How to equalize your headphones: A Tutorial - Page 2

Yes. This will, of course, also work with hardware parametric equalizers. The same technique can be applied, only you will be playing with real knobs and sliders instead of software knobs and sliders. In fact, this may even be preferable because then the response would be applied to all outgoing audio data, not just audio played through the media player.

Certainly. I will add them to the OP as well.

The settings I use for my STAX Lambdas are:

Mode: Linear Phase
Peak type: S-plane type II

(I always use the same mode and peak type, BTW)



-8.5dB at 7400Hz with a BW of 0.750
-9.7db at 13100Hz with a BW of 0.950

Note that this response will probably work with any Lambda series headphone, from the original Lambda all the way to the SR-404, because they all have the same shape and hence, the same transducer-to-eardrum distance.

The settings I use for my ER4S are: (These settings are not final, I am still working on them.)



-2.5dB at 2500Hz with a BW of 1.500 (In this case, the 2500 peak is actually built in to the headphones and is not due to resonance. Its actual amplitude is 5dB. I find that not taking the peak completely out sounds better than removing all of it.)

-5dB at 6000Hz with a BW of 0.850 (Notice that the resonance is lower with the ER4S than with my full size headphones. I thought this was a little odd, but I have verified with sine waves and pink noise that it is indeed 6000Hz. I think the notch may need to go a little deeper, though.)

-6dB at 12550Hz with a BW of 0.950 (The ER4S has just a little too much high-frequency boost. This helps to make it sounds more normal.)

Sometimes I will add a small bass boost on certain songs. I use an S-plane type II peak at 100hz at 2dB with a BW of 6.00. Both of the above responses, while greatly different from each other, give nearly the same perceived response with both sine waves and pink noise.

But that's just the problem. Even if the headphone had absolutely equal output at every frequency, there would still be resonance peaks for anyone who used it. That is why we need proper equalization.

Yes, the blue screen with the white curve is from Cool Edit Pro 2.0, one of the best audio editors ever released. The picture is the log sweep sine wave in viewed in spectral mode.

I really wanted to include a link to the Cool Edit software. I figured it would be OK since it is no longer on the market, but I'm really not sure what Head-Fi's rules on that are, so I left it out.

As for the soundcard issue, no, I did not take that into account. To be honest, it did not occur to me that one soundcard could have such a drastically different frequency response from another that it would make a perceptible difference.

I just noticed something you said that I overlooked before:

Funny you should say that, because that is exactly what I used it for! My job involves recording sermons for churches and editing their content (doing things like removing mistakes, pops and clicks, noise, etc) and mastering them for eventual broadcast on radio stations. For the longest time, I only used Cool Edit Pro 2.0. I only upgraded to Adobe Audition 3 when I read about the Healing Brush tool, which I now view as utterly indispensable.

I just found a freeware program that could prove exceptionally useful for equalizing your headphones. It is called "SineGen" and it can produce a sine wave that is smoothly sweepable throughout the entire frequency range. Using this program, it should be exceptionally easy to detect where there are response peaks in your headphones! I will have to update my tutorial to reflect this discovery.

http://www.softplatz.com/Soft/Audio-Multimedia/Other/SineGen.html

It is too bad the master control of the program isn't properly interfaced yet. If it was, you could set up a macro that would play a selected group of frequencies automatically, with each frequency having its own amplitude setting. This would make it exceptionally easy to quickly discover your personal headphone response, completely bypassing the first part pink noise process. (Which, while valid, is a little tedious.) The pink noise will still be needed when actually equalizing, because you will need to hear all frequencies at once so that you can have something to reference the problem frequency to. In the meantime, the program will still be exceptionally useful and can replace the current sine wave technique.