The most reliable/easiest way to EQ headphones properly to achieve the most ideal sound (for non-professionals)
Jun 20, 2016 at 5:19 AM Post #181 of 316
By testing audio according to the Harman curve, it gives somehow weird results in the 3-4 kHz field by boosting them too much. After flattening according to my hearing, I get logarithmic flat, pink noise actually. It looks like Harman curve isn't appropriate for every ear or one must implement it logarithmically (with -3db slope).
 
Jun 20, 2016 at 7:32 PM Post #182 of 316
By testing audio according to the Harman curve, it gives somehow weird results in the 3-4 kHz field by boosting them too much. After flattening according to my hearing, I get logarithmic flat, pink noise actually. It looks like Harman curve isn't appropriate for every ear or one must implement it logarithmically (with -3db slope).

Make sure you're not confusing the measurement graphs of the "with compensation" and "without compensation." For example,Tyll's headphone measurement graphs are "after compensation," which means the ideal target curve should look more like a very steady and gentle slope of -10 going from 2 KHz to 20 KHz. There is no big rise at 3 KHz shown in a compensated curve. Compare the measurements shown on this page, where the blue and red curves are compensated, and the rough looking gray curves are before compensation: http://www.innerfidelity.com/content/headphone-measurements-explained-frequency-response-part-two#lQj6g6oGKtuyso3M.97
 
Jan 17, 2017 at 2:25 PM Post #183 of 316
FYI, Tonebooster (creator of Isone) released a very useful headphone plugin called Morphit. It corrects your headphone's frequency response, simulate other famous headphones, and more.
 
Check out the thread I just created for it here: http://www.head-fi.org/t/832543/tonebooster-morphit-correct-your-headphones-frequency-response-simulate-other-famous-headphones-and-more
 
Jan 20, 2017 at 10:02 PM Post #184 of 316
I posted something in the computer audio forum last month that I thought people would find use for it on here. 
 
http://www.head-fi.org/t/829549/equalizing-headphones-to-your-ears-with-in-ear-binaural-microphones
 
Jan 21, 2017 at 2:11 AM Post #185 of 316
  I posted something in the computer audio forum last month that I thought people would find use for it on here. 
 
http://www.head-fi.org/t/829549/equalizing-headphones-to-your-ears-with-in-ear-binaural-microphones

That is really cool. I didn't know binaural microphones were readily available for purchase even on sites like Amazon. That really helps simplify the process of measuring and EQing headphones. 
 
Jan 21, 2017 at 2:00 PM Post #186 of 316
That is really cool. I didn't know binaural microphones were readily available for purchase even on sites like Amazon. That really helps simplify the process of measuring and EQing headphones. 

Yeah it's a fun thing to experiment with. Although equalizing a headphone to be flat right at the ear canal entrance is probably not the most accurate thing to do, since what we perceive as neutral doesn't measure perfectly flat sound at the ear. Now an interesting thing to test would be to measure a flat speaker with these microphones, preferably in a room with little reflections, then match the response of the headphones to that.
 
Jan 21, 2017 at 2:31 PM Post #187 of 316
Yeah it's a fun thing to experiment with. Although equalizing a headphone to be flat right at the ear canal entrance is probably not the most accurate thing to do, since what we perceive as neutral doesn't measure perfectly flat sound at the ear. Now an interesting thing to test would be to measure a flat speaker with these microphones, preferably in a room with little reflections, then match the response of the headphones to that.

Measuring speakers that way would have the benefit of including each individual's HRTF, instead of using just a measuring microphone placed at the listening spot (which is what IK Multimedia's ARC System does). 
 
I wonder if there's a way to measure the individual's ear canal's resonance and then factor that into the overall measurement?
 
Jan 21, 2017 at 5:14 PM Post #188 of 316
Yeah it's a fun thing to experiment with. Although equalizing a headphone to be flat right at the ear canal entrance is probably not the most accurate thing to do, since what we perceive as neutral doesn't measure perfectly flat sound at the ear. Now an interesting thing to test would be to measure a flat speaker with these microphones, preferably in a room with little reflections, then match the response of the headphones to that.

 
This is essentially what the Realiser has you do: measure speakers and headphones using the same binaural mics; it then does all the grunt work of making filters to change the cans into the speakers. You can also use the mics to "turn" one set of headphones into another using the same procedure.
 
 
I wonder if there's a way to measure the individual's ear canal's resonance and then factor that into the overall measurement?

 
You would need a probe mic. But since the effect of the canal is (as far as I've read) independent of direction, it would seem you would only need to measure it for those cans inserting into the canal.
 
Jan 21, 2017 at 6:08 PM Post #189 of 316
I'm curious about the mic that the Realiser uses, I can't find anything like it available. Apparently it is surrounded by foam so you can easily insert it into your ears, and probably get better measurements than what I'm getting with my mics.

I'm also wondering if something like a Stax ear speaker would benefit from EQ more than the HD650. Sure it's got better distortion measurements, but do you guys think EQing an HD650 and either a planar or electrostatic headphone to have the same frequency response would show noticeable differences?
 
Jan 21, 2017 at 7:04 PM Post #190 of 316
I'm curious about the mic that the Realiser uses, I can't find anything like it available. Apparently it is surrounded by foam so you can easily insert it into your ears, and probably get better measurements than what I'm getting with my mics.

I'm also wondering if something like a Stax ear speaker would benefit from EQ more than the HD650. Sure it's got better distortion measurements, but do you guys think EQing an HD650 and either a planar or electrostatic headphone to have the same frequency response would show noticeable differences?

 
I tried finding specs on it, but couldn't find anything in the A8 manual. It almost seems like they put them together themselves. I guess I should ask in one of the Realiser threads. I use the same binaural mics from your link. I find I have to use the windscreens to get consistent placement, else they slide around like hot butter.
 
Yes, I think EQing any headphones with significantly different FR will result in immediate differences. If the headphones are low distortion and their response is minimum phase, then a minimum phase EQ should technically be all you need to turn one into the other. There are practical limitations, since you can get some pretty steep dips in the FR due to interactions between your cans and your ears.
 
Jan 21, 2017 at 8:48 PM Post #191 of 316
clearly we all buy a Realiser just for the mics. ^_^
 
for the ear canal resonance, it seems to be mostly a matter of length, at least that's how many people treat it. not that I have any idea how to measure my ear canal without ending up with a hole in my ear drum.
now some of the newest dummy heads can have ear canal bend, so maybe it's not as insignificant as many made it to be? IDK. we need some of the guys who played with cadavers and couplers to get that kind of answer. so I'd rather just fool around in the 2.5/3khz area with an EQ and settle where it feels right to me. not very objective but I don't know what else to do.
 
Jan 22, 2017 at 10:04 AM Post #192 of 316
  clearly we all buy a Realiser just for the mics. ^_^
 
for the ear canal resonance, it seems to be mostly a matter of length, at least that's how many people treat it. not that I have any idea how to measure my ear canal without ending up with a hole in my ear drum.
now some of the newest dummy heads can have ear canal bend, so maybe it's not as insignificant as many made it to be? IDK. we need some of the guys who played with cadavers and couplers to get that kind of answer. so I'd rather just fool around in the 2.5/3khz area with an EQ and settle where it feels right to me. not very objective but I don't know what else to do.

 
It's not that it's insignificant, it's whether its affect changes based upon the relative position of the transducers. For transducers external to the pinnæ data suggest that the effect of the canal is constant. For IEMs, which insert into the canal, you would need a probe mic measurement at the eardrum to get the true story. This sucks because I think we'd all like to be able to get proper HRTF effects with IEMs, but probe mic measurements are a different bear than measurements at the meatus.
 
Mar 6, 2017 at 1:13 PM Post #193 of 316
Hi, I just registered after reading up on all of this for a while. Main purpose was to flatten my new DT990 pro's a bit since they are kind of bright in the high end.
 
Managed to pinpoint the exact spikes, at least the main ones, and level them out.
 
 
For those who have the DT990 pro and wish to slap some EQ on, these are the settings that will eliminate the three prominent spikes in their freq. response :
 
 
filter 1 : 150Hz, -3db, Q=1.2
filter 2 : 6450Hz, -3db, Q=4.1
filter 3 : 8875Hz, -7db, Q=5.2
 
 
It looks like this on the parametric : http://imgur.com/LAZ0r3D
 
 
The way I pinpointed these was first by following this guide, which led me to these big spikes plus some smaller ones.
By adjusting the Q and levels a bit I was able to sum those few little spikes within these three adjustments.
 
I noticed only the one fairly gentle (read: low Q) boost of about 3-4db between 75Hz and 500Hz for the lows and lowmids.
In the highs above 2K there were these two very big spikes between 5K and 7.6K, and between 8K and 10K.
 
I used the Frequency Response Data from Goldenears to verify that what I was hearing and finding corresponded to their data, which it did rather perfectly.
 
 
My conclusion, personally, is that in the lows and mid lows, there isn't that much of an issue at all,
and representation between say 100 and 500Hz is quite flat although a bit boosted across the band.
 
But then starting @ about 5-5.5K two huge spikes, which to me explains away their overly brightness...
 
 
You can set these filters in f.e. Equalizer APO or any other kind of EQ to have a more 'true' flat response.
 
I found that by using only these three filters the end-result is even better than when EQ'ing all the little spikes and cuts, ending up with almost 15 cuts.
 
 
 
Try it if you have these Beyers, I thought it a nice idea to provide these for others, since nowhere could I find it for the DT990 pro.
 
 
Cheers.
 
Jan 13, 2018 at 12:52 PM Post #194 of 316
This method seems very interesting. Would anyone who has the Sony xba-n3 be able to follow these steps and report back the results and the best way to eq these headphones as per this method? I can't do it for now because I only have the iem with a balanced cable that has a 4.4mm male plug and can only use it with my DAP. I'm not able to connect these to my computer!

Thanks!
 
Jan 18, 2018 at 1:07 PM Post #195 of 316
I haven't read the entire thread, but I'm actually big proponent of using sine tones, although it's definitely necessary to use a sweepable generator and not static tones. Static tones are next to worthless for this, I'd say.

I have this one bookmarked: http://www.szynalski.com/tone-generator/

One point about the FR alterations you posted for the DT 990s... in my experience you can't necessarily trust peaks >5Khz as manufacturing tolerances can cause variation between individual headphones of the same model. So I would not do that without at least verifying they have a peak there using a sine sweep.

For the 990s probably they're manufactured with good enough QA and tight enough tolerances that you won't see wild variation in the treble, but you never know. Cheap headphone transducers are practically each a unique snowflake in terms of all the craziness you get in the upper treble range.
 
Last edited:

Users who are viewing this thread

Back
Top