These measurements look quite a bit different compared to the ones Tyll posted on IF, particularly the 6khz dip which shows up even on the uncompensated curve.

http://www.innerfidelity.com/images/NADVISOHP50.pdf

I have a pair as well, and from listening so far I don't hear any siblance at all and no lower treble emphasis, in fact it sounds a bit shelved starting from the upper mids. It's presentation is a lot warmer than say my Paradox, which some people say have a slight low treble peak. Even my HD580 had more treble presence all throughout.

I actually wonder if they've been retuned a bit since the earlier batches. Pretty certain I'd be able to hear that 5khz peak, considering the ringing. I'd provide a serial number but I can't find any. Thanks for posting!

OH I just remembered something.

@ultrabike: What cables were used when performing the measurement? Was it the stock cable?
Because I remember switching cables, and it produced strange effects due to the headphone being dual entry TRRS. When I used a normal TRS to TRS interconnect with the pico and on my clip+, the mids got recessed but the sound was still alright. It was only when I really pressed the plug in with significant pressure that the sound normalized.

Right now I'm using the V-Moda one button cable for TRRS jacks, and the not very good quality stock cable for normal 3.5mm jacks like those on PMPs and amps because the only TRS to TRS 3.5mm interconnect that works fine has been a really stiff audioquest cable.

You're sure you would? This peak is 4 dB, less if you take 500 Hz as reference, and less than bass boost.
The peaks are wide resonances, so it should sound less like razors of death, more like room reverb.
The mid hole disappears after a short time delay, so does the bass boost.

Do you have any pics of those?

Also, I haven't seen any midrange blip in my Paradox - there is a high treble "less cut" region around 6k.

Pics of what? Good point regarding the wide resonances at 5khz, I should test that out first because I'm not sure what that would actually sound like. I imagine that dip at 3k combined with the peak at 5k would result in a distinct sound for the headphone though, one that would be easily audible.

I did a quick comparo again, and the sibilance is pretty minor actually. Probably somewhere in between HD558s and KSC-75s. though not offensive seems actually higher than my KSC75s. Dunno if its the combination of roll off around the upper treble and mids that accentuate this for me. The bass on these is really nice and tight however. I did feel it had some mids depression. But IMO these are not lower treble lacking. A little emphasis there, but I think it adds excitement for lack of a better word.

Here are the impedance measurements which seem to agree well with IF. Seems mostly resistive in the audio range:



I unfortunately only have two cables available for these (the ones w and w/o the volume control) and using either did not change the FR much. However, what did change the FR a little was which cup the cable was attached to. This was repeatable (multiple times, even if I use one or the other cable). Is a fairly subtle change between ~250 and 600 Hz of about 1 dB. Cup re-positioning might give more variation that that.

FR of right channel with cable attached to different cups (light blue is right entry, light yellow is left entry - both are right channel):



Compare FR with KSC-75 (right channel yellow):



Compare FR with Alpha Dogs (right channel yellow):



Compare FR with HD558 (right channel yellow):

Nice comparo !

Thanks

The levelplot differences with the cable swap is a bit suspect....

The 1.5dB difference around 400Hz (as well as 40Hz) cannot come from the cable as:
A: the impedance is almost linear there.
B: The calculated level difference between the 2 cable 'sections' is around 0.025dB, would be between 0.022dB and 0.0255dB depending on the frequency.

Yup. I don't think it's the cables either since I got the same results independent of the two cables that came with the HP50s. It's likely something in the headphone end... connectors, internal wiring, smurfs, dunno.

Here is a new set of FR measurements (like 10 minutes ago - right driver only) where all I did was go back and forth between connecting the cans through the right cup or the left cup (the NAD in it's awesomeness allows you to use either cup to plug the cable):

2nd time around:



3rd time around:



4th time around (zoom in):



5th time around with the other cable w volume control (zoom in) - noisiness on the left cable entry is background noise:



It looks a little worse in the original post above (#5) because the FR plot is a little taller there, but the magnitudes seem consistent.

It's really splitting hairs different, and really don't think this is an issue at all.

Puzzling right? If I find something I'll post about it. Could be wrong, but don't think it's a problem.

Anyhow, here is the original impressions linky:

http://www.changstar.com/index.php/topic,1211.msg32563.html#msg32563

I think Anetode is right on the spot on the 3-4 kHz weirdness. Definitive good bass extension, and impact is there. I also felt instrument separation was quite good. The clamp is tight, but my head is probably on the narrower side of things, and me likes creamy leathery pads. Again, very good set of closed cans.