EQing IEM to Headphone By Ear Experiment

[yoda]

(This was originally posted on reddit, but didn’t get much traction, so I’m posting here to see if anyone here might explain the science behind my findings).

…

I tried EQing Galaxy Buds 2 Pro to HD600 by ear, here are the results:

Disclaimer: People here are pretty well informed about HRTF stuff, basically you probably won’t get the same experience as all our ears are different. Not to mention tips affect the galaxy buds sound, etc, so I’m using the default tips provided. I did make sure there was no previous eq on before doing my tests. I’m writing this down just to describe my experience, your mileage may vary.

Filter: ON HS Fc 1500 Hz Gain -15 dB

Filter: ON HS Fc 5000 Hz Gain 12 dB

Filter: ON HS Fc 6900 Hz Gain 5 dB

Filter: ON HS Fc 7100 Hz Gain 14 dB

Filter: ON HS Fc 9500 Hz Gain -28 dB

So I’ve been really struggling with EQing by ear, as without direct a-b I find it hard to distinguish volume changes. Then I realized, wait, I can literally just feed the hd600 into one ear, and the galaxy buds 2 pro into my other. Then all I had to do was play any frequency and eq it until the image was centered.

I started by matching volume at 1000 Hz, then jumping 500 hz at a time. Obviously, the smaller the jump the more precise you can be, but I was just doing this to mess around, not get an exact EQ. The first thing that jumped out at me was how much quieter 2000 Hz was on the hd600 vs the galaxy buds. Yes, it really ended up being -15dB until I really felt it matched. (speculation: maybe this is part of why we don’t get a sense of stage in iems? Also, both hd600 and galaxy buds 2 pro match their respective targets well. What’s happening here?). Then around 5000 Hz suddenly the hd600 was much louder (I applied high shelves as I went). Finally, at 9000 Hz and so on the hd600 really quieted down.

How does the eq sound? Just from a glance, you got to admit it looks weird. However, subjectively it is excellent. Let’s get the predictable stuff out of the way: Any sense of technicalities is not improved, nor stage. I think you really need to be more fine-grained to get those. However, I do think tonally it sounds much, much more similar to the hd600. Vocals are richer and more natural, which you really would not expect based on the graph. Turning off the eq, there is a sense of glare apparent. Just really, really grainy in comparison. I’ll just say, this does not sound dark to me, despite the eq only bringing stuff down. You could probably get rid of the last filter if you want some more spice though.

Overall, I’m glad I tried this, it worked better than I imagined. And I truly think this is an actual, “easy” method to eq headphones to each other, at least to a relative approximation. Next, I might try eqing either galaxy buds or hd600 to susvara and see how it goes.

…

I’ve been listening for a couple more hours or so, and man the audio quality difference is stark. The galaxy buds have always been a “listen only when it is the only option” thing for me, and now I’m listening to it voluntarily. Yes, it is still miles away from the hd600 technically, but the tonality is essentially perfect for me. Yes, it does sound almost exactly like the hd600 tonality. I must emphasize, it doesn’t sound dark at all, nor bass boosted. But I still don’t understand how dipping 3k by 12dB would make anything sound natural. Is this type of variance possible due to HRTF? Can my ear like Harman for headphones, but vary this much for iems?

[Fc-Construct]

This is a pretty interest approach to EQing by ear. My kneejerk reaction whenever I see things like this is that if it sounds good to you, it sounds good. But give it a week or two, try different things, then come back and see if you like it.

I think what we’re missing is information the settings of your Galaxy Buds 2 Pro. What is its original graph/EQ? Secondly, I’m not sure how you’re feeding in the HD600 for the EQ, but if you’re using the Buds 2 Pro under the HD600, you’re hearing some of the HD600 still i.e. the Buds 2 Pro isn’t completely isolating.

Another method for doing this I might suggest is to pick a very simple track, e.g. a slower solo piano piece and then trying to match the sound of the Buds 2 Pro with the HD600. Once you’re satisfied, move to say a solo acoustic guitar and so on. Then go back to the starting track and see if you’d change anything. Once you feel like you’re at a good spot, then you can try against a wider range of music to see if you like it to the HD600. I’d be curious to see what you find.

[yoda]

Thanks for the response. The galaxy buds pro 2 is stock with no eq in the app or samsung settings applied. I don’t know the exact graph for my unit as I don’t have any measuring tool at hand. The right galaxy bud was simply placed into the ear, then the hd600 was put on. I set the hd600 to be left channel only, and started matching.

The problem with even things like simple piano pieces is I don’t know what frequencies to adjust. I’ve never thought of music in frequencies, ever, and even if I did hear some difference between the buds and hd600 when playing a track, I couldn’t tell you what frequencies differed. I even struggled to hear peaks and dips in a frequency sweep, the fact that both the volume and frequency change at the same time is too much for me. That’s why I find this method so easy, once I hear an uncentered frequency I can immediately adjust it, and that’s the only variable.

Subjectively, I’m still having a great time, just amazing how natural the sound is, with a lack of glare/grain present in the original tuning. But what still sticks in my mind is how crazy the drop in volume from 1k to 2k was, that I had to EQ it down 15 dB. For two apparently well measuring products, that seems like too much.

I’m really hoping others can replicate this method and post their results. Have you or anyone else at the headphone show tried this method before? Or tried EQing an iem to a headphone in general? You guys have a lot more equipment available than I do so I’d really like to see what results you all get. Particularly how an iem and headphone compare in that 1-5k range.

[Fc-Construct]

The problem with even things like simple piano pieces is I don’t know what frequencies to adjust. I’ve never thought of music in frequencies, ever, and even if I did hear some difference between the buds and hd600 when playing a track, I couldn’t tell you what frequencies differed.

So the funny thing is that traditionally, this is how EQ is done. You take individually instruments and you tweak it a little bit here and there to get the desired sound you want at certain frequencies. For example, with acoustic guitars adding a light touch around 300 - 400 Hz gives it a warmer, richer sound. There’s all sorts of charts for this, including this one from Sweetwater that’s pretty popular.

But even if you aren’t a mixing engineer, learning to EQ at frequencies (individual instruments are the best starting point) is pretty critical. The Harman Research’s “Trained Listeners” had to go through a program that taught them to identify changes in frequency. You can try it yourself - it’s called Harman How to Listen: https://harmanhowtolisten.blogspot.com/

Anyway, have you tried swapping it so you have the HD600 on right channel, and GBuds 2 Pro on the left? Curious to see if you get the same result. 15 dB downwards is definitely quite significant. I would expect only about a 5 - 8 dB difference.

[yoda]

Just tested and the imbalance is definitely less when swapped. 8dB or so between 1k and 2k. That ended up being around the difference between 1k and 2k in the final eq with all the filters so that makes sense why it sounds natural to me. Did also retest the original config and it still is a 15dB imbalance. Pretty bizarre stuff. Regardless of my poor ear imbalance, that dip between 1-5k is still there. Unfortunately I don’t have other iems to test.

[Auris]

This would be a good candidate out-of-the-box. Although, the treble and just the feeling of IEMs and headphones are so different that my brain can’t see pass that.

[EminentOne]

hylo

wao
i try setting on my b2p

it sound like i have pillow on my head