LOL (to the whole thread), sorry.
It was so depressing and entertaining at the same time. esp the earliest pages.

(hurray for the outspokenness of the pirates and a bottle of rum).

To say thank you to all (for everything),  I will share with you a very sweet and hot tip: Take your ESP-950 down (from the shelf) and listen to it with  a 5-6 dB reduced level around 1 kHz, the broadness of which should be some 1½ up to 2 octaves (depending on your tastes), and also turn the level of the bass slightly up. (do what you want with the treble, personally I do nothing )
Now, you say, who would care what an obscure fellow (of utterly negligible fame)  says about improving some phones through EQ-ing?  I agree completely.  Who would care? Indeed!  However,  this is different!
 Not only will the sound from these cans suddenly seem much less muddled,  but one will easily achieve a certain kind of highly sought after  "canvas of sound" on with big and small details seem to be painted in good proportion to each other, and with all the delicacy, color and depth one could want.  A kind of sound it is easy to get lost in ( in a good way, I hasten to add).  Combined with their famous low weight and top-notch comfort, it is difficult to think of anything better  (at least for classical and stuff).
 Btw, I use it with transformer step-up units, but it is not essential for getting  a good result.

A few words about the need for precision in EQ-ing. Achieving perfect flatness of response is a great thing! Fantastic and  gobsmacking etc. But one does not really need it. Recordings are all over the place, so are our ears.
Take  a big collection of records. I use to think of the individual records and their tonal balance as represented by stars in a globular cluster. If you have  a frequency response  from your  headphones which most people would tell you was flat, this would make perhaps 60 percent or the recordings in the collection sound very well balanced (  Since we do not notice small deviations. Only perhaps some 2-3 %  would be really close to the ideal in absolute terms).  Now think of  these 60% of  recordings as  stars being inside an imaginary  circle drawn on the globular cluster with its center in the middle of the cluster (never mind the black hole in there somewhere).  Now take a slightly less perfect EQ-ed headphone. To represent it you will have to move the circle  a few light years in some direction, say for example to the right. It turns out that millions(!) of stars that previously were outside the circle on the right side now have miraculously turned up inside it! Of course, an ever bigger number of stars are lost on the left side, since that area is closer to the  more heavily populated central region, but still! As a result, say that now  perhaps 58% of the recordings will sound well-balanced instead of 60. Only a couple of % less, big deal! (Still a star performance so to speak). And the nearest 15-20% will sound quite OK anyway. The cirle would probably be somewhat deformed (in real life), but I left that out, for simplicity.
I have also looked away from the distribution of cosmic dust in the cluster. Only with LP records dust might become an issue, I think.....

Well, this is the truth about EQ-ing, in astronomical terms and as I see it. I have written this because I think that many are afraid of getting lost in EQ, not having access to measuring instruments and such.  But don't be afraid!  Any move in the right direction is a good move. If you could live with a bad result, you can always live with a better result.
So, in the case of the ESP-950,  if you turn the level at 1 kHz down  5 instead of 6 dB, it does not  really matter. Seriously.
Cheers, Olaf




 

So... in the end it appears the good old HD600 was already close to the newly discovered Olive-Welti target ?
Assuming the K812 is close to the O-W target of course.
This means a headphone should measure as flat as possible (after the needed corrections) and not have lifted bass or rolled-off treble.

Based on my own observations, that a 'flat' measuring headphone sounds the most realistic, this doesn't come as a surprise for me.

Hm, yeah, with the extra harmonic distortion and more elevated, less smooth treble of the K812 vs HD600 (kinda hard to tell with that scale), I could see the K812 being problematic for some people.

The OW curve is basically JBL's synthesis demo room with a touch higher bass and a tad lower treble. That doesn't necessarily mean tilt (as you correctly pointed out for the 1M time, Marv). As measured at the eardrum it looks like what Tyll posted in his article. As measured on the gear we're seeing measurements from on this site? Who knows what that looks like. I have no idea if we need a mids hump or a 7kHz trough, or what have you.

The two biggest deterrents for me with the K812 (which I haven't heard) are price and Mupps saying it's too easy to listen loud. I'm a quiet listener and want to preserve my hearing.

As it stands, I have tuned the EQ for my other cans to closely match the curve and I love it. It still needs a few tweaks, but it's really good.

Corrected measurements (due to calibration issues):

1) http://www.changstar.com/index.php/topic,1482.msg39235.html#msg39235

2) http://www.changstar.com/index.php/topic,1482.msg39288.html#msg39288

Only if you consider 5db from 5kHz - 12kHz "close".

The AKG certainly sounded a little hotter and harsher to me (DT990 still takes the cake). Here is a little closer look using the latest AKG measurements (red-K812 / yellow-HD600):



As in the previous measurements, the channel imbalance in the treble region (right louder) was present in this second set of measurements for these particular cans: