[medium jim]

JohnR:

I have never professed an expertise in anything, especially not audio.  I spent a lot of time setting up my present system and find it to do everything I want from it.  The walls disappear effortlessly into a beautiful soundstage.  Notes rise and decay as they should (to my ears), there is clarity, coherence, detail, both macro and micro.  No suck outs or peaks and it takes me to a happy place when I need it. 

I was reminded that this is a forum, yet it appears to be anything but it at times and that is sad.  I'm sorry that I have caused so much controversy for expressing honest (to me) opinions.  Too much spy v spy!

Yes, the cost to measure is a non-sequitor, and once again, when I move this system into a different environment, I will invest in measuring technology (as I have said many times before).  It would have saved me many hours of trial and error!

Jim

[medium jim]

Not trying to beat a dead horse or rile anyone, but here's a thread were another user measured his system, had a suckout at 125hz, got some great advice and all was well in audiophilium:

http://www.audiocircle.com/index.php?topic=98500.20

Again, this is not intended to be confrontational or inflammatory, just a fact.

Jim

[Ethan Winer]

With a swept-sine program, you get a result that is not only more accurate but (as Ethan notes) you get it much more quickly. Each time you move the sub, you just press a button (or two). Before you know it, you have a dozen locations done and can pick the one most likely to give the best result.

Exactly. It's not whether one should choose measuring over listening, or vice versa. Rather, the purpose of measuring is to more quickly improve the sound for when you listen. I've seen people spend literally years tweaking speaker placements etc, when suitable software would have gotten them to their nirvana in an hour or two.

--Ethan

[medium jim]

Exactly. It's not whether one should choose measuring over listening, or vice versa. Rather, the purpose of measuring is to more quickly improve the sound for when you listen. I've seen people spend literally years tweaking speaker placements etc, when suitable software would have gotten them to their nirvana in an hour or two.

--Ethan

Ethan, I agree with that....while it didn't take years, it did take a couple of months.   Funny thing, those with planars often do spend years fussing with them to get them just right.

Jim

[jackman]

Hi Guys,

Interesting thread.   I'm not going to say that measurements are more important than actual listening.  If it sounds good to you and measures poorly (or not flat), more power to you.  This isn't about living up to someone else's definition of good sound.  In my experience, there is no way I could have dialed in my subs as good as they are today without the ability to measure with a microphone and program.  The process of measuring was pretty simple, just align the mic, push the button and wait for the results.  The next step required adjusting the DCX and measuring again but when you see the results, immediately, that part is pretty straightforward and easy. 

In the end, you can always go back to the original setup if the flat (or flatter) response is not to your liking.  There is no way I could have acheived the end result without measurements.  My system is full range but if it only went down to 40 or 50hz, I would not have worried about measurements as much.  Dealing with deep bass is a hassle without the ability to measure and EQ, IME.  Other people may be able to achieve great sound without measurements but I am not blessed with this ability. 

[rollo]

  With room setup and or bass setup measurements are accurate and less time consuming. Funny thing was the subs were pretty good as is by ear. However after adjustment from measurements better yet. I'm in both camps.
    Don't laugh but positioning the subs was done by crawling around the room to listen for strong bass output then placing the sub there. My Wife was laughing her arse off. But it worked.



charles

[David C]

interesting thread to follow ...... same line of a different question (rookie here so be kind). How do I set up a measurement, which software (cheaper is better) and how do I get the audio signal from a laptop into the system. My pre amp only has RCA inputs?

David

[Davey]

Of course the original query is somewhat ambiguous and rhetorical, yes? 

If a person is being serious about the process, there is no way to design a speaker system and/or set up a system in a room without some form of objective measurement.  Your audiophile ego might persuade you to believe that measurement "by ear" exclusively is possible, but that's incorrect. 

Cheers,

Dave.

[jackman]

interesting thread to follow ...... same line of a different question (rookie here so be kind). How do I set up a measurement, which software (cheaper is better) and how do I get the audio signal from a laptop into the system. My pre amp only has RCA inputs?

David

If you have a laptop, you can get an Omnimic from Partsexpress.com.  It has a test CD and microphone.  You run the CD on your laptop and plug the mic into your USB.  It plots out the in room response on a graph for you to see.  You can position your speakers to see if that helps but will need an EQ to adjust your system.  I only use this on my bass/subs.  Once you have things dialed in, you don't need the mic anymore, unless you want to mess with things again or make a change to the system.

There are lots of other options.   PE has a new microphone setup that is very cheap, Dayton Audio UMM-6 is only $88.  This combined with Room EQ Wizard freee software and your laptop will allow you to measure your system.  It is useless if you don't have a way to adjust your system (active crossover or EQ) and I don't like EQ's on the mains, just the subs.  I run my subs active, using a Behringer DCX, through the second set of RCA outputs in my preamp.  If you only have one set, you can get a splitter and run one set into an EQ for your bass.  I'm sure there are other ways but this is the one I'm most famliar with.  There are also some pro amps with built in xover for subs but I'm afraid I can't help with those.  I use a pro amp (Crown K1) with the Behringer and it's more power than I can ever imagine needing in my room.  The Crown runs cool and is very quiet and powerful. 

Good luck,

J

[andy_c]

Let me add a bit to what Ethan was saying about 1/3-octave smoothing.  Here's a couple of examples of real-world measurements of my own system at the listening position.  The speakers are JBL LSR6332, which roll off pretty quickly below about 60Hz.  They are designed to be used with a subwoofer, so that's why the response does not go very deep.  Below is the response graph from 40Hz to 200Hz with 1/3-octave smoothing, 5dB per division.



Other than a moderate downward tilt with decreasing frequency, it looks pretty decent above 70Hz, right?  Below is the same measurement without smoothing.



Hmm, that is not so good.  At 124Hz, there's a dip of about 10dB that's almost completely obscured by the 1/3-octave smoothing.  By the way, 124Hz is the frequency at which the distance from the woofer to the back wall is 1/4 of a wavelength.

Before I did these measurements, I had planned on just using two subs and crossing them over to the mains at about 60Hz.  But the system is clearly in need of help above 85Hz, and I would have never been able to quantify this if I hadn't done the measurements.  I decided instead to do a four-sub configuration, something along the lines of the Geddes technique, operating at least some of the subs up to about 160Hz to try to fill in some of the dips.

Room EQ Wizard has a sine wave generator feature that I tried out to get a subjective impression of the difference in loudness between the dip at 124Hz and its return to the roughly correct level at 119Hz.  The generator can be set so its frequency tracks the frequency of the cursor on the active graph.  I moved the cursor from the dip at 124Hz to the normal level at 119Hz.  The perceived level difference was dramatic.

1/3-octave smoothing is kind of like looking at the measurements through rose-colored glasses.

[medium jim]

See post #61, it contains a link to a thread where another AC member measured his system via 1/3rd octaves and found a suckout at 125hz...go figure.

Jim

[JohnR]

Yes, if it's bad enough you will see it with a smoothed/third-octave measurement.

Nyal Mellor and Jeff Hedback recommend using both 1/3 and 1/24 smoothing for examining/assessing the FR in the bass region. It's in section F of their report:

http://blog.acousticfrontiers.com/whats-new/2011/10/13/acoustic-measurement-standards-for-stereo-listening-rooms-pu.html

The other sections provide examples of analyses that can be done with e.g. REW but not with pink noise or test tones.

[andy_c]

See post #61, it contains a link to a thread where another AC member measured his system via 1/3rd octaves and found a suckout at 125hz...go figure.

Having the woofer radiating surface 27 inches from the wall must be pretty common .

What I hope to do is have the subs back against the wall so the quarter-wave notch frequency for the subs is above the cutoff frequency of the subs' low-pass filter.  Then the subs will fill in the quarter-wave dip of the mains, and the quarter-wave dip of the subs themselves won't enter into the overall response (because it will be filtered out).

[Soundminded]

I've never had much luck correlating speaker measurements with the way they sound. It seems to me the problem is very complex, there are so many variables. One is how much output is harmonic distortion. For really deep bass you need speakers capable of producing loud low distortion sound at high levels. Another is speaker placement. Even a few inches can alter perceived response substantially. Also listening position, very great variations over small distances at very low frequencies. Another is the room cutoff frequency, most home listening rooms are not large enough to acoustically reinforce deep bass. Another is program material which varies in spectral balance all over the place. Perceived spectral balance is also affected by how loud music is played. The Fletcher Munsen curves not only show how insensitive hearing is at the frequency extremes at low level but that relatively small changes in actual loudness at those frequencies can have substantial perceived effect.  As a result, I'm forced to conclude that the only way to get the bass to sound balanced is to adjust it for each recording.

Most hi fi systems seem to me to have entirely inadequate bass compared to live music. I've also noticed that as an upgrade to their $22,000 a pair Revel Salon Ultima 2, they've added $20,000 18" subwoofers to their product lineup. If you buy 2 that's $40,000 to produce bass for speakers half that price. But if you take Floyd Toole's advice you'll need 4 so a $22,000 speaker now becomes a $100,000 speaker.  That seems like an awful lot of money to me for speakers. But then I'm not an audiophile. 

[andy_c]

Most hi fi systems seem to me to have entirely inadequate bass compared to live music. I've also noticed that as an upgrade to their $22,000 a pair Revel Salon Ultima 2, they've added $20,000 18" subwoofers to their product lineup. If you buy 2 that's $40,000 to produce bass for speakers half that price. But if you take Floyd Toole's advice you'll need 4 so a $22,000 speaker now becomes a $100,000 speaker.  That seems like an awful lot of money to me for speakers. But then I'm not an audiophile. 

With a combination of Geddes (or similar) multi-sub technique, DIY subs, pro amps and miniDSP, one can get the job done at a reasonable cost.

[medium jim]

Here's what 124hz is all about, "Rhodococcus; Sexual dysfunction (Isochronic Tones 124 Hz) Pure Series ."...enjoy!

http://www.youtube.com/watch?v=_R60xM2DqP0

Jim

[medium jim]

Or maybe 124hz Binaural Beat Gamma Brainwaves is your thing!

http://www.youtube.com/watch?v=ZYaZvzZ0fos

Have a nice day!

Jim

[kevin360]

Of course the original query is somewhat ambiguous and rhetorical, yes? 

If a person is being serious about the process, there is no way to design a speaker system and/or set up a system in a room without some form of objective measurement.  Your audiophile ego might persuade you to believe that measurement "by ear" exclusively is possible, but that's incorrect. 

Cheers,

Dave.

Indeed, Dave, the original question is rather ambiguous. I 'measured' my speakers before I purchased them – to be sure they'd fit in the room I was planning to build. I 'measured' that room in a similar way – to find workable dimensions with relatively benign calculated modal behavior. In fact, I pretty much painted myself into a corner with all of the preliminary measurements and the dimensions/layout which arose from them. Before adding the first piece of gear to the room (before there was a room to which to add gear), I attempted to address as many potential problems as my wallet would allow – that's being serious about the process, isn't it?

I suspect your assessment of our audiophile egos is, at least, on the dart board. I might temper the statement as Ethan did, although I wouldn't say that it only takes an hour or two – there is a learning curve. After a bit of diddling around (as time permitted in the last month, or so), I'm getting a feel for RoomEQWizard, but I haven't done anything meaningful yet. Of course, to move things about, I'm going to need longer leashes on each speaker/sub.

In my case, I want to measure because I'm curious. Things are pretty darn great right now – after a good bit of experimentation. I really have no 'audiophile' ego to bruise, but I do think I have fairly well developed 'ears' – started playing music at an early age. I could hear my bass problems in that 'by the numbers' designed room. In practice, the room didn't seem quite as well behaved as it did on paper.

Right or wrong, my approach has been to keep adding bass treatments – by the numbers, as it were. To redress the booming/overhanging notes and suck-outs with EQ, I'd need more specific information, but being in the ballpark is sufficient for the trap method. Likewise, It was painfully obvious (without the quantifying measurements) that my initial corner treatment was a disaster – at least, for planars/dipoles.

At the end of the day, the system just releases music into the room. I find it hard to drag myself away from it, and the experimentation has been fun too. Those are both positive things in my life. Likewise, it has been entertaining, thus far, to dabble with the measurement stuff. This may sound silly, but my initial 'graphs' were too clean – I've figured out how to get 'real' values. Now I want to see the impact of various treatments that are easily removable/relocatable/tunable.

The problem, however, is that every time I fire up the system, I get lost in music. Although it is entertaining and enlightening to do the measurements (and I am quite curious about it), listening to music is simply more enjoyable.

To measure or not to measure – there is no sine qua non in this hobby, except to enjoy!

[andy_c]

To redress the booming/overhanging notes and suck-outs with EQ, I'd need more specific information, but being in the ballpark is sufficient for the trap method. Likewise, It was painfully obvious (without the quantifying measurements) that my initial corner treatment was a disaster – at least, for planars/dipoles.

That brings up some really important points.  First, suckouts should almost never be addressed with EQ.  The reason for this comes from a theoretical concept originated by Bode (of "Bode plot" fame).  This concept is known as "minimum-phase" behavior.  One result of Bode's work is that if one takes a minimum-phase system with frequency response errors, and corrects those errors with EQ that is also minimum-phase, the transient response problems associated with those frequency response errors are also fixed up as a beneficial side effect.

But this gets complicated.  Real-world audio systems in the bass region behave as minimum-phase systems only in certain frequency bands.  Suckouts are almost always non-minimum-phase, which means trying to EQ them out will almost always do more harm than good.

So this brings up the question of how to determine the frequency ranges for which the system behavior is minimum-phase.  The REW author, John Mulcahy, has found an innovative way to do this, which is described in the REW manual.  If anybody's interested in this, I can go into more detail tomorrow, as it's getting late here.

[JohnR]

I'm interested I had assumed that reflections would generally make the system non-minimum phase, but this sounds worth looking into