Still fighting... (Patreon)

There are some complaints starting to appear that my Patreon is turning into an Earfun only project. I blocked this dude immediately and refunded his pledge as I was hoping most of you would understand my intent that I am trying to achieve. 

My initial goal was to create a sound which would rival most high-end headphones (I am not even talking about IEMs, as here it's pure luck if the tuning will get along well with your ear canal resonances) and I wanted it to be as affordable as possible for everyone and chose the Earfun Free Pro 2 for this task, as it's really great hardware-wise and thanks to the accessible internal EQ it should be possible to bring the sound to a really high level. I think I proved with my last video that the most important aspect of a sound reproduction system is a correct frequency response and there will be some surprises for sure when I reveal the results which headphone was actually which. All hyped aspects like "decay, attack, resolution, detail..." blah blah are just secondary and a direct result of a correct or should I say smooth frequency response without any obvious flaws. If a driver can reproduce all frequencies at the desired amplitude without any audible distortion the only remaining errors will be the resonances created by your ear or the device, which need to be corrected to get a smooth natural response at your eardrum.

It's been 3 weeks since I started my work on the Earfun Free Pro 2 and I must confess I underestimated the entire process. The more I listened, the more time passed by, the more issues I started to hear and the more I tried to improve on them. The brain will quickly adopt to any given sound signature after some minutes and will simply try to "filter out" all issues not noticing them anymore. But I wanted the sound to be "as perfect as possible" within the given limits. 

Although I really enjoyed my previous tuning attempts, after some time of listing I started to notice the remaining resonances, which bothered me, be it from the acoustic design of the Earfun Free Pro 2 or due to my individual ear canal resonances or simply both interferring with each other. I used the last days trying to optimise my settings and reduce all resonances that I was still hearing. Mainly the areas at 6, 9, 12khz seem to be responsible for most issues I could still notice.

Keep in mind that the sound hitting your eardrum when listening to a speaker from front will not look like a Harman target curve. I have never measured my own HRTFs as it would have cost me 500€, I will still try to do this during a research project at the acoustic research institute in Vienna as soon as it will be announced, but as for now I can only offer you an HRTF measurement from a friend of mine who provided it to me and plotted a graph showing the response with a sound source coming from front at an angle of 0° and 30°:

Comparing this to the Harman target with a similar range you can notice that it's something completely different: 

It's just a smoothened averaged curve over various listeners with added preference for bass and treble...

Everyone who tried my strictly Harman tuning probably noticed that it didn't sound good. It was not horrible but not really great or "correct" sounding. I rather tried to capture the timbre of a speaker playing in front of me and this is where the problems started. To be able to capture that character I had to depart more and more from this smooth target curve and just  like seen from the HRTF graph from my friend there is a noticable dip at 10kHz which can be 30dB in amplitude and will of course vary between different ears etc. 

With all my previous tuning attempts I was still hearing some ringing in the 9-10kHz range, but also 12-13kHz seems prone to this, thus I had to reduce these areas considerably while boosting the entire range around to get something that resembles the timbre of a speaker playing in front of me in a better way. It is still not perfect, but I think it sounds much better than before without those ringing issues I heard before. Here you can see a measurement of my previous tuning in blue with the current one in orange:

You can notice that I had to reduce all areas at 6, 9 and 13khz which at the same time changed the overall balance a bit. To my ear the current result sounds much smoother than before without any ringing or resonances at least not that obvious anymore. Comparing the perceived sound to a neutral speaker it is still not 100% there, but I am closer than I was before. It is extremely hard to capture this character in a realistic way. 

What bothers me are those dips I created which may either be the result of my own ear-canal resonances together with my own HRTFs making my brain expect a different response than simply a smooth curve like the Harman target.

Here you can see the current tuning overlaid with the raw response of the Earfun Free Pro 2:

Notice how the dips I created are exactly opposite of the peaks the unit produces at the given measured resonance. I am not sure if the dips I had to create in order to get a smooth response were because of my ear canal resonances, thus 6.1k would be the fundamental, with 12k being the 1st harmonics, and 9.5khz being probably the result of my HRTFs simply expecting this dip in order to make my brain believe the spectrum it was receiving is natural.

I am not quite sure how to deal with this if I need to tune for other resonance frequencies than mine. The raw response will change a lot between 6.1, 6.5 and 7kHz as seen here:

 You can see that together with the main resonance also the upper harmonics will shift from 12.7kHz  which I am getting with 6.1kHz to 14.4 with a 7kHz resonance. I am not quite sure how to exactly deal with this to really get the desired effects, I assume I will need to create the same dips I created for myself at the given resonane frequencies while keeping the 9.5khz dip constant throughout the tuning-range. Unlike with my tunings for the Free Pro, where I tried to create exactly the same response for every single resonance, I will probably need to adjust the tunings towards the new resonances creating dips at different positions than I hear. I am still not quite sure how to deal with all this. And I cannot verify the results on my own. 

My current tuning which sounds fine for me (right now) with a resonance of 6.1kHz will sound completely off for someone with 7kHz resonance as seen here:

Therefore I would ask everyone with a resonance close to mine to check if the new tuning I created offers better results for you as well. Maybe I am overcompensating for my own ear/HRTFs which would again give different results to someone else with different ears. Listen to pink noise, as this will reveal the differences in the best way. If pink noise sounds "decent" any music will sound decent as well. With music some issues do not become that obvious, but pink noise being a continuous spectrum with all frequencies represented to the same degree, you should be able to judge which one sounds "more correct".

Here my current tuning for 6100Hz:

Here my previous one (you can find all older versions in the archive):

Please let me know if the new tuning gives an improvement also to you with a smoother response without any obvious rining or other artefacts which I definitely perceived from the old one. It doesn't make sense for others to try it if your own resonance deviates more than 200Hz as the result will be wrong anyhow like shown in my measurement above, but I would welcome some honest comments from everyone with a similar ear canal like mine to be sure I am on the right path and not going around in circles whole the time.

Sorry that it takes so long, but I cannot be satisfied with half-baked results and want the final tuning to be as perfect as possible which of course needs time and effort and I am thankful to you supporting me with all this.