Re: Is there an 'adaptive' EQ plugin available, or what do you call them?

kludge@xxxxxxxxx (Scott Dorsey) writes:

Randy Yates <yates@xxxxxxxx> wrote:
kludge@xxxxxxxxx (Scott Dorsey) writes:

DeeAa <aepheikki@xxxxxxxxx> wrote:

What do you call those EQ units that can be used to correct or 'level'
a speaker's frequency reproduction?

Fraudulent.

Are you claiming it's not possible to equalize a speaker's on-axis
response? That is, for the most part, false.

No, but it's not _useful_ to equalize a speaker's on-axis response, because
the off-axis response contributes greatly to what the listener hears. On
top of this, most of the problems that you hear on-axis are things like
horn resonances... and you can equalize the response so that a sweep test
looks good, but the impulse response will still be screwy because it's not
the right solution for the problem.

I mean the ones where you stick a measurement mic in and play a test
sound array and it makes corrections based on that. Room correction?

Room correction and speaker correction are different things. Both are
more or less misguided in different ways.

I can see why you might make such a statement regarding room correction
since this type of processing is listener position dependent, but why
would you make it for speaker correction?

Because most speaker problems aren't on-axis issues... most of the frequency
response problems with speakers are due to narrowband cabinet, driver, or
horn resonances. We've come to the point finally after a century of research
where we _can_ actually build something flat on-axis without too much work.
The problem is that this is only a tiny part of the problem of making something
sound good.

Also a _lot_ of the response issues you hear in speakers today are due to
crossover issues.... since the drivers aren't in the same location in space,
the crossover has to be configured so that the response is flat at only one
angle. The closer you can get the drivers together, the easier that problem
is to solve. Equalization _can_ solve it for one angle (really, equalization
is what crossovers are all about), but you don't always listen at that one
angle.

And let's not forget extremely uneven decay times across the audio spectrum in most
typical rooms, particularly below, say, 300 Hz. EQ really won't address this in any
meaningful way, as Scott says. It's tough (impossible?) to undo the timing kinks of
a room that might have an RT30 of 200 mSec in one place and 3.5 Sec bubble a 1/4
octave up.

One of the things in our room that took some sweat (and hammering, as Mike says),
was getting very even and short decay (< 150 mS) below 1K, and then a smooth and
intentional rise to around 1.2 second > 10K.

I continue to smile at the jaw-drops I get from several different engineer friends
when they hear their stuff in this room, including a mastering engineer who shook
his head in pleasant disbelief.

YMMV.

Frank Stearns
Mobile Audio

--
.
.

Relevant Pages