"Musical" response curve

[TKoP]

I can't remember when, and can't find the thread, but there was mentioned a speaker/room response curve that was "better" than a flat speaker/room response curve.

Not sure why this one percolated up to the gray matter, but it happens.

The questions related to this:
1) What does that response curve look like?
2) Why is this better than a "flat" response curve?

Intuitively, at least to me, is that a flat response would mirror the engineer's design for the recording.

 

[Zing]

20Hz-22kHz +/-0dB is flat. I don't know how that can be bested, or even be possible. But I believe the accepted tolerance for "flat" is +/- 1.5dB, meaning that there is never more than a 3dB swing between the lowest SPL and the highest SPL of given frequencies (with 3dB as the threshold of being able to notice the increase/decrease in SPL).

Anything defined or perceived as "better than flat" would have to be subjective. What you find better, I may not and vice versa.

I think it's important to get the flattest response possible as a basis. It won't matter if you like it, you just need it for a foundation. From there you can tweak the response to taste, thus making it "better" for you. Therefore, I don't see how there can be just one, predefined curve.

 

[Botch]

A lot of people like their bass boosted, and their treble boosted; many graphic equalizers hence exhibit a "smiley face" with their sliders; that may be what you're remembering.
No, it's not accurate, but a lot of people prefer it.

 

[Flint]

Alter and I both waxed poetic about the natural and musical sounding curve was one which was tilted down from bass to treble at a gentle slope.

I prefer about 0.5dB per octave from 100 Hz to 15kHz. The straighter the line of the slope between bass and treble, the better. The SAL at 15kHz should be about 4dB lower than at 100Hz.

 

[DIYer]

Alter and I both

You mean Altec formerly known as soundhound?

I took up on that advice from you guys and set up my stereo that way. It looks like real global temperature chart from 1998 to 2013. I like the sound of it.

 

[Flint]

Yes, Altec or Soundhound,

The most musical frequency response, in room, is one which slopes down as the frequency rises. We debate the rate of slope, but the bass should generally be louder than the treble.

This is what speakers with a boost in treble, which may sound more "detailed" or "exciting", cause listening fatigue and less natural or musical.

 

[TKoP]

Yep, it was the Altex/Flint that I remembered.

So, to confirm, the reasoning behind the slope down from bass to treble is to minimize listeners fatigue by damping down the treble a little?

 

[DIYer]

^ That's what experienced listeners agree to be "natural" sounding frequency response.

 

[Flint]

It is also the standard target THX and DTS response measured at the listening position.

A flat response --- linear from bass to treble --- almost universally sounds brittle, harsh, and overly "analytic" with people who record music for a living because it fails to take into account the realities of the room acoustics, our physical auditory properties, and the true nature of sound as it passes through space before it reaches our ears.

So, most hard core naturalists, like me, have come to adopt the downward sloping response as most ideal.

 

[Flint]

Here's a chart showing my actual, real life, home speakers' response curve for the left and right channels. Notice the downward slope of the curve:

 

[Deleted member 133]

In the belief that there might be much more to this subject, I have just ordered a copy of http://www.amazon.com/Sound-Reproduction-Acoustics-Psychoacoustics-Loudspeakers/dp/0240520092 which should arrive this week.

Here is an interesting summary / review of the book. http://www.stereophile.com/referenc...production_loudspeakers_and_roomsi/index.html

I met Floyd Toole as a result of my having participated in some of his listening tests at NRC in 1989-90 (if I recall the dates correctly.)

Jeff

 

[Flint]

Hey! I did some work with Dr. Toole at NRC around that time. I was a nerd kid working with Belden on Cable testing trying to determine if there was any validity to the audiophile cable claims. NRC offered a great place to run some experiments.

 

[Deleted member 133]

Small world!

 

[DIYer]

For some reason, I can barely remember the last time IG/COF/Flint posting frequency response graph. It used to be a daily routine... :eusa-whistle: