Equal Loudness and Harman Curves

I'm trying to wrap my head around the relationship between the equal loudness (Fletcher Munson) and Harman curves. The Harman curve for headphones seems to be a purely subjective sound preference curve derived from listeners being given eq controls. But in the articles I've read so far, I have not seen a reference volume. The equal loudness curves suggest that listeners would tend to prefer more bass boost as volume goes down, so a reference volume would seem to be necessary. So, is the this target headphone response curve just compensation for equal loudness response at a given reference level?

 

From a little bit I read, I'm guessing they feel the response of headphones actually shouldn't be quite flat, in order to deliver a better listening experience. Maybe so. My loved AKG K240DFs were "free field equalized" to sound kinda more like speakers in a room, more open, and to me sounded way better than the K240s. The big problem is that headphones' response depends on the source impedance and the loading from the head/ear they are mounted on. Therefore you unfortunately cannot measure them and say "this is correct/flat!"

As for reference level, something like Yamaha's variable loudness should ideally yes be calibrated to a reference volume. I don't think they do that or talk about it, probably figuring that is way beyond their customer's comprehension/level of wanting to fool around. I think Harman likewise is probably just using average listening levels and not getting so super detailed.

 

The F-M equal loudness curves are all measured using single tone sine waves. This is far removed from the human perception of music.

Back in the 1970's and early 80's a loudness function was incorporated in most integrated amplifiers, the idea being that at lower listening levels, the set up tried to mimic the equal loudness curve by selectively boosting the lower and higher frequencies. It was hardly ever used, and was eventually dropped from products from the mid 80's onward.

Harman (as in the company Harman) have been publishing a range of different curves for regular headphones and inner ear phones over the last five or so years. This is a good write up about this Harman tweaks its headphone target response .

 

Yes, already read that one. Note that in the first paragraph it refers to the tonal balance preferred by listeners.

I don't agree with this statement at all. The curves ARE based completely on human perception.

 

85 dB (slow, C-weighted) says post #22

Theory: If I EQ headphones to Harman AE/OE curve, what other attributes should I look for?

 

"The basic methodology is the playing of two pure tones using sine waves at set frequencies and at incremented sound pressure levels (volumes) while having a listener report when they perceive the two as being at equal volumes."

From Fletcher Munson Curve: The Equal Loudness Contour of Human Hearing

And even wikipedia "An equal-loudness contour is a measure of sound pressure (dB SPL), over the frequency spectrum, for which a listener perceives a constant loudness when presented with pure steady tones."

So sure they give a threshold of perception under those circumstances. Musical perception is an entirely different kettle of fish.

 

From what I am looking at, this Harman curve just seems to be the same as equal loudness curve at a set frequency. So, any headphones designed per the Harman curve would just have the effects of a loudness button designed into it.

I didn't really expect that this was going to have to be debated. I think the majority of folks hear this change in tonal balance as volume is changed. And we all know, because of equal loudness curves, it is critical that volumes are exactly matched when a/b ing equipment.

 

That seems to be a non sequitur. You are saying (a) that tonal balance shifts with volume, and then (b) that volume should be matched. Either statement is true in and of itself, but the two are not necessarily connected. And tonal balance with volume is subjective, age related, and any other hearing issue related.

Anyhow nothing is absolute here. FM etc were averages of very many tests, which is right and proper in any subjective test involving auditory perception.

 

they are related because many people prefer more bass and treble than is natural because our ears are less sensitive to these frequencies and more sensitive to midrange frequencies among other reasons.
they do not represent or accomplish the same thing but are certainly related.