[
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2008, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]
121.222.225.150
In Reply to: RE: why doesn't mid-fi harshness show up on FR tests? posted by BillyBuck on April 24, 2008 at 22:36:01
Frequency response plots are done by sweeping a tone slowly from 20 Hz to 20 kHz. Only one frequency is being played at a time and what is being measured is the output at that frequency. Distortion occurs at other frequencies than the test frequency.
In addition, music rarely consists of a single frequency. Even a solo instrument playing a single note is generating sound that includes more than a single frequency because the sound produced contains the note being played plus a series of overtones of different strengths which are what gives that particular instrument its tonal character. Music is a much more complex signal than a test tone and our ears respond in different ways to measuring instruments.
There's also a personal taste factor regarding what type of high end response different people like. In a hall at a live performance the high end gets rolled off slightly because air actually absorbs high frequencies and the hall is large enough for that absorption to become noticeable. It's unlikely you'll notice high frequency absorption by air in rooms the size of most listening rooms so that makes the top end sound a little hotter anyway. A speaker with a genuinely flat top end response may well sound a little bright in many rooms, especially if you're listening on axis and there's a lot of hard reflective surfaces like glass in the room since they reflect more high frequency sound than most walls and much more high frequency sound than curtains and soft furnishings which are absorptive to some degree at high frequencies. Toss a recording that is overly bright in the top end into the mix and you'll certainly have lots of people complaining about brightness, and some even with normal recordings in the wrong room.
So those are the factors that come into play:
- single frequency tones for the tests vs music which is a more complex signal;
- how test instruments respond vs how people respond;
- anechoic chamber tests, or simulated anechoic chamber tests vs listening in a room which can boost the top end if it's overly reflective at high frequencies;
- some recordings are overly bright anyway and make any tendency towards top end brightness in a component sound much worse.
The tests are fine as far as they go, but they're not made using music, they don't show what the room is adding to the sound, and they don't take bright recordings into account either. How a speaker measures in an actual room will always be different to anechoic or simulated anechoic chamber measurements and every room is different so even if the tests used actual room measurements those measurements would only show how the speaker worked in that particular room and not in yours. Results in your room may be better or worse than the results obtained in the room the tests were made in, but they certainly would not be identical.
Frequency response plots provide some useful information but they need to be interpreted carefully and you need to take the room the speaker will be used in into account, as well as the sort of recordings you like and your own preferences for a warm sound with gently rolling off top end or a cooler sound with a more extended top end.
David Aiken
Follow Ups:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2008, Audio Asylum, all rights reserved ]