Sennheiser HD650 Review (Headphone)

[Robbo99999]

Built-in from his Roon player.



They're all the same for all intents and purposes, one has a little more bass than the other, but they're all virtually the same when listening to them. The comfort though is a tad different with clamping forces and such.

I couldn't blind test which is which between 650 600 6XX, while the 660S was just a glorified build of the 58X to my ears which had the most bass (didn't really feel like bass, just felt a tad more muddy tbh). I can't fathom boosting lows on any of these, idk if it's placebo (didn't blind test "distortion after EQing tbh"), but I really don't like what comes out of these when I boost the lows, I go the opposite, and try to high pass filter anything out under 55Hz tbh.

I don't believe in "blind testing" headphones aside from identical SKUs seeing as how they all have different cups and external housing, so when you put them on to adjust them, they instantly give away what they are (we had to put tape on the parts I was going to be touching the headphones by, as to not give away which they were, and also had to avoid seeing the headphones at all as to not know which were which concerning clamp force.

Their HD800S headphone also suffers this bass distortion issue which leads me to believe all they're doing is constantly creating the same slight variation of the same old driver with some tuning here and there, and a new housing that does most of the "soundstage" heavy lifting due to cup size and shape incidentally. But it's obvious they don't take it deathly serious as they wouldn't have that idiotic treble rape spike, unexcusable in a $1,000+ headphone in my view.

For $150-$300, the 500-600 line from Senn are a good buy (honestly mostly due to the solid choice of weight and comfort, and not being morons making circular earcups, instead having a more ergonomic ovular shape). I'd really like to see them move on to planar's just to see if they got the chops of making something sonically exceptional, as their current crop is just "good" (the 800 is just bad, sorry). They have all the tools, yet I've seen their amps for example measure terribly. Doesn't look like they do an insane amount of validation, as I don't know how they could have something like that awful treble and horrendous sub bass distortion on their flagship line. They strike me as a company resting on it's laurels, not really moving much. Same old variations, nothing really new.

If you boost the lows below 100Hz, I think you gotta take out some of the 100-250Hz otherwise it's likely to sound more muddy - which is actually the shape of the Harman curve in relation to these two headphones (HD600 & 650).....& muddiness is excess between 100-300Hz IIRC.

 

[Robbo99999]

"Harman curve"? Seriously. It's like there is/was no other company in the world that actually bothered to test gear? Give me a break.

It's so mindnumbingly boring. They are (were) a small company. Not remotely influential in the scheme of things. Sidney, and Bernard, sure, absolute legends, but ultimate authorities they and their company were not.

You missed the research then? Along with the "standardised" measurement gear that continues to be made & supports it.

 

[Robbo99999]

Actually they don't. The difference between Harman and diffuse field is mainly in the lows. At 20Hz it is 10dB off from Harman (and 5dB off from my target).
Funnily enough the HD600 with Oratory EQ the 20Hz is still -5dB off from Harman.
Oratory EQ for the HD650 is following the Harman curve down to 20Hz.

Actually they do, I'm talking about his Preference Score or Deviation From Target, which prove the point I made, it's listed at the bottom of his pdf files. I also think by eyeballing the frequency response of the HD600 that it would acoustically match the Harman Curve more accurately than the Diffuse Field, even though they're not marketed as such.

 

[phoenixsong]

I'm feeling somewhat conflicted- in some cases, speakers are preferred to have bass roll off or a high pass filter applied to clean up the distortion in the low frequencies, but in headphones, is the opposite desired? (High distortion>roll off)

 

[uwotm8]

"Harman curve"? Seriously. It's like there is/was no other company in the world that actually bothered to test gear? Give me a break.

It's so mindnumbingly boring. They are (were) a small company. Not remotely influential in the scheme of things. Sidney, and Bernard, sure, absolute legends, but ultimate authorities they and their company were not.

Shhhh. There was a lot of "scientific" work and you/me/us all have no rights to have any doubts on that remarkable research. Finding the Holy Grail of sound. Again. Like never before. The result... well, it sounds like typical bassboosted garbage but now garbage is not JUST a garbage - that's "science"! LOL

Typing this with an experience of "oratorizing" ("harmanizing") a few pairs of differnet headphones.

P.S. It's actually fun to read sometimes, one of the best parts what I like is circle... circlesciencing like simulating different THD levels or freq response on the same (STAX-009 iirc) headphones and extrapolating the results to real cans.

And then "reference" AKG 371 is born

 

[wemist01]

Hard to hear any differences.
Have a try...

https://www.npr.org/sections/therecord/2015/06/02/411473508/how-well-can-you-hear-audio-quality

Perhaps you would like a check on your headphones before starting...

https://www.audiocheck.net/soundtests_headphones.php

Thanks for this- It's interesting to document that there IS a difference in audio quality between the tracks, but answering the question regarding which is higher quality is difficult. Initially, I assumed that uncompressed would mean sharper treble, but ended up picking the two lowest-quality tracks in the first two tests. Then I switched to picking the track that sounded the smoothest and got the uncompressed file in 3/4 of the remainder tracks, while picking the worst track on the hiphop one. I look forward to repeating this test with my better cans in a day or two when I've forgotten what order the files are in.

 

[Helicopter]

I'm feeling somewhat conflicted- in some cases, speakers are preferred to have bass roll off or a high pass filter applied to clean up the distortion in the low frequencies, but in headphones, is the opposite desired? (High distortion>roll off)

I usually want the most frequency range I can get without problems, either electronic or audible. Headphones generally have less distortion than speakers, so I am going to use fewer high pass filters. If a headphone or speaker has audible distortion or other problems, then I will correct it, but if it is still performing well enough, then I will take the bass extension.

 

[Robbo99999]

Shhhh. There was a lot of "scientific" work and you/me/us all have no rights to have any doubts on that remarkable research. Finding the Holy Grail of sound. Again. Like never before. The result... well, it sounds like typical bassboosted garbage but now garbage is not JUST a garbage - that's "science"! LOL

Typing this with an experience of "oratorizing" ("harmanizing") a few pairs of differnet headphones.

P.S. It's actually fun to read sometimes, one of the best parts what I like is circle... circlesciencing like simulating different THD levels or freq response on the same (STAX-009 iirc) headphones and extrapolating the results to real cans.

And then "reference" AKG 371 is born

There is no "Holy Grail" for headphone sound....but there is for speaker sound! ;-) The point is that an average best sound is as close to the Holy Grail as possible when it comes to headphone sound that has not been personalised to your own HRTF data......so in terms of measuring a headphone against a 'benchmark', then the Headphone Harman Curve is as close to that "Holy Grail" as possible. Until we have ways of easily measuring our own HRTF's and buying headphones designed/personalised off the shelf to our own personal HRTF, then that's as good as it gets.

 

[dfuller]

Of course the last part of the quote means they're screwing with their own (possibly scientific) findings to appeal to global audiophoolery...

More like they're tuning it to roughly that target curve, then playing with it til it suits their tastes/their "house sound".

 

[MZKM]

Just saw this from Oratory on the frequency resolution for the preference rating formula:


Maybe this (partially) explains the discrepancies you're getting between yours and his calculated scores.

I got the 12PPO (121 data points) information from him.

 

[Tks]

If you boost the lows below 100Hz, I think you gotta take out some of the 100-250Hz otherwise it's likely to sound more muddy - which is actually the shape of the Harman curve in relation to these two headphones (HD600 & 650).....& muddiness is excess between 100-300Hz IIRC.

I do, it solves it somewhat, but not all tracks. Oratory's EQ is pretty good in this regard if you mess with it on your own a bit as well.

I want to say real quick,

40Hz and below on virtually all dynamic drivers on headphones I've heard are basically nonexistent. They're there if you boost volume high enough but are then in ear damaging territory.. The fact that they all seem to heavily distort under 50Hz makes me rest easy decimating those frequencies and lower. Less driver stress to carry over garbage into the audible band (by what phenomena I can't really be sure of course). For non electronic music this isn't really a problem, since you're not usually getting sub bass impact along with miss and highs spiking as well. But with electronic music, everything simply sounds cleaner at higher volumes when you don't have drivers having to kill themselves keeping up a tune that has sub bass and a ton of mids and highs rapidly.

 

[Robbo99999]

I do, it solves it somewhat, but not all tracks. Oratory's EQ is pretty good in this regard if you mess with it on your own a bit as well.

I want to say real quick,

40Hz and below on virtually all dynamic drivers on headphones I've heard are basically nonexistent. They're there if you boost volume high enough but are then in ear damaging territory.. The fact that they all seem to heavily distort under 50Hz makes me rest easy decimating those frequencies and lower. Less driver stress to carry over garbage into the audible band (by what phenomena I can't really be sure of course). For non electronic music this isn't really a problem, since you're not usually getting sub bass impact along with miss and highs spiking as well. But with electronic music, everything simply sounds cleaner at higher volumes when you don't have drivers having to kill themselves keeping up a tune that has sub bass and a ton of mids and highs rapidly.

(It's got nothing to do with "ear damaging territory" if you use EQ to boost up the lows (to Harman levels), but does indeed relate to distortion that you mentioned).

 

[dfuller]

I have to agree with others here. I did not like boosting lots of bottom end into these headphones. 2 or 3dB as a broad shelf at 20Hz? Sure, whatever. But this +8dB or more business just makes them fart at any appreciable listening level.

 

[bobbooo]

I got the 12PPO (121 data points) information from him.

What did he say exactly? I have a feeling he meant he smooths the graphs to 12PPO when displaying them, but the 500 data points from 10 Hz to 20 kHz is referring to the preference rating calculation. The reddit thread I quoted him from is specifically about the preference ratings.

 

[ZolaIII]

There is no "Holy Grail" for headphone sound....but there is for speaker sound! ;-) The point is that an average best sound is as close to the Holy Grail as possible when it comes to headphone sound that has not been personalised to your own HRTF data......so in terms of measuring a headphone against a 'benchmark', then the Headphone Harman Curve is as close to that "Holy Grail" as possible. Until we have ways of easily measuring our own HRTF's and buying headphones designed/personalised off the shelf to our own personal HRTF, then that's as good as it gets.

Hire you go pal and have some cheap sunglasses.
https://github.com/jaakkopasanen/AutoEq/blob/master/compensation/loudspeaker_in-room_flat_2013.png

 

[sacguy231]

I really enjoy this headphone, but sadly I had to sell my pair because the pads are just shy of the depth they need to be for me. I'm sensitive to my ear touching things. I since got both the HD700 and the HD800S, but I could have been very happy with the 650 as my only can if the comfort had been better for me. :/

 

[PeteL]

I'm feeling somewhat conflicted- in some cases, speakers are preferred to have bass roll off or a high pass filter applied to clean up the distortion in the low frequencies, but in headphones, is the opposite desired? (High distortion>roll off)

I assume that when speakers are not able to reproduce the full band, the remainder is to be assigned to a subwoofer. Now this is as with anything, up for debate.

 

[YSC]

There is no "Holy Grail" for headphone sound....but there is for speaker sound! ;-) The point is that an average best sound is as close to the Holy Grail as possible when it comes to headphone sound that has not been personalised to your own HRTF data......so in terms of measuring a headphone against a 'benchmark', then the Headphone Harman Curve is as close to that "Holy Grail" as possible. Until we have ways of easily measuring our own HRTF's and buying headphones designed/personalised off the shelf to our own personal HRTF, then that's as good as it gets.

I think the ear difference for most ppl will be compensated by the brain, the HRTF of our heads differs less than the individual ear shape especially the canal shape (hell, from my UERR my left and right ear differs A LOT). but then I still feels the diffuse field curve where the dummy head would measure with a flat speaker is placed would more represent a neutral sound than the Harman Preference, which is for ppl's taste and we know that most ppl prefer boosted bass (see how Sony XBS series and Beats sold with massive bass boost..) especially in short sections.

I tried personally using the 8030C's dip switches, when placed in my room very close to the wall and listen at ~60cm nearfield the -4dB bass tilt actually measured and sounded more neutral, but when I undo that t-4db tilt making it having wall loading bass boost by 4-6db I enjoyed it for most of the normal tracks in short period, until I came into some pop with kick drums doing the beat stuff and the bass muddies and overwhelmed the vocal. If I didn't go to that track I might actually prefer the not attenuated bass.

I have a feel that the extra bass shelve at ~8-12db from diffuses field of the Harman Preference Curve came from this also and the diffuse field would more represent the neutral speaker in the holy grail of speaker

 

[Robbo99999]

I have to agree with others here. I did not like boosting lots of bottom end into these headphones. 2 or 3dB as a broad shelf at 20Hz? Sure, whatever. But this +8dB or more business just makes them fart at any appreciable listening level.

I agree with this to some extent, but it depends on the headphone. My NAD HP50 closed back is fine with Harman Curve all the way down to 20Hz. My Sennheiser HD600 farts a bit like you say - so instead I use a peak filter for the bass boost rather than a Low Shelf, that way it has natural roll off:

And the same reason I use a peak filter on my AKG K702 (another open back headphone), but my K702 deals with bass boost better than my HD600 in my listening tests:

 

[uwotm8]

There is no "Holy Grail" for headphone sound....but there is for speaker sound!

Again IIRC one of brilliant Olive ideas was to bring headphone sound to preferrable* variation of good**-speaker-in-proper-room sound.
*true and fair are different words, right?
**imagine if Harman ever buit one
Not a natural sound but speaker sound. Brilliant!