[RPGWiZaRD]

Quote:

 
There is a reason why more expensive gear tends to be flatter proprtionally to price. Have you tried to equalize your 'phones flat? If not, please give it a try, say, 3-4 days and then judge. I liked the music I had on my headphones and speakers but after applying some equalization I like it even more. Why? Because I get more detail, sounds are more accurate to each other, better soundstaging aswell. After seeing what was lost in the colouration I now enjoy my music much more, sometimes I turn EQ off and get a feeling "Heck, how deaf was I to enjoy music this way, bottom is boomy and highs are screaming, I loose sooo much of the music itself".

Yes I've tried EQing my headphones flat and I didn't find that optimal sounding for my taste and slightly boring to listen to. I do usually balance it out yes but don't make it completely flat as with a flat response I'd find bass slightly lacking vs highs being slightly too prominent and mids/vocals possibly not quite forward enough in the music how I personally for my taste find it optimal sounding. I agree about the improved detail recognition though, that's mostly why I bother with EQing for starters to take care of excess upper bass bleeding into mids or shrilly/excessive sounding highs that makes it somewhat thin/analytical/cold sounding or recessed mids that makes the music sound like it's missing "weight" etc. My preference has even changed over time from when I just got my first somewhat decent headphone until today, at first I preferred a very V-shaped frequency response as that's how my EQ settings used to be, today I can barely understand how I used to think those settings could sound nice.  
 
Quote:

You prefer to hear music the way you like but what about hearing it the way it was intented to, how can that be overrated? Nothing really personal but your method has a weakness that disqualfies it from the very begining- it has no reference sound you could tune to. You are blind in your tuning since what may seem to you desirable at first is usually just adding more colouration. That is why it is so important to first get rid of biggest overexaggerations and only later optionally apply taste correction. I myself did this on my 'phones: I first adjusted the reponse to as it should be according to sinegen test and after listening to music changed it slightly.

 
That's where you are wrong, I'm not blindly tuning, after lots of EQ experience and different headphones I've tried with very different sound signatures I've now got a very good sense of how I personally want it to sound like. That's why I like Sony XB500 so much for example and would pay up to like $200 for it as it's already got that kind of sound signature that is to my liking, a somewhat slanting response curve from lows to highs where bass has the biggest emphasize and mids have a decent boost and highs are slightly veiled. However it's too exaggerated in this sound signature, bass is too dominant (especially the upper bass that bleeds over mids) and mids and highs are quite a bit veiled in comparision to the bass so I boost both mids and highs to make up for it. But highs would require significantly bigger boost than mids to make it completely flat but I still don't boost the highs more than the mids or just a bit more than that but with my EQ settings the highs are still not quite on the same level as mids are and mids aren't quite on the same amplitude as strong bass tones are but this is still how it sounds optimal to me, if I make it more flat sounding than this it'll start sounding worse to me. Some spend $1000+, others are equally satisfied having spent less than $500 if they happened to pick wisely and picked up something that is closer to their personal preference in sound.
 
Frankly I don't care how it was intended to sound like, I'm just looking for the greatest possible satisfaction which I get when it sounds the way I want it to sound like. Why would you "pretend" to like something else better just because some1 else says that's the way to listen? I thought we all had in common to try and get the best satisfaction out of our hobby which I personally believe is when we truly find a headphone that sounds the way you prefer it to sound like and why there's such a wide range of headphones with very different sound signatures and that itself is a good point about my way of looking at this, since we all have our slightly different taste of how it should sound like. If it wasn't like that then we'd all be using the same headphones and there wouldn't exist such a diversity of headphones as it does today.
 
Quote:

 
Believe me as I went through this myself, your ears as well as mine and many other fokls' lie. You never really know what is bad untill you hear it right, which is why giving more or less flat equalization a try is so important.
 
PS. All the above mentioned arguments apply generally to music with real instruments. If your preference is towards techno/trance/house or alike then yes, you don't really gain much if anything.

 
Yes your ears will often lie until you've become experienced enough and tried several different headphones and EQing, it's definitely often the case that you wouldn't think something could sound better until you hear it. However when talking about EQing we're mostly talking tonal balance, tonal balance can improve SQ in the sense if there's too big derivations from the flat response especially in certain ranges but headphones can still sound bad even if they are balanced which is what you especially pay for in higherend headphones, the sound quality is just better regardless of their frequency response balance, in fact many expensier headphones aren't really balanced either, many expensier headphones still exaggerates the highs especially and take for example Ultrasone Pro 900 which isn't particularly expensive but still expensier than most headphones and is very unbalanced with exaggerated bass and recessed mids and exaggerated highs yet it doesn't sound very bad at all.
 
I disagree about the real instruments vs computer generated sounds. EQing will benefit any genre, I often even listen to pop or trance when EQing my headphones but I usually also check with other genres with real instruments but I've EQ'd for years and probably spent more time on EQing than most people around here cuz I'm a tweaker that just needs to fiddle around with all settings whether it's audio or video equipment and want the best possible result I can get. For example I spent like 2 years finetuning Sennheiser HD212 Pro with very small changes, XB500 I've kept tweaking for like 8 months etc. Doesn't really matter which genre I listen to I'm still usually able to tell how it will sound more or less to my liking (which then sounds great with the other genres too).
 
PS, I love arguing. So just try me, I will always find counter-arguments as this topic we're discussing is very subjective so there's no definite rule to apply here. ^^

 

[MaciekN]

Personally, subjectively, for me change from unEQ to EQ towards (yes, not fully) flat was almost as big a jump as moving from outputting sound from sounblaster to an external DAC. You have your taste, you have some experience as you have tried some headphones and settings, flat included, so you know how it can sound and how you like it. What can I say more, to each his own then.

 

[Vkamicht]

Just wanted to add that I've finally (after a couple months of trial and error, constantly tweaking) arrived at a point where everything sounds right. After spending hours away from my headphones, I come back and sit down and it still sounds perfect, the sweep is near perfectly flat, the tonal balance of bass, mid, high is perfect, and... wow. That's really all I can say. Audio bliss every time I put these headphones on. I think this is about the closest sound you can get to a direct feed into your brain, bypassing your ears (isn't that kind of what we're doing? fixing all the flaws in both the headphone as well as our physical hearing?) Everything sounds so smooth, better-than-real-life, not fatiguing in the least, not boring in the least. Just amazing. PiccoloNamek is my savior for starting this thread.
 
I've also learned a lot about what different frequencies do to the sound. My EQ has gone through many phases, I've saved "snapshots" of various dates for when I felt like I had a really good EQ curve, but now when I go back and listen to them I can hear many different flaws. Some of them sound much too "soft", which would be 4-6khz range dipped down. Some of them sound smileyfaced. Some are very fatiguing to listen to because I tried to flatten frequencies above 14khz but you have to be careful here, because it should never be uncomfortable to listen to. It's a tough process - you have to make a judgement call if you compare, for example, 1000 hz vs. 8000 hz on whether they sound equal. Higher frequencies tend to sound louder when they really aren't, it's just a sensitivity thing.

 

[jerg]

Quote:

Just wanted to add that I've finally (after a couple months of trial and error, constantly tweaking) arrived at a point where everything sounds right. After spending hours away from my headphones, I come back and sit down and it still sounds perfect, the sweep is near perfectly flat, the tonal balance of bass, mid, high is perfect, and... wow. That's really all I can say. Audio bliss every time I put these headphones on. I think this is about the closest sound you can get to a direct feed into your brain, bypassing your ears (isn't that kind of what we're doing? fixing all the flaws in both the headphone as well as our physical hearing?) Everything sounds so smooth, better-than-real-life, not fatiguing in the least, not boring in the least. Just amazing. PiccoloNamek is my savior for starting this thread.
 
I've also learned a lot about what different frequencies do to the sound. My EQ has gone through many phases, I've saved "snapshots" of various dates for when I felt like I had a really good EQ curve, but now when I go back and listen to them I can hear many different flaws. Some of them sound much too "soft", which would be 4-6khz range dipped down. Some of them sound smileyfaced. Some are very fatiguing to listen to because I tried to flatten frequencies above 14khz but you have to be careful here, because it should never be uncomfortable to listen to. It's a tough process - you have to make a judgement call if you compare, for example, 1000 hz vs. 8000 hz on whether they sound equal. Higher frequencies tend to sound louder when they really aren't, it's just a sensitivity thing.

You, sir, have walked the rite of passage, and turned into a master equalizer.

 

[Vkamicht]

Quote:

You, sir, have walked the rite of passage, and turned into a master equalizer.

 

 

[Zida]

care to show us an image of the curve you're running? I think it would be very interesting to see

 

[Vkamicht]

Yeah, here you go. It should be somewhat different from the last one I posted
 

 
Despite my claim of "perfection", nothing is perfect, I'm just ultimately satisfied with it. There are still a few things that I could fix: above 14khz could still use tweaking, well-mastered music that doesn't roll off the highs can sound brittle sometimes (but not often) It's somewhere in the really high frequencies where you feel more than hear them. Also a channel imbalance around 10khz where the sound suddenly skews to the right side then comes back around 12khz.  Aside from those things, however...
 
The huge peak up to 20khz wasn't something I did manually, it's just a byproduct of lowering everything else. And remember Voxengo automatically adjusts curves 0 line so these aren't the exact numbers.

 

[soldermizer]

Interesting tutorial.  I am still very new at this (headphone EQ-ing).  I found an outside link that has a lot of information -- very eclectic.  Assuming it's true (always a risky proposition on the Net), the interesting point is that even experts disagree over such things as loudness curves and the proper way to measure the response with a real or dummy head.   What I get from both the above tutorial and the cited link, is probably the best thing to do is to tame the obvious peaks (as cites here) and then perhaps, if this is a valid goal, to equalize subjectively equal loudness with per-octave (one-third-octave?) pink noise.   One trouble I already see with either a swept sine or pink noise is your own hearing and/or phones may drop off considerably at the very top or bottom.  my hearing seems to poop out above 13K Hz for example.  Perhaps the old rule of thumb (ear?) is to bring down the peaks before (indeed, if ever) trying to fill in the valleys.
 
This hobby offers many fantastic time-wasters.  For example, if you have built your own mic's, how can you be sure they're calibrated?   For those of us ("me") whose patience is on a short fuse, the resonant peak search & destroy and (one-third) octave subjective loudness matching neatly sweeps the problem of objective measurements under the rug [grin.]
 
http://www.slideserve.com/presentation/23730/Binaural-Hearing-Ear-Canals-and-Headphone-Equalization

 

[MaciekN]

You can remove resonances either by EQing them, or improving/modding your headphones's housing. Stiffen them and add some damping and you are done with most of the EQing probably...

 

[AlexRoma]

             Sennheiser HD600  (HD650 cable , no mods applied)
 

 
Freq    Gain(db)      BW 

  30          8          1,5 
 100        -1            1
2000       -1            1 
3000      -4,5          1  
4300        1            1
5700      -5,5         0,3 
7000       0,5          1  
11111     0,2         0,5 
16000    -0,8         0,5 
20000     3,5           1
 
This EQ is supposed reducing HD600's sweetness (that leaves an impression of the singer smiling all over the time).
After applying it, HD600's become more analytical and correct, but still, warm and pleasant.   

 

[Assimilator702]

The HD600 looks pretty good up to 3k.
 
Not exactly flat down low but thats neccesary for any low end presence.
 
If you look at most high quality speakers past 2k thing get VERY erratic.

 

[AstralStorm]

 
Quote:

The HD600 looks pretty good up to 3k.
 
Not exactly flat down low but thats neccesary for any low end presence.
 
If you look at most high quality speakers past 2k thing get VERY erratic.

 
Not true, good speakers try to stay linear there, within 3 dB at the very least.
The very best are flat with excellent extension both sides.
 
The main problem is that you gave to measure with a corrected capsule in an anechoic chamber to really know, and that tells you nothing about response in a room.
However, in case of headphones, higher mids and up are affected by head response.
In case of IEMs, insertion depth has a profound effect too - pushing the ear canal resonant frequency higher.
 
Usually pinna has *cancellation* in range ~3-5k, ear canal has *resonance* between 6-10k. Unfortunately, nobody makes linear headphones or IEMs, so it's pretty hard to find this out precisely.
 
Quote:

You can remove resonances either by EQing them, or improving/modding your headphones's housing. Stiffen them and add some damping and you are done with most of the EQing probably...

 
Yeah, except how do you know what frequencies will your treatment correct? It might actually worsen something else instead. Also "stiffen" how, "add some damping" how.
EQ is predictable and simple, even in analog passive circuit form.
 
---
 
By the way, don't trust headroom graphs below about 100 Hz. They're completely inaccurate there.
Best way to equalize is by ear, with many tones. I find pink noise not precise enough. Of course finding the correct curve shape given point measurements to minimize number of parametric bands takes practice.
 
My preferred way is to drive ElectriQ with Foobar2000 tone generator at -6dB, set volume so that 1 kHz tone is just slightly annoying (should be equivalent to 60-70 dB).
Takes a long time, but totally worth it. Find maximum boost first, readjust volume in dB (before EQ) to avoid clipping, adjust analog output volume to compensate so that 1k tone is still the same volume.
 
Mode in ElectriQ doesn't matter all that much, unless you happen to build an unstable filter. Smooth changes in phase are very hard to notice with headphones, especially at higher frequencies. Might be more noticeable with multi-driver designs.
For me, flat-sounding equalized doesn't sound boring in the least.
 
Sample eq by me, for one pretty mediocre Sleek Audio SA6 (single wide-band balanced armature IEM). No, I have very good fit and seal with those.
As you can see, they're pretty rolled off.

51105106

 

[MaciekN]

Quote:

 
 


 
Yeah, except how do you know what frequencies will your treatment correct? It might actually worsen something else instead. Also "stiffen" how, "add some damping" how.
EQ is predictable and simple, even in analog passive circuit form.
 
51105106

You can not go wrong by improving the drivers housing. You remove/reduce resonances and distorting backwave reflections, things that would require very precise EQing, even though some things just can not be done by frequency adjustment. When you equalize by ear all your work is dependant on the exact gear you used to EQ. Change cable/amp/pads and all your work is wasted.
Stiffen by adding mass/replace with stiffer material. Dampen by putting soft, breathing, yet dense material (cosmetic pads are just fine if you want to try) behind the driver, exact location depends on type and construction.

 

[YTCrazyTieGuy]

Cool stuff!
 
When listening to flat speakers are we supposed to hear a flat response? maybe I just didn't understand it correctly but aren't different frequencies of the same amplitude in dB supposed to be percieved in different amplitudes?
 
Thanks!

 

[jerg]

My M50 settings are getting crazily complex! It's always small tweaks for the better though. I've gotten all the small precision detailed changes, now it's all just broad tonal shifts for the ultimate goal of perfecting the sound to my own ears.