Quote:
Originally Posted by gorman
A sound at 58dB is perfectly audible, unless you have hearing damage. And it may well be that the sound engineer that mastered that particular CD was using equipment with a -12dB response at that frequency. In that case, the more faithful reproduction is that with the -12dB response.
|
Indeed, a very good point. That goes even more to making the case that flat curve doesn't necessarily matter though, doesn't it? If you can get a good audible detail along with some coloration that you personally would prefer one way or another, does having a flat curve really matter?
I think another consideration is that no recording is really mixed with flat curve in mind. Given how few audio equipment out there is flat, a flat curved reproduction equipment is not what the recording are mastered for. If everyone had equipments are that essentially flat, and all recordings were mastered to that, then of course a flat curve will preserve probably is the most faithful rendition of the music itself. That's just not the case though.
Now, to get even further into discussion on why curves means very little to how a particular equipment sounds. Curves are simply measured by using test tones.. it basically tells you what the volume of a particular tone is across the entire spectrum. It does not take in account all the other aspects of sound except for producing a tone.
It does not measure decay, pacing, reverberation, warmth, soundstage, sound imaging, or any of the characteristics that's crucial into making good audio equipments.
It does not tell you that UE-10 looks like it has a more enhanced bass when it's just producing a singular tone, but the lack of a sufficient decay makes the bass much leaner and less satisfying than the bass that 2X-S produces. It also doesn't tell you that even though with the test tone, the bass seems to be mighty fine, with a reduction in volume, UE-10's bass falls off dramatically faster in comparison with some other headphones. Which means you have to crank the UE-10 up to a higher volume level before a satisfying bass is achieved.
Frequency response chart is a very, very small part of the overall equation towards the evaluation of a headphone. Not to mention the different measuring methods that can screw up the test. Especially with IEM's, difference in the ear canal shape of the dummy head can dramatically change the test results. However, the results will probably not be dramatic to us, as our perception of overall sound should be pretty consistent. When measured with a machine though, the curve will look drastically different, won't it?
For example, if my ear hears 60db versus 65db... hmm.. there is a difference, but the overall sound impression doesn't change that dramatically. When a microphone measures 60db versus 65db... omg... look at how the curve on the chart just fluctuated.
What if Ultimate Ear's testing method isn't really the most optimal way to conduct those tests? Who can say their dummy head is the most accurate as opposed to another dummy head? What is an acceptible, universal testing parameter? If Ultimate Ear's testing method is a little bit off... say.. by a few mm of difference inside the ear canal, then what have they been calibrating their earphones to this whole time?
A lot of reviewers and magazines do not use frequency response chart in their evaluation of a particular audio equipment. The testing methods and such are just too much in flux in itself. Frequency response chart can be helpful in identifying a particular flaw or characteristic you've heard from a particular equipment, but only after you've heard it first. It cannot be used the other way around to judge the qualification of a particular piece of equipment; it's equivalent to judging the usefulness of a book by looking at its glossary.
