Flat Response

[Deward Hastings]

Re: Flat Response

Still zero evidence/facts/data to support your claims/opinions?

Still zero evidence/facts to support yours ? ? ?

No cherry picking, remember?

[Zilch]

Re: Flat Response

Still zero evidence/facts to support yours ? ? ?

No cherry picking, remember?

What do you want him to do, reproduce every speaker on the list?

I believe the evidence is clear -- the central tendency of modern loudspeaker design is flat axial frequency response, and most are doing a respectable job of it as evidenced by these independent evaluations. We also have credible research indicating that this practice is well correlated with listener preferences....

[Guest]

Re: Flat Response

What do you want him to do, reproduce every speaker on the list?

Nah, he just want's me to remind him of one in particular....:D

"Flat response" sells (to the ignorant)

LINKWITZ ORION

[Deward Hastings]

Re: Flat Response

All stereo recordings are artificial spatial constructs. ALL.


"Fake" spaciousness with an artificial spatial construct? The whole thing is fake!
The main spatial difference with dipole vs monopole is depth perception. 2D vs 3D. Otherwise, not much different at all.
Pinpoint imaging?? That's as unrealistic/"fake" as it gets


"Compromising" Imaging? Accuracy??? To what?

Ah, something we can (mostly) agree on.

All I want is a sound image that's a plausible replica of what was there, and enough "accuracy" that I can distinguish an Oboe from a Bassoon's upper register (or a Clarinet).

[Deward Hastings]

Re: Flat Response

What do you want him to do, reproduce every speaker on the list?

He's the one who specified "no cherry picking" . . . and would be the first to howl if I picked only the ones that slope down . . .

[Deward Hastings]

Re: Flat Response

It's the same conundrum; the paradigm is wrong

What "paradigm is wrong" ? ? ? "Acoustic" music? Reproducing music performances? Accurate recreation of live performances?

[Zilch]

Re: Flat Response

Ah, something we can (mostly) agree on.

All I want is a sound image that's a plausible replica of what was there, and enough "accuracy" that I can distinguish an Oboe from a Bassoon's upper register (or a Clarinet).

This is a rationale for cranking down the high frequencies, "there" being in an acoustic concert hall in lieu of any and all alternative venues?

Lynn Olson called it -- "Dull, dull, dull." :p

All stereo recordings are artificial spatial constructs. ALL.

Fine. Accurately reproducing what's there by definition implies flat frequency response. The result may or may not be "realistic." Take that up with the program producers ... and the paradigm....

What "paradigm is wrong" ? ? ? "Acoustic" music? Reproducing music performances? Accurate recreation of live performances?

Nowhere close, and we are near the dead end of this one. Talk to Ken Kantor about this; we haven't yet even properly framed the questions.... :o

[stephed]

Re: Flat Response

All this posts and in my opinion one very big thing about flat response was missed.

We are all effected by noise induced hearing loss and age induced hearing loss. The problem with either one, they don't reduce our hearing evenly across the full 20-20K bandwidth. As a general rule high's always go first than midrange then lows.

This means that what a speaker sounds like to you in our teens is not the way it will sound to you in your 20's, 30's, 50's or 80's. It's common that hearing loss is not even from left to right ear.

Theoreticlly to build the perfect speaker we would need to take a full hearing loss exam and compensate for noise & age induced losses.

This would definately not lead to a speaker that measured flat, but would sound good to you.

Plus, were are all effected by what we have gotten used to in the past.

This is just a theory of mine. Just another thing to think about.


Ed

[auracle]

Re: Flat Response

All stereo recordings are artificial spatial constructs. ALL.


"Fake" spaciousness with an artificial spatial construct? The whole thing is fake!
The main spatial difference with dipole vs monopole is depth perception. 2D vs 3D. Otherwise, not much different at all.
Pinpoint imaging?? That's as unrealistic/"fake" as it gets


"Compromising" Imaging? Accuracy??? To what?

Very true. Complete accuracy is a tail chasing exercise with a playback system that cannot possibly create the exact spatial sonic image of what one would hear listening to a live performance due, at a minimum, to inevitable differences between each ear's proximity to an actual live event source and that of two speakers to each ear in your living room. Even if you were to duplicate what a person actually hears at a live event with a mannequin head recording of it - you can't reverse or undo the process with speakers that are several feet on either side of a person's forehead. The crosstalk pattern going into the recording cannot be preserved exactly as the crosstalk one would experience coming back out - particularly if one is walking about a living room with speakers placed several feet apart at one end. And with that consideration, we haven't even addressed the impact of room acoustics where wall, floor, and ceiling reflections - particularly in the lower frequencies (300 -1500hz) are prevalent. Linkwitz's concepts are valid given the obvious flaws inherent in a stereo playback system - his approach asks - "what methods can we use to fool the listener into thinking they're hearing something real"? From this perspective, the concept of direct/reflecting approaches like the dipole and to a lesser extend - bipole, make perfect sense The goal in general is to reduce or eliminate multipath in the vicinity of time zero so as to maximize the speaker's control over spatial cues. Then add enough reverberant effect to fool the listener into thinking there is something really happening in my living space. This obviously will work better for some recordings over others depending on the listening space. A close mic'd guitar will sound more natural with a little backwall reflection because that is precisely what you'd hear if someone were playing an acoustic guitar in your living room. Trying to re enforce the precise reverberation of a huge concert hall in your living room - now that's what separates the skilled from the wannabees.....

[dlr]

Re: Flat Response

All this posts and in my opinion one very big thing about flat response was missed.

We are all effected by noise induced hearing loss and age induced hearing loss. The problem with either one, they don't reduce our hearing evenly across the full 20-20K bandwidth. As a general rule high's always go first than midrange then lows.

This means that what a speaker sounds like to you in our teens is not the way it will sound to you in your 20's, 30's, 50's or 80's. It's common that hearing loss is not even from left to right ear.

Theoreticlly to build the perfect speaker we would need to take a full hearing loss exam and compensate for noise & age induced losses.

This would definately not lead to a speaker that measured flat, but would sound good to you.

Plus, were are all effected by what we have gotten used to in the past.

This is just a theory of mine. Just another thing to think about.


Ed

This is true, but with gradual loss, we aren't fully aware of the top octave loss. We judge a speaker's reproduction by what we hear live today unless loss is extreme. I can't say that I'm aware of my inability to hear 18KHz today vs. when I was young. I'm aware primarily when I play an 18KHz tone that I did as a test in college and I heard it easily then, but cannot hear it now. Otherwise, I'd have no idea.

I'm not going to get into this thread much more, some folks aren't contributing as a debate, only as a hammer. But as far as the top octave being flat or not, my (anecdotal) experience with the folks who attend DIY NE surprises me. I find a tendency towards preference for (what I consider) more exaggerated highs from the younger guys. What many of them like would drive me out of the room over time. Maybe this has to do with preference of music, maybe the way music is produced today, maybe something else. It is also likely why a lot of system designers create flat on-axis. They've got to stay in business. What sells in stores is boom and sizzle more often than not and only less exaggerated in the "high end" I think. Comparison of what sells in a store is, to me, not representative of what people might prefer in a controlled test.

I've got some designs of mine flat on-axis. I've got some with a 1-2db downward slope. I try to achieve the same sound as I perceive it. Variables such as dome tweeter diameter (my only work so far) seem to be an important factor as well as how much felt I've applied that alters the off-axis. Were I to listen to a system at higher volumes than I do now, I suspect that I'd have to roll off the highs a bit more. As it is, I don't crank it up often, so less rolloff tends to sound better. There's no way around Fletcher-Munson.

Were I to try to make a living rather than produce the "most accurate sound", I know I'd design closer to flat to get the most sales. :( I doubt that I would design for what sounds most accurate to me, I expect that I'd have to go for what sells or fail in the business. That's one reason I'd never consider trying to make a living selling speakers.

dlr

[biff]

Re: Flat Response

All this posts and in my opinion one very big thing about flat response was missed.

We are all effected by noise induced hearing loss and age induced hearing loss. The problem with either one, they don't reduce our hearing evenly across the full 20-20K bandwidth. As a general rule high's always go first than midrange then lows.

This means that what a speaker sounds like to you in our teens is not the way it will sound to you in your 20's, 30's, 50's or 80's. It's common that hearing loss is not even from left to right ear.

Theoreticlly to build the perfect speaker we would need to take a full hearing loss exam and compensate for noise & age induced losses.

This would definately not lead to a speaker that measured flat, but would sound good to you.

Plus, were are all effected by what we have gotten used to in the past.

This is just a theory of mine. Just another thing to think about.


Ed

Not too sure I agree completely with your theorem - Your idea assumes a perfect audio memory that would let you compare what you heard at 17 at a given event with what you are recreating, say 35 or 40 years later. I think that our references are, for the most part, more recent, or at least should be! If you go to a concert tonight or even to a friends house and hear him play that should overwhelm what you "think" you remember decades back and if you use that as a target for your setup you will have acheived the goal. And most gradual hearing loss is a diminishment, not a brickwall cutoff, so it is unlikely that you would design a system that would have a nasty peak that is outside your threshold and yet be very noticable to someone who has a little less "experienced" ears .

This is assuming that as a music lover you do have a more recent memory than Deep Purple in a gym as your last listen and current target. That said, you touch on something very relevant, and that is taking care of your hearing so we can continue to appreciate the music all around us by wearing protection when appropriate, periodic cleaning and maybe testing to see where you stand. Last spring I felt a small loss of amplitude in my right ear in the upper mids and local DR shrugged, saw nothing, gave the usual response (you can't be young forever). I booked a test at an out of town ENT specialist where the young lady administering the test opined that she wished her hearing was as good as mine (at 54) but saw a drop in response of a db or two as I noted, and combined with an exam they determined my eustachian tube was constricted and not allowing pressure equalization and gave me an inhaler that took care of it in a few days. Point being that hearing is relative, as you say, but your reference should be updating also so your perception should be mostly accurate for a long time. If you have a loss of hearing that you perceive you should get an exam or cleaning and let the music flow.

And, for everybody - not sure where it started, but "definately"? - c'mon, let's not lower the bar too much, eh?

[Xyrium]

Re: Flat Response

Sounds about right.
For an incredible demonstration of how dry a living room sounds, Denon recorded a symphony in an anechoic chamber. It's so dry, it feels like your ears are being sucked from your head, and my room is "wet".
http://www.amazon.com/Denon-Hi-Fi-Wo.../dp/B0000034M9

Yes, that's how a dead recording sounds, but not a dead living room (reproduction), correct?

I personally like a fairly dead listening room, because I hear what the recording sounds like, and not my room. Of course, without 12" thick absorption on every wall, I'll never get there (anechoic chamber). I'm sure some recording engineers dial in a little EQ to ensure compatibility with dead and live rooms. To me, a reverbant room causes me to lose the timbre of the instruments FWIW.

Any recording engineers here to chime in on what you try to predict? I also suppose it depends on the genre.

[Xyrium]

Re: Flat Response

You make a good point here, because ultimately, it's what sounds good to the individual. Flat response to me sounds terrible. That's why it's never good to buy gear based on someone else's opinion.

This thread is probably going to hit the limitation on posts before all of the variables could be discussed. Toss in THD of the inner ear (don't forget, every canal is shaped differently), and even the reduction or count of TRPA1 proteins if we're going to cover the physical attributes of human hearing.

The thread has already gone into the realm of the subjective, though some of the theories put forth may contain facts.

All this posts and in my opinion one very big thing about flat response was missed.

We are all effected by noise induced hearing loss and age induced hearing loss. The problem with either one, they don't reduce our hearing evenly across the full 20-20K bandwidth. As a general rule high's always go first than midrange then lows.

This means that what a speaker sounds like to you in our teens is not the way it will sound to you in your 20's, 30's, 50's or 80's. It's common that hearing loss is not even from left to right ear.

Theoreticlly to build the perfect speaker we would need to take a full hearing loss exam and compensate for noise & age induced losses.

This would definately not lead to a speaker that measured flat, but would sound good to you.

Plus, were are all effected by what we have gotten used to in the past.

This is just a theory of mine. Just another thing to think about.


Ed

[MSaturn]

Re: Flat Response

I don't think there are post limits.

I've never seen such a lengthy debate about optimal FR - or one with as much actual reference citation. This is a good thing.

[Guest]

Re: Flat Response

All this posts and in my opinion one very big thing about flat response was missed.

We are all effected by noise induced hearing loss and age induced hearing loss.

It wasn't missed. Personal taste/HF hearing loss is a variable, not a constant.
Hence, design flat, linear, undistorted amplitude response...then eq, "treat", etc, etc. to your hearts/ears content. That's why cars have adjustable seats and volume/bass/treble controls on the stereo.
There is no way to design a fixed distortion of the on axis (or sound power) to suit the whims of a few individuals. The needs of the many outweigh the needs of the few (who assert that their personal TF should supersede all others) ;). Check out the Soundstage link and erase any doubts (or just look at all the graphs I provided. You don't even have to get past the "A's" to see that reality). Real researchers/designers know this and produce accordingly.

cheers,

AJ

p.s. BTW, even basic, $200 HT receivers now have multi band eq as part of their (auto option) system calibration/adjust ability. It's only the grumpy old Luddites that don't have tone controls on their equipment