Harman Curve Poll (Public)

Discussion in 'Blind Testing and Psychoacoustics' started by ChaChaRealSmooth, Nov 9, 2020.

?

What do YOU think of the Harman Curve

  1. Strongly Disagree- The Harmon Curve sounds bad and it's potentially dangerous for good sound

    28.7%
  2. Somewhat Disagree - I like some things about it, but I'd rather not be the target

    38.5%
  3. Somewhat Agree- I like more things about it than not

    25.3%
  4. Strongly Agree - Now, this is how music should be heard!

    2.3%
  5. f**k me in the ass with the IE 2017 treble boost! (Harmon Curve is for kids with hearing loss)

    5.2%
  1. dubiousmike

    dubiousmike Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    269
    Trophy Points:
    63
    Location:
    Los Angeles
    Great thread, and so nice to know I'm not alone in hating the ubiquitous application of this garbage FR curve to modern IEM's. A rise around 3k just kills me and is why (along with ergonomics), I often continue to reach for my old Westone w4/40's for casual listening, notwithstanding their obvious lack of detail, because they don't hurt me every time a high hat is in the mix.

    Back in 2006, when I first found my way on the audio boards, I'm pretty sure all of the entry/mid-tier IEM's I bought actually had a minor (or in some cases major) dip in this region, including Ety Er6i, NuForce NE7m, Senn IE8's and Westone W4's. I used nothing but silicone tips because I didn't need Comply's to take the edge off the upper mids. Now the trend seems to be just the opposite, and Comply's are a necessity for me with most of what is on the market.

    Maybe it is just my ears, but the whole IEM pinna gain concept rings false in my experience. I find my '06 Beyer 250 ohm dt880's 3.5k peak problematic, but an equivalent peak in IEM's even more so. By contrast, the minor 6k peak on my OG hd800's (which has a minor dip at 3k), isn't really an issue for me, especially with the sd resonator in place. Similarly, I suspect part of why I still continue to really enjoy my Fostex Thx00's, from the first batch sold, is that they also have a dip in this range. I also find it funny that AKG's own very popular and decent (IMO) y50 on ears have a sizable dip at 3k while its parent co. continues to push Olive's conclusions.

    Does anyone have a link to any of the HK / Olive resources that specifically discuss upper mids in headphones? The basic articles I am finding seem to lump listeners together based on +/- db's over 1k and under 300hrz. See: https://www.soundstagesolo.com/index.php/features/217-where-are-we-at-with-the-harman-curve The youtube video on Olive's blog also seems to suggest that his listeners were only permitted to eq the top and bottom of the FR curve in expressing preferences: https://seanolive.blogspot.com/2014/01/the-perception-and-measurement-of.html
     
  2. Thad E Ginathom

    Thad E Ginathom Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    14,129
    Trophy Points:
    113
    Location:
    India
    Life is so unfair on those of us who don't even have $100k.
     
  3. Mont789

    Mont789 Acquaintance

    Joined:
    Oct 9, 2020
    Likes Received:
    47
    Trophy Points:
    18
    Location:
    UK
    The harman IE target is pretty much making a V shaped ER4, The ER4SR i use only needs a -2.5db at 3.5k. No idea why he even did this since pinna gain was to fix that fact the Ety ER4 goes in very deep, He doing this on shallow fit IEMs with peak/dips everywhere?.
     
  4. Rthomas

    Rthomas Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    966
    Trophy Points:
    93
    Hi Guys,

    I understand that senior members here disagree with the Harman target.

    What puzzles me is this.

    We have two sources jaakkopasanen and Oratory1990 who use some math that I don't understand to compare how close headphones get to this target in stock form.

    1. jaakkopasanen - this guy has a headphone ranking purely based on how close a headphone gets to the Harman Target.

    https://github.com/jaakkopasanen/AutoEq/blob/master/results/RANKING.md

    Near the top of the list (First 30 or so entries) I see: HD600, Sonoma Model One (which Tyll praised for its tonal balance), Utopia, Focal Elex, Verum One, Senn HE-1 etc.

    All well regarded headphones for tonal balance.

    2. Oratory

    Most of his highest scoring headphones (Scores above 80) are universally acclaimed.

    HD600, Focal Utopia, Senn HE-1, Senn HE-90.

    If their math is correct major headphone manufacturers seems to mostly agree with the biggest differences being in how much the bass is boosted and the research says that bass boost must be adjusted to personal preference.

    Sorry if this has been covered already but I just don't see how the target can be anti-hifi based on this info. If I'm missing something big please point it out to me.

    I would love to have an SBAF neutral target for the Utopia or the SR009S to try to see what I'm missing.

    I'm not trying to start an argument as I'm not qualified to judge neutral sound. Just trying to learn.

    I understand disagreeing with the bass boost but the rest of it seems legit?





     
  5. Lyander

    Lyander Official SBAF Equitable Empathizer

    Pyrate Contributor
    Joined:
    Feb 25, 2017
    Likes Received:
    10,961
    Trophy Points:
    113
    Location:
    Philippines, The
    https://www.superbestaudiofriends.o...topia-focal-utopia-revisted.9750/#post-313715

    Just using MiniDSP EARS but the sentiment is more or less there. Can't personally say anything about the Utopia because I don't even remember my own very brief listen but as much as I love the HD600 it can still be somewhat hot in the upper mids, and painfully so. Also feel it's vaguely bright but my upstreams have all been geared towards taming the HP-3 so... hah! :D

    Still counts as a preference thing of course.

    Also, apologies if I'm missing anything but how did Oratory and jaako manage to get proper Harman compensations on their measurement rigs? Same equipment and calibration process? And it's not just the bass itself that many seem to disagree with as far as the Harman target goes, it's the weird lower midrange skewing and just how fatiguing it can be over extended listening.
     
    Last edited: Jan 22, 2021
  6. Vtory

    Vtory Audiophile™

    Pyrate MZR
    Joined:
    Sep 12, 2016
    Likes Received:
    10,831
    Trophy Points:
    113
    Location:
    East Coast
    Let me call out the most fundamental things.

    1. If you don't fully understand and acknowledge pros and cons of specific scoring formulations, DO NOT GIVE ANY TRUST ON single number score. Translating frequency response to a single number means informational loss of one whole continuous dimension. The number is very subjective to the way they used.

    2. Focus on false negatives and false positives.

    Please note that even harman haters agree that some good headphones may look good in harman-compensated coordinates (particularly open-back ones as sub bass targets physically less meaningful).

    The fact people disagree doesn't mean harman target is anti- or counter- predictive. It's more like that harman target isn't practically any better than other ones.
     
  7. yotacowboy

    yotacowboy McRibs Kind of Guy

    Pyrate Contributor
    Joined:
    Feb 23, 2016
    Likes Received:
    10,682
    Trophy Points:
    113
    Location:
    NOVA
    Home Page:
    But in my understanding, the Harman target is really intended to be a marketing tool, and to that point we don't have any reference for its effectiveness aside from Beats by Dre sales numbers. The Harman target isn't intended to be a way to gauge how "good" a headphone sounds, it's intended to be a way to determine how close a headphone will perform according to a preference that appeals to the largest portion of a population. Match this curve and you've got a good chance at selling your headphones to a majority of the population. It has nothing to do with pursuing "the absolute sound" (to steal a phrase).
     
  8. Donald North

    Donald North Friend

    Pyrate
    Joined:
    Oct 4, 2015
    Likes Received:
    932
    Trophy Points:
    93
    Yes it does, in their minds
     
  9. Tekker

    Tekker Facebook Friend

    Banned
    Joined:
    Aug 1, 2020
    Likes Received:
    133
    Trophy Points:
    43
    Location:
    Netherlands
    After I stuff the nozzle with cotton (dried alcohol swab is not strong enough), remove the stock mesh, put micropore tape over the nozzle to seal it off, and with some HD650 driver cover felt in Comply T-600 large eartips, the Harman tuned, overhyped Blon BL-03, sounds actually great.

    Somehow, the lower-mids and vocal forwardness stays intact, while the huge boost in the upper-mids and treble gets tamed. The timbre itself of the Blon is amazing, it’s just the stock FR-response that obsecures the timbre potential.
     
  10. Stereo Skunk

    Stereo Skunk New

    Joined:
    Jan 22, 2021
    Likes Received:
    6
    Trophy Points:
    3
    Location:
    Dallas, TX
    I like to EQ headphones myself but even though I used my ears as the final arbiter, I am still unsure if it's right! How silly. Also, it's dependent on the recording so I have found there is quite the risk of 'overfitting' where adjustments are necessary for each song. I have yet to try to take an average of my settings across all songs, which I could do but damn am I lazy.
     
  11. tomtrp

    tomtrp New

    Joined:
    Jan 13, 2021
    Likes Received:
    7
    Trophy Points:
    3
    Location:
    Australia
    As a physical science student, I read Sean Olive's paper and I am disappointed at the research quality as well.
    Besides the weak statistical inference from the 268 sample to the population, the paper has no repeated experiment. It is a single experiment case study. When you try to research human's preference of sound through headphones, how can you draw a deterministic conclusion from a single 268 people sample experiment with little randomness. There should be at least another 200-300 people that is tested with same method applied and yielding similar correlations. They even draw conclusions from each strata like nationality, age while each strata only has 50 or less people.
    Any other physical science results or even social science results will require way more stricter validating process to be accepted, applied and promoted.
    And it appears more and more to me that Harman utilizes and exaggerates
    the limited results of their research to be a rule maker for marketing purpose.
    All in all, AES is not a pure academic entity and has little authority in the actual academic world.
    Let's not put the same confidence in such research as in true physical science which is carefully constructed, mathematically sound, with its predicted results tested by numerous experiments with stunning accuracy, and reviewed by top peer scientists around the world.
     
    Last edited: Jan 25, 2021
  12. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    ^ Sean Olive needs to justify his job, and his quota of one point five AES papers every ten years.

    Thank goodness headphone manufacturers, and companies like Apple (AirPod Pros) are not following the Harmon targets.
     
  13. Rthomas

    Rthomas Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    966
    Trophy Points:
    93


    Ok. So you're saying that their math may be questionable. Even saying that I can simply use the eyeball test.

    Oratory has a free website with his measurements.

    1. HD600 - Long regarded by everyone and their grandmom as a neutral reference. It seems to align almost perfectly with the Harman target other than the bass boost. How do you explain this? Does the HD600 sound wrong or does Oratory not know how to measure a headphone and compare it to the target? I really don't see any options other than these two.

    https://headphonedatabase.com/oratory?ids=226

    2. Focal Utopia - again same story as the HD600. Slightly brighter than the target without the bass boost.

    https://headphonedatabase.com/oratory?ids=178

    3. Orpheus HE90 - Considered one of the greatest headphones ever made. Midrange follows the target almost perfectly. Again no bass boost.

    https://headphonedatabase.com/oratory?ids=224

    4. Sennheiser HE-1 - Senn's cost no object flagship - This thing looks like it was designed by Harman! Even includes the bass boost FFS

    https://headphonedatabase.com/oratory?ids=72


    Ok I've given you 4 headphones that are considered the greatest ever by most. I've read that most music lives in the midrange. These 4 seems to follow the target almost perfectly in the midrange. Bass boost can be adjusted to taste of course.

    How can the target be crap/anti music/ BS? Either the Target has some merit or the best designers don't know how to tune a headphone or I'm missing something big time :drunk:
     
  14. Tekker

    Tekker Facebook Friend

    Banned
    Joined:
    Aug 1, 2020
    Likes Received:
    133
    Trophy Points:
    43
    Location:
    Netherlands
    HD600 is too bright in the upper-mids, and has quite a bit more lower-mids emphasis than the Harman target. Coupled with a lot less bass, the sound is very different.

    So no, it doesn’t follow the Harman curve. It has more differences than similarities. Only similarity is in the upper-mids, and generally this is the area people find the HD600 too emphasized in, even though it’s refined more than most other headphone that have the same boost there, and has more lower-mids presence than the Harman target to counter the upper-mids emphasis.

    Let alone if it had the same lower-mids presence as the Harman curve. In that case, it would definitely be considered too bright and thin.
     
  15. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    The reason the math is questionable is because it's based on a simple regression fit and then average of absolute error. As such, this method unfair punishes headphones which have a smooth response (highly valued) but dark or brightly tilted. Same goes with dip, which are often more difficult to hear than peaks. Also, a headphone which conform largely to the Harmon Target, but has narrow spikes in bad areas are not punished enough. This is the problem with simple math.

    OK, let's do an eyeball test, but also be careful with the details.
    1. Be sure that you are using the Harmon 2018 Target, not the Oratory Optimum HiFi Target, which is actually pretty good. I'm glad to see Oratory finally get to his senses and not blindly follow Olive's shit curve. The target is still a bit too bright for my tastes though.
    2. Little things matter. 3db-5db difference over a large swath from 4kHz to 10kHz are a big deal. Don't miss the little details! According to Harmon, we are supposed to boost the area from 4kHz to 10kHz 3db - 5db for the Utopia and HD600. I think not!!!! Be wary of compressed Y-axis: upload_2021-1-26_8-35-16.png
    3. HE90, the "greatest headphone ever made" is actually a brightly tilted headphone, but with smooth response. If Harmon 2018 says it's perfect, then it prefers bright headphones. Also Harmon 2018 says the HE90 is bass deficient. You can't conveniently ignore this. The HD600 is great, but has boost upper mids around 3kHz. Harmon 2018 says HD600 is perfect around this area - Harmon prefers a mid-forward and bright frequency response (which is the bane of SBAF).
    4. You are cherry picking stuff from the RANKING list. Up there at top is the HD560S (which is complete shit, mainly because of where the peaks are located) and the HE4XX (treble spike). While HD650 is ranked below the DT990-250 (a U-shaped disaster of epic proportions) WTF?
    1. I think that's a misunderstanding of the term that "music lives in the midrange". When people like me say this, I am not only referring to the frequency response. I am referring just as much to the timbre, harmonics, distortion, smoothness of response, and cohesiveness with the highs and lows around the midrange. Oftentimes, it mentioned in the context of priority in respect to tradeoffs. You are making a straw man argument here.
    2. Even if we do consider the midrange from a purely FR perspective, we cannot ignore the highs. Boosted highs, even slight on the order of 3b-5db on top of the HD600 or Utopia per the Harmon 2018 would be a huge deal breaker with these headphones.
    3. When we do consider the mids, the mids of the HD600 and Utopia are considered to be not totally optimal (per SBAF). The HD600 is slightly boosted at 3kHz, the Utopia has a slight honk around 1.5khz and a small boost around 6kHz.
     
    Last edited: Jan 26, 2021
  16. Rthomas

    Rthomas Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    966
    Trophy Points:
    93
    Hi Marv,

    Ok, thanks for explaining the issues in detail. I think I understand a bit better now.

    Yes, Oratory's optimum curve is without the bass boost.

    And yes when my headphones are EQed to HT 2018 I guess I could use a bit less brightness. I find the bass boost satisfying although on some music it is a bit much.

    It would be cool if an SBAF optimum target could be added to Oratory's website for a few key SBAF approved headphones. I wonder if he would mind.....
     
    Last edited: Jan 26, 2021
  17. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Oratory's data is all there. It wouldn't be that hard to tweak his data to an SBAF curve and apply a different formula to get numeric scores. His measurement methodology / human processes look to be solid. Much better than Amir's. Actually, Amir shouldn't even be mentioned.

    Might be an interesting side project.

    However deep secret: most people really into the hobby do not give a shit about measurements, unless they are used to crap on expensive gear to appeal to the masses with < $200 chipamp gear feel better. Call it "populist" measurements if you will.
     
  18. Hands

    Hands Overzealous Auto Flusher - Measurbator

    Staff Member Pyrate MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    12,285
    Trophy Points:
    113
    Location:
    Colorado
    Home Page:
    Honest question. Is this sarcasm?

    While not quite the Harman curve, I do see a trend towards companies like Apple and Bose tuning in such a way that matches the RTINGS curve. And that curve seems to be more similar to the Harman one than not.

    Maybe you were being just that specific in what you said, or maybe I'm missing something obvious.

    Either way, I think it's only a matter of time before the Harman curve starts to be a dominating force in the market, given the influence of Reddit, ASR, and all the Youtubers.
     
  19. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,771
    Trophy Points:
    113
    Location:
    Padre Island CC TX
  20. Rthomas

    Rthomas Friend

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    966
    Trophy Points:
    93
    It would be super interesting to know which 1 or 2 headphones measure closest to the SBAF target in stock form.

    Of course I mean leaving aside all other qualities and focusing only on the frequency response.

    Thanks!
     

Share This Page