[Dixielandjazz] Recording Jazz (long -- delete if don't like tech talk)

[Edgerton, Paul A]

Bob Ringwald wrote:
"If it is a 9 foot grand, even the drummer ends up 15 feet away."

In his reply, Jim Kashishian said:
"I have tried to point out that there can easily more than a  second of
delay in the live sound from the other guys in the band."

Ahem...

Jim, I'm sure you realize that the speed of sound is roughly 1100f/s. To
have a one-second delay the other instrument would have to 1100 feet away.
If the drums are only 15 feet away, that's a delay of about 7 milliseconds. 

This is an example of the Haas Effect, which says that for delays less than
roughly 20 ms (the exact threshold varies among individuals) the delayed
sound cannot be separated from the original. The ear hears it as one sound.
Delays of more than 20 ms become two distinct sounds.

This is easy demonstrate: stand facing a hard surface (like a brick wall)
and clap your hands. When you're 15 to 20 feet away from the wall you can
hear a distinct echo. Now keep clapping and slowly walk toward the wall.
When you get about from the wall, the echo will seem to disappear.

Your hearing will merge sounds that fall within about 20ms into one sound.
Sounds more than 20ms apart are perceived as being distinct sounds. That's
why so-called early reflections sound good. (Think of a saxophone player
warming up by playing into a wall.)

You can do the same thing using a digital delay, except that if the delayed
signal is mixed with the original, you'll get comb filtering. But if you pan
the original signal to the left and the delayed signal to the right and mix
them at the same volume, then increase the delay time from the shortest
available setting, first the apparent source will move from the center of
the two speakers toward the right, or delayed side, getting wider and wider
until at 20ms the two signal seem to come unglued.

So the problem of hearing the rest of the band is not caused by delay. It's
a different problem: masking. That's the tendency of the ear to be
insensitive to softer sounds falling in the same frequency range as louder
(or earlier) sounds.

In Ringwald's example, he could probably hear the drums just fine -- as long
as he isn't playing. Then the sound from his piano would mask much of the
sound coming from the distant and therefore softer drums. The cure for this
is, as you both have said, to move closer together or get the distant sound
source in your monitors.

Back to Jim:
"Without a full monitor mix, you can wander around the stage and you very
well may have to decide who's rhythm you wish to stay with."

The other solution is to stand halfway between the two. This won't help the
two widely-spaced guys, but at least you'll be getting a relatively normal
balance between them with equal delays.

I have expounded on the subject of PA systems before, but to summarize,
there are three workable approaches: 

1. Set everybody up close together and forget the PA. In small rooms this
works very well, and I'll bet that when most of you rehearse, you don't use
a PA system, other than for vocals.

2. In a larger room, you have the problem that some of the instruments
project much better than others and may need a little help. So you might mic
the piano and the clarinet and have the bass player use an amp. The brass
and drums can usually hold their own in a setting like this. You're still
going for a nearly acoustic sound.

3. In really big rooms, you're going to need to mic everything. This
includes the drums -- though not necessarily the bass since he's probably
using an amp. You'll also need monitors so you can tell how close to get on
the mics. The monitors don't need to be loud and don't even need to be
full-range speakers, but they have a reasonable representation of the total
sound so that each player can hear his place in the mix. Since the bass goes
all over the place anyway, it's usually wise to leave it out of the monitor
mix. 

And for God's sake, once the monitors are set, don't change their mix during
a performance! Even a bad mix is more workable than a constantly changing
balance.

The idea here is to have the lowest possible volume on the stage. The closer
you can get to the sound of situation 1 above, the better the results will
be.

Also note that I'm only talking about sound from the musician's perspective.
The Balance needed for the house is a whole 'nother thing with its own
special problems. The real trick is finding a setup that meets both sets of
needs.