rcd@opus.UUCP (03/24/84)
<+>
I have a manual for a piece of audio equipment which criticizes loudness
compensation circuits in preamps "...because they are based on the obsolete
Fletcher-Munson curves and produce excessive mid-bass boost..."
[Background, for general interest: The "Fletcher-Munson" curves are a
presentation of empirical data which show curves of equal perceived
loudness on a plot of sound pressure level (acoustic power) vs. frequency.
(Say wha?...) For example, using a 1KHz reference they show that a 60 dB
sound at 1KHz sounds as loud as roughly 70 dB at 100 Hz or as 57 dB at
3KHz - thus one constant-loudness curve passes through these three points.
Each curve has a characteristic shape which shows that the ear is most
sensitive around 3KHz and loses sensitivity at higher and lower frequencies
according to a particular pattern. The family of curves shows how the
differential sensitivity to frequencies changes with sound level, becoming
less pronounced at higher levels. End of digression.]
Now the question: If these curves are obsolete, can someone please tell me
WHY they are obsolete - is the data wrong, was it presented wrong, or has
it been consistently misinterpreted?
Here's the reason I'm asking - I'm wondering if people have built "loudness
controls" with contours which mirror a F-M curve for some particular
loudness level. If so, they've obviously misinterpreted the data; the
control contour should be based on the variation in shape of the F-M curves
for different levels. Example: If you have music recorded at a typical
90dB SPL but you want to play it back at a typical 60dB level, the contour
you want is one which will boost bass and treble so that the system
response (system = loudness control + ear) has the shape of the 90 dB curve
at 60 dB - NOT one which will give FLAT system (control+ear) response at 60
dB.
On the other hand, if there is something fundamentally wrong with the F-M
data, I'd like to make some careful notes about it in my copy of Beranek's
"Acoustics" book.
--
{hao,ucbvax,allegra}!nbires!rcddmmartindale@watcgl.UUCP (Dave Martindale) (03/25/84)
Typical "loudness" controls introduce some amount of bass and treble boost in the lower 1/3 or 1/2 of the volume control setting - above that there is no effect. So, even if the compensation is ideal over the range in which it operates, it is compensating to make the sound appear to have the same tonal balance at lower volumes as it does at some specific higher volume where the compensation reduces to zero. This may be at quite a range of sound pressure levels at the listener, depending on sensitivity of the power amp, speakers, and the room. So it can't be more than a very approximate compensation anyway. I've seen one brand of receiver (Yamaha?) which had both volume and loudness pots. You set the loudness controll full clockwise (which was "no effect"), then set the volume control to what you considered a realistic full-volume listening level, and then you turned the loudness control counterclockwise to reduce the perceived loudness to where you wanted it. This reduced the midrange more than high or low frequencies; the intended effect was that the low-volume sound should seem to sound the same as at the full-volume level you picked. It seems to me that this sort of two-control system at least has a chance of working the way that loudness compensation is supposed to work. If so, why is it so rare?
rfg@hound.UUCP (R.GRANTGES) (03/26/84)
It is a very old idea. The first preamp I ever built had it (circa 1952). It is rare because it costs more, using more parts, and mfrs feel, probably rightly, that most folks wouldn't have the foggiest idea how to use it. hound!rfg
ron@brl-vgr.ARPA (Ron Natalie <ron>) (03/26/84)
I used to have an old sylvania receiver that had both a loudness pot (this is what you were supposed to user to make it louder and softer) and an "acoustic level" control that you were supposed to set once for the particular room and leave. -Ron
keller@uicsl.UUCP (03/27/84)
#R:opus:-27300:uicsl:2700008:000:602 uicsl!keller Mar 26 19:58:00 1984 From my Crown IC-150A preamp manual: "The work of two research teams has been widely accepted throughout the audio field. The findings of Fletcher and Munson (1933) has been used as a basis for loudness compensation for many years. More recently, the I.S.O. (International Standards Organization) has accepted the work of a second team, Robinson and Dadson (1956), as being more accurate. As a result, the trend in the audio industry is to use the I.S.O. curves as a basis for loudness compensation." My Crown equipment has worked flawlessly for 8 years, but would anyone choose Crown today? -Shaun
keller@uicsl.UUCP (03/27/84)
#R:opus:-27300:uicsl:2700008:000:602
uicsl!keller Mar 26 19:58:00 1984
>From my Crown IC-150A preamp manual:
"The work of two research teams has been widely accepted throughout the
audio field. The findings of Fletcher and Munson (1933) has been used as a
basis for loudness compensation for many years. More recently, the I.S.O.
(International Standards Organization) has accepted the work of a second
team, Robinson and Dadson (1956), as being more accurate. As a result, the
trend in the audio industry is to use the I.S.O. curves as a basis for
loudness compensation."
My Crown equipment has worked flawlessly for 8 years, but would anyone
choose Crown today?
-Shaunandrew@inmet.UUCP (03/31/84)
#R:opus:-27300:inmet:2600062:000:1082 inmet!andrew Mar 28 11:33:00 1984 The problem with conventional loudness controls is that the volume control setting has little to do with the actual sound level - the obvious variations in speaker efficiency, room size, and program level make it impossible for a single loudness control to provide accurate compensation. Yamaha (and others) have added a separate control to allow fine-tuning of the amount of loudness compensation. It's still comparatively rare, probably because the marketing types find useless bells and whistles easier to sell than truly useful ones. Speaking of loudness - has anyone else noticed that Harmon-Kardon products tend to overdo it? On three H-K products - my old 330c, my parents' 230e, and my wife's 430 - the loudness compensation produces an unlistenably boomy sound even at the lowest listening levels. Andrew W. Rogers, Intermetrics ...harpo!inmet!andrew 733 Concord Ave. ...hplabs!sri-unix!cca!ima!inmet!andrew Cambridge, MA 02138 ...uw-beav!cornell!esquire!inmet!andrew (617) 661-1840 ...yale-comix!ima!inmet!andrew