What's going on? (Patreon)

Just wanted to keep you updated what I am currently working on. 

I am in the process of creating compensations for the new microphones I got recently.

I was not satisfied with my first results and I started to work on it completely from scratch. I want to eliminate my own HRTFs as good as possible from both my recordings and measurements. In order to achieve this, I need to separate the HRTFs from the rest somehow, to get a real "neutral" kind of sound with as little errors as possible. 

For this task I measured a neutral speaker playing from front and the HD600 and I was surprised how well both align above 3khz (measured nearfield to avoid too much room influence and averaged for left and right ear):

I am not sure why the speaker measurement has a much higher 3khz peak compared to the headphone, but everything above looks really solid to me.

The area above 3khz is the one which varies most between different listeners, different headphones, but also different listening angles, I was surprised that the HD600 seems to follow the spectrum I get from a fronally placed speaker that well, which proves that the HD600 is still one of the most neutral headphones out there. But you can also notice that there are some slight deviations in the response, which I am currently trying to correct for. Right now I am listening to the equalised HD600 like this trying to get used to this new sound. It sounds different than OptimumHifi for example, but at the same time more true as soon as the brain adopts to this, treble becomes more open and more realistic sounding, although I can still hear some edginess in the response, it is still not perfect and I will try to equalize the affected areas manually in order to check if I can get better results from a selective manual adjustment than some autoEQ. But if I verified the result my compensated recordings should basically represent "neutral" well with very little added coloration. 

The problem I had was that when listening to my recordings over the Free Pro 2, I could notice some peaks which obviously could have been remaining artefacts from the HD600 response. If I manage to eliminate the influence of my own HRTFs completely, this should also make my recordings appear smoother over other headphones.

An overlay of the speaker with Free Pro 2 and neutral tuning shows how both dips at 9.5 and 13.4 align perfectly. Keep in mind that the Free Pro 2 is measured with the couplers, while the speaker was measured with my ears at the ear reference point, thus the measurements are not directly comparable:

I tried to add the dip at 7.4k externally with a filter and it definitely improves the response for me, it was obviously still one area that bugged me, which I couldn't quite track down precisely just by listening:

Filter: ON PK Fc 7450 Hz Gain -8 dB Q 8

I wonder if adding this single filter will improve the response for others as well, as it's definitely part of my own HRTFs. These are all factors that will affect how well others perceive the sound that I created "for my own ears".

Of course this will also mean that once finished I will have a completely new headphone reference for "neutral" which at least should work well for my own ears, I am also planning to compensate for my left and right ear separately. Keep in mind that we are talking about nuances here, and given the fact that people are listening and enjoying headphones with responses that are far more off, it questions the effort I am doing here, but I just try to make it as solid as possible to ensure that my data (either binaural recordings or measurements) are correct (to the most possible degree). 

All this will take some time, on top of this we have semester holidays starting next week which I will use to visit my parents together with my son, therefore please apologize the lack of real content, but I promise to offer a completely new tuning this month, I am still planning to create an HD800S simulation next. 

So stay tuned, and thanks so much to everyone believing in what I am doing.