Home Artists Posts Import Register

Downloads

Content

I was trying hard to finalise my EQ for the HD600 and make it sound as close to the speakers I was hearing and named it "x" standing for the Nubert X3000 speakers I was using as reference. As the sound of the speakers is not perfectly flat especially in my room I cannot guarantee that this EQ is perfectly flat either, but I think at least I managed to make it sound as close to the sound I was hearing from the X3000 as possible. In reality the area around 1500Hz is perceived way softer through the speakers, but there is no other way to model that frequency range through headphones more convincingly and I think I manaed a compromise which sounds pretty realistic and should be an improvement over my previous versions.

You will notice that the differences to my previous versions appear rather minimal from the graph, with blue being V2, green V2.1 and black V3, but it's simply the overall balance between all frequencies which is important to be correct to be also perceived as natural sounding:

In reality the human ear perceives lower mids and upper bass a bit different than "flat". Usually I hear a resonance at around 530Hz with the part at 300hz and also lower with less amplitude, which also affects the entire upper mids and treble range if missing especially compared to the speakers. I left these details out as I tried to model the entire rest to sound already realistic without those adjustments.

These are all nuances which contribute to the realism of the sound and I tried to make it as good as possible at least to my ear. You will also notice that I adjusted the upper filters a little bit in frequency. If the filters around  8, 12.7, 15 are not perfectly aligned you will notice a disharmony in the treble. Everyone should try adjusting them for himself. Move the filter down or up by 1hz and hear how this will mostly effect upper mids and treble. I am not 100% convinced that I achieved a perfect balance even for myself let alone for other ears. But you can try playing around with these to maybe achieve an even smoother treble response.

There is so much uncertainty in all this, that I simply cannot guarantee perfect results to everyone, although I think it should be much better than the HD600 stock anyhow. Also the condition of the pads will affect the final result a lot. I just mounted rand new pads and the sound changed considerably. I haven't done any measurements of my HD600 with the other pads yet, but I will definitely do them next and also try to create extra settings for those which will be available for the Hifi-level only, it's simply way too much effort honestly and I can only waste my time with this if I get enough support from you.

Below you can see an image of all pad shapes I currently own, left being the brand new ones, in the middle my current semi-worn ones and at the right completely worn ones.

I will try to fine tune some areas for myself in future, but those changes probably won't work equally well for others let alone be visible within the graph. If experimenting a little bit with all filters you can maybe manage a smoother sound for yourself as well. Just try to adjust them bit by bit like lowering one filter by 0.1dB and hearing what it changes. With the current balance between all filters I think I achieved a pretty realistic speaker like spectrum, also playing sine sweeps the amplitude remains smooth with just those areas apear boosted which also appear boosted in real life. If sweeping from 1000hz up, you should only notice a slight raise in amplitude around 6khz with a distinct drop at around 7, then slightly increasing towards 8.5 with more boost close to 11 and 12.7. But the transition should be smooth and not appear like some peak. Every ear may need different adjustments though, I cannot say for certain how you perceive this in reality.

I will use this current target for all my upcoming headphone EQs. Also my own binaural recordings will be optimised to be rendered as fairthully as possible if listened with a headphone equalised to this target. I readjusted my compensation to get an as flat as possible response when listened through the headphone, and although the result is quite real, it is not 100% the same thing. I tried hard to bring it even closer, but it's close to impossible to achieve the same sound I am hearing in real life to that of the recorded source, but nevertheless I guess I am closer to anything else I managed until  now.

Below you can see the reference spectrum in grey overlaid with the binaural recording of this spectrum already compensated in blue. I had to adjust the balance quite a bit otherwise the resulting headphone target to render these recordings correctly would be way too basslight and harsh and won't represent the real spectrum I am hearing. In orane the same recorded spectrum played back over the HD600 equalised to my current target.

You can see that at least according to the graph the equalised headphone should reproduce the original recording pretty faithfully, but I deviate from the recording in some areas which are also the areas where I do not follow my reference spectrum strictly as shown in the measurements above, but had to depart or model the sound around those areas otherwise the resulting sound would simply not sound the same as in real life, thus this is the best compromise I could achieve between keeping my binaural recordings as neutral as possible and keeping the sound of the headphone as neutral as possible and also be able to render my own recordings as faithfully as possible to close the entire recording chain as good as possible.

Remember: this works pretty well with my own ear and my own headphone. As soon as I mount other pads etc, the sound will change too much. I am still not sure about the influence of other ears, how much this will affect the sound, but these ears will also affect the sound around you to the same degree. It has just to be measured how much the delta will be between what the ear make out of the sound around you compared to the sound the headphone is offering. I will try performing some experiments in this regard as mentioned before.

Coming up next is an optimised setting for the Qudelix 5K which I want to finish within the next days, then I will also try creating settings for my other pads and this will finish my work on the HD600, therefore I would like to jump to the HD58X and the HD6XX as next headphones to be covered for May. I hope for further support to really justify all this effort which goes into all this. But having achieved a final graph which seems to work pretty well at least for myself, it shouldn't be such a hard task anymore to transform this graph onto other headphones.

Please let me know if this setting is an improvement for you as well and let me know your thoughts. If any grainyness remains, you probably need to fix it by fine tuning those filters on your own. If you have brand new pads mounted, try pushing your headphone a bit closer to the head to simulate how it should actually sound.

Update: I made some changes as a result of pad swapping, performing new measurements etc I had to adjust the sound to get away with all harshness. Although the harshess is hearable from speakers as well, it's rather or probably a side effect of various frequencies creating phantom frequencies, like if playing an interval not in harmony you can sometimes hear a 3d frequency which does not exist in reality. It's actually quite complex and I have no experience but I tried to recreate these frequencies I heard from the spectrum, while in reality they form on their own if all other frequencies have a particular ratio to them. Therefore this update which I think improved the overall result quite a bit. 

This is based on completely new measurements I did of my HD600 and I will talk about all details maybe later if there some interest. But here you can see V3X in green and V3.1x in blue. 

I corrected for a bit more lower bass as my previous measurements were not that precise, there was obviously some low frequency rumble from outdoors captured as well, and you can see that I lowered some frequencies while boosting others to still maintain the same character but get a smoother result overall. It should appear more transparent and simply more real compared to before although I somehow depart a bit from the sound character of the speakers. I am sorry I was bothering you will all those versions, but it's extremely hard and frustrating getting something really decent especially with so much variance by just swapping pads. You can see the HD600 with my updated settings and with my current medium pads in black, new pads in blue and worn pads in green. There is more than 10dB difference in some areas between new and worn pads which is impossible to control and predict, let alone what will happen with other ears. Everyone will hear something completely different, sample variation of the headphones not even taking into account.

I will still wait for your feedback, although I am not sure you are really getting the same result as me at all, nevertheless I wonder if it's an improvement for you as well or if some areas got worse. I may tweak this a little bit more although already my own recordings sound smoother as well with this new setting, but next I want to prepare a setting for brand new pads and I still need to create one for the Qudelix 5K as well, which is also quite a hassle.

Comments

oluvsgadgets

It seems the higher frequency the more the filters start to differ

Anonymous

So Oluv, recently I am busying working so don't have time to listen to your newest EQ, and today when I try to listen to it I forget something improtant and then same shit got me again. For some reasons, your EQ since v1 always sound good with low-middle volume, but when every time if the music is slightly louder(I am listening to spotify some song just randomly louder than others), IT BECAME REALLY SHOUTY, BRIGHT THAT HURT MY EAR. So weird, when I listen to low volume it make the sound stereo and realistic, but when it is at a higher volume it is too bright. I don't know, maybe this is the cost of being realistic sounding. The wearing position and pads condition also impact the sound a lot, maybe someday you should even show us how you place the hd600 on your head; Where is the exact place of the headphone drivers, should we place it higher than our ear or lower or just center; And to use a ruler to show how thick the pads are. I have tried to listen to pink noise from pink noise generator every time your new EQ comes out, they always sound differ from what I listen from Uboom L, and every time I try to EQ it toward Uboom L with my listening, the EQ file always end up become very alike to your V0 EQ, but your newest EQ is no doubt always better than it! How interesting, really wonder how you do the EQing, it must a hard work, thanks you for your continuous passion with EQing.

Anonymous

I am inclined to agree, every version except V0 to me, seems to be a bit harder to keep a consistent volume with. V3X is certainly better to me than V1 and 2.1, but the volume thing is still faintly there. I am able to listen consistently with a lower volume, and do think it sounds better than V0, but also can't help but feel that strange volume issue. It's probably highly affected by the mastering quality of any given track, so I agree that it could be a side-effect of the neutrality that the EQ delivers. I will say that this is the V3x is the only version where Oluv's gallery recording of the wood floor, actually sounded like people stepping on old, creaky floorboards, rather than rustling through leaves. I wish I had the proper equipment to accurately measure the decibel level playing, so that I would be able to tell if I was actually listening safely, since the sensation of loudness you mention can be concerning at times. I do think it sounds really good, and better than any of the previous versions, but the fear of damaging my hearing does have me playing at a lower volume than I believe I normally would with other headphones, also HD600 stock or V0. With the perfect levels for safe listening, I am convinced that the brain would be able to adjust, to the point of the loudness sensation disappearing over time.

Anonymous

Greeting, friend, we are the two guys who love posting a long comment haha, and we usually have the similar feeling, which is good, it means that I am not special. I don't understand why the eqs are this special, is like they have very good performing but high required to get hit by the sweet spot. When we talk about hearing protecting, I used to borrow friend's iphone and airpods to test what is 85 db sound like, and the result is I usually listen to music with 70 db in first few minutes for excitement, then tend to stay at 65 or even 60db. One thing I also do is to put a db meter right beside my ear when I listen to speaker, and the result show my comfort region is also 65-70db.

oluvsgadgets

I also think 85dB is way too loud for listening although it’s regarded as standard in the studio, it’s simply too loud for me as well. I will also do precise SPL measurement to have some values and also give you an exact playback volume for my own recordings.