Ah, and what about the other headphone, the open back one, which one is that if I may ask @Torq ? 
1 Like
I completely agree. This is hands down the best closed back I’ve heard, even though it’s not my preferred tuning.
1 Like
The Stellia would be the closed-back choice.
Gun-to-the-head choice … the Utopia for open-back.
But I’d be frequently thinking about the ZMF Vérité or the Abyss AB-1266 Phi CC. Choices are much harder with open-back. My mood when asked could swing it.
It’s a bit like that age old condurum …
Wilma or Betty?
And of course while I might go with Betty … when I did, I’d be thinking of Wilma …
4 Likes
“Young men’s love then lies not truly in their hearts, but in their eyes”
~Shakespeare
When ever I think of this subject lol…men are such fickle beasts, especially when it comes to the fairer sex lol…or our favorite toys…
I know that my own eyes have betrayed me, of their own volition and under no control of mine own brain/heart lol 
2 Likes
Hmm… the unit I’m measuring for some reason isn’t quite as flat in the lower mids and the bass bump looks maybe a bit more dramatic.
Any thoughts on this?
This is raw (mic only), calibrated at 300hz
Also, the sub-bass measures a bit differently as well. The rest looks fairly close though. I tried this at a number of different positions and there is a drastic change (so getting a good fit is key), but this was the most common shape I got.
1 Like
A couple of thoughts come to mind, but they’re just possibilities.
It could just be the smoothing band selected. I typically post using 1/12th octave smoothing (you can always see it in the graphs I generate) and only choose something else if there are meaningful differences or issues/anomalies visible that I think warrant highlighting after comparing them.
Such differences were a lot more common and/or significant, when I was just using the EARS on a desk “as is” (i.e. with no vibration treatment or acoustic isolation involved). Though I’ve never posted measurements done this way.
Sub-bass measurements seem to be the most volatile, for me, on the miniDSP EARS. So many outside influences can impact that. I found to get something consistent enough to be happy with I had to isolate the entire rig from vibrations. Which took a bit of doing. It wasn’t expensive, but involved a fair bit of trial and error with materials and design etc.
Having done that it, while it helped quite a bit with consistency, especially for the lower frequencies, also exposed some issues elsewhere. So, I wound up putting together an isolating acoustic enclosure. The inside of which absorbs energy emitted internally, so it’s not reflected back to the microphones, and the outer section attenuates external noise.
With those two things done, the end results have been quite a lot more predictable and consistent. And some specific cases that I felt were down to headphone placement issues proved not to be. Though, that said, placement still is an issue … it does have an effect - sometimes a big one - but there were some oddities I’d ascribed to being placement-related that did turn out not to be.
I don’t know that all the above is necessarily worth it with the EARS, but I did it because I personally wasn’t happy enough with the results I was getting just plonking it on a table and measuring away - and until I’d done all the above I didn’t actually post any measurements (which is why the earlier reviews/impressions I’ve done done’t have them). Of course, now I’m not happy with the additional setup time when I want to do a measurement!
Beyond that, I do multiple placements (five positions) of the headphones, and each position is measured and averaged several times (four passes), before averaging those results across the positions.
2 Likes
That’s good to know. I definitely just have it on my desk. It’s reasonably quiet but certainly far from perfect. And that completely makes sense with the averaging. I’ll give that a shot when I get home.
1 Like
I think most people use it that way. As long as there isn’t anything on the desk like a spinning hard-disk or something with a fan in it, averaging will likely be a good enough solution.
1 Like
Following up on @Resolve’s post, here, this is how my measurements look, using both RAW and my custom calibration, when plotted with 1/12 and 1/24 smoothing.
The different smoothing settings do not make a difference in the fundamental shape of the plots here, so I don’t think that’s the difference. Makes me more curious as to what is …
4 Likes
I tried it on a few different sources, as well as with a bit of extra vibration padding setup to no avail. There are minor differences but they all still resemble the same general shape, with the main difference between mine and yours being more bass presence past 200hz on mine. Could be miniDSP EARS imperfections, but could also just be slight variations in headphone units.
2 Likes
All this does is reenforce my belief in hearing things over graphs…graphs are nice, and tell you a general baseline, but can be different between users…kind of like people lol, we all snowflakes floating on the wind 
1 Like
Yeah but it’s still useful to be able to have an idea before getting a chance to hear it. Also, if the difference in graphs really does come down to differences in the units themselves, that might be a bigger deal. Personally it’s not a problem because I’d EQ the bass down a bit for both, however with Torq’s unit I probably wouldn’t need to use as much.
2 Likes
I’m not a scientist… but from what I’ve researched and been told, power supply, isolation, variation between machines, any interference etc. Can cause variation in measurements…
But I agree good baseline going into the headphones you’re looking at buying.
2 Likes
Absolutely. That’s the other thing that could be the likely culprit. I’ve used it across a number of sources, and while the shape is generally similar, there are notable differences.
2 Likes
Given where the differences are, I am inclined to think it is down to differences in output impedance in the amp being used. I’m not sure what you’re using for your measurement source, but if it’s the Cayin IHA-6 then its ~10 ohm OI (SE, I assume that doubles if balanced) into a 35 ohm dynamic load is almost certainly going to cause some audible (and thereby easily measurable) FR shifts.
2 Likes
No I only use the balanced output on the IHA-6, which is <1 ohm output impedance. The SE outs are awful and in my opinion totally pointless. For SE I’ve been just using the output on the idsd micro black label, which is also <1 ohm output impedance. For other sources, I’ve tried a modi 2 magni 2 stack, an odac & o2 combo, a hifiman tube hybrid amp (for fun), and an essence STX, which uses a burr-brown as well.
2 Likes
In which case, I shall track down another pair of Stellia and throw them on my rig and see how much variation there is there!
3 Likes
Maybe you can try measuring this one on your rig? I’d be interested in the results, because it could very well still show a difference in the EARS rigs, even though the calibration profiles are provided.
4 Likes
@Resolve has posted his video review of the Stellia for those looking to sit back, relax and watch something.
5 Likes
To those owning a pair: How are the cables? I read somewhere they are a bit stiff and quite microphonic for on the go? Is it really that bad?