Edited by Rico613 - 6/29/15 at 6:33pm
This link worked for me: http://download.freewarefiles.com/files/SineGen.zip
I'm having some trouble understanding parts of this guide. I think I understand how SineGen is used, that is: finding peaks, adding different amplitudes of that peak frequency, and finally comparing them to some reference frequency (@ 0dB) to 'calculate' the 'correct offset' for the peaks. Once you've established this, transfer these offsets onto an EQ program.
Ok, so if the above is correct, I've done that. However, I don't understand how the pink noise file is used in the audio player to check your EQ adjustments. I can understand how I might use SineGen to test the EQ offsets, by sweeping the wave past the adjusted peaks, however I don't see how a static pink noise audio file can be used to check the peaks. Can anyone fill me in here?
FYI, I'm using:
Also, is the correct way to adjust Electri-Q in foobar to go Preferences>Playback>DSP Manager>Electri-Q (as an active DSP)>Configure selected? As shown below:
And how do I know which preset is active in Electri-Q? Is it the last preset I had active before closing the window? Do I need to keep the Electri-Q window open?
Just discovered this thread, so I thought that I should present some information.
The idea of equalizing headphones is good since every ear is different. I have been equalizing my headphones for quite some time now, but I actually do it by analog filters between the amplifier and the headphone. This is done in a tiny box, which I can carry with me.
There are some things that should mentions regarding equalizing headphones.
First we must know a little about how headphones are measured.
The frequency response are measured at the eardrum on an artificial head. That frequency response shall not be flat as it includes the outer ear & ear canal response. This is known as head related transfer response (HRTF). The problem is that HRTF is not only one response. It is a set of responses dependent on what angle the sound arrives at the ear. So what HRTF shall be used/is used?
The main result of the HRTF is that we hear treble better (louder) when it comes directly into our ears than then it comes from ahead of us.
When the measurement of the headphone (at eardrum) is done, the chosen HRTF response are subtracted. We then should end up with a flat response if we have an ideal headphone, or ...
Well, it is now it starts to get complicated.
The HRTF response is individual for each one of us, so the measured response will only be correct for the "average" man/women.
With headphones we get the sound into our ears at a different angle than we do then listening at loudspeakers, and because of this we get too much treble at our eardrum because of HRTF. I have not seen that this has been compensated for in measurements, and this might be the reason what Headroom recommends that the frequency response above 500Hz should roll off some 3-5dB and that from 1000Hz to 20kHz there should be a gradual roll off to about -10dB at 20kHz.
When listening with headphones we do not get cross feed, and this makes the sound contain even more treble (than compared to listening on loudspeakers). This can be fixed with cross feed in an headphone amplifier, but the cross feed shall not have a flat frequency response because of the HRTF. The cross feed also needs to add a delay corresponding to appr . 15-17cm to be comparable to listening to speakers.
This extra delay introduces some cancellations (1000-1250Hz, 3000-3400Hz, ..) and amplification (2000-2300Hz, 4000-4500Hz, ..) of certain frequencies when listening to speakers, which is used for mastering the recording. So when listening to this recording over headphones with no cross feed it will have too much energy at 1000Hz and 3000Hz, and so on...
Another thing is how should a headphone like the Sennheiser HD800 be measured, as the diaphragm is angled towards the ear...
When it comes to the resonance, which happens between the eardrum and the headphone, the measurements will show this, but only for the "average" head. So the response on your ears will be different.
So to the equalizing.
To equalize the headphone "by ear" is very difficult. The reason is the loudness of different frequencies is perceived different at different volumes. This is know as the loudness contour curves. Basically if you tune the headphone frequency by ear, you will only get it right for one specific volume, and that is the volume you run the test tones/pink noise at. But there is nothing that can be called "correct" when tuning the response. If the sound engineer doing the recording/mix is doing so at 90dB with loudspeakers with flat response, the same song will sound thin when played over the same speakers at 70-80dB due to the loudness counter.
So you should tune the response according to your preferred listening level.
An example: I have a pair of Sennheiser PX200, and after reading Linkwitz pages I did the EQ. by ear ending up with a notch reducing 2,5kHz about 2dB and 6,5kHz 7dB, together with increase of bass and high treble. For long I thought this sounded very good. But after reading measurements of the PX200, I changed the filters to only reduce an area around 1kHz, increasing high treble and bass/lower midrange, and this setting was a lot better! The PX200 has a very nasty peak/drop in the 7-9kHz reagion, and I found that this problem could be reduce a lot by modifying the headphone. The trick was to add a 7mm felt pad on the plastic protection cover in front of the membrane. This felt pad actually reduces the resonances between the headphone and the eardrum and flattens out the treble response.
Headphones having foam pads tend do have less resonance because the foam itself reduces the resonances. The Sennheiser HD238 & 239 are headphones that for me sound really good right out of the box, and they have foam pads. But this by no means always like this.
What would I recommend.
I would recommend taking the measurements as a starting point.
Then I would try to find out if the resonances shown in the measurements corresponds to my hearing, or are they lower or higher in frequency in my ears.
Then I would make an EQ, which flattens out the peaks according to my center frequency of the resonances, but using the measurements when deciding how much to reduce and how broad it should be.
When this is done I would EQ the bass (in most cases that means increase it) based on the measurements to get it more flat.
Then I would EQ the midrange/treble so that the frequency response starts to "drop" at 500Hz, -2dB 1KHz, -4dB at 10kHz, -6dB at 20kHz. This last EQ you should try different fall off curves. It might be that you need more/less roll off.
Also if you use the headphones outdoor, on buss/train/boat/plane..., the fall off should be higher since the headphone otherwise will sound thin under those circumstances. This is because there is so much noise that the frequencies below 500Hz are masked by the noise. My PXC250 noise canceling headphones sound good in silent surroundings, but thin when I'm in a plane or in a car.
Hope that this can help you getting better sound of your headphones, and that it was not too complicated.
Not sure about sinegen, but you could try this,
and for pink noise:
Also check out the rest of the site, there are some really cool tones there
very useful links to many head-fiers. excellent resources to check even HPs and earphones! thx for pointing out.on behalf of Head-fi community.
I felt compelled to join so I could say this:
Wow. Just wow.
Thank you. Songs i've long gotten sick of now bring a smile to my face again. I can almost double the volume and still stay in the realm of comfort. In general, everything now has rich, very satisfying bass. Very bass heavy songs, like trance, positively tickle the ears. Why, I stayed up till 3 in the morning last night because I couldn't stop listening to it!
I mean, who knew the bass guitar was supposed to be audible as an individual instrument at all times? I always thought it was just there to add depth to the sound, even if you couldn't pick it out. Silly me!
It wasn't until a couple months after I got the album Who are You that I noticed most of the songs had an acoustic guitar in them. Where did that come from? I thought. Why was it there? You had to strain to identify it, much less hear what it was playing. In fact, the whole album was a chore to listen to. The melodies weren't enough of a payoff on their own, and listening intently enough to pick out how all the accompaniments interacted was exhausting. I wondered why Townshend had added so many instruments to his songs that they ended up a muddled soup of sound.
I also wondered why the distinctive Queen vocal harmonies sounded the way they did, almost like a single "voice" (even though no human voice can sound like that) with multiple overtones of indefinite pitch. If I didn't know better, I would have said the vocals sounded like some kind of strange instrument they somehow manipulated into saying words, maybe through a talkbox. I always assumed this was a result of the multitracking technology they used. They did after all break records with the sheer number of voice samples they put into their songs.
Well, last night I followed your eq guide and boy, don't even get me started. No wonder Who are You was hard to listen to! Pete Townshend's bloody uber distorted electric guitar was drowning everything else out! With eq, I could actually hear everything, I could count the instruments, and it sounded like what I had only caught glimpses of in the past with intense concentration: all the instruments united in a complex tapestry of interwoven parts. Additionally, Queen now sounds more like actual people singing in harmony instead of, whatever that sound before was.
In fact, a whole bunch of songs are less annoying now, whether due to shrill guitars or shrill voices. A lot of Queen songs, especially later ones as Freddie's voice declined (hey, why don't YOU try geting AIDS and vocal nodules and remaining one of the world's best singers for 20 years!), are kind of hard on the ears, as it's quite obvious he's straining to sing in the same range he did when the band started, but it's just too high for him. At least now his voice is less strained and piercing and more natural.
My buds now sound like tin cans without the EQ.
Anywho, enough with my latherings of praise for this guide. Let's get down to business.
I'm using a pair of Sleek Audio SA-1s with the more bass-heavy tuning tips.
Now I don't want to give the impression that these headphones are shrill screaming abominations because they're not. Since I really got into listening to music with high-quality headphones I've had a pair of Sony XBA-1s, Shure se215s, a pair of Skullcandy something-or-others that I didn't care what were called before I threw the box away and now can't find the name of, and these.
The first song I listened to on the XBA-1s was Stairway to Heaven, and it brought tears to my eyes. Everything was just so beautiful. (I'm assuming that's a standard reaction to a first encounter with good audio equipment.) I really liked those. Everything was...crisp. Overtones were positively buzzy. Anyways, due to utterly retarded cable design, where the left ear cable went directly into the combined cable, while the right ear cable was about three times as long and was supposed to go around the back of the neck, which meant every time anything jerked on the headphone cable, 100% of the stress was transferred to the left earbud alone, the solder connecting the left cable to the left earbud wore out and eventually shorted. That's why I'm only buying headphones with detachable cables from now on.
The se215s had much better bass response and were more balanced, but felt less... "alive". I don't know if that's because the magic of newly experiencing good music wore off or because the dynamic drivers were less precise than the Sonys' balanced armature (I actually bought the se215s under the mistaken impression that they had BAs as well...LOL), though I have a sneaking suspicion it was the latter, as I noticed the difference when I bought the things and only discovered what drivers they truly had after they were already dead and gone and I was shopping for a replacement. Anyways, those could have lasted me 10 years, but then my family got a new dog, who, while I was asleep, literally on the first night, grabbed them off the nightstand, shredded the cable, swallowed one of them whole, and cracked open the other one like a walnut, as well as destroying my brand new laptop's power adapter (why aren't BOTH cables removable?! whyyyyyyyyyyyy?!) and a USB cable charging my phone. It was a tragedy.
I got the Skullcandies as a Christmas present from a family member who didn't know that I already had a pair of better headphones, but I didn't mind. They made good backups (SPOILER) in case something happened to my se215s. These did turn out to be screaming abominations. Don't get me wrong. I was surprised by how good they sounded at first based on past experiences with cheap headphones. (My broke self, skeptical of how headphones could possibly be worth hundreds of dollars, bought a pair of cheap no-brand headphones at a department store on a whim and upon listening to them immediately threw them away in disgust. Now to be fair, since these were a present, I have no idea how much they cost. They might have been pretty solidly midrange.) The bass and clarity were acceptable. I even questioned the notion of shelling out a Benjamin to get another good pair to replace the se215s when I could just keep using these. But what got to me was the noise. Have you ever been to a concert that was so loud that (paradoxically) you couldn't hear the music because the resonance/attenuation/interference in your ears turned it all into noise? It was kinda like that. Plus they were very uncomfortable.
Because of the XBA-1s, I wanted another pair of BA headphones to replace the se215s, but the cheapest ones were the se315s, a little out of my price range when combined with the remote control cable, mandatory because the included cable could only go over-the-ears, which interfered with my glasses. I considered buying another pair of se215s but honestly I was sick of them and wanted to try something else. I didn't like how they were upside-down, or how they inserted at a weird angle, or how they were designed to be held up by the over-the-ear cable, so people who don't wear them like that had to constantly fiddle with them, or how they fill the entire earlobe, or how the plastic seams and injection molds were at the extreme edge of the casing, so that those little sharp bits were guaranteed to scrape the earlobe during aforementioned fiddling, until they gradually wore down over the months against an increasingly calloused earlobe.
Making the se215s transparent was a bad move, as it displays to all the world how tiny the drivers are, how much empty space is behind them, how small and comfortable you could have made the headphones, and how you opted to put them in what appears to be exactly the same case as the se315s, which have much larger drivers and seem to be what the case was designed for, perhaps to save money by using the same mold for another product, instead of something less cumbersome.
Honestly, the only justification for the use of this case for drivers this small I could come up with was that it was somewhat more comfortable to sleep on your side with. Unfortunately, though, the weird angle of insertion, when put under pressure, say, from a head resting on a pillow, just happened to drive the tip of the earbud into the side of the ear canal and close it off, effectively muting it. (At least for me it did. Maybe I have weird ears.)
Anyways, the only headphones I could find which had detachable cables and were designed like God meant them to be, i.e. a simple tube you stick in your ear that has a cable hanging DOWN from it, were a pair of Panasonic something-or-others that got rave reviews but were no longer being manufactured and were impossible to find, and the Sleek Audio SA-1s, which seemed to get mixed responses regarding audio quality. This turned me off, since I got the impression that they sounded worse than the se215s. I also didn't like the idea of swappable tips, as it implied you had to choose between super treble and super bass, instead of just having a balanced sound. But, when I saw they were on sale for $40 on Amazon, it was too good a deal to pass up.
As it turns out, they indeed have better bass than the se215s and are even more balanced with quieter highs (at least with the bass tips on). They are less precise and "deader" than the se215s, though (there is one song in particular that has "buzzy overtones" that are readily noticeable on the XBA-1s, just detectable on the se215s and completely inaudible on the SA-1s). Keep in mind I only tried the treble tips once when I first got them, though the difference between the two tip types isn't very big. Also, I used the screamin' Skullcandies for about a month between the destruction of the se215s and the arrival of the SA-1s so this isn't exactly a side-by-side comparison.
I quite like the mellow sound, though I would give up the balance for some more clarity, especially now that I can eq everything, and what sounded mellow before last night now sounds like a pair of tin cans.
And that was my description of what the SA-1s sound like, which somehow kept expanding itself until it became the entire history of my headphone experiences. Jesus. I'm expecting all of you to skip all that, but now I'm too lazy to delete it all.
Lemme get to the point: here is my calibration for my SA-1s. Important to note that first off, I'm using the bass tips. Secondly, the little mesh insert things that are supposed to keep debris out of the drivers fell out recently, so that may or may not change what they sound like. I don't know. I wasn't paying attention to see if it sounded different; I was more preoccupied with "where the hell did it go?!" Third, I calibrated these on my computer, which is really loud (significantly louder than an Xbox 360. Maybe half as loud as an average server rack, if you've ever been around those) and sitting right next to me. It isn't terribly bad with headphones on, but it's definitely quite audible. Fourth, I was on my ADHD medicine yesterday, and that gives me mild tinnitus whenever I take it. Fifth, my headphones are hooked up to my computer through a pair of desktop speakers which give the signal a not insignificant hiss. Loud enough to hear over the computer, but not too bad. Sixth, yesterday's barometric pressure in my area was- do these even have an effect at this point?!
[Uh, crap. This is embarrassing. I was going to insert a pic of my equalization here, but apparently my account doesn't have "permission".]
As you can see from the above picture, I found three peaks. I had a bit of trouble doing this, so I used an unorthodox method to equalize these babies. The download links for the pink noise and sine wave sweep are both dead, so I used a pink noise file linked by someone else in the thread. Since there was a significant delay between making an adjustment in Electri-Q and actually hearing it, the pink noise file was only a couple seconds long, and I couldn't figure out how to get Winamp to keep playing it on a loop, I couldn't calibrate with the pink noise file. (Also, what is pink noise ideally supposed to sound like? It seems like you're supposed to fiddle with the peaks until you can't hear prominent high pitched noise, but if you do that, aren't you in danger of making the highs too quiet?) What I did was, after I found the peaks in SineGen, I selected those frequencies and adjusted their volumes to match several lower-pitched reference tones. After that, I calculated the dB difference renormalized to zero to get the amount in dB I should lower the peaks in the equalizer. With that, I had both the location of the peaks and their exact offsets. I used the pink noise file and some music to roughly determine the widths of the peaks and to make sure it sounded alright. I don't have a sine wave sweep file so I can't do further experimental verification of my results.
There you go. Not terribly exact, but the music sounds good, so there you go.
Thus concludes my very first "cool story bro" on this forum.
P.S. I noticed in SineGen that at the very low frequencies, the volume drops off steadily to a severe degree. I can't remember the number, but I got to total silence before I got to 20 Hz. That's normal, right? Or should I try to adjust it so the bass is the same volume all the way down to as low as possible?
P.P.S. I was taken aback by the sheer harshness of the unequalized headphones as compared to the equalized ones. Like, it was almost painful. This got me thinking. Does anyone think these shrill resonances might have something to do with hearing loss among less careful headphone users? I seem to recall Wikipedia saying that it tends to start out in those upper high frequencies. Any correlation? Would equalization help prevent hearing loss among those that aren't just plain reckless with their volume levels? I know I have, desperate for more dynamic range, occasionally cranked up the volume past the point where those high frequencies become unpleasant. Maybe it works the other way around, though. Maybe by making music much more pleasant to listen to, it actually increases the danger by allowing people to comfortably raise the overall volume further. I dunno. What do you guys think? I think it's kind of interesting.
^Un. Be. Lee. Vah. Bull.
That was one hell of a read. I actually read the whole of it.
Sadly, I can't answer your questions. Great way to bump up the thread though!
Thanks. That's really cool that you found my novella interesting. I'll try to get that picture of the eq up. Do you know how I can get permission? Do I have to make a certain number of posts or something?
Hmm...yes, as far as I know. However, I only started posting pictures after I began posting more frequently. I think you'll have to be a 10+ or 100+ Head-Fi'er to post. You might wanna ask Currawong, an admin. He's usually active, but you could ask around.
I am trying to get Electri-Q to work in Foobar...
I have installed and enabled the VST wrapper....
I have then installed Electri-Q and put the .DLL in the components directory....
I have tried selecting every possible folder in the VST wrapper options....
But I will NOT find Electri-Q so I cannot use it....
Does anyone know why this is please?