Measurements: Charts, Graphs, Software & Methods



Nuraphone is weird, because when I first heard their pitch (“We stick a thing in your ear and measure to get better sound”), it made perfect sense to me: personalization of headphones to individual HRTF is almost certainly the future (Smyth’s Realizer being an extreme example), and so I figured that was what they were doing. Looking closer at how they say it works, however, apparently they use otoacoustic emissions as their major (sole?) test. I’m not an audiologist, so perhaps there’s some nuance I’m missing there, but my understanding was that otoacoustic emissions can be used to determine hearing damage, rather than HRTF. I guess compensating for hearing loss is a pretty useful feature in and of itself, but it’s not what I’d expect.


Thanks for the annotated exposition. I found it both useful and interesting. As I’m known for being entirely irreverent (and yes irrelevant too) on this forum, might I suggest that there may be a market - as there is in certain other products - for some celebrity ear models?

Imagine testing on a molded set of Beyonce’s ears. Or, for that matter Eric Clapton’s?

I’ll crawl back into my box now.


In a sense, our ear sims are already based on celebrities - albeit within this nerdy field - as Gunnar Rasmussen and Per Brüel’s ears were used to make the impedance target for the first IEC711 sim. Still, I think pinnae probably have more market appeal, particularly for the binaural folks - forget “as the artist intended”, now you can have “exactly as the sound entered the mastering engineer’s ear”!


Great post @Mad_Economist! I really appreciate the post, data you’ve accumulated, and citations.


I’ve been playing around with a Dayton IMM-6 and Audessey Calibration mics for IEMs using 2mm wall thickness vinyl tubing and REW.

I’ve been making some calibration curves for Diffuse Field and Harman Target IEM curves and here’s my current results:


Moondrop Kanas Pro

I used MiniDSP’s Diffuse Field calibration curve to normalize the Dayton data to Diffuse Field and then used ER4 data to normalize to Harman Target. Probably not exact in anyway, but it’s a start. I’ll probably be tweaking it a bit more when I have time.

Obviously, I made a lot of generalities of the Harman curve response.


I compensated incorrectly to ER4-XR instead of ER4-SR. I fixed that in latest photos.