I was under the impression that the preference score is a predicator for how much the market is going to like the product in comparison to a contender. It aims to maximize on average the individual person's willingness to pay and keep the product over the other. As far as I was told a difference between 80% and 95% is provided, but it is only due to rank; the bigger or lesser difference doesn't make sense as a distance. (see:
https://www.statology.org/levels-of-measurement-nominal-ordinal-interval-and-ratio/). We are on an 'ordinal' scale. You don't say otherwise, whilst handing out quite accurate numbers. That said only for clarity.
A more crucial question came to mind. The Harman is a curve again addressing preference on average. It is smoothed and generalized so much, that must-be-there features are suppressed. In short, because it addresses the average, the Harman is decidedly an optimum for no one. It is always more or less wrong because people have different head and ear shapes besides just taste. Do you agree?
If (not) so, what about finding a personal preference to be exact? (A) There could be an optimization for the 3kHz region. The amplification maximum could be at a different frequency, the amplitude be different, and its sharp or broad shape may differ (objectively) for individual people.
(B) Same with the 10kHz trough, which is told to be a must btw, and the 14kHz something peak, again a must.
Did you ever try to identify your optimum by preference, after miticulously (OTT as you say) equalizing the IEM to Harman as a starting point? Leave all corrections in for Harman, and then find a deviation that fits your hearing the most by setting additional filters?
I suggest you read the links in the preamble of my EQs.
[...]
- The EQ Score is designed to MAXIMIZE the Score WHILE fitting the Harman target curve (and other constraints) with a fixed complexity.
This will avoid weird results if one only optimizes for the Score, start your journey here or there.
There is a presentation by S. Olive here.
- The EQs are starting point and may require tuning (certainly at LF and maybe at HF).
- I sometimes use variations of the Harman curve for some reasons. See rational here and here
[...]
Apologies for the tone of the to last links but there is some information.
Some quotes:
[...] Nov 2021
This study, demonstrated according to S. Olive that:
•A 64% (the majority) of listeners prefers headphones with a frequency response that adheres to the Harman Target Curve.
•A 15% portion of listeners prefers 3 to 6dB more bass, usually younger males and less experienced listeners.
•A 22% portion of listeners prefers 2 to 4dB less bass usually older population, and biased towards females.
Based on this data one can roughly derive preference range, neutral being the default Harman curve:
•-3dB/Neutral/+3dB/+6dB as EQ for bass preference;
•-1.5dB/Neutral/+1.5dB for as EQ for treble preference;
That means that you can still be within the “taste zone” of the Harman target while not being completely on the “majority” curve.
You just change the likelihood of the EQ to be optimal for the user, but still rather close to it...
A number of people will actually prefer that, it's also a major piece of knowledge that is often ignored.
A marketing department might even tailor the default curve to cater best for a certain audience…
@amirm is probably +3dB@LF -1.5dB@HF as I have already casually observed, but the listening material also have an influence…
Again these EQ are starting point as clearly stated, for about a third of the population it is not the ideal curve...
The map and the country, or if you prefer, model vs reality.
The users must not comply with the target but the target can be adjusted to the user.
All the models are good
approximations so most probably, in the details, every user has his/her own slightly different preference target.
Again from the study at least a third of the population exhibits significant deviation from the default target.
The whole "taste zone" is the "Harman target curve" for the complete population (with "normal hearing").
It may not matter if we are only ranking devices against each other (although it may, see hereafter) but it will matter when trying to EQ to the exact target. Readers need to understand that and I wrote it in my EQ preamble.
As a matter of fact, to me, one of the greatest outcomes of the study is that yes, differences exists between users but they are quantifiable and trends are clear to be seen.
How long before AIs will automatically adjust the EQ based on the material been listened to and your Facebook account?
Note that measurements for physical (anatomic, fitting etc.) variations is very much possible already (otoacoustic emissions and self/adaptive EQing products)
[...]
My preference is basically on the average target, with maybe a tad less HF.
As quoted I have devised a simple strategy to get even closer:
Based on this data one can roughly derive preference range, neutral being the default Harman curve:
•-3dB/Neutral/+3dB/+6dB as EQ for bass preference;
•-1.5dB/Neutral/+1.5dB for as EQ for treble preference;
So the base EQ (or OTT) + two shelves + a lot of blind A/B listing = very close to your ideal curve.
That all you can safely conclude from the available results. Finer or larger adjustment may be related to outliers or subject with abnormal audition.
"(B) Same with the 10kHz trough, which is told to be a must btw, and the 14kHz something peak, again a must."
Do you have hard evidence for that, in in particular in an IEM?