Frequency analysis on speakers

  • Thread starter Thread starter Rock Star 87
  • Start date Start date
R

Rock Star 87

New member
is there any cheap, or prefereably free way to know what frequencies are boosted and/or cut in my speakers. I know a CD of test tones would be a start, but how much is a spectrum analyzer? Any suggestions?
 
The long and the short of it

Short answer: No. :cool:

Long answer: Even if you test your speakers in an anechoic chamber, the frequency response/radiating pattern of your speakers varies by relative position; i.e. the freq response of the speakers will be different 3 degrees up and 5 degrees to the left than the will be at dead center and at 0 degrees vertical and 10 degrees right, etc. Even quality near-field monitors will measure different just in the movement of your head 10cm to one side or the other as you sit in front of them.

And that's just in an anechoic chamber. When in your room, you are not only measuring your speakers, you're measuring the room as well. Seperating the coloration from the speakers from the coloration of the room is impossible in such a coupled system. Taking these factors together, measuring a speakers response is an exercise in futility.

The answer you're looking for: It can't be done both cheaply and accurately. The high-resolution spectrum analyzer is easy; these can be gotten from even basic editing software such as Sound Forge and Adobe Audition. The hard - and expensive part - is getting the sound from the speaker to the analyzer. This requires a specially calibrated microphone designed for such testing *and* a clean, colorless signal path from the microphone to the analyzer (which, BTW, is usually an external hardware one and not PC software for the signal path just given.)

G.
 
is there any way to get relatively close, assuming I'm not quite that far into recording. I'm still at my dad's house, but when I get my own place, I'll start calibrating it. Before that, how much would a spectrum analyzer with a specially calibrated mic cost?
 
What's type of room are you recording in?

What kind of gear are you using?

What type of mixing problem/issue are you trying to correct?

Some questions can be answered without an analyzer.

Wish
 
RockStar,

A professional spectrum analyzer with matching calibrated microphone and pink noise generator can cost as much as a couple of thousand dollars.

On the other hand, a spectrum analyzer program for a PC can be had for fifteen bucks. Add a $39 Radio Shack microphone and you can have the whole deal for under $60.

The problem is the results for the $60 package will be absolutely useless. The Even Bigger Problem is the results from the three grand pro model will be virtually just as useless.

I'd advise you that you are simply and erroneously barking up the wrong tree when you start asking about measuring the freq response of your speakers. Don't even worry about that.

First and foremost, let your ears do the measuring, skip the analyzer. Take your favorite CDs and play them on your speakers. Pay careful attention not to the music, but to how the speakers sound. Then listen to the same CDs on a couple of other typical playback systems where you expect most of the listening to be done (is it a home stereo, a car stereo or an iPod?, etc.). Again, listen carefully to how they sound different from your monitors. This will giive you a "real world" idea of what your monitors sound like and give you a good base as to how to record and mix your music. You'll know that if it sounds like A in your monitors, it will sond like B in the real world. That's all you need to know. Frequency response is just a number that's not worth remembering.

Second, if your looking to make these measurements with an eye towards using an EQ to "calibrate" the response of your speakers, forget that idea too. That is an old idea which has long since fallen out of fashon and been largely discredited. EQ can add more phase distortion and such than the coloration that it takes out. The problem is the induced distortion cannot be erased or mixed for the way the natural coloration of the speakers can. Just knowing from listening that you speakers may, for example, have a crossover peak at around 250Hz as compared to other systems, gives you the ammunition to know what to expect. If you hear a bit of a peak in your music at 250Hz, you'll know that it's your speakers and not the mix that causing that peak, and you'll know therefore not to adjust for it.

Third, if your looking to use the analyzer to tune your room, you're in over your head. That is something that requires expert training to do with an freq analyzer and a microphone. Yes you can make treatments to your room to make it sound better (see the forum on desiging and building studios for help on that), but making any sense out of a freq analyzer's readings is futile if you don't have the knowledge and training on how to properly interpret those readings. I've been at this game for about 25 years now, and I still wouldn't even pretend to know how to best tune a room with such a setup.

Sorry to sound like such a bummer on your picnic, Rock, but I'm sure there are more than a handful of pro engineers on this board who'd back me up on each point. Just trying to save you some money and travel time on a dead end road is all.

HTH,

G.
 
SouthSIDE Glen said:
First and foremost, let your ears do the measuring, skip the analyzer. Take your favorite CDs and play them on your speakers. Pay careful attention not to the music, but to how the speakers sound. Then listen to the same CDs on a couple of other typical playback systems where you expect most of the listening to be done (is it a home stereo, a car stereo or an iPod?, etc.). Again, listen carefully to how they sound different from your monitors. This will giive you a "real world" idea of what your monitors sound like and give you a good base as to how to record and mix your music. You'll know that if it sounds like A in your monitors, it will sond like B in the real world. That's all you need to know. Frequency response is just a number that's not worth remembering.
Click to expand...


Wow. Nice explanation.
 
in response to wish14, it's not really a problem I'm attempting to correct so much as trying to get a feel for new speakers.

For glen, I've been doing some of that, and it sounds like a relatively good idea, I just have to put more trust in my ears. We all know that mixing through headphones is a bad idea, so I finally got speakers that work, and sound good to boot. I have a feeling that these speakers have a natural bass boost though, so I've been boosting the treble on my board. The other good point is not to eq the speakers. I see the point that numbers aren't so much what matters.

Tonight I'm going to pick a few songs that I know very well, and I'll set the speakers to that. thank you everybody for posting, and helping me to realize the best analyzer, my ears.
 
Rock Star said:
I have a feeling that these speakers have a natural bass boost though, so I've been boosting the treble on my board.
Click to expand...
Rock, it sounds like you have a pretty good handle on the situation and are off to a good start.

I'm not sure just what you meant by the above quote, as to whether you thought you were wrongly or rightly boosting the treble because of a possible bass boost in your speakers, but just this last bit of advice, if I may presume...

If, as an example, you have indeed ID'd a particular bass peak in your monitors, boosting the treble in your mix to "compensate" is exactly 180 degrees the *wrong* thing to do. If you think about it, what that means is that when you move to a playback system that does not have that bass peak, not only will the bass sound weaker because you haven't adjusted that at all, but the hi freqs ("treble") will be artificially boosted. You'll wind up with a mix that does not have enough bass and has too much high end.

The key to remember is that you need to mix to make your recordings sound good "in the real world", not necessarily to make them sound good on your monitors. If you know your monitors have a peak somewhere in the bass, then you'll know that a mix that sounds bass-heavy (at that peak frequency, anyway) on your speakers will sound "just right" on speakers that don't have that peak in them.

It's like trying to adjust the colors on your TV screen while wearing colored sunglasses. If you're wearing rose-colored sunglasses, you don't want to adjust the picture so that it looks right with your sunglasses on; if you do, then to everybody else who is not wearing the same sunglasses, the picture will be lacking in the reds and heavy on the greens. Once you know that a perfect picture looks a little rosy-tinted through your sunglasses, then you'l know to adjust the picture so that it looks rosy-colored. It may not look right through your sunglasses, but to everybody else it will look great.

Make sense?

G.
 
i wasn't turning the treble up in my mixes, rather my mixer.
 
Rock Star 87 said:
i wasn't turning the treble up in my mixes, rather my mixer.
Click to expand...

Like the others said, I don't think this is the best way to go about it. Ideally, you should just leave the EQ flat and learn to adjust the mix for the bass boost you are hearing. If the bass response is just overbearing, you could cut the bass a little at your mixer (though you'll still have to learn how your mixes will translate with this adjustment). Boosting the treble to compensate to too much bass just seems that it would be adding to the problem, rather than correcting it.
 
Does anyone use those little shelving EQs on the back of some nearfields? JBL even has a fairly extensive set of EQs on the back of one of their more recent nearfield models. Then there's the Alesis digital DSP models that allow EQ - I passed those by (purchase wise) but gave a listen anyway...I didn't listen to Dynaudio Air models cause I don't have the $$$ they have some EQ for example also...

I might actually begin to adjust mine a little since I have brought my nearfield triangle in closer (moved to a smaller studio space) and I don't monitor as loud as I used to. I'm still learning the sound of the room and my references in the new place though...

As far as measuring stuff goes (if you want to go there) I use ETF software and a cheap Behringer omni measuring mic to give me a picute of the performance of the speaker/room combination. Besides Frequency and spectrum analyzers theres also speaker damping (among various models) and room mode exitation which can affect your RT60 (reverb) times. ETF software has a waterfall spectrum that shows that kind of stuff. Some people like to see stuff like that on graphs and charts - other people just walk into a room, hum and clap for a few minutes and go "move the couch over there, hang Auralex here, and put the speakers there...".

There's a lot more to it (I just mess around with it myself, semi-pro guy here) and if you follow Ethan Winer around and get connections from the studio building website here and other places you can pick up some things. Mostly it costs a lot of time and experimentation to get the best plan - like all things audio...listening to the sounds in the room, listening to the sounds of audio folks in the forums.
 
You must log in or register to reply here.
Back
Top