We Built it, We've measured it, HELP US TWEAK IT - acoustics of the BLACK CAT - Page 2

It is a sizeable investment in time and money to purchase adequate test gear, pay for the course, keep proficient, etc., etc. There is an enormous amount of information provided in the class. It takes weeks or months to sort out what was presented to begin to fully grasp the material. As with anything, anyone worth their salt continues to grow beyond what they learned in the class. The best video calibrators are this way as are the best audio calibrators. That requires a lot of time and dedication. More than many people may be willing to give. In addition, some folks look at the training as a hard sell to their clients. It is much more profitable (they think) to sell boxes than try to learn this "hokey magic" stuff, and explain it to a client. They aren't sure how it fits into their business model. When in fact, it should be the central core of their business model.

REW is very easy to use and view. The following video will walk you through the set up and how to view the graphs.
http://www.gikacoustics.com/video_re..._tutorial.html

No the advice is "know your limitations and the limitations of the equipment"

Dennis is the pro, and he knows his limits and the limits of his equipment. I don't know mine.

In the words of Arthur C. Clarke, “The limits of the possible can only be defined by going beyond them into the impossible.”

I don't mean to belittle Dennis's advice, as his opinion and perspective is invaluable to me - I've bookmarked a number of his posts on this and related subjects - but is this a hobby you enjoy? Did you set out to do this because you figured that you could just cobble it together and it would be perfect? Did you set out to do this because you wanted a terrific learning experience and a great home theater in the bargain? Do you hope to have the best theater someone else can build for you or do you want the best theater you can build?

That's just the other side of the story, from my perspective - clearly this is Morph1c's project - his money and his theater.

Hey all,

First off I want to sincerely thank everyone's contribution to this thread so far. There doesn't seem to be many threads on measuring rooms on this forum (a quick search for titles with "measur" in them only found 7 hits, 2 of which are related to room acoustics, with this one being the only one where someone posted results)

Next, I want to state that my theater in its current form already sounds amazingly good to me, hands down the best I've ever owned and perhaps the best I've ever heard. I could button it up today, but I want to know what little things can be done to make it even better.

As Dennis noted earlier, there's a limit to what can be reasonably changed now. The following are seemingly fixed: the room's layout, bass traps in the corners, the furring for the fabric panels, the carpet, the fabric chairs, and the 4 brand new Triad silver surrounds. But some things are still controllable - for instance, placement of a 2nd subwoofer, replace some linacoustic with batting, place 6-mil plastic or scrim over some linacoustic, move front sub around behind the screen, put an extra inch of linacoustic behind the screen, etc. There's also a list of things that I really rather not change now, but could if I had to, such as the amp (Marantz SR9200), the LCR speakers (M&K 850s), and the sub (M&K MX-150).

Regarding a calibration specialist, the closest Level-II professional to me said that I was a bit out of his driving range for a day trip. I wrote back saying that I could put him up for the night, but he never replied. I was hoping to get an estimate from him, but he never provided one. That said, I have two issues going this route: 1) my amp isn't capable of implementing equalization (presumably a key aspect of the service) and 2) it's a difficult to justify >$1K given I already think the room sounds great.

This brings me to conclude that I should continue taking some measurements, tweak the few obvious things mentioned above, and be happy. If nothing else, I'll at least learn more about it all.

I want to thank BIG for signing up to take this journey with me. I wasn't sure if he would want to or prefer to just wait until I'm ready to put up the GOM fabric. Well, he showed up this morning with printouts on how to get better measurements and more knowledge on using the Omnimic software. We spent a couple hours measuring stuff and learned quite few more things about the room (I'll let him report the findings).

One question: we don't currently have any of the molding up (see pics in BIG's first posting to this thread). The molding (5-8" base, 4" chair, and 7" crown) are going to all be hardwood (Cherry). Since the molding is not up, the sound waves are hitting the DD+GG walls instead. How significant of an impact will the molding have? Is drywall any more absorptive / less reflective to some frequencies? If so, which frequencies might be affected and how?

Thanks again!

It seems plausible that 7" crown at a 45-degree angle to the wall and ceiling might make a measurable difference, since it presents a different angle of incidence.

I'm sure that when measured with sufficient detail, you'd find those materials are different. I don't expect that you will hear or measure a significant different (if any at all). Even if I did hear a difference, it wouldn't change my plan for decor.

Just in general (bearing in mind that I am a total beginner when it comes to this, but I am certified to teach high school physics - which means very little, directly - but I did stay at a Holiday Inn Express last night ) - keep in mind that reflected sound is going to have to come directly to your ears for you to hear it with enough gain to be disturbing to imaging. If it reflects off two or more surfaces, it's probably both sufficiently delayed and sufficiently damped to be a non-issue. So, if the chair rail is below your ears, the sound is on it's way down and will continue down, hitting your chair or the floor - or on its way up if the rail is above your ears (assuming the speaker is at ear level). Snell's law acts in 3-dimensional space, you know? Judging by your pictures, the chair rail is largely irrelevant - especially for the second row and for the surround channels. I suspect the LCR are high enough and the chair rail is low enough that it won't matter.

Aside: Further, and to some of Dennis's points - We don't know what spectral content is likely to reflect off the chair rail in the first place, as we don't know what the off-axis performance (or the on-axis performance for that matter) of your speakers is. We don't know if your speakers are toed-in or not (are they - this is free optimization! check out this pdf explanation for when and why to toe-in). Additionally, we don't know the ability of your mic to pick up the difference in the first place. Please don't let me mislead you into thinking I'm trying to talk you out of measuring this and making changes as needed, I'm the last one in this thread to do that , but this is one of those limitations Dennis was referring to where you can chase non-existent problems down rabbit-holes.

You teach High School Physics? Where in Atlanta?

I don't currently - I left teaching to be a mechanic (in Buckhead). When I taught, I was at Northview High (Fulton County Schools - John's Creek). I think, technically, my certification has expired.

In an effort to give you a target to keep in mind I want to post some information I was lead to at another forum. A paper was prepared by Nyal Mellor and Jeff Hedback (of Acoustic Frontiers and HdAcoustics, respectively) to help guide the hobbyist is measuring and designing high quality home listening rooms - "Acoustical Measurement Standards for Stereo Listening Rooms" (2011). This was done with a 2-channel audience in mind, but there are plenty of similarities, IMO. I have a pdf of the paper, but I don't know how to make the whole thing available (or if I have the license to actually).

Right at the top, they summarize all their recommendations. These goals or standards will help to keep all the various aspects of acoustics in mind. Many of the standards have to do with room construction, so obviously now is not the time for those... anyway, on to the standards!

Summary of acoustical standards
Criteria Measurement Target Acceptable Range*
A: Noise Control RC

• RC20
• RC30

B: Reflected Sound Energy Time Curve, 0‐40ms

• L & R speakers visually identical
• ~10dB reduction in energy level by 40ms
• Clear decrease in energy over the 40ms
• Peaks smooth in pattern and density

C: Low Frequency Decay Times

• Resonances from 35Hz 300Hz should not extend beyond 350ms before decaying into the noise floor or reaching a level of 40dB.
• Below 35Hz this standard is relaxed to 450ms.

D: Midrange Decay Times T20, T30, T60

• T60 between 0.2 and 0.5s
• T20, T30 deviation <25% across one‐third octave bands

E: Midrange Frequency Response

• L speaker within +/‐3dB at 1/3rd octave
• R speaker within +/‐3dB at 1/3rd octave
• No deviation greater than 3dB between L/R speakers

F: Low Frequency Response

• Within +/‐10dB at 1/24th octave
• Within +/‐5dB at 1/3rd octave

G. Room Size and Construction

• Physical dimensions: Size of room 224sq ft to 475 sq ft, Volume of room 1,750 cu ft to 4,750 cu ft
• Dimensions lack common divisors
• (2x) layers of 5/8 gypsum board
• Surfaces are constructed using similar methods

* Not all measurements have a lower level performance target defined.
** All measurements should be carried out for each speaker separately except the low frequency measurements which should be carried out with both speakers playing together. It is important to ensure that the measurement levels are consistent when comparing left and right speakers otherwise erroneous conclusions may be drawn from examination of the data.

Ignoring the LCR situation for now, let's look at the bass.

I don't know what smoothing may be applied to the response graph, but I'm guessing it's not more than 1/24th octave smoothing. If that's true, I think you've made the standards I cited. I imagine you can hear the peaks if you play a sine wave sweep, but can you hear them with normal content? I bet you're OK. Thumbs Up!

But what's my opinion worth? How does it sound (feel )?

The caveats: You've only got one measuring location here; thorough optimization should include all the seating locations. I have a hunch that with subs only along the front wall, you're not doing anything to mitigate the modal response of the length of the room and the mic was probably near the center of the room, where that resonance has a node (null) accounting for the 31-32Hz dip. I think that locating a sub near the midpoint of the side wall would help that. Alternatively, nudging the seats away from the center of the room may help, if it's troublesome with normal content.

(don't let me monopolize the conversation - feel free to tell me to shut up)

Don't use white noise! Use pink noise!

You'll get misleading information with white noise. Geeze, those omniMic guys.
Each octave going up the scale has twice the number of frequencies. Pink noise is equal energy per octave vs equal energy per frequency.

Imagine a pen with two cows mooing (first octave). The next pen has four cows mooing (second octave). With white noise the second pen (second octave) will be twice as loud as the first. With pink noise the each cow in the second pen will moo half as loud as those in the first pen...each pen will be equally loud. If you're measuring FR vs SPL, you must use pink noise.

Yeah, but we also noted that it may not matter due to the phenomenon happening below the center-channel's frequency range - that is, the second peak is around 60 Hz and gone by 70 Hz. From the near-field measurement, it looks like the center-channel isn't kicking in until around 100 Hz...

Is that the normal thing to do? I've seen others focus on the low-frequencies first, do you know why that is?


I believe it was 1/12th


Here might be my weakest point, I don't know what great bass is suppose to sound like. The sine wave felt smooth, but it didn't rattle my bones either.

A couple days ago I configured my amp to know about my speaker's distances and evened out the SPLs to 75 db, including the bass. The amp also had a "bass peak" test, which I took up to -12 db, that is when I felt like I started hearing some distortion. Since then, I've watch two movies and played some samples from the Reference Blu-Ray Demo Disc
. In all this time, I haven't yet been tempted to crank the volume or the bass - that is, reference listening volume sounds pretty good! It's a little strange for me, really, as at the beginning of a movie, I feel that maybe it's a little too quiet, but then I get lost in how clear everything sounds and forget all about it.

I think I should listen to the "bass" segment of the Blu-Ray Demo Disc more carefully, walk around the room and what not...



Your comments have been some of the most helpful so far - please DON'T shut up!

Can we get a quick run-down of the processors and loudspeakers that are connected?

I'm not very confident that the speaker measurements you made are very close to reality. I don't know how omnimic does it's processing to generate the graphs, but I think it's getting a lot of reflection and room interaction in the measured response. I imagine that the speaker was measured installed in the false wall, to the side of the screen. It looks also like the measurements were taken at 75dB (according to the dB scale in the posted graphs). If both of these things is true, omnimic would have to be doing some pretty fancy gating or other filtering to avoid measuring the room response - so if we know what model of speakers you've got, we may be able to find measurements from a review or other online user and avoid having to make careful first-hand measurements. Your own measurements would be ideal, as then we could confirm that the speakers are in good condition as well (it happens that assemblers wire tweeters backwards or install bad capacitors, and listening casually, you may never realize). Your best bet for making your own measurements would be to drag the amp and the speakers outside - set the speaker up on a stand (a ladder should work fine, or maybe a handrail on a deck) and get everything away from the speaker so there are no nearby reflective surfaces.

One of the reasons that I am dubious about your response plots is the substantial variation below 500Hz. Things like baffle step compensation and the proximity of the side walls will really change the response in that range - this is the range of frequencies where room boundaries can dominate the response and positioning is the best recourse for correcting those interactions. I realize that you've built everything and set it up with the L/R outside the image. Did you guys install an AT screen with the center behind, or is it below? Reworking that would be a significant pain, I am sure, but that may be the only real option for smoothing the low end - don't do that on my suggestion! We need a second opinion - based on evidence and experience (neither of which I have on my own). (This is me reaching out of my comfort-zone of knowledge)

Some thoughts about your description of the sound:

It looks like (at least for the one position you've measured) you've got enough dB for bass, with the possible exception of the 32Hz null, which I still think is position related. So the other thing to optimize in terms of bass response is the reverb time, which I think looks reasonable. Descriptions of good bass are hard to make clear - bass drums should kick hard, but not sustain, explosions should really get your attention but not overwhelm other sound effects. Can you compare your experienced sound levels to some reported in the movies with bass thread (some of them are on the demo disc, I think)? If you have no idea how that should be done, let's talk about it. We should be able to assess the headroom in your system and see what it really produces relative to reference.

You've calibrated speaker output to reference (75dB at seating position is produced by 75dB signal), right? Most rooms (I find anecdotally, reading here) sound harsh at reference level. Generally, this is because their system distorts when pushed that hard (another reason we need to know some specs on your speakers and amps). If you find it sounds clean at those levels, that great news, but I think it may have something to so with the missing high end from your speakers.

I read the other day that your AVR doesn't have any kind of auto-EQ, and I don't think it has any manual adjustments either, so if that icing on the cake won't exist, we'll do what we can with positioning and treatments.

(Sorry this is so much to read at once. We should perhaps focus on one aspect at a time - you chose.)

Much of what the treatments do cannot be measured. However, some of it can. IF you really want to know what the treatments have done, you need to remove them all, get a baseline, then methodically bring in treatment to see what effect it has had on the response. The bottom line is, you will not be able to fix the remaining items in the FR (provided it was measured correctly) without a parametric EQ. Provided the positioning was correct to begin with (that's one of the biggies of the plan in the first place), positioning at this point likely won't help you very much. If they are currently mispositioned, then it can assist in some things. I deal with the low frequencies last. Deal with the L/R first. You have to remember why surround sound was created in the first place. It all begins with the L/R speakers. Then moves on from that point.

Reference level is only harsh IF the room is not calibrated, poorly laid out and not treated properly. In a properly designed and calibrated room, reference level is quite invigorating and send that little chill up your spine resulting in one big smile! Cheers!

Absolutely

• the three fronts are M&K LCR-850 (same as the LCR-750, but with a wood veneer)
• all four surrounds are Triad OnWall Silver Surrounds
• the one and only bass is a M&K MX-150
• all connected to a Marantz SR9200

I'm not sure either, but I'm working on getting REW set up - just posted my first post on the REW forum. I think using REW will help a lot - not only is it more well known, but I've read that it's more tuned for room-measurements (i.e. its namesake), whereas the OmniMic software is more tuned for speaker calibration, which I think is why it's missing some features found in REW...

One thing that might be relevant is that all the measurements were made using the center-channel only and it, the center-channel, was behind my Seymour Center Stage XD screen - the drivers are about 9" behind the screen. I figured this would be OK, as it might only affect the highest frequencies, but I read that AT are mostly for <400Hz :dunno:


I *think* so - my SPL meter measured 75dB when the pink noise was running (C weighting, slow response). That said, it seems my amp is set at -12dB, though I haven't touched it since I calibrated it (set speaker distances and loudness). I'll double-check in a little bit...


Thanks for the suggestion! I will do this, at least for the center channel. I'll try to superimpose the graphs of it being behind the screen, same location but with screen removed, and outside. Given that I haven't ordered the mic or sound card, it might take a week or two...


The center channel is behind the screen, the left/right channels are not - is this a red-flag? Still, I'm not following you, what is your suggestion? - to put all 3 behind the screen? Regarding significant pain, did you catch the part where we opened up the ceiling to install a plenum to reduce air-turbulance noise caused by the HVAC (see here)? - *that* was a royal pain, but it paid off, now we can hardly tell when the HVAC is running


You mean the position of the sub, right? It's off center behind the screen - kind of at the 1/3-rd position with it's front driver facing sideways - 90-degrees from facing the screen.

BTW, took my wife out to lunch today and we discussed the theater (she was pretty happy about how The Ghost Protocol sounded last night). I tried to explain why the delay for putting fabric up - that there was an opportunity to make some small improvements to make it sound even better, such as, blah, blah, blah, put another subwoofer on a sidewall to flatten out the nodal frequencies, blah, blah, blah - she didn't even blink! Of course, she probably doesn't realize how much subs cost, but it doesn't matter now - that she didn't bother to push back means its got WAF With that, I spent a little time today checking on if I could find another MX-150 anywhere - there are none on Ebay or anywhere else. Any idea if its OK to use mismatched subs?


Holy smoke, that thread is 131 pages! I read the 1st page and the last three pages. Mostly movie-recommendations, I'll have to come back to it later...


Yes, the Reference Blu-Ray Demo Disk has a "bass" section with the following:


I'll give it a special listen in a little bit...


I may be wrong about it being at reference level. As mentioned above, it seems the amp is at -12dB (not 0dB). That said, when we played BIG's OmniMic test-tones disk, my SPL meter measured 75dB pink-noise coming from the center channel. I haven't touched the volume level since, so I *assume* that it's still at that setting


I think I'm going to start calling you "HelpfulFred"

For reference: a review of the 850 Series was posted at hometheater dot com and includes the following image.

The purple trace is the LCR-850 with 1 watt of power: "All passive loudspeakers were measured at a distance of 1 meter with a 2.83-volt input and scaled for display purposes." You can ignore the other traces as they correspond to other speakers in the series.

Quoted from review, "The LCR850's listening-window response (a five-point average of axial and +/-15-degree horizontal and vertical responses) measures +3.64/-1.26 decibels from 200 hertz to 10 kilohertz. The -3dB point is at 96 Hz, and the -6dB point is at 80 Hz. Impedance reaches a minimum of 3.49 ohms at 230 Hz and a phase angle of -50.43 degrees at 129 Hz. Sensitivity averages 88.5 dB from 500 Hz to 2 kHz."

I don't know how to reconcile the image with the text. Maybe I'm reading this wrong, but that graph looks to me like 1 watt at 80Hz generates a 74dB response. I have trouble understanding how that's -6dB from 88.5dB - I suppose I'm not well versed in reading these specs.

Regardless, my hunch from this is two-fold: first, a higher than standard crossover may be in order; second - these things will need a whole lot of power for 105dB peaks at 3m distances. I haven't done any math to support that claim.

Also, note that the omnimic measurement differs wildly from this above 10KHz.

I'm a little confused by your first sentence. For instance, if a room is measurably too "dead", then can't it can also be measured to be better after replacing absorbers with diffusers? Is this the "some" part that can be measured?


I'm actually not interested in knowing what's changed from baseline, so much as what can be done to make it better. Or are you saying that knowing how to make it better can't be done without having a baseline?


Fair enough but, assuming the measurements are correct (we'll know in a week or two when we get REW up), you see that there are some issues with the FR? what are they? are they major? do you feel that a parametric EQ can correct them, or just make them better? Given the difficultly I'm having finding a Level-II nearby, would you think that Audyssey would get close enough? (yikes! I didn't mention needing a new receiver to my wife - better take her out for dinner )


I'm not following you, what does "positioning" refer to and why wouldn't repositioning help very much now?


Interesting! When you say L/R, do you really mean L/C/R? - note, this room is strictly for movie-watching (no two-channel music)


So either my room is awesome or else I'm not listening at reference levels after all. I'm guessing it's the latter

Re: positioning.^^

I'm pretty sure Shawn is referring to placement of speakers.

For giggles, I transcribed the data from the hometheater.com review onto the center channel pink noise (1 meter on axis) graph from above. It's better than I thought, but done by hand on a Friday night, it may not be terribly accurate - no that it needs to be, ultimately. Not sure what to make of the differences. REW may prove more illuminating... ?

I ran across this in another thread, but if you are still getting a strong reflection off the ceiling, you may be able to angle your LCR down a bit to help as well. It sounds like they have a pretty strong off-axis response. You might also try raising them off the floor and angling them towards the listening position which increases the distance to the listening position to help even out the seat to seat response. However, raising them may worsen your reflection problem.

That roll off is either:
1. Bad tweeters;
2. your measurement kit is broken;
3. screwed Xovers in the speakers.