oluvsgadgets
patreon
Sennheiser HD600 - natural target V1
🕑 Added 2023-04-11 14:17:10 +0000 UTCSince my return from holidays I have spent all recent days optimising my target, as I was not satisfied with the previous result (V0) especially when directly compared to speakers and also when listening to my own recordings. I thought my "universal" target was actually somehow closer, but not smooth enough. That's why I tried to get both variants a bit closer to each other and come up with an all-in-one all embracing "natural" target. It took me several approaches to get the sound closer to what I hear from speakers and get the overall balance more correct and hopefully also render my own recordings more faithfully. Keep in mind it's always a 2-way process to get everything right. The idea is still to hear pink noise played over speakers to sound as close as possible to pink noise played over a headphone and also pink noise recorded when played the speakers to still sound the same as the speakers when listened to this recording over the heapdhone. Right now I was mostly concentrating to match the sound of my headphone to that of a speaker as good as possible. And although my recordings sound already way more realistic than they did with V0, I will need to readjust my compensation for my binaural recordings next. My compensation for my binaural recordings is already departing from fully neutral in order to be rendered faithfully, otherwise the headphone necessary to render these faithfully would be way too bright and too basslight, therefore I applied these corrections to my recordings as seen below. I will need to fine tune some parts even better to close the entire circle and will concentrate on this during the next days.
First of all: it's close to impossible to recreate the spectrum of frontally placed speakers by a headphone exactly. There is some discrepancy between how we perceive the balance of bass, mids and treble from speakers to that over a headphone which is also represented in my compensation to some degree. Besides there are some phantom frequencies which I perceive in real life, which I cannot quite reproduce by the headphone, it's probably due to the fact, that with a traditional 2 speaker stereo setup there is no sound coming from exactly front, thus correlated and uncorrelated pink noise sound completely different. It would be much easier to deal with this if sound was coming from the entire front or simply by an additional center speaker, nevertheless I tried to capture that "character" as close as possible and came up with this final result which I think is the closest I managed so far regarding "the sound of speakers over headphone".
Keep in mind I will try to fine tune the result as soon as I receive the Neumann KH80 speakers, together with their built in EQs I can hopefully manage for a truly flat response at listening position which will help me to compare pure pink noise over headphones to that played over speakers without too much coloration added by the room etc. Right now I have way too much bass energy together with several room modes which influence the character of pink noise too much, as I am always trying to compensate for the missing bass I am hearing from the headphone.
Below you can see the measured spectrum of the Nubert X3000 at listening position and in black the final spectrum created by my HD600 with my final adjustments (left and right averaged):
You can see that I am following the original spectrum quite closely in most parts above 2khz, but need to depart in some areas. I tried filling up that dip at 3.4khz, just like the dip at 8.2khz, reducing that peak at 10khz while adding another one at 12khz to get closer to the original spectrum, but that did more harm than good, also raising the dip at 1.5khz or changing the balance resulted in a sound that differed too much to what I was hearing from speakers. Keep in mind with music all those adjustments were hardly noticeable, but as I used pink noise whole the time, any small adjustment became hearable, and I tried to adjust it to such a degree, that I can securely switch between the speakers and my headphone playing pink noise and get the feeling that I am listening more or less to the same thing. As mentioned the speakers still contain some kind of "Hhhh" and "Ffffff" which I cannot quite reproduce by the headphone properly, therefore I boosted the 2khz area a bit more, although I tried to keep closer to the original spectrum for longer, but this was the best balance I could achieve, any deviation from this made the final result appear "worse" or let's say more different from the speakers.
Compared to V0 in pink you can see that I have more upper mids and more treble now and I adjusted the bump at 2.5khz which fits much nicer into the overall spectrum now, otherwise it would stick out as disturbing frequency.
I also added a low-pass filter to tame that treble peak of the HD600 at 18.5khz, this filter can also be used to control the airyness by adjusting the Q-factor, although I think with the current value it sounds most correct to me. With this current adjustment treble should follow the natural way you would also hear from a speaker with just a dip at 17khz, which you can fill up if needed by the optional 3 filters at the end, which reduce the peak at 10khz a bit, raise the dip at 8khz and also fill up that 17khz area, but I am not sure if those will work for everyone to the same degree, therefore adjust them to your own personal perception. I can hear the changes but they do not improve the result for me that much except the 17k filter, but I can live just without the other 2 adjustments and I wonder how others may perceive them. Also not sure how that dip at 8khz will behave for others, as it will probably move in frequency with other ears, therefore maybe leave that filter just off or adjust it by listening. Adding the peak at 12khz does more harm to the sound for me, therefore I didn't include it here.
Here you can see the difference with (green) and without these HF-filters unsmoothened to see the difference better:
V1 compared to HD600 stock in blue:
and compared to the HD800S:
Actually this is how the HD800S should have sounded, if you remove all the peaks and adjust upper bass and mids slighlty, you would get a pretty similar response.
Keep in mind, I have no idea how my setting may work for others. I asked my son and he preferred my setting when compared to the speakers, I also tried putting some putty inside my ear and this changed the response completely for both speakers and headphone, but I still liked the equalized one more, although it lost all its quality with modded ears. I will do further experiments also with comparative measurements and maybe I can come up with a final averaged target-curve.
I haven't readjusted my settings for the Qudelix 5K yet, are there any Qudelix 5K users out there who need those settings? But I will sit down tonight and try to match the settings for the Qudelix 5K as well and maybe also contact the company about the issues with their filters.
Please let me know if V1 is an improvement over V0 for you, I also attached Oratory1990's Optimum Hifi profile for the HD600, I wonder which you prefer.
PS. I attached an alternate version which I created independently and then tried to merge both variants, but in the end came to a slightly different result which to me sounds even smoother, not sure about others, but mids seem to be cleaned up better, pink noise appears smoother to me which may be a result of the slight reduction at 800hz along with a smoother upper midrange around 6khz. You probably won't notice much difference with music, but if you listen to pink noise you may notice some slight chanes. Below you can see V1 and V1a side by side here. Please let me know which you prefer.
Comments
pokketa
BTW I try to EQ my q30 with pink noise, by using soundcore's APP and a free EQ app from google app store since soundcore APP's filter is not enough, I don't know how to tune it, so my way to tune it is to make a filter to highest db and then reduce it slowly until the special specturm disapear, And I quite happy with the result even I know that that can't be the best EQ, But still I think pink noise is super useful for tuning.
pokketa
So I try the V1 for a while, I choose V1A as the version since I think they are no big differences and you said that is the best. And it is actually so hard to exactly hear and tell the differences and fault in all these eq, In some recording it has big differences some just slightly. So yeah it must be a hard work for you to keep listening and adjust the EQ. When I use V1A to listen some vocal song, most of them sound not that big differences, with V1A the vocal seems more stereo and a little bit more airness, I guess the more treble and the sibilance from the EQ make the effect, meanwhile some song seems much more stereo and much more sibliance, to a certain degree I would call it harsh, but also give me the feeling that the singer is at the stage and I am right at the concert, hearing the song from the big speaker. I guess different recording and mixing effect it a lot. Then I try some classical music, I know HD600 is known to be good with classical, but this Eq make every instruments so stereo that I can actually tell where is the flute and where is the violin, it gives me a feeling of I am listening inside a small concert hall, with all those extra airness and echo-like feeling, but same again, every instruments sound little brighter than the stock, and I don't know if that actually depart from neutral sound. So the conclusion part, I think the original HD600 is a quite safe headphone, you won't hear much peak and fault with it, and oratory is also using a safe way to tune it by adding it's bass and remove some peaks, and meanwhile V1A indeed make some music sound like it is from a frontal speaker or instruments, but also making the sound more bright. So yeah I still don't know if we should choose a frontal speaker as the reference to tune a headphone, since a speaker is speaker it is not perfect and headphone is headphone, I am not sure if I like the feeling when using this EQ to hear a studio-made music, but for sure if I want to listen to classical or concert recording I would use this EQ.
