[hyperreal]

The whole point of the process is that you can't base anything off of the headroom head-mic because of its inherent non-linearities. The goal of my equalization was to attempt to smooth out the peaks/nulls throughout frequency sweep, meaning that to my ears the frequency sweep keeps the steadiest "perceived" level throughout the range. I'm still not quite sure what pure, flat-response pink noise is supposed to sound like, so that didn't factor in as heavily when creating the curve. I still think there is room for improvement on this curve and once I get as close to linear response as I care to go I will then attempt to find a curve that sounds "good" to my ears. Who knows? Maybe I might end up thinking that a near-linear response sounds best to my ears, we'll see.

 

[Ashirgo]

Surely, you will do as you wish, but I would not dismiss HR graphs so easily. What is more, I would not trust line sweeps, for they do not tell the truth, as does pink noise. For instance, it is unlikely to hear ear resonances with line sweeps, but it is explicit with pink noise.

Listening to it, you can direct your attention to following things: no freq. range should seem prominent or distinctive, and the overall balance should slightly emphasize lower frequencies.

 

[PiccoloNamek]

HR graphs have always been a little questionable, and furthermore, they represent the actual response of the headphone, not the perceived response, which is what is important.

Sine wave sweeps are actually better for determining peaks or dips than pink noise because is easier to hear a change in volume when sweeping through a range of frequencies than it is while listening to all frequencies at once.

Pink noise is more useful during the actual equalization process, when you already know where the problem frequencies are and can then equalize them and listen to their volume relative to other frequencies. The goal is to make every frequency sound equally loud and have a perceived uniform response to sine waves.

 

[Ashirgo]

Quote:

Originally Posted by PiccoloNamek /img/forum/go_quote.gif Sine wave sweeps are actually better for determining peaks or dips than pink noise because is easier to hear a change in volume when sweeping through a range of frequencies than it is while listening to all frequencies at once.

That is right, these are great for doing that, but the only way I can hear resonances or if any freq. dominates others (in a form of peak) is by listening to pink noise.

Oh, and I little doubt if we really should seek "perceived" response so badly, since our hearing is more sensitive to say 2000-3000 Hz range regardless to our equipment and environment. If we achieved a perfectly flat response of equipment and room, the perceived response would still peak in that register, even if the measured was flat. I believe that is just a natural order of things.

 

[fufula]

I'm a newbie when it comes to this stuff, kudos for the guide, it really helped me find a few peaks here and there, but I have a question.

Around 4.1kHz the volume in left headphone seems to be going more and more quiet as I go up. Then the right headphone goes quiet as well. A little later (still going up slowly), when both headphones are really quiet, I can clearly hear that the volume is getting louder in the right headphone until it gets back to normal, and then the left one goes back to normal as well and all is fine again at around 4.6kHz. There's a huge peak after this.

I know it's not in my head 'cause I've asked for other people's opinion without even hinting what's wrong and they've noticed it as well. Are the headphones borked or is it something else?

 

[PiccoloNamek]

Quote:

Originally Posted by Ashirgo /img/forum/go_quote.gif That is right, these are great for doing that, but the only way I can hear resonances or if any freq. dominates others (in a form of peak) is by listening to pink noise.

Well, maybe that is just you then. I don't see what is so difficult about sweeping the sine wave by hand and hearing where the volume peaks at, though. With my headphones, if I start the tone at 1kHz and begin moving the slider up, the response is nearly flat until around 6kHz, where it gradually starts to become louder and louder, until it reaches maximum volume around 7.4-7.5kHz. Then it goes back down again. So now I know that is where the resonance is. Not only where it is, but roughly how wide it is as well. I can also determine its amplitude by comparing it to other frequencies that I know do not suffer from the same problem.

Quote:

Oh, and I little doubt if we really should seek "perceived" response so badly, since our hearing is more sensitive to say 2000-3000 Hz range regardless to our equipment and environment. If we achieved a perfectly flat response of equipment and room, the perceived response would still peak in that register, even if the measured was flat. I believe that is just a natural order of things.

Our hearing is more sensitive in the 2000-4000hz (approx) because of the effect the pinnae have on the sound. With headphones, this effect is nearly eliminated. Anyway, it wouldn't be an issue to equalize your speakers to a response that would be perceived to be flat. You could use the same methods described here and end up with a response that actually sounded totally flat (even if in reality it wasn't).

 

[leeperry]

Quote:

Originally Posted by fufula /img/forum/go_quote.gif I can clearly hear that the volume is getting louder in the right headphone until it gets back to normal, and then the left one goes back to normal as well and all is fine again at around 4.6kHz. There's a huge peak after this. I know it's not in my head 'cause I've asked for other people's opinion without even hinting what's wrong and they've noticed it as well. Are the headphones borked or is it something else?

apparently, all the phones suffer from FR imbalance...even the most expensive ones.
if there's a really HUGE difference, maybe you could RMA them...or use different EQ on each driver, a stereo EQ is always a compromise

 

[CompressionalFlagellation]

What's George Yhong's VST wrapper for Foobar? I read a description on a couple sites but i still don't understand. i we tod it.

 

[CompressionalFlagellation]

tucows download for sinegen 2.1 seems broken -- sinegen 2.5 available at softpedia

 

[PiccoloNamek]

George's VST wrapper allows you to use VST plugins with Foobar. VST plugins are usually special kinds of effects and tools you can use in audio production programs like FL Studio or Cubase, but the wrapper allows you to use them in real-time in Foobar.

 

[Ashirgo]

Oh, I meant only the resonance and large range hills (say 4000Hz to 9000 Hz), do not mind my half conscious writing

To elaborate, if I (I, and it seems true only for me for now) want to eliminate the resonance, I listen to line sweeps and equalize it as long as I get a "flat" response. Then, going to a pink noise, I still hear a distinctive "action" and "stir" of these frequencies around 7300 Hz and equalising them down while listening to that noise is more helpful.

 

[hyperreal]

I went through the process for the Shure SCL-4 and it was a much more simple process than the HD-280s.

Here are the results:

Attachment 15756

SCL-4s are somewhat bass shy so I put a gentle rising curve down to about 23Hz leveling off finally around 130Hz or so. They're VERY flat in the midrange until a slight peak in the 8.2KHz range then they fall off gradually after about 9KHz until about 13-14KHz, so I tried as well as I could to compensate for that. I decided not to include specifics for the peaks this time around. The only one you need to worry about is the one at around 8.2KHz, and I used a 0.4 octave peak 7.5 dB down at exactly 8.2KHz. Here ya go.

 

[Ashirgo]

.

 

[K3cT]

Is it just me or does that Electri-Q PMEQ introduce some distortion in the treble region? I've been comparing the sound with other EQs and it seems that only Electri-Q has this problem.

 

[PiccoloNamek]

What mode are you in? I've never heard anything out of place.