EQ conversion between Harman and Optimum Hifi

[Jose Hidalgo]

Hi everybody,

I asked Oratory1990 about this conversion, and he basically said : "create a Low Shelf filter, Fc = 105 Hz, Gain = -5.5 dB (adjustable to your liking), Q = 0.71".
He also said that to perfect everything, we could slightly attenuate some higher frequencies (around 9 KHz), only by about 1 dB and again, to your liking.
So as he implied, 3 or 4 filters total should do the job.

Using REW and both Harman (AE-OE 2018) and Optimum HiFi curves, I came up with these 4 filters that convert the first into the second really closely (look at the vertical scale) :

• Filter 11: ON PK Fc 80 Hz Gain 1.0 dB Q 0.45
• Filter 12: ON LS Fc 110 Hz Gain -5.8 dB Q 0.75
• Filter 13: ON HS Fc 6000 Hz Gain -1.2 dB Q 0.707
• Filter 14: ON HS Fc 11000 Hz Gain 0.8 dB Q 1.25

I'm happy with the result (plus everything can be adjusted to one's liking). I just wonder if there a supposedly "better / official / authorized" way ?

 

[Jose Hidalgo]

For an intermediate result between Harman and Optimum HiFi, one could try these four filters instead :

• Filter 11: ON LS Fc 110 Hz Gain -2.8 dB Q 0.75
• Filter 12: ON PK Fc 180 Hz Gain 0.3 dB Q 1.00
• Filter 13: ON HS Fc 6000 Hz Gain -0.6 dB Q 0.707
• Filter 14: ON HS Fc 11000 Hz Gain 0.4 dB Q 1.25

 

[Propheticus]

Why?
If they're essentially the same, apart from the bass region, why not adjust the Harman presets to you liking by adjusting the low shelf filter?

To quote Oratory1990 on what the Optimum HiFi curve is:

[...]
A headphone version of B&K‘s 1974 Optimum HiFi curve.

Turns out that it essentially confirms Harman‘s research: Same results at mid and treble, and slightly different results in the bass (which is well within the scope of Harman‘s research. Bass needs to be personalized, as the Harman papers make abundantly clear).

I just wonder if there a supposedly "better / official / authorized" way ?

Yes, the way you like most. They are preference curves.

 

[Jose Hidalgo]

Hi Propheticus. Yes, that's exactly what I said in my first post. They both are preference curves. Actually EVERYTHING is preference if we think about it.

My point here is that we could simply have a slightly more "scientific" approach :

• First, match the Optimum HiFi curve as close as theoretically possible
• Only THEN, adjust what we want to our personal liking, simply by adjusting the various filter gains.

I'm certain that the difference will be hard if not impossible to hear, but I prefer this approach intellectually. It satisfies me more.

Actually it's the same approach as what I consider to be the "Hi-Fi" approach : first try to have as much "fidelity" as possible to the original (even if often we can't hear the difference), and only THEN, adjust what we want to our personal liking (add EQ, add tube simulatos if that's your thing, add whatever you want).

I mean, arguing that something is "good enough" is not a reason to not try and do "theoretically better". That's also my point.

 

[Propheticus]

Actually it's the same approach as what I consider to be the "Hi-Fi" approach : first try to have as much "fidelity" as possible to the original (even if often we can't hear the difference), and only THEN, adjust what we want to our personal liking (add EQ, add tube simulatos if that's your thing, add whatever you want).

I think I can agree here up until the EQ part. With the exception I won't call tube simulated sound Hi-Fi in todays state of the art.
If tube sound is your thing, go ahead. Whatever floats your boat, just don't call it better or higher fidelity than my 'cheap' AVR with Class-AB amps.

"theoretically better" is something in the end only you can determine. You've taken a good starting point for preference: the Harman curve (or the other one.. potatoes), but it's a generalisation. Further tweaking for your specific preference, which might be an outlier, is not something anyone else will be able to tell you. So try out and double blind test yourself to decide between tweaks. I'd argue you need to limit yourself in this regard... what is the improvement you're getting with diminishing returns in ever smaller tweaks? Use some of that time to enjoy the music instead.

 

[Jose Hidalgo]

So we agree in the end Don't worry, I do a lot of listening too. As for tube sound, it's not specifically my thing, it was just an example. The examples above didn't take me much time, half an hour or so.

My point is like yours, "whatever floats your boat", as long as we know where we're coming from (a "Hi-Fi" basis to begin with). I just think it should be the same for EQ curves. Harman and Optimum Hi-Fi are two good starting points, based on well-documented statistical studies (one in 1974, one more recently).So why not try to match them both as closely as we reasonably can ? And then adjust things to our liking, knowing where we're coming from.

If we can easily match Optimum Hi-Fi from any Harman preset by just adding the 4 aforementioned filters, I see no reason for doing otherwise.