24.13.232.33
In Reply to: RE: An Honest Question. posted by thetubeguy1954 on May 03, 2009 at 10:08:27
So here is the flip side of your question.
Have you ever heard two different types of speaker that sounded the same, either in a room or even outdoors (quasi anechoic)?
My guess is no, even though you feed the exact same signal to each right?.
So at least anechoically, one can focus on what the speaker does relative to the sound reaching your ears or more accurately what the speaker does to the sound in addition to the original signal, because if it didn’t add or subtract, the two would be identical.
The issue takes an ugly turn once you start looking at what a speaker does relative to what it is told to do. It screws up in so many ways It is hard to know where to start.
The most important thing to keep in mind is your dealing with several dimensions in an unseen space.
For example, what good is flat response?
It means the speaker produces all notes within that range at the proportional loudness as on the recording. On a graph this is loudness vs frequency
Flat is obviously ideal but less obviously, if one has home speakers which actually measure flat, they tend to sound “bright” and or sibilant in a living room.
Most home speakers have a deliberately rolled off high frequency response for that reason. At the low end, you ears are progressively less sensitive as the frequency falls, at 20Hz, a signal has to be about 80dB just to be detectable, a hundred million times more acoustic pressure than what you can detect around 3KHz.
Flat to DC is the ideal but the acceptable cost and size prohibit actual reality.
What if you had “flat” speakers when measured at a low level but found that the response changed as you changed level? This is normal power compression response (change in response due to heating of the voice coils and crossover parts, the only solution is headroom.
What about distortion?
So what if the speaker is flat, but in addition to producing any tone you fed in, the speaker generated a quieter set of tones that were 2,3,4,5 etc times the original tone.
With music, these “free sounds” can either sound bad if they are mostly odd or sound nice if mostly even (being a musical chord interval away), or to the passer by sound “loud”.
Keep in mind this effect as well as some other problems speakers have is related to loudness and the problems get louder faster than the intended signal, the only solution is headroom.
What about time?
So what if a speaker is flat but when you put in an instantaneous impulse which represents a wide frequency span, the sound that comes out occupies a span of time 10 or 100 times large, in effect spreading the impulse out by frequency?
How much and at what frequency is acceptable is a matter of debate.
What about space?
Obviously where the sound goes in a room makes a large difference.
While modes are usually not that audible close reflections, especially to the sides can absolutely kill imaging. To extend the direct field as far as possible, directivity is your friend, unfortunately meaningful directivity is only possible via physical size, a horn or large panel etc.
I can tell you this, that all the speakers we make I use a “flat (with a small hf roll off)” target for all the alignments, I set the time target to be what is appropriate for the target response curve.
As much as that is so, the different models do not all sound exactly the same. For example, an sh-95 and sh-50 sound very similar until you reached loudness where the sh-95 ran out of gas first and the sh-95 has a higher low cutoff etc.
The sh96 and sh64 sound similar but not as much like the 50 or 95.
I do think that, as you get closer to producing only the actual signal, in time with the actual signal and as little else as possible, that it is inevitable that systems sound more and more similar.
Best,
Tom
Follow Ups:
Post a Followup:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2009, Audio Asylum, all rights reserved ]
[ Go Green ]
[ Take5 ]