FR graph doesn't always = what the headphone will sound like. It gives a good distinction, but won't prove a headphone will sound like X. It can confirm subjective findings though. Just can't prove them on its own.
It does give a good indication of how something might sounds, but it isn't an in concrete rep of what it actually will sound like. The aggressive treble you stated as an example. What you thought was aggression on the SR60 graph might have been something else in the real world setting. Reading an FR graph is still not easy, we still don't know everything. We know how different sounds come about most of the time. However, there are other ways to create the same type of sound. Sometimes different properties of sound have a same general shape and size to them (while other occurances may differ as well). There is no guaranteeing that you're looking at an orange at a distance when it could be a small grapefruit.
If you look at a PFE232 graph, you'd think the treble was spiked and hard and aggressive. It's actually soft and more reserved to many people's ears.
EDIT:
I do want to make a statement about the objective vs subjective. They both try to do the same way. They both try to draw a graph. Your "objective" statements were as subjective as your subjective ones. We still try to do the same thing though. We try to draw an FR graph with our words and what we hear. We don't hear tones, so the FR graph is a little harder to do. We listen to music. Music has properties that relate to the FR in different ways. Each property we list helps the reader realize what we're talking about in relation to the FR graph (with respect to our ears). Then the magnitude is normally shown as a numerical word (good, fair, strong, weak, present, etc) to give an idea with the depth or amount of the property we hear. Both of these together give readers and idea of what frequency range we talk about + the relative amount it will be boosted or reduced. Some examples below, of your objective and subjective statements (which really were all subjective):
- Bloated is not a specific shape. It can actually have lots of forms. Bloated normally comes for the boost of the lower bass or even the higher mid-bass. It's a bump somewhere that takes away from any other bass property (like punch, texture, and impact) by creating too much quantity.
- No treble. This can mean a few things (it's not specific enough). It could mean that the higher highs are lacking. You end up with a lack of sparkle and extention to the point where they are so recessed they rarely show up. It could also mean a loss of treble as a whole. No snaps can be heard, and you can forget about extension.
- Accenuated bass. This is subjective as well as your ear should also pick this up in the real world. If not, there is something up with the objective data being used. It means some part of the bass spectrum has been raised. Once again, like the no treble form, it isn't specific enough. It's as bad as saying bloated bass or no treble. It isn't specific enough. I find it shameful you'd call this objective.
- Reduced treble. This is as bad as no treble. It gives no details on what area of the treble (like the no treble). It's the same as no treble to a lesser extent.
- Notch cut out of midrange. The funny thing about our ears is that we won't hear this notch unless we listen to an FR sweep. So this can't come up in subjective data, ever. It is objective though. But will only come up 100% if we listen to an FR sweep. In natural music, the notch will just end up creating a recessed sound that may create artificial sound stage. The off tones will make up for the notch and "hide" it.
You had one statement that was objective. Notch. That's the only thing that could be proved objectively. I'll give you that. However, it doesn't create recessed vocals as you say. It creates a lush sound to them taking away from the main pitch but also keeping the energy (I made my own notch at 1k with EQ; notched down about 6 dB).
EDIT:
A nice teaser pic
Edited by tinyman392 - 9/1/12 at 10:07pm