[harley52]

Danny and others,

 I was looking into the Fletcher-Munson curves earlier and finally realized why I'm so sensitive to the 1-4k hz. range. According to the curves, we humans are much more sensitive in that range, yet, generally speaking most A-philes and speaker builders call for a flat frequency from their speakers. I'm I crazy in my interpretation of the F-M curves or does everyone else have it wrong? Just kinda generally speaking of course.

Thanks for any insight anyone can provide.

[FullRangeMan]

After 2 years of marriage wives speak only in this range.

[paul79]

Well, a flat frequency response IS what we want. This will create the sounds as they would be in real life. No?

[Danny Richie]

Think of it this way.

Consider any musical instrument and how it sounds. Would we alter it around the sensitivity range of our hearing? No, not really. It is what it is.

So in reproducing it true to original then accuracy from the speaker is essential.

Such things might be more essential to the guys mixing and mastering our music to reach a balance that we enjoy more.

[charmerci]

The audio equipment should re-produce exactly the sound that comes from the instruments. Then we will hear/interpret it as the M-F curve shows.

[FireGuy]

Well, a flat frequency response IS what we want. This will create the sounds as they would be in real life. No?

As I understand it, yes.   And that neutrality should be one of the highest design characteristics for speaker manufacturers. 

[THROWBACK]

I believe the topic is more complicated than what has been discussed above. Yes we want flat response, but that means actual flat response, not necessarily measured flat response. From what I have read and heard, tweaking your system to give measured flat response using a standard sound-level meter will result in the system sounding too bright to your ears. AFAIK mics and ears do not respond exactly the same to sound. Many studies have shown that a measured fall-off in measured sound above one kHz (one dB per octave or so) not only sounds better but is actually more accurate to the source.

[Danny Richie]

I believe the topic is more complicated than what has been discussed above. Yes we want flat response, but that means actual flat response, not necessarily measured flat response. From what I have read and heard, tweaking your system to give measured flat response using a standard sound-level meter will result in the system sounding too bright to your ears. AFAIK mics and ears do not respond exactly the same to sound. Many studies have shown that a measured fall-off in measured sound above one kHz (one dB per octave or so) not only sounds better but is actually more accurate to the source.

It is really hard to say that a dipped response is more accurate, but it can be more pleasing. The real key is not to look just at the on axis response of a speaker, but look at the horizontal and vertical off axis as well.

An overly bright or lively room is much more of a problem than a neutral speaker.

[stereocilia]

It depends on what you mean by "sensitive.". Hearing sensitivity traditionally refers to how loud a sounds have to be in order to detect them.  By the way there is no such thing as better than normal hearing.  (Any hearing sensitivity better than some agreed-upon standard for normal is still called normal since there is no such thing as super-human hearing.)

When sounds reach the point where they are uncomfortable at lower volumes than normal then we call that hyperacusis.  It's not too uncommon to have poor hearing sensitivity and hyperacusis at the same time.

[coke]

What about people who have high frequency hearing loss?  Would it be better to tune things so music sounds like used to, or is it better to tune it flat so music sounds as it does to them now? 

[BobRex]

What about people who have high frequency hearing loss?  Would it be better to tune things so music sounds like used to, or is it better to tune it flat so music sounds as it does to them now? 

The mind is pretty good at filling is some of the missing "details".  Compensating is tricky - do you adjust according to a hearing test curve (which may or, more likely, may not follow any type of possible EQ curve) or do you "season to taste".  In other words, how do you get back to the ideal of "flat".

[nickd]

I have had this discussion many times with audio designers / engineers over the years. Most engineers I have met lean toward the "speakers should measure flat". I on the other hand, am of the opinion that flat is a fine goal, but not if they don't sound good camp.
The speakers I have have seemed to love the most have almost all had a bit of a "Fletcher Munsion" suckout in the midrange.

It takes a computer to design a good speaker I completely understand. But, the ear should ultimately be used to confirm desired results.

[harley52]

 Ok, so if a flat response is the ideal  then could it be said that the BBC got it wrong and still does get it wrong to this day? And just to cause trouble, why is it that not a review by the paid people, for what they're worth, has ever said the Harbeth's, Spendor's, Chartwell's and on and on has not found them to be musical.

 When we go to a live event we don't dissect the music as A-philes often do and as I used to do in years past . The thing is music is to be enjoyed and if the speaker makes a high hat sound like an air hose then that would be very distracting and I wouldn't care for that myself. But if the sounds of an instrument or what we think the instrument sounds like, is close enough then why get your shorts in a wad over miniscule things. I know musicians know how  some instruments should sound but, the largest percent of listeners are NOT musicians no matter want they tell themselves.

 As soon as I figure out how to cause more angst and turmoil I will post it here!

[ricardojoa]

Not an expert, but you can have two speakers with the same drivers and measure flat on axis but differ at off axis, by having different crossover point. A mismatch directivity can cause a dip at xover point which may give that pleasing sound. 

[charmerci]

Ok, so if a flat response is the ideal  then could it be said that the BBC got it wrong and still does get it wrong to this day? And just to cause trouble, why is it that not a review by the paid people, for what they're worth, has ever said the Harbeth's, Spendor's, Chartwell's and on and on has not found them to be musical.

At this point, no one has made the perfect speaker so designing some that are not perfect in ways that people like is an admirable goal.

[mlundy57]

I don't know that I would want a perfect speaker that was absolutely 100% lifelike. After all if the music coming out of my speakers sounded exactly like the symphony what would be the point of going to the symphony?

[stereocilia]

In almost every case it's better to think of high frequency hearing loss as a shaped noise floor rather than an eq curve.

[Guy 13]

In almost every case it's better to think of high frequency hearing loss as a shaped noise floor rather than an eq curve.

Hi stereocilia.
Interesting, however, can you re-phrase that for me and other members with same or similar low level understanding. I am sure I am not alone.
Sorry, look like my English vocabulary and knowledge is a little limited.I am 66 years old, therefore, my hearing is slowly going on the fritz...

Guy 13

[JLM]

The recording/mastering process requires absolute flat response.  The challenge as expressed by the F-M research is that humans are less sensitive to lower and higher frequencies at low sound pressure levels.  That's what loudness circuits were designed for, but now virtually all audiophiles are "too pure" for that.

I'd question the accuracy of the headphones used in audiology testing (at least compared to what is used in studio or audiophile systems).

Proper bass reproduction at home requires the application of multiple sources (read "swarm" here at AC or Floyd E. Toole's "Sound Reproduction").  This really cannot be argued against because low frequencies behave like waves and can result in +/- 20 dB responses that vary by listening location and frequency.

To season for individual taste you can try EQ for the higher frequencies. 

The real goal of audiophiles is to achieve emotional satisfaction, not perfection.

[Guy 13]

The recording/mastering process requires absolute flat response.  The challenge as expressed by the F-M research is that humans are less sensitive to lower and higher frequencies at low sound pressure levels.  That's what loudness circuits were designed for, but now virtually all audiophiles are "too pure" for that.

I'd question the accuracy of the headphones used in audiology testing (at least compared to what is used in studio or audiophile systems).

Proper bass reproduction at home requires the application of multiple sources (read "swarm" here at AC or Floyd E. Toole's "Sound Reproduction").  This really cannot be argued against because low frequencies behave like waves and can result in +/- 20 dB responses that vary by listening location and frequency.

To season for individual taste you can try EQ for the higher frequencies. 

The real goal of audiophiles is to achieve emotional satisfaction, not perfection.

Hi JLM.
The last sentence:
The real goal of audiophiles is to achieve emotional satisfaction, not perfection.
Well, that's what I have been looking for for years and now that's what I have obtained with my combo Decware-Omega 7F at a price I can afford.
The purist will say that the sound of my system is colored and not flat,
but I don't care, because I like what I hear and that's what's important.
Thanks for mentioning that important phrase.

Guy 13