If you find the Elexs a little sharp at times I would try a couple things before a planar, because planars have that very revealing quality to them as well.
-Try a simple treble roll off first, reducing all that sibilance presence and air by a few dbs can soften things up without losing detail. When Im using my EL8s without OOYH I bell 4khz by a couple dbs and roll off over 8khz by a couple and it effectively reduces some fatigue while still being very detailed
-Put that revealing quality to use with binaural rendering, try Waves NX and Out of Your Head, with OOYH you really have to flatten out the headphones response and then give it some volume though. I use OOYH 99% of the time listening to music, glorious.
-Try an older Audio-GD amp with wolfson chips or r2r/multibit, as it may be the amplifier(feedback distortion of transients has that sharp etched sound).
You may like the Mobius though, I find it to be very detailed without the hyper revealing qualities of my desktop setup. Whether using the binaural rendering or not, but i only use them via ldac bluetooth.
The periapt cables can be a little bulky, but the higher gauge is probably good for a longer run.
Alot of what we hear in headphones as more transparent, more detailed, revealing, etc. is either a lack of ringing, resonance, high frequency distortion, and/or an improved response, how quickly the driver returns to a resting state. Are these qualities actually masking samples in the audio signal or compounding with them? If we can hear these qualities amidst the audio signal that would also seem to indicate that our brains are able to perceive more information. But I agree, I mean really, 40000 samples per second already sounds like an absurd amount of information. For recording, some of the best recordings Ive heard I didnt attribute to a higher sample rate but due to the use of ac mains isolation for the recording equipment ;]
I dont have any bias towards sample rates though, no difference, fine, yes difference, fine, something else to try and enjoy. One instance that suprised me was someone mentioned using dsd512 upsampling with OOYH on the headfi thread, and when I tried it the presets with more reverberant rooms became much more natural sounding, less fatiguing, and more externalized. I tried to share this observation in the Smyth Realiser thread on head fi not knowing that dsd was a trigger word for faux objectivists, and was met with the usual regurgitated tripe (like brownies[re sprites lol] with tiny spears pouring out of the woodwork), so I linked them to archimago blogs just to occupy their time confuse them and save me from trying to have a discussion with someone firmly planted in belief.
Like look at ASR, ooo soo objective soo factual. Yet Ive seen several occassions now where Amirs measurements were innaccurate compared to measurements from no less than 5 other sources. When you actually read his posts you see that he'll put a balanced unit in high gain and test the single ended outputs at max volume, alters graphs, etc. etc. Theres a reason hes banned from here, headfi, and every other av forum over the last 20 years. People would try to instruct him on how to measure certain types of noise, and how that noise effects audio rendering, which doesnt show up as distortion or jitter, and still affected the audio despite a low SNR, and he would just keep repeating "output of the dac output of the dac" Its like comparing dlp to lcd. Sure the gamut measures the same on a static pattern. But the actual video rendering is drastically different.
On that forum they frequently accost people who mention liking non feedback or multibit as liking distortion more then accurate audio(separate of an actual house sound where 3rd order harmonics are tweaked). Yet theyre completely ignoring feedback distortion, it doesnt even exist to them. Audible distortion doesnt exist and inaudible distortion is for the audiofools. Makes no sense. I dont know what the deal is, steering people towards specific equipment? a bias towards other equipment? But now the way its presented and the rabid fan base its heavily influencing the markets and trends, which leads me to conspiracy, you have a group of people who intentionally yes men each other and it eventually leads to ad revenue and sponsorship.
But that level of blatant supremacy and absolutionist behavior and at the same time they present themselves as the good guys, you know who else did that? the nazis. =/ Ive seen to much contradictory "fact" from these types over the years on changstar, whats best, here(no the behringer uca202is not the end all be all of digital audio... !!), asr, headfi, etc, even toms with their pc motherboard audio is just as good as a benchmark and a fully double blind ab test, just no. I also dont really read any audio reviews or impressions, mainly just technical papers. I look at the technology and then try what interests me.
Im more in the extended listening and acclimation camp. AB with AV is very difficult due to our memory and several other variables specific to how the test is actually conducted. I was experimenting with power supplies on a component way upstream(the uptone usb regen that was in front of a breeze duu8 that sent optical to an audiogd nfb11) and was able to hear very clear differences when listening over time. One supply was detailed yet fatiguing(meanwell SMPS), one was less detailed and non-fatiguing(ifi ipower a low noise smps), one was detailed and non-fatiguing(jameco lps). These observations directly correlated to the different properties of these supplies. The meanwell smps is high noise low output impedance, the ifi ipower is low noise high output impedance, and the jameco lps is low noise low output impedance. A quick ab and "trying" to hear a difference didnt yield such observations, only in just listening over time. It sounds stupid, a 0.5a power supply on an over-engineered usb hub, yet I went back and forth for months and the results were consistent. The noise caused the sound to be fatiguing despite a very low snr, and the higher output impedance masked detail, despite a 12khz fft looking exactly the same. It was the same with quadrapole cables and digital components, due to the varied current draw a noise generating voltage develops across the inductance of the cable, solved with $1 of wire, doesnt show up on any measurements outside of putting an isolated probe to the actual cable, should be a non issue due to SNR, but really improved imaging and transparency.
Heres another one, that I briefly mentioned about electrically connected components, and a very anecdotal but interesting observation, I think.
So when you have an electrical connection between two or more components, leakage loops form through the AC mains, power supplies, DC output of power supplies, interconnects (analog and digital) to other power supplies and back through the AC mains, this is separate from ground loops. This current generates noise, and higher impedance connections(as simple as surge suppression) increase that noise. With my mutli channel system, putting the entire system on a low interwinding capacitance isolation transformer, with filterless distribution and using mm om3 fiber to break electrical connections to the pc and projectors, did a few things.
Technically, it reduces common/traverse mode noise by 140db / 70db, provides movless surge suppression, the filterless distribution provides low impedance connections for all the components in the sound system(avr, 2ch amp, 3 mono amps) reducing noise from leakage loops, and the fiber extenders break leakage loops to the pc and projectors.
Subjectively, imaging and transparency improved dramatically, especially atmos tracks so much more height depth and localization. What was interesting though, my gf watches stuff on youtube during the day, and specifically content like people walking around a city alot of the backround noises, road/traffic, wind, crowds, would all sound very nondistinct, brittle, fatiguing, and she would keep the volume very low and use alot of range compression. After the power tweaks though she uses a much higher volume, uses the surround upmixer, less range compression, and frequently commented(completely unsolicited I might add) on how much more detailed and comfortable the audio was. When you look at ultra high end recording/mixing studios, screening/listening rooms etc, they all use some type of isolation from ac mains, fiber for digital signal transmission, and filterless distribution.
Holy crap this got long... !