a room that has this floor and ceiling reflection kind of gets its way and you'll see there's kind of two in there it's a little bit misshapen in the in-between area you'll see that like here you'll see it should actually just completely turn into nothing here it's actually pretty large and even starts to enlarge again here that's why I like these it's pretty easy to see the effect and if you want to show somebody or even show yourself the effects of your treatment this is an easier way to visualize it this is 250 but you can pick different yeah you can pick different the reason I pick 250 for this but it's telling me that there's something at that frequency that's wrong acoustically in the room yeah look at every one okay yeah we're not doing that right now cuz we'd be here all day if I do this for myself for a client like I can easily spend eight hours I mean I might spend two hours taking measurements and another four to six hours analyzing it okay so we're gonna look at a case study now I know you guys probably want to see the subwoofer thing I did a separate presentation for the subwoofer integration it's easier for the purposes of illustrating these concepts if we just stick with a bookshelf speaker and go sure so here is a case study so I was sent these Dayton audio MK 402 s for review I stuck him on some stands in my office something kind of like that they were probably about six feet apart six and a half feet from me it's an office system I was you know this is how yours would be but the concept remains the same scale it up right and that's what the response looked like when I measured it so this is the kind of notorious Dayton audio and - high-frequency lift but here's what I was talking about before I mean it's actually bass you can see like pretty solid out to 40 Hertz here 50 Hertz is the point where it's roughly equal with all of this but there's a big dip right there at around 80 Hertz and there's some noise over there and peaks and dips and it doesn't look good at all so we took the speaker outside and we measured you outside this is what speaker characterization looks like this is across the frontal plane of the speaker only the horizontal plane and you can see that the response has that upward tilt I was talking about but interestingly enough it's pretty even like even though the response isn't flat or smooth it doesn't change a lot with angle and that's good and we've seen in the polar pattern that that remains true as well that's what that's a polar map so it's that bad it's that means it's a cubile I can do something with this so I took this I looked at it this is a different piece of software than Roo but I took measurements with Roo to do this and I decided that well first I need to maybe tackle that low-frequency problem so I looked at just that portion of the response and looked at these dips these peaks and decided what I wanted to do to EQ them so Roo has an ability to automatically generate EQ for you nothing wrong with that I would just say go ahead if you've never done a lot of EQ and go ahead and use that I actually don't use it a lot for low frequencies only because it helps keep me from doing what I think Roo can do - a lot of systems and that is if not done correctly and the Routan with the system not set up correctly Roo could potentially apply like 20 EQ filters it would just go to town and the system will look really flat in Roo but it's going to add potentially a lot of group delay it can actually make those if you look at those impulses it can cause the way that they ring over time - actually not they're supposed to decay out really quickly and they won't that's ringing actually and it doesn't actually sound very good you'll do something and you'll you'll get it all set up and you'll measure and it looks good to sit in your primary listening position you know this doesn't sound it sounds like the bass is gone and and then people will tell you oh it's because you didn't put it in a house curve it can actually be because you basically over process the signal so if you can learn to eyeball DS so that you can say that oh yeah it looks like it's about 48 Hertz and the Q looks like it's something like around 2 you can actually avoid that tendency and you get it flat enough I mean our resolution of our ears below 100 Hertz is not even as good as 1/3 octave so you don't need to be that precise but because I know you guys are gonna do this and I certainly do it sometimes too I did go ahead and let Roo Auto calculate some so here you can see how I did that I said a response curve for that speaker and applied the EQ and this was the response after everything was done now the high frequencies are flattened out I should be really careful so you saw that I characterized the speaker ahead of time I did not EQ anything above 500 Hertz based on in room measurements when I applied EQ in order to make that graph you saw there was some manually apply DQ's that was because I in that other piece of software I generated the EQ filters that I wanted to use above 500 Hertz in the system and then I manually put those into Roo so that it wouldn't try to compensate for those when I did the a queuing of the low frequencies yeah should I think they should both be 112 I usually use 112 it's actually finer than your ears can hear most low frequencies so that's what that that's the power response of the system that's what it looks like so this was this is not actually measurements this is what it looked like based on simulation but it's close enough and so you can see that the response still remains relatively even but now it's flatter and that's nice so here you can see a little animation that shows the difference in the low frequencies this is the waterfall here's a member I told you I don't like to use them but I showed them for a reason because I know this is what everybody's used to seeing so I'm showing it to show you how it affected the ringing but here's the thing about this you can see that everywhere where there's big ridges there also was a big peak in the response in the first place so I didn't need to look at the waterfall to know there was gonna be ringing or a ridge I only needed to look at that steady-state and here are filtered impulse responses so notice what EQ did look at how much it knocked out the ringing that was going on so in case anyone's ever told you or you read that EQ has no effect on the dampening or ringing of a system that is absolute proof that it does here they are overlaid again look at the massive difference that green is the old one and cutting out the ringing in the system so the approach that I showed you here makes it look like EQ and measurements are the greatest thing ever and that everybody should go out and do that but I need to tell you that there's some things that I kind of fudged here to make it look better because I was trying to make a point that is that I really only I said I took one but I took a couple of measurements but I only did them in one listening position in an office system if you do it in a theater like this there's gonna be too much variation to get that level of improvement you could make that level of improvement for that seat but it wouldn't apply anywhere else so that my other talk deals with a solution for that acoustic treatments and a multiple sub approach is only the only way you can get rid of that variance so the e queuing approach unless you're using multiple spread out low frequency sources there's only going to be valid in one position and you run the risk it's a very real risk of making it worse in the other positions it can make things worse generally so if you've looked at some of Floyd tools writing he actually did some writing for audio holics that touches upon this he mentions how EQ that was being applied in the heyday before his multi sub approach was developed and sound field management tended to make things worse so I was telling you earlier one of the ways that can happen just applying too much EQ to a system in the wrong way based on faulty measurements can make a system look better and actually sound worse if you use a lot of boost which people tend to do and you don't have to use like you might say oh well I made sure I only used three decibels of boos but if you've got a bunch of filters that actually overlap each other and they all have boost that compounds and that boost goes up you can run the risk of overtax in the amplifier so remember every three D B's requires a doubling of amplifier output so like think about how much power mic system needs now imagine adding six or ten or 20 decibels of boost at low frequencies now his system is extraordinarily efficient and he's got I don't even know how much power over there so it's not a huge issue for him but for a lot of us especially if you're using subs that don't have big giant separate amplifiers you could be talking about taking your subs from its limit already - like so far beyond the limit that you're going to destroy him so you got to be careful you don't do that and then as I said 700 Hertz here you can see there there isn't an exact number 500 Hertz 700 Hertz there's a certain point though where you just shouldn't be applying any more EQ the 700 Hertz was because a company I was working with at the time had said that was their limit so I was trying to be fair to them but I myself actually tend to keep most of my a queuing below a hundred Hertz and just play it just a little bit usually low Q so broad filters above that point and I applied nothing above 5 to 700 Hertz at all and less like I said I fully characterize the speaker outside and based it on that polar response so room treatment as I mentioned one of the problems is the EQ doesn't fix all of the problems that you'll pick up in your measurements and so when you're trying to look at the room and figure out what you're supposed to do with that room treatment does need to be something that you would think about so one of the things that people tend to kind of get wrong is they think that the room is really really lively and echoey and they need to bring their RT 60 down the reality is that they do tend to have some flood reko and problems but most rooms if you took all of these treatments out of Mike's room and you measured this room it would sound echoey but the RT 60 time would probably still average about 500 milliseconds most rooms of this size typically always measure in that range I measured my parents had this room that I thought for sure was going to be like the perfect example of a domestic space with a really like high decay time because it's all wood the floors were wood it's a timber house so it's got these big wood beams it's got tons big glass windows really tall ceilings 500 milliseconds on the on the nose when I measured it and that's because drywall is actually an absorber at low frequencies leather seats are an absorber up to mid-range frequencies carpets as an absorber at high frequencies and this room itself is not very big so combined all that together and you're going to see a relatively low decay time so what you need is actually selective treatment in certain positions to primarily get rid of some of that flutter echo and you need a lot of low frequency absorption so base traps and like I said the walls actually are a base trap so accounting for that or designing the wall to be a better base trap can be one of your techniques so one of the things to keep in mind why I use eq and multiple subs is because short of like two-foot thick I said one foot or more but like massively thick low-frequency velocity absorbers which would be this is called a velocity absorber like pink fluffy insulation that comes out from the wall this far without that on every wall you're not going to observe low frequencies these are not absorbing anything below about 200 Hertz and so if you're trying to fix that you really have to use other techniques and I I can just tell you I have measured a ton of rooms I have applied I mean somebody you saw last year or if you went to Expo no they're the base trap that I built this thing was 24 inches by 24 inches by 36 inches at one point I had two of them that I was using in my tiny office that's like 10 by 12 feet and it made a very small difference in the measurements so adding a lot of base trapping really doesn't do as much as as EQ and multiple substitute so really you're focused on other things I consider them complementary though and like I said in my rooms I design the walls to be based traps I do add velocity based traps and I often add pressure absorbers so here's an example of the measurement before and after I did so this isn't my office so here's what I got to tell you about this because I think this is impressive notice the Schroder curve here that's that black line how that changed an angle notice the differences in the peaks and dips obviously the initial didn't change at all but everything after that did like I said it's changing a little bit but it's not changing a lot here's what you got to know about this there were two treatments on the sidewalls and there was one base absorber that was all I did and that was the significance of the difference in the measurement so I know you're not seeing anything like a frequency response that you could think about and understand what that means but remember that these are not affecting the steady-state very much they're primarily affecting only the stuff that happens over time and that's what this is a measurement of so you can see that it did have a pretty dramatic effect given that it was too small wall absorbers and small base trap so here's an example comparing the art EDT sorry the early decay time before and after treatment so you can see it's lower it's also mostly flatter it isn't here and to be honest that's probably the effects of the treatments are fabric-covered and so they tend to be reflective excuse me at high frequencies so as you can see I used a free tool I mean I did a lot right with that free tool I took a speaker outside I measured it I came up with an EQ keq curve that turned a 50-dollar speaker and the one that had a pretty listenable flat response I figured out what kind of acoustic treatments I needed and I applied them I got rid of the effects of room mode so that the bass response was flat and it really didn't cost me much of anything the the acoustic treatments were something I made from materials I bought at Menards so you can see that this tool is really powerful and you can do a lot with it to make a system sound better so I call it a free tweak because even though the tool itself doesn't cost anything the dramatic improvements you can make to your system are far greater than what is typically true of other I mean I know you guys make fun of this stuff so you know where I'm coming from but people spend big money on new cables cable Isolators stands amp stands speaker stands you name it this makes a massive difference by comparison to those those probably don't even make any difference so I don't review cable saw so the point is this is like this is a real tweak this is something we should be paying a lot of tension now this is something that's worth our energy and it doesn't even cost that much so that's really the whole thing and you

Channel: AV NIRVANA

Views: 79,622

Rating: 4.8379121 out of 5

Keywords: room eq wizard, REW, guide

Id: -A8LjsOx6d0

Length: 41min 57sec (2517 seconds)

Published: Mon Sep 10 2018