Nectar Hive Electrostatic Headphones measurements and discussion

Let me first start from a set of disclaimers

• The measurement was taken from my own pair of Hives. I paid the full price with the self-justification to support small makers (pretty convincing excuse, isn't it? haha)
• Due to the small scale of handcrafting manufacture, these headphones can be subject to product to product variation by nature. While Sajeev ( @qu1en ) seems to try his best to maintain consistency, it's a better practice to take all my arguments with a grain of salt not ruling out I happened to get a unicorn by chance. Cool thing is Sajeev doesn't hide measurements on his side at all. Buyers will have some technical conversation with him while placing an order. He will show the measurement for your specific pair then.
• The primary purpose I re-measured myself was to display in a little sbaf-friendlier way as Sajeev adheres to use raw compensation with his rig.
• Any subjective points made in this post is based on the pairing with the Kgsshv carbon (See my sig for the full chain info) built and loaned by a hobbyist builder. I also heard that several advanced suggestions (details omitted as clearly not my bag) incorporated in this amp to max out. As I do think Hives' sound is highly dependent on how they're driven, readers may pay more cautions.
• Again, I really and really like these headphones (at least with the current pairing). So, feel free to adjust/disregard my over-enthusiasm if any.

Notes on frequency response

• As mostly shown in my measurement posts, I don't necessarily agree with what EARS results suggest to me. This time, however, at least for FR, I think I hear Hive very closely to the measurement as displayed in the plot above.
• Slight and wide peak by 1~1.5db around 30hz on the left channel is due to imperfect seal. The above plot is computed from the average over 5 data points (a little more tries than usual as I didn't believe the result at first -- much better than expected) and I couldn't give enough care for the first two measurements. In later tries, things were better matched across channels, which converges to the right average.
• I achieved very similar bass response in the HE6SE measurement. And interestingly, I also found some subjective similarities between the two products. I'd say Hives can reproduce 85-90% of HE6's incisiveness, transient, and slam in lows -- although I don't think such virtues measurable this way. And doubtlessly, there are also a few things Hives do better than HE6 in bass performance -- such as delicacy, articulation, details, nuances, etc.
• No or minimal (at worst) peaks around 1~1.4 khz. All the Stax estats (and some dynamics) I heard had nasal and tinny sounding which I suspect associated with this region. Even Koss 95X had a little bit such thing. This, combined with very flat response up to 1khz, seems consistent with very well seasoned vocal presentation (neither muddy, boxy, cuppy, nor hollow) among my subjective findings.
• 3 khz is a little nuanced and so is 5 khz (to less extent) . And they're hearable -- positively to my linking. Tastefully bright, clear, and moderately edgy. Maybe a little northen to what I truly prefer, but well below my comfort boundary.
• I believe one of the reasons why I took ^ very affirmatively is related to small dips at 2k and 4k hz. While I am not 100% sure if I like flat response in those regions, I'd definitely like to have slight dips rather than the opposite. Subjectively, Hives' presentation is a little on sweeter and thicker side to me.
• Maybe the most (potentially) bipolarized part -- good to me but I think some may hate -- could be a narrow peak at 8khz surrounded by deep narrow dip at 9khz and low-q one at 6.3khz. This works for me like a magic. Spot on sharp and resolving yet all the sibilance on upper highs well recessed. Nice sense of dark tone without sacrificing treble energy, too. I also believe having 9khz dip better generally as the most compensation curves do not address actual human HRTF in that region.
• I don't quite give much credit to EARS and any Dummy heads/HATS beyond 10khz. Subjectively Hives' top end and air were pretty good though (not very unique as it's one of typical estat strengths).

Notes on harmonics

• I don't seriously take harmonic measurements from EARS for a number of reasons. I am still finding out how to reasonably address 4.5khz measurement artifact. Moreover, as I said in recent measurement posting, my living place (PA) got way much noisier than TX. The algorithm how REW detects harmonics often fails to get solid results with noise floors above a certain threshold.
• That said, nothing negative really stood up in either plot. Ignoring a couple of spurious peaks at 30 and 105hz, everything else looks low enough.
• H3 also well below H2 all over the spectrum, which could be good -- for your reference, in this regard, stax estats behave well with flat couplers but not very well with dummy heads with artificial ears to my knowledge although I regard all such as measurement artifacts though.

Notes on spectrogram

• Also not very much to talk.
• One thing I noticed was ESP95X sounded like having more reverbs and slower decays than Hives (making 95x a little more dimensional and life-like). But my spectrogram result indicated the opposite (95X a little faster in decays).
• I couldn't find any good objective explanation for my subjective finding yet. Maybe things I want to figure out could be more complicated byproduct from time and frequency components together. But I digress for now.

Closing thoughts

• Overall, this is one of the best measuring *estats*.
• With the exception of Senn estats which I neither enoughly heard nor measured to date, all the popular models Stax launched didn't measure well (than they sounded to me) -- It applies to even SR007 that I thought fantastic subjectively.
• Hives are a little more consistent between good objectivity and subjectivity. Glad as a measuring dude. Great job. Well done Sajeev@Nectar!
• Will try to drop some comparative plots later. Struggling to find previous measurements after the backup storage was dead. Oh well.

Related links:

• Official site
• Subjective impression and discussion
• My own impression
• Sajeev's own measurements of sample Hives/Pollinator pairs

 

@Vtory didn't mean to dislike and it won't let me change it for some reason.

Great post and great measurements! I need to get more estats to compare to the hive but I have a suspicion the hives will pull ahead for me

 

Dropping a few comparative plots with subjective notes. Be aware of potential memory and confirmation bias as many assertions come from my memory.

To begin with there are three specific models I've owned before and I found them strongly similar with the Hive sounding respectively.



Like I said in the OP, Hives reproduce fairly close bass to HE6s (grill-off + focus-a pads) in quantity as well as quality. The plot above indicates FR doesn't much differ up to 400 hz.



For highs,quite a lot patterns shared in both headphones. The plot suggests Hive has a little more recession at 2 khz and 9.5 khz. Subjectively I found Hive's high was a little more likeable and agreeable. Note that HD600s will be closer to Hives as their pads get used and worn.



I also found some similarity with Elex in highs. But Hive doesn't have tinny and honky sounding in upper mids to me, which also shown in the plot. The proximity isn't that surprising after HD600 plot as Elex was once claimed as super hd600 (very arguable and not very accurate wording I think though).

It's interesting to see all these three models above have 9k dip in their own ways (but not to the Hive's extent). I'd repeat the assertion that I prefer to have fairly strong dip in this frequency region.

 

Headphones compared below aren't similar with the Hive by any means, but could be used as helpful data points to imagine Hive's tonal balance.



The invincible and long-beloved Verite. I still love my pair a lot and give constant and robust head time.

Plot-wise VO is a little more mid-bassy and more recessed over lower and middle highs. Interesting to see VO also has very strong 9k dip that I like to see.




Without any EQs, SR1a is a lot brighter, sharper, and ediger in typical configuration (I believe most users hear them with wings 30 to 40-degree open in azimuth). Note 0 degree means wings fully closed and 90 degree the opposite (wings fully open). As the product does not follow circumaural structure at all, comparison may require more careful caution regarding measurement limitations -- do not take too seriously.




The two products shown above have better looking downtilt in highs. And I indeed think both could be preferred in terms of darker tonal balance and timbre. However lack of 9 khz dip (and along with some other reasons) makes me feel a little awkward with either product (more so with Verum and less so with 95x). Verum-1's peaks and dips patterned in a little similar way with the Hive but not as much as HD600 and Elex I presented in the previous post.

 

Been a while since I posted my measurements and thought.

Updating this thread with an interesting additional data point. Below are my EARS+SBAF compensation and GRAS+IEF compensation results respectively. And I believe this shows some interesting points regarding similarity and discrepancy between the two combos (or add measurer errors/biases too!).





To my understanding, IEF compensation is basically to use Harman curve without bass compensation. Correct me if I am wrong.

Hive was particularly chosen because it was the most recently taken results on both my end and his.

Observations below.

• Both results mostly present similar and consistent peak and dip patterns. But amplitudes may look fairly different to the point face-value takers conclude differently. I'm using Hive as one example, but I've seen such cases very often.
• I hope nobody seriously takes 10khz and above. That's simply "nobody knows what actually happens" region.
• If one product is measured SBAF-neutral, than it is very likely to have boosted peaks in midrange and lower/mid trebles. Also IEF likely reports it has more dips around 2khz and 4.5khz. Of course, things can occur inversely for IEF-neutral products.
• This actually illustrates one reason why I always additionally adjust in the brain when I look at gras+harman/ief measurements. Personally what I perceive is closer to sbaf+minidsp. People like Sean may have vastly different perceptions.
• Both graphs have different tick scale in y-axis. Be careful.