This isn't a very thorough test, and I don't really have test equipment beyond the mic, so bear with me.

The Process

My signal chain was White Noise Generator -> Digi001 -> M-Audio SP-5B -> (room) -> ECM8000 -> Digi001 -> Recorder.

The White Noise Generator & Recorder being Cool Edit Pro. I don't know how to turn off monitoring of what's being recorded, so there was probably some low level feedback, but I don't think it skewed my results. If anyone knows how to turn this off, let me know.

I took out my Radio Shack SPL meter and raised the volume until the meter read A Weighted 70db. I chose 70db because it was bearable to listen to. Other volumes would've revealed somewhat different results.

Then I stuck my mic around the listening position, perhaps a few feet in front of it, and hit play/record. Here's a chart of the results of 30 seconds of white noise.

The numberes on the right are arbitrary, I manually centered it around 0db  just so it would be easy to read, but I think it's actually centered on 3db. Remember, according to the Radio Shack meter, the sound in the room was around 70db.

Looks Good at a Distance...

So this frequency response curve looks pretty good, fairly linear. Now let's zoom in and see what's really going on.

Okay, the swings are very much exaggerated visually because I zoomed in vertically but not horizontally, but this tells a very different story. It's important to note how graphs can be used to lie, or at least cover up the truth. The first graph looked fairly linear, and at first glance I would be impressed with the results, but on closer look the scale is so small that the tiniest movements turn out to be really big movements.

This graph is a composite of about 30 second of white noise that basically kept all of the peaks. Notice that broad dip around 200hz leading into that hump centered around 700hz, and then that dip around 1500hz leading into the rather jagged upper frequency range with a definate trailing off the higher you go.

So what does it all mean?

The jaggedness is caused by the measurement method, see that "Linear View" checkbox at the top? There are two ways of measuring frequency response. One involves giving each octave or doubling in frequency equal strength, the other gives every 100 hz equal strength. I chose each octave, so as the scale become smaller in the upper frequencies. Clicking the Linear checkbox would've made the bottom even intervals, much more like a ruler.

Notice the space between 100hz and 200hz. Each time that much space is taken up on the bottom, the number doubles. So if it's an inch between 100 and 200, then it's in inch between 200 and 400 and 400 and 800, and on the other end, between 10,000 and 20,000. Equal weight is given to each octave, not to each set of 100hz.

I'm ignoring that sharp rise around 100hz as probably an anomoly of the equipment, but now that I think about it, my sine-wave sweep (which should be my next measurement) did reveal a strong buildup around 110hz according to my ears & my SPL meter... Is it the room or the M-Audio's? Something tells me it's the room & a big standing wave problem. I also have a flutter echo problem in the untreated area of the room right near the speakers.

Update April 28, 2003: My recent article Bad Acoustics In My Living Room showed me that that sharp rise around 100hz was actually a serious room mode. Take a listen for yourself.

My girlfriend and I are arguing about the merits of heavy drapes. Acoustically they'll help deaden the critical wall behind the speakers & a large untreated area of room that currently has a strong flutter echo, but she thinks heavy drape are passe. I told her they didn't have to be dark, but I don't think she's going for it.

And is that good?

The Event 20/20bas monitors are rated as "50hz to 20khz +/-3db, ref 500hz" meaning that from 50hz to 20khz the response is within 6db, with 500hz being the center of the curve. How does my room measure up? On the chart it goes from that spike to +10 down to let's say -5. That's quite a range. Though let's visualize much smoothed out lines that run though the center of all the jaggedness though. In this case the peak is around +7 or +8 from around 500 to around 1,000hz.with a dip to around +1 around 2500hz. Then it rises up to +4 or +5 for a while before beginning it's descent.

So we can say that between around 110hz and about 12,000hz, the range is 6 or 7 db (+/-3), but if you went up to 20, then the range would be more like 10 or 12 db. (+/-5).

How much of this is room & how much is monitor, I can't tell. I would've stuck the microphone right next to the speaker, but I got terrible feedback any time I tried. I'll try again once I learn how to turn off the record monitor mode or I find a different frequency analysis program with white noise generator.

The Sound of Silence

I also recorded "silence." Well, my room is far from silent, the computer CPU fan is very noisy and needs to be replaced with a more quiet one before any recording with a mic takes place. Here's the graph, again, zoomed in.

Looks like Pink Noise with a big spike around 15-16k. I didn't notice anything interrupting the experiment to cause such a big spike, so I have to assume it's the computer fan that has a specific frequency at which it causes a lot of noise. Actually, if I move my head a bit so that the couch doesn't block the computer from my line of site, I can hear the high pitched whine. I just took out my acoustic guitar, it's an "A" and when I look at my computer's motherboard monitoring software, I notice that the CPU fan speed is around 4600 - 4800 rpm. 4800 divided by 60 seconds is 80 cycles per second, which is closer to "E." Maybe I'm hearing a harmonic of it, or a different fan... I'm looking forward to buying quieter fans.

Remember, this graph is centered so that 3db is about 70, so simply subtract (technically, add) any number on the righthand from 67 and you get the actual number. So that spike is 67 - 10 = 57db... damn loud. The low end noise around 200hz is also about that loud. When the computer store clerk told me I had purchased one of the loudest fans on the market, he wasn't kidding me! Time to replace that thing. I'll take more measurements after that. The thing is extremely noisy and whenever I'm not using the computer, I turn the thing off so I can get some quiet!

The Final Analysis

Well, monitoring in my living room has it's flaws, as does buying the $300 monitors rather than the $800 or $1,200 ones. I can't treat the room for acoustical problems in any way that I want, and the speakers themselves have a less that perfect response to boot. Though the graph that came with my M-Audio's, and the graph on the M-Audio website is a more linear than my graph, especially at higher frequencies, so the room must be at least partially to blame here. Remember, the monitor really can't be evaluated apart from the room. The room plays a tremendous role in the sound of your system.

The whole idea behind near-field monitoring is that you eliminate or reduce room anomolies by being so close to the speakers that any reflected sound is miniscule compared to the direct sound you're receiving. If, like many a studio, my monitors were at arms length, the room would be less important, especially as I turned up the volume of the monitors & if I deadened the room somewhat more than I already have. However, I'm using my monitors for mid-field monitoring, which I really shouldn't be, but I'm fairly used to the sound of them at any place in the room, so I don't think it makes a tremendous difference where I sit, my brain will compensate.

I was mostly interested in testing the whole package of room & equipment for monitoring. Of course, I should be a lot closer to the monitors than I am, I'm all the way across the room because that's where's most comfortable at the moment. These are the tradeoffs of building a home studio. Unless you have some space that you can dedicate to the task of building a studio, and you know what you're doing, you'll have to live with comprimises. I know what I'm doing, but I don't have a dedicated space. Hopefully, one day I will and I can address each and every acoustical problem until I have a great sounding space.

Discuss this article in the Forum.