Easy DIY way to test headphone accuracy relative to one another

Subscribed.  Good topic, I look forward to hearing other's thoughts on you trying to dethrone the syrupy-ier-ness. 

Sean Olive, of Harman Kardon's research arm, maintains an interesting blog that has a number of informative takes on comparative testing of speakers and headphones. In short, HK employs a squad of trained listeners in a reference room with a visually obscured speaker turntable. While blind testing wouldn't be very effective with headphones due to the obvious differences in their weight/feel/etc., the listener training program does provide a creative and empirically tested solution for SQ testing (http://seanolive.blogspot.com/2010/12/how-to-listen-course-on-how-to.html). They have an app available that could benefit a DIY SQ reference-based testing regiment @ http://harmanhowtolisten.blogspot.com/2011/01/welcome-to-how-to-listen.html .

http://www.linkwitzlab.com/reference_earphones.htm

The "compare by contrast" is a great idea.  The eq method you propose seems too have a few too many possible notching passes for me. The article suggestion of random tracks is great, it lets any recording or source (even bad ones) have the role of a reference. 

Hmm, very interesting. Wonder if there are other threads like this one...

Very interesting as well. Bias, could you publish your test results in more detail. Thanks.

I just discovered this thread.

bias has a special place in the origin-story of the Orthodynamic Roundup thread, so I thought I'd throw out some very random thoughts here.. which will probably be all too familiar to longtime orthonauts.

The process bias describes is what I've been calling "educating the ear-brain". With real sound in real rooms as a lifelong reference, we all know what reality sounds like-- doesn't matter if our ear-brains are golden or leaden, we all have the same reference. Eventually, over time, all these colored transducers we've been hearing all our lives will, even if only on a subconscious level, start to sound wrong somehow. But the process is slow. It takes doing a lot of listening, remaining skeptical even of your own evaluations and reactions, staying humble because you know how (and how easily) you can be fooled, and constantly testing, probing, poking everything. Eventually, slowly, painfully, in a kind of seeking oscillation around the line of flat response, we realize bassy is wrong, bass-lite is wrong, bright is wrong, "lush" is wrong, "full" is wrong.. no matter how pleasing they may sound at the moment. Eventually we seek the middle ground where the headphone sounds more like the recording than itself, and we keep on getting better at recognizing this middle ground, this center point, without being fooled (by ourselves, by anyone) so damned often. If we can keep ourselves from investing our emotions heavily in an ammo box of strong opinions, the process moves faster.

Eventually, slowly, we can tell when a recording's been manipulated to sound a certain way, as nearly all of them have. The headphone that lets us tell this most easily is perforce the most accurate headphone, but it's not something that we can use to shout from the rooftops or clobber one another on forums with. That headphone we defended to the death five years ago got quietly left in a drawer last summer without another thought. You've already listened to it for the last time and you don't even know it.

Call it the evolution of personal taste if you want. To me that sounds too elitist, because the process is pretty much universal, far as I can tell. It will go on whether we know it or not, whether we act on it or not.

I should point out that it's not necessary to be aware of it or act on it. There's a place for the inaccurate but pleasant headphone, just like there's a place for macaroni and cheese. Not everyone needs or wants a more-accurate headphone, much less a most-accurate headphone. Most people don't need or want to hear everything that's on a recording, because they're not professionally involved. They seek an emotional experience, not an accurate one, and anyone can get that from a medium as limited as AM radio. So we shouldn't get vehement or adamant about any of this. All I'm saying is that if we DO seek accuracy, we can train our ears to hear it, but it's a very slow and inconvenient process with plenty of public embarrassments along the way, and it's good to have made recordings oneself and to realize that most people don't even want to know it's a recording.

Having said all this... I know we want quick ways to establish accuracy because measurements and the ear-brain training are so @#%$! much of a PITA, but wouldn't the notch at such a high frequency, where cavity resonances are starting to show up, and where the ear is less sensitive to level variations, simply favor the headphone that has more output at the notch's center frequency?  ... Anyway, thanks for reading all this. These are only my opinions, based on my own stupidity and facepalming over the decades.

Interesting will go around conducting the test with my mic in the future.