WTF Revisited: Sennheiser HD700 Stories, Review, and Measurements

thats a eq curve not the frequency response of the headphones. That oratory guy likes to use eq to flatten headphones, lol. I use his measurements to look for imperfections and quality of the headphones mainly not eq or harmon curve reasons. i usually used sonarworks measurements but they tricked me a couple of times so i had to look for more measurements.

 

compared to 700 looks kinda super duper smooth. im very well versed with measurements this may be a bigger challenge than you think, lol.

 

https://www.reddit.com/r/oratory1990/wiki/index/list_of_presets

Those are his measurements try to find smoother, lol.

your measurements of the 250ohm and 600ohm are negligible since beyer headphones change in sound over time and the graphs are almost identical in smoothness once pads are worn down. your measurements show a 6khz peak which goes away entirely after "burned in" fully. keep in mind this change only takes a couple of months.

 

I would love to figure out the mythical "timbre" thing. - I grew up before the metal dome tweeter detail craze took over. I'm a huge believer in your BWC (big woofer club) and natural treble extension that just exists in the air, vs. being forced through the air at you. This is one place that I agree and disagree with you regarding the HD800 (Face tweeter) and maybe it's just the way I interpret what you are saying.

Option 1: - it is literally a ring radiator tweeter. I have Vifa XT'25's as the main tweeter in my home 5.1 setup and that is closer to the HD800 treble presentation.

Option 2: - You believe the timbre of the treble is incorrect on the HD800. I personally think it is a lot closer to my preference in timbre, it's just the rest of the driver / enclosure can't give you the full-range experience. I personally prefer the top end of the HD800 over the Utopia.. and if the Utopia did it the same as the HD800, I'd have 3 pairs of Utopia, because I love everything else going on with that thing.

Edit: For clarity's / transparency - I'm not saying the treble TUNING is correct - I am saying the sound of the motor is more correct. - Car people know what a small block chevy / mopar hemi / Nissan/Infinity 3.5/3.7 whatever they are at now kazoo.. The "timbre" of the motor is made up of materials, geometry, etc. (design) and can be "tuned" with different length "ports" (anything in the path of the inlet and outlet of the air in the motor) - But you can always tell a Wankel from a Corvette from a G37. Removing the WRX rumble with equal length headers... it still sounds like the motor, it just presented different.

I believe we are currently limited by physics with driver technology. You can only move air so many ways and the best way is with big cones and well designed motors. Focal is on the right track and they probably are going to have to go back to the drawing board on the motor for all the n00bs who have driving them into Xmech and making up words that aren't what is actually going on. I've heard "clipping" used way too many times. (No you fools.. you are driving the very fragile former-less coil into the backplate with your 20hz 120db attempts)

Sorry to side step the thread - I am still honestly hoping to get a recommendation for a headphone that has that natural effortless presentation.. (I believe we are just call it HD6X0 + HD800 Tech ?? HA HA HA HA)

 

I completely forgot, the innerfidelity data already uses a quite different compensation curve for the left and right channel to begin with:

From 7-10kHz the ID curve is quite different. Between 8-9kHz almost a 5dB difference.

It seems he uses a different compensation curve for IEMs, but some of the over-ears that still manage to get good matching despite the different ears are:
HD800, JPS Abyss, HE1000, etc.
Smaller planars generally seem to get less treble with the right ear:
HE560, STAX Lamdbas, etc.

However the K701 gets very good matching despite traditionally bad imaging. The drivers aren't angled and the FR doesn't lend itself to a large stage, I'd say.

On-ears seem to get notoriously bad matching from what I've seen so far. Look at the Urbanite! I'm assuming they had good matching to start with, but on the HMS II.3 it doesn't.

I think Marv is on to something with the matching despite mismatched ears. Untuitively it also makes sense that headphones which get good matching despite mismatched ears (when using the correct compensation curve) will have good imaging.

 

Yup someone pointed out the different comps used between IF channels and that explains why blue trace always seemed slightly more pronounced.

Now I'm easily distracted and prone to misunderstanding but it seemed to me the implication was that different results on different pinnae were more indicative of imaging tightness, as such implies greater interactivity with the auditory system. Dunno, just seems to me now that I'm looking at IF graphs I have that headphones I've found had great imaging do seem to measure with more of a discrepancy between 7~10kHz, but again I've not tried very many headphones

As to why this doesn't result in perceptive imbalance I'm in agreement that the brain does a lot of background processing (oversimplification) based off living-in-environment that compensates for mismatched pinnae, which basically everyone has to varying degrees. Still want to look more into this but it's not in any of my usual online databases lol, I'm a bit far afield.

Same goes for on-ears. Speaking from limited experience with clip-ons and the like I tend to make minute adjustments without really thinking about it, I've noticed. It would make sense that even watching square waves to ensure good coupling there's a profound difference between experiencing imbalance for the self and guesstimating based on readings.

 

Well, given that the compensation curves are different for each channel and those are based on a farfield source in an anechoic chamber I'd hope that headphones with more realistic interaction would indeed give better matching.
With RANDY where such compensation curves don't exist (yet) a larger difference may be better.

Honestly if you find it to be the other way around we both might just be projecting our own placebo onto noise. Maybe there isn't so much to it after all.

As an aside: I find it weird that there's such a large difference in the compensation curves sub 1kHz. Unsurprisingly virtually no headphone is able to achieve perfect matching in that region as a result and if it does I bet it's coincidence.

 

The way I see it good channel matching even allowing for different transfer functions between ears seems to coincide more with the in-head-presence than out-of-head, but I'd like to confirm for myself with models I've not heard or seen measurements of yet.

Yup, very likely personal biases colouring how we're reading similar data, haha. Funny how there's a plethora of information on how sound localisation works in farfield systems but there's a relative dearth of the same for personal audio, but I guess when you bypass as many of the same cues we've grown to rely on from an evolutionary and developmental standpoint it's more difficult to determine things with any fair degree of empiricism.

It's not happening anytime soon but I'm hoping to save up for some fake pinnae that are reasonably similar to actual ears in feel, texture, and asymmetry plus a flat plate system to check against, but maybe in-ear mics would be a great stopgap for the interim. I vaguely recall you use/d such a system, @Serious? Would be curious to see if those could help validate imaging delineation when compared to how they'd measure in free air or on a flat plate.

 

Is the word being repurposed? (I used to hate that word but learned to live with it)

Timbre is a simple word that means a flute sounds like flute, a violin like a violin and... the rest. All this is very real, and the definition we all know.

Isn't the audio/hi-fi word "flavour?"

 

This is what I was getting at before in terms of combination with other elements of the headphone.

Since driver is off axis with ear line, psychoacoustically one could postulate that the 6khz peak could draw brain's attention to the fact driver is off axis/not in line with ear and not directly to sides of ear. As mentioned earlier brain pinpoints original sound location of higher frequency tones unlike lower frequency tones. Since the driver is off axis located you could postulate that it would make brain perceive the borders (if u want to call it that) as farther away or positioned differently than headphone drivers usually are. If you use frequency too low it will be less locatable due to omni directional perception , if you use frequency too high it will be less audible/noticable - as high frequency gets higher, audibility of sound lowers.

I don't claim to have all the answers as I don't have access to Sennheisers engineering documents, but I find it hard to believe they released 4 headphones over the course of a decade with very similar variations on the same treble curve for soundstage-optimized headphones, and that this is somehow an unintentional design, accident that has nothing to do with soundstage, or something one of largest headphone companies in the world wouldnt be able to fix if it wasn't there purposefully (when people who don't like it "fix" it with a $1 piece of foam).

Also re: use of measurement models of perception another poster mentioned, models are just that - models. New engineering discoveries result in original model being improved over time. It still is not measuring the brains actual perception of audio. If we could actually measure the brains perception of audio than all other measurements would be irrelevant because the perception and interpretation of the raw sounds by the brain is the primary thing that matters in terms of how we hear audio. If there were no perception of audio by the brain one would be deaf.

Already did this with sonarworks 4 yr ago with both hd700 and hd800 [original] when I got them. EQing to harman curve or try to flatten treble made both headphones soundstage sound significantly smaller and less impressive. You can't move the drivers around though so the EQ won't tell you how the perception of FR will differ in these bands in hd700 vs a traditional on axis with ear line driver headphone.

 

So this is sort of HD700 related but also HD800 related

Back when I had the HD800 a while back, I only had solid state gear. Pretty good stuff, but not as delicate w/ the treble as something like LP.

The one thing that always confused me is that the HD700 never bothered me on any content, but the HD800 bothered me a bit on some (not all) content with that solid state gear. Meanwhile other people seemed to have opposite experience.

Recently, I noticed that i seem to have a bit of treble sensitivity @ 2-3khz range specifically when I was testing tubes in the LP.

I was looking at some comparisons of HD800 vs HD800S and stumbled upon this graph:
https://soundontime.com/wp-content/...s-HD800S-A-Comparison-Of-Treble-Resonance.jpg

I was just thinking.. I wonder if possible the reason the HD700 is so appealing to me is that it has a dip @ 2-3khz? Simple function of the area where my ears are super sensitive, the HD700 dips - which isn't perceived as too much of a dip for me as i am sensitive in that area to begin with. Makes me also wonder on my current equipment how the HD800 would sound vs the HD800S, and if the elevated distortion in said graph is audible and perhaps what was irritating with HD800. The HD820 is a more enjoyable listening experience for most of the music I listen to, but not quite as epic in scale/air for certain types of music (i.e. broadway for instance) as I remember the HD800 being on the same amp/dac; but I never got to hear the HD800S on my own equipment like HD800/HD820, only on HDVD800. Very interesting stuff.

 

Honestly, aren't we done with picking over the rotting carcass that is HD700 discussion on SBAF? I mean, I thought this was done once no one could successfully do any sort of mod to at least help in fixing the issues to anyone's liking on here. HD800 at least had a bunch of mods to bring it back from the edge.

Hopefully here's to another 5 years ahead when this is brought up again.