Flat Response

[dano]

I would like your opinions on actual vs. perceived flat frequency response. Some say a flat response sounds to forward or even harsh when auditioned in typical listening rooms,
yet it seems a flat response is an important design goal for many designers.

Do you like or dislike the way speakers with measured flat response sound and why? People who have actually listened to measured flat speakers can add worthwhile insight into this. I?m certain some speakers sound nice with a flat response while others do not. Let?s examine the causes for this. I believe we can gain valuable incites into speaker design that?s not possible just studying books and learning theory. It has been said Experience is the greatest teacher.

I would like to keep this informal and friendly if possible. Anyone who wishes to participate is welcome and encouraged to voice their opinions. I don?t expect to solve this debate here. I think it will be interesting to hear everyone?s views. Disagreements are good, and I?m confident everyone will be courteous and respectful.
________
LAMBORGHINI BRAVO

[Pete Schumacher]

Re: Flat Responce

A flat on axis response in a room may result in a non-flat response at the listening position when all reflections are factored in. Flat power response along with a flat on axis response is preferred for that reason.

A well treated room is as important as a good system.

[MSaturn]

Re: Flat Responce

He got everything right. Forgetting about your room is a bad, bad thing.

FR at both the assumed listening position, and the power response (off-axis) must be accounted for. A 'bright' sound is almost certainly a lot of upper-midrange reflection (in a well-designed system.)

[DDF]

Re: Flat Responce

"I’m certain some speakers sound nice with a flat response while others do not."

That just about sums it up.

It depends on the tonal balance of the off axis sound, how lively the room is and how big it is. Generally, the smaller it is and the livelier it is, the more the off axis response affects tonality.

Very often, getting a nice flat line on axis requires launching a poorly balanced sound into the room at other angles. The smaller and livelier your room, and the more noticeable this will be as a tonal imbalance.

Examples:
1. Height and frequency of diffraction peaks from baffles are very dependent upon listener location. Designing an xover to perfectly flatten this out at one observation by necessity calls for the response at other observation points to suffer. In a smaller, livelier room, its been my experience that the design should allow some error on axis to get a better balance with other listener locations and the indirect room sound contribution.

2. Tweeter off axis energy can be excessive with a cookie cutter LR4 designs (for example), IME. Sometimes its better to cause a little dip on axis where the tweeter cuts in, to compensate. Many times IME its better to go with a lower order over which may not be as hyper flat as an LR4, to reduce the tweeter's off axis energy into the room near xover.

3. Different radiation patterns for different drivers may call for the design to be voiced non flat. Good example is Wayne and Curts speakers with the dipole mid. They tuned it differently on axis to compensate for the extra rear firing energy of the mid

4. Room placement near boundaries. Response measures are invariably taken "quasi anechoic" without room effects. Place a "flat " speaker against the wall, and the upper bass will be too prominent. Lots of folks here design for 2 to 4 dB BDC when placement is near a wall, to compensate. "Quasi anechoic" this would show up on the graph as too low a level in the lower mids/upper bass.

5. If used for home theater only, it may be beneficial to put a small peak near 3 kHz on axis, to aid voice intelligibility

6. Very low frequency subs will see "room gain", a boost below the frequencies where standing waves no longer generate. Its better in this case to overdamp the response so that it prematurely rolls off, to match the room gain

7. A room with too high a reverb time in the high frequencies (ie too little high frequency absorption) will benefit often from a gentle high frequency roll off, to counter the strident excess high frequency energy in the room

8. A speaker with a very restricted low frequency will often sound too bright if the high frequencies are too flat. It may sound fatiguing over time. In this case, it often helps to start a very slow high frequency roll off above 5 kHz or so.

9. Listening level. If the speakers are intended to always be used at low SPLs, you can often get away with a some low frequency boost, to compensate for the ear's lower sensitivity in the bass at low SPLs. This takes the tonal balance and adjusts it closer to what it would sound like at realistic (louder) listening levels.

10. If the speaker is meant only for home theater and not music, a few judicious peaks in the response at carefully chosen frequencies can make it sound more "dynamic", but at the cost of fidelity with music.

11. Listening time. If they are intended to be used at night when the kids are asleep, it may be better to restrict the low frequencies and roll them off early. My computer speakers are this way and I can play them louder without complaints from upstairs, since bass travels through walls and ceilings much more readily than mids.

12. Listening outdoors at largish distances results in more high frequency absorption than listening in the house. A bit of treble boost could help.

13. Radiation pattern. Single driver (cone type) systems beam at higher frequencies. IME, having a very high frequency rise helps make these sound more balanced in room.


Dave

[Zilch]

Re: Flat Response

Consider what flat anechoic response means -- that the speaker is accurately reproducing the spectral content of the source at the measurement angle. Axially, Job #1 by most metrics, but not the entirety of it by any means or measure, and also not to be confused with in-room response....

[MSaturn]

Re: Flat Responce

3Khz? The human voice is fundamentally from about 80Hz-800Hz .. I don't believe this business about 1-3Khz. Listen to a sine wave at those frequencies and tell me you've heard someone speak that high :eek: ... The harmonics in my voice probably don't get above 320Hz, besides things like glottal noises and fricatives.

Vocal intelligibility is based - at least in english - in vowels; a dipthong, for instance, is not going to include high harmonics like the "ess" or "t" noise, which are the things that get amplified in the 1-3Khz range.

[Guest]

Re: Flat Responce

I believe we can gain valuable incites

I'm with you on this

A well treated room is as important as a good system.

...and I would say, a good system importantly needs no room "treatment".
(Or as Sean Olive might say, why waste heat in these energy conscious times?)

cheers,

AJ

[Pete Schumacher]

Re: Flat Responce

3Khz? The human voice is fundamentally from about 80Hz-800Hz .. I don't believe this business about 1-3Khz. Listen to a sine wave at those frequencies and tell me you've heard someone speak that high :eek: ... The harmonics in my voice probably don't get above 320Hz, besides things like glottal noises and fricatives.

Vocal intelligibility is based - at least in english - in vowels; a dipthong, for instance, is not going to include high harmonics like the "ess" or "t" noise, which are the things that get amplified in the 1-3Khz range.

Well, then you haven't been listening to vocals while watching a spectrum analyzer.

I have, and the harmonics reach well past 1KHz. And some female vocals have fundamental notes over 800Hz, with the obvious harmonics above that.

[Pete Schumacher]

Re: Flat Responce

...and I would say, a good system importantly needs no room "treatment".
(Or as Sean Olive might say, why waste heat in these energy conscious times?)

cheers,

AJ

I wonder how a system would sound in a tiled bathroom?

[MSaturn]

Re: Flat Responce

Perhaps I'd expand it to 1Khz. That's about the 'soprano C'. Keep in mind .. well, I'll say everyone speaks lower than they sing (tessitura-wise.) When we're talking about speaking intelligibility, I'm sticking with definitely below 1Khz. C7 is a silly-*** note, and doesn't happen very often at all.

Extraordinary examples (women like Mariah Carey, for instance, who can get up to 3Khz, even if that's basically a whistle and not a voice fundamental ..) are excluded here. In that case, I'd have to say the voice gets to 10Hz, too - looking for backup on that, apparently you can get to sub-1Hz. beee-essss

[Guest]

Re: Flat Responce

I wonder how a system would sound in a tiled bathroom?

Reductio ad absurdum this early?
C'mon Pete, let's let it play out a little longer before pulling that card eh?

So how did your system sound in the bathroom?
Did you listen to it in the refrigerator also? ;)

p.s. Remember, this is another "A" flat response thread

[MSaturn]

Re: Flat Responce

Well, AJ has a point - reducing, as much as possible, interactions with frequency modes in a theoretical room is a desirable. I also agree that a system is best-designed when it could work in more than one space without a bunch of hassle.

Dipoles, of course, help greatly in this regard

[dantheman]

Re: Flat Responce

MSaturn, check this out: http://hyperphysics.phy-astr.gsu.edu...sic/vowel.html
and here: http://hyperphysics.phy-astr.gsu.edu.../vowel.html#c3

Not necessarily fundamentals. Oh man, a lot of posting since I looked for these links.

Dan

[Zilch]

Re: Flat Responce

I wonder how a system would sound in a tiled bathroom?

Like Villchur and Allison wanted.... ;)

http://www.classicspeakerpages.net/library/acoustic_research/special_sections/additional_ar_documents/the_sound_field_in_home_lis/

[dano]

Go

Thanks everyone. This is better than I hoped.Don't stop.
________
SEXUAL HEALTH FORUMS