Actually, I'm not sure anyone will be able to answer this, but I'll give it a shot anyways.

I'm hearing a song that sounds very chaotic and washed out, because it's a poor quality recording and it wasn't eq-ed properly.

On speakers or with headphones on, it's hard to tell some sounds apart, they all sound so blended into eachother. If I take the headphones off and put them around my neck, I can hear everything very clearly and evenly leveled out. Obviousely because the high frequencies travel to my ears while the lower ones die out.

Knowing this, which frequency range suffered the most loss from taking off the headphones? In other words, I want to know in which frequency range there's the biggest noticable difference between listening to the sound with the headphones on and listening to the same sound with the headphones off.

64hz, 256hz, 512hz, 1024hz, 3500hz, 5000hz or 8000hz?

Seems to me that the obvious answer would be 64hz, but that doesn't seem to be it. The 64hz frequencies aren't really that noticable to begin with. So it's neither of the ones I mentioned, but rather something in between.

It's low frequencies, definitely. Depending on what style of music you're listening to, this may or may not be helpful:
Lowest "A" on a guitar is 110 Hz.
Lowest "E" on a guitar is 82.41 Hz.
Lowest "A" on a bass is 55 Hz.
Lowest "E" on a bass is 41.2 Hz.



This might be more useful to you, if you have a full-length keyboard to compare to. There's probably one particular note or set of notes that's noticeably louder than the others. (Hopefully, it's not the bass drum.) If you can match in on a keyboard or some other instrument (or maybe in Fruityloops or something), you can get a good idea of what frequency range to turn down, and maybe what slope(s) to set it at. Anyway...
Middle C is 261.63 Hz. This is the 4th octave. I'll call it "C4."
C3, then, is 130.81 Hz.
C2 is 65.41 Hz.
C1 is 32.70 Hz.

I can't really make a suggestion of what to turn down without actually having heard the song, but I'd guess that you turn down everything below (and including) 256 Hz. If you can get more EQ bands, do it, and shape it like a parabola (to start out) with the lowest frequency band completely pulled out, and play around with it from there until you get something satisfactory.

Or if you have Audacity, try running a highpass filter around Middle C (261.63 Hz). Again, this is just my guess without actually hearing it.

Or, you can send it to me and let me tinker with it.

Also keep in mind that changing the overall EQ will change the tone quality of pretty much everything in the recording (although probably not drastically in this case).

That was quite helpful actually.

This would mean the range I'm looking for is somewhere between 256hz and 512hz.

I typically do the parabola thing with the sliders, but in this case it's just one particular frequency flooding the rest. I'll probably have to bring it down about 1/4th or so and see what it does.

I'll have to try that when my ears stop ringing! :/

This is my EQ bible - http://www.recordingwebsite.com/rwtip/archive/rw15r.html . Get to know it. Love it. Print it out, and staple it to your forehead. Don't ever lose it.

Also, notes do not increase in a linear fashion, but in a parabolic one. There is a wider frequency range between two notes in a high octave than the same two notes in a lower octave. Like A increases like this: 55, 110, 220, 440, 880, etc. I can't seem to remember the formula, but it seems like it had something to do with the square root of 12. Meh.

And I love you Shintock. *kiss*

>.>
<.<

*awkward moment*

*runs*

Aside from that, I've found that using a one-octave Q when equalizing with those frequencies usually gives the best results. Try different types of EQ-ing too, for the 50hz for bass, I use a hi-pass instead of peak. The key is to experiment and to learn what sounds right with your stuff.

Yeah, but not all instruments have the exact same curves. I know they always use the example of the piano, where an actual piano will generate a wide slope of frequencies in a bunch of different ranges, but a synthetised piano will generate a much more limited array of frequencies, so it's curves are sharper. If you cut all of the real piano's predominant band, you still hear the note, but as a kind of echo. If you do the same with the artificial piano, it kills almost everything.

Oh and shintock, that rules!

The EQ bible or the dog? :p

Both! :p

I think what you're referring to is tone quality. The dominant frequency of any note is what determines the note. There are also harmonics that occur when a note is played on a real instrument that give it it's "shape." This is why a certain song may sound better on one instrument than another, or why a certain key on the same instrument may sound better than another. A single stringed instrument can offer different tone qualities. The high open E string on a guitar has more high frequencies than the same exact note & octave played on a larger string farther down the fretboard. There, it will have less on the high end and a more "round" sound.

The big thing with synthetic sounds is that you don't have all the frequencies that occur with a real instrument, which lends it to not sounding as full or whatever. When you EQ a certain frequency out (like the dominant frequency) on any non-synthetic instrument, you are still left with all the frequencies that make up the tone quality of that note, resulting in the echo you were talking about. With synthesizers, you don't have as many of these frequencies, so it gets changed a lot more in EQing.

Here's a graph I whipped up showing the curve I was talking about. Try to imagine it as a single line.