I'm discovering a new world of headphones. My ATH-AD700 arrived today 
 
This is the raw measurement data of the headphones:
 

 

 
 
And after EQ:
 

 
These are temporary notch filters while I'm planning to build a headphone amp (any recommendations?). But they're good enough I think.
 
The notches are here: http://gainphile.blogspot.com.au/2012/05/audio-technica-ath-ad700-correction-eq.html
 
Later on I will DSP or build an ASP of the filter.

its not exactly smooth, at all, could behave strangely

Gainphile! I recognize you from your speaker website. I've been reading it a lot lately as I'm preparing to make a simple 2-way OB speaker as my first speaker build.
 
Have you considered the match to your HRTF when it comes to EQ'ing the headphones? If you make a simple in-ear mic from the same Panasonic capsule you'll be able to use your excellent speakers to measure your HRTFs in free and, by power averaging, diffuse field. You can either directly EQ the headphones or make an individual cross feed from the data.
 
I've already experimented with measuring my HRTFs but I only have rubbish speakers and need to limit volume in my apartment building. Room EQ Wizard is very convenient because you can measure the speaker at some distance (mic in air) and make a faux 'sound card calibration' from this. Then when the mic is in your ear the mic+speaker system has already been calibrated away.
 
I wholeheartedly recommend the O2 headphone amp by NwAVGuy. Extremely well thought out and tested. The design is so good that my amateurish build holds up well to spec (measuring through my sound card line-in).

Quote:

wir3d said:

its not exactly smooth, at all, could behave strangely

Click to expand...

 
This was my first time of measuring headphones, and I must admit I did not expect such non-flat response above 1khz.
 
However a quick glance to other independent measurements it seems this is quite normal. Perhaps due to them being basically 2" fullrange speakers with expected cone breakup mode.
 
Here is Stereophile's measurements on AKG K530, Audio-Technica ATH-AD700, Beyerdynamic DT440, and Grado SR80. They all have typical breakup response above 1khz.
 
http://www.stereophile.com/content/between-ears-art-and-science-measuring-headphones-page-3
 
e.g. Grado SR80
 

Quote:

gainphile said:

 
This was my first time of measuring headphones, and I must admit I did not expect such non-flat response above 1khz.
 
However a quick glance to other independent measurements it seems this is quite normal. Perhaps due to them being basically 2" fullrange speakers with expected cone breakup mode.
 
Here is Stereophile's measurements on AKG K530, Audio-Technica ATH-AD700, Beyerdynamic DT440, and Grado SR80. They all have typical breakup response above 1khz.
 
http://www.stereophile.com/content/between-ears-art-and-science-measuring-headphones-page-3
 
e.g. Grado SR80
 

Click to expand...

whaaaaaaaaaat, now that is just crazy bright. Like not even funny bright. Thats certainly not good for you
 
I know almost all headphones go nuts at the higher end. Whats strange to me is how nuts yours are going at the lower end. I am quite interested to see how it sounds, but like I suspect, it could be strange.

Quote:

symphonic said:

Gainphile! I recognize you from your speaker website. I've been reading it a lot lately as I'm preparing to make a simple 2-way OB speaker as my first speaker build.
 
Have you considered the match to your HRTF when it comes to EQ'ing the headphones? If you make a simple in-ear mic from the same Panasonic capsule you'll be able to use your excellent speakers to measure your HRTFs in free and, by power averaging, diffuse field. You can either directly EQ the headphones or make an individual cross feed from the data.
 
I've already experimented with measuring my HRTFs but I only have rubbish speakers and need to limit volume in my apartment building. Room EQ Wizard is very convenient because you can measure the speaker at some distance (mic in air) and make a faux 'sound card calibration' from this. Then when the mic is in your ear the mic+speaker system has already been calibrated away.
 
I wholeheartedly recommend the O2 headphone amp by NwAVGuy. Extremely well thought out and tested. The design is so good that my amateurish build holds up well to spec (measuring through my sound card line-in).

Click to expand...

 
Thanks. I have not considered HRTF, just a mere correction of Frequency Response. I'm pretty new at this.
 
But for sure HRTF and also Equal Loudness Contour plays a big part on how the sound is perceived. Not exactly an easy subject to tackle at once too ! 
 
I'm interested in having a microphone at the ear.... how exactly did you do this ???
 
I will study the O2. Sounds like a popular product ?  I just found out an op-amp is enough to drive the headphones (cMoy etc.). Interesting as it makes the EQ-amp integration very easy.

Quote:

gainphile said:

 
Thanks. I have not considered HRTF, just a mere correction of Frequency Response. I'm pretty new at this.
 
But for sure HRTF and also Equal Loudness Contour plays a big part on how the sound is perceived. Not exactly an easy subject to tackle at once too ! 

 
I'm interested in having a microphone at the ear.... how exactly did you do this ???
 
I will study the O2. Sounds like a popular product ?  I just found out an op-amp is enough to drive the headphones (cMoy etc.). Interesting as it makes the EQ-amp integration very easy.

Click to expand...

 
The kind of setup that I've used has been detailed by bluemonkeyflyer in this post,
http://www.head-fi.org/t/452404/just-listened-to-some-fostex-t50rps-today-wow/8025#post_8354423
We've exchanged a lot of communication in our Fostex modding. The mic is just a WM61a capsule in a foam earplug (or IEM tip). Blocked ear canal measurements have been shown in the literature to be more repeatable than open ones and should be reliable for open headphones. 
 
Here's an example of how the AD700 sound on my head to the capsules. 

The dip at 3 kHz certainly shouldn't be there. I should see a smooth rise.
 
 
As far as HRTFs go, the important parts are - 
 
- In the free field there is gain from ~1.5 kHz to a peak of ~15 dB at ~4 Khz. Depending on the individual this drops to a dip at 7-10 KHz caused by outer ear reflections.
- The positions of peaks and notches at high freq are angle dependent, so a power average tends to wipe them out. The more diffuse the sound field the more bass-midbass and the less pronounced the high freq features. I can post later on what the FR equalization of dolby headphone looks like.
 
RE: amping, I can measure the bass response of the AD700s from my O2 amp and straight from my (Xonar) sound card, if you would like to see how an amp will clean up their FR.
 
Psychoacoustics are crazy fascinating. I think you'll have a fun ride.

It turns out that the headphone can be equalised with passive EQ. This makes things very easy.
 
http://gainphile.blogspot.com.au/2012/05/audio-technica-ath-ad700-correction-eq.html
 

Quote:

gainphile said:

It turns out that the headphone can be equalised with passive EQ. This makes things very easy.
 
http://gainphile.blogspot.com.au/2012/05/audio-technica-ath-ad700-correction-eq.html
 

Click to expand...

i wonder how that affects SNR and THD

Quote:

wir3d said:

i wonder how that affects SNR and THD

Click to expand...

 
SNR:
The inductor induced 0.5db loss which is quite insignificant considering the headphones' sensitivity. This loss would affect the both the 'signal' and 'noise' so SNR would be the same.
 
THD:
This is quite a meaningless number by itself.
I have a full plot of harmonic distortion profile from 2nd to 4th order below. Except few areas where the resonances takes into affect they are all 40db below fundamentals.
This is way above what normally active electronics like amplifier induced (let's say -90db or -100db).
 
In summary, the distortion caused by the inductors are negligible.
 

Very very nice. Well done, and I'm guessing it also sounds good, and the soundstage, is it still intact?

Call it the AD700 sound rectifier, sell to people who complain about little bass.

Sent from my HTC HD2 using Tapatalk 2