Some of the contents of the pages on this site are Copyright © 2016 NJH Music
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Wy bones sound late (stuff to read)
- To: brass-band@xxxxxxxxxxxxxxxxxx
- Subject: Wy bones sound late (stuff to read)
- From: "Chr.J. de Ruiter" <deruiter@xxxxxxxxxxxxxxx>
- Date: Mon, 17 Jul 1995 12:57:09 MET-1METDST
- Comments: Authenticated sender is <ruiter@hp750.fgg.eur.nl>
- Organization: Erasmus University Rotterdam
- Priority: normal
- Reply-to: deruiter@xxxxxxxxxxxxxxx
- Return-receipt-to: deruiter@epib.fgg.EUR.NL id <01HSZ79B1HCG00IMXY@HEARNVAX.nic.SURFnet.nl>; Mon, 17 Jul 1995 12:55:50 +0200 (MET-DST) Mon, 17 Jul 95 12:55:56 +0200 (1.38.193.4/16.2) id AA01165; Mon, 17 Jul 1995 12:59:53 +0200
This article is
fairly lengthy and has been accepted for publication in the ITA
Journal
(International Trombone Association). Consider its appearance on this
list as an opportunity for "peer review" and as a chance for those of
us who use cyberspace to have an advance view and a chance to put in
our 2 cents. Lawrence Borden.
April 27, 1993
Lawrence Borden
721 Boscobel Street
Nashville, TN 37206
Fon (615) 255-4191
Fax (615) 259-2753
Internet: Bordenll@xxxxxxxxxxxxxxxxxxxxx
Why Trombones 'Sound' Late
All trombone players have had to hear about being late, we
have done battle with this demon and have blamed a wide
variety of causes. "You're late!" is still heard even as the
players struggle to correct the problem. Of course this
problem is not unique to trombones, it is common to many of
the instruments in a band or orchestra and the problems and
solutions are often similar.
In order to understand these problems and suggest some
solutions it is necessary to examine some of the factors
that go into the production of sound and human perception of
musical tone. Any examination of the fascinating set of
problems relating to lateness covers a wide variety of
fields including physics, acoustics, instrument design,
psychology, conducting technique, tradition, and methods of
practice and pedagogy. Because musicians find themselves
using the explanation "It's the distance!" the first stop
should be the simple examination of the actual delay time
caused by distance.
The speed of sound at sea level, 68 degrees F is 1129
ft./sec. The distance from the trombone section to the
podium in the Nashville Symphony Orchestra was measured and
found to be about 35 feet.(1) Conversations with
colleagues in other orchestras reveals that this distance
typically varies between 25 feet and 40 feet. The two charts
illustrate the effect of distance from the conductor. In
Chart B the length of various notess (at four different
tempi) is calculated in both traditional notation and in
terms of milliseconds. One quarter notes at a tempo of 60
beats per minute is equal to 1000 milliseconds. In Chart A
the delay time from the trombones to the podium is
calculated. This is the calculation for direct transmission
and is the minimum time required for sound to travel from
the trombones to the podium.
There is really nothing that can be done about this time
delay except play with some clairvoyance. Having to
anticipate the rest of the orchestra consistently by even
this small amount would certainly result in many trombonists
looking for other work! This amount of delay is actually not
a very great component of the lag time typical in many
orchestras. At a distance of 35 feet this delay is about 31
milliseconds and at a tempo of 120 beats per minute this is
almost exactly equivalent to a 64th notes.(2) Fortunately
for us there exists the phenomenon of the 'precedence
effect'. This is a neurological effect where the brain is
capable of gathering data about a set of impulsive sounds
being heard and combining this information into a single
perception. The precedence effect allows us to hear similar
types of sonic events with fast onset that occur as much as
35 milliseconds apart as a single event.(3) As a result of
the precedence effect a 31 millisecond delay might not be
heard as inaccurate even though the precedence effect is
less effective in most practical situations where there is a
mixture of types of sounds (for example, string sounds and
brass sounds).(4) The problem is made worse when this
unavoidable delay is magnified by combination with other
factors. What are some other factors that might contribute
to making it so hard to play "on time"?
Imprecise or inconsistent conducting
A beat pattern that is too large, too small,
inconsistent in pattern, upbeats that are not in the same
tempo as the downbeat, beats that occur low enough so it is
impossible to see the rebound point, a broad sweeping beat
without a rebound point, erratic beat size, accellerandi and
ritandi that are sudden or lumpy; these make any
anticipation almost impossible. Without the ability to
depend on the conductor to deliver consistent placement of
the beat the players cannot play by sight and must wait
until after the beat to play.
Excessive distance from the conductor
A deep seating arrangement makes it difficult to react
quickly. It increases the basic time required for the sound
to travel both to and from the back of the orchestra and
increases the difference between what is heard as the beat
and what is seen as the beat. Communication within the
orchestra becomes more difficult as the orchestra occupies a
larger and larger area.
Excessive distance from the back wall
At low and medium tessituras the trombone has an
extremely high component of its sound which is radiated
toward the back wall from the bell. Up to about 400 hertz
(G4) the sound is essentially omnidirectional.(5) When the
back wall of the shell is far away a great deal of distance
is added to the total sound path and causes the listener to
hear reflected components of the sound arrive much later. If
the back wall is 10 feet behind the low brass then by using
Chart A we can see that the minimum delay is still only 31
milliseconds for the direct first arrival of sound, but the
onset of the first reflected sound arrives at the front of
the orchestra about 18 milliseconds after the first direct
sound. Since the precedence effect will accumulate
information for a small fraction of a second for integration
in the brain, an echo will probably not be heard in this
case, even though the effect is for the onset (or attack) to
be 'spread' over a time of 18 milliseconds. If the distance
between the brass and the back wall is great enough the
listener might hear this added time component as part of the
unique sound of the concert hall or perhaps lateness on the
part of the brass. This is a case in which the precedence
effect would fail to cover the differences between the first
arrival of sound and subsequent echoes. Distance added to
the path of reflected sound blurs the arrival of the
complete orchestral sound.
Playing by ear rather than by sight
When not playing with the conductor's gestures (by
sight), but instead by what we hear (by ear) it is necessary
to add the time it takes for the sound of the rest of the
orchestra to reach you to the raw time of transmission to
the front of the orchestra (plus the time necessary for you
to react to it).
Lack of a concert 'home'
It takes time and effort to learn to compensate for the
various factors that cause an audible delay. It seems
reasonable to theorize that it takes much more time if you
have to do it in several different concert venues. Having
the same 'home' for the orchestra's concerts and rehearsals
makes it possible for the players to experiment with timing
adjustments. Especially valuable in making these adjustments
are recordings made by the orchestra that can contribute to
an understanding of the orchestral sound in its primary
acoustical home. However, many recordings are made with the
microphones placed quite close to the orchestra and not out
in the hall where the audience hears it. Although this is
good recording technique there is a difference between what
the audience hears and what a stereo microphone records.
Care must be taken in studying the orchestral sound based
only on recordings. It is quite possible to learn the
appropriate compensations for several different acoustic
spaces and experienced players are able to adapt well known
timing solutions to new situations.
Use of sound shields I
The use of large sound shields (usually clear acrylic
plastic) between the brass, strings and/or woodwinds
increases the quantity of sound reflected from the back wall
because of primary reflections from the shield. The larger
the shield the greater the reflection. This elongation of
the sound path changes the quality of the sound and blurs
the arrival time. Using the previous example of an attack
spread over 18 milliseconds you might now add another time
component (a much weaker one as it will be reflected at
least twice) for sound leaving the bell of a trombone
bouncing off a plastic shield ten feet in front of the brass
and then off the back wall of the shell ten feet behind the
brass. This adds 40 feet of distance to the sound path.
Chart A shows us that at a total distance of 75 feet the
delay time is about 66 milliseconds. Now a direct component
arrives in 31 milliseconds as before, the echo delay would
be 35 milliseconds. Since human hearing can still only
accumulate and integrate the first approximately 35
milliseconds of difference the remainder will be heard as
aggregate lateness and/or the sound of the acoustic space.
This secondary reflected component of sound is very possibly
obscured by the length of the played notes itself, its loss
of energy by two reflections, and the inverse square
law.(6) It is nevertheless there and has a role in
producing a sense of 'lateness'. An added problem is that
although the echo delay for the low brass would be 35
milliseconds in this case it is actually 66 milliseconds
later than the arrival of sound from the first desk of
violins! 66 milliseconds is a very long time in music. Chart
C shows us that 66 milliseconds is a little longer than a
32nd notes at a metronome marking of 120 beats per minute.
It is very important to understand that the ability of
the human ear to pick out a single sound or timbre from as
complex a system of sounds as an orchestra is dependent on
the first arrival of the sound direct from the source.
Perception of this 'first sound' is the basis on which we
are able to correctly interpret the series of echoes and
reflections that follow.(7)
These shields also give a false impression of balance
within the orchestra and can produce destructive
interference phenomena. Large flat plastic surfaces also
reflect the sounds of other players on stage in a manner
that is very unnatural.(8) The shields used in the
Nashville Symphony Orchestra are particularly large, 4 feet
wide and 4 and 5 feet high. When placed in front of the
brass and angled so that the shields are not perpendicular
to the axis of the brass instruments, nearby string players
are often exposed to not only the direct sound of the brass,
but also additional reflected sound!
Use of sound shields II
Shields in front of the brass section prevent some of
the sound of the orchestra from reaching the brass. What can
be heard is a mixture of direct and reflected sound. Simply
stated it means that the time required for the sound of the
rest of the orchestra to reach the brass is increased. This
reduction in aural intimacy strongly implies reduced
communication on stage. Although sound shields may be clear
barriers that attempt to protect the hearing of the
orchestra members seated in front of the brass they also
make it more difficult than it already is to hear the subtle
nuances in the strings (and to pick up on that already
optimistic cue in the 2nd trombone part attributed to the
3rd stand outside 1st violin in a tutti orchestra passage!).
The sense of isolation produced by plastic shields is a door
that swings both ways and simply does not help the orchestra
play together.
Do sound shields actually protect the hearing of
players already sitting at least 10 feet in front of the
brass from damage over the long term? It is probably too
soon to give a definitive answer to that question in the
context of a professional symphony orchestra, but it is
possible to speculate that the easy and uncontrolled
availability of such shields guarantees their use in spite
of any problems they might cause. We need to remember that
only small shields are needed if they are properly placed
and that shields are not very effective if placed more than
a few inches from the ears of the person(s) being protected.
After all it is ears we wish to protect from excessive sound
pressure, not feet!
Improper practice with the metronome
Since it is generally desirable to spend some fraction
of practice time working with a metronome it is common to
hear brass players tend to play just behind the 'tick' of
the metronome. Playing right with the sound makes it almost
impossible for brass players to hear the 'tick' unless it is
amplified. We have learned to play in time, but behind the
time by a discrete interval.(9) The parameters of this
discrete delay and its effects are under investigation.
Fear of 'biffing in'.
This one is called 'keeping your job' and it is to be
expected that we want correct entrances with our colleagues
in the brass section. Experience and confidence, what
teachers call conviction, help minimize this factor, but all
players want the reassuring feel of being 'in' the sound
rather than risk being ahead of it. If players are 'in' the
sound, time is being added to the interval it takes for
sound to arrive at the front of the orchestra. It helps
greatly to be 'on top of the beat' and therefore at the
front of the orchestral sound around you. If the entire
brass section is playing together it is, unfortunately, not
a guarantee that they are anywhere near the 'top of the
beat' so it is clear that a brass section must both be
together and 'on top of the beat' in order to on time.
Shape of the sound envelope
Poor sound production technique, such as a delayed
start to a notes, a weak attack, or one that is essentially
legato in nature, is perceived by a listener as blooming;
full volume achieved well after intended beginning of the
attack. Efficient and immediate physical function when the
brain demands that a notes be played and a good front to the
attack are necessary to counteract this. Attacks of
trombones are generally less immediate and energetic than
the attacks of trumpets. Instrument design criteria, such as
the formulation, thickness and tempering of the brass used
in the bell as well as the treatment of the bell edge, also
affect the response curve. (A substantial change in response
characteristics occurs depending on whether the bell wire is
pressed in place or soldered in place!) In the same vein,
different mouthpiece and leadpipe (mouthpipe) designs and
combinations also can cause substantial variance in the
shape of the attack envelope.
Trombones, tubas and horns are also often heard as
behind the trumpets for psychoacoustical reasons, even if
tones are actually started at the same instant. The time
required for the buildup of a complex tone is only one of
several highly technical physical and psychoacoustic
phenomena that enter into the problem of lateness.
Precursors
Precursors are those things that we do in order to feel
ready to play, but have little or nothing to do with
actually playing. Several examples might include conducting
oneself for a downbeat, moving the slide back and forth
several times just before playing a notes, or licking the
lips 2 or 3 times before we play. If you are playing alone
who are you conducting? If your slide does not work then it
is too late to fix it the instant before you play. One lick
might be necessary to moisten the lips but the rest have no
practical purpose. These are simply unnecessary habits that
take time and make quick reaction feel very uncomfortable
and unnatural if omitted. When precursors are ruthlessly
eliminated so that the player can demand and get a notes
started with no delay, a major factor contributing to
'lateness' is eliminated. Elimination of precursors allows
for extremely quick responses to the wide range of stimuli
we must react to when performing, including our spontaneous
musical inspirations.
Late entries after breathing
It seems a fairly normal thing to hear brass players
lose time when they take a breath. This is a lack of
awareness by the individual player that breathing takes a
discrete amount of time combined with the ease with which we
are able stretch our sense of internal time. It is necessary
to minimize the time spent taking in air and it is not
always possible to add extra time to the flow of the music.
Sometimes we simply grow careless and other times we have
demanding music that allows us very little chance to
breathe. In any case, it is important to manage the
inhalation so that the following notess are not late.
Aural discrimination I
There are at least two factors here that contribute to
"lateness". The first is that below a certain minimum
threshold the human hear hears just one sound event, even if
there are actually more than one. This minimum threshold of
time is about 1-2 milliseconds when click stimuli are used
in laboratory conditions.(10) With longer stimuli (such as
notess) the echo threshold can be 20 milliseconds or
longer.(11) A 30 millisecond figure is given as lower
limit for the aural discrimination of cardiologists
listening to heart sounds. Even though some of the heart
sounds are very sharp (mitral valve snap),(12) it is a
less than perfect listening environment. For practical
purposes a value of 25 milliseconds has been chosen as a
practical lower limit to echo threshold discrimination,
especially if we are going to ignore frequency and volume as
factors that will alter this result. This limit is
represented in Chart C by the grey shaded area. The second
factor is that at lower frequencies the ear is less
sensitive than at higher frequencies.(13) In the first
case we actually get a bit of help since tiny errors in
"lateness" might fall below this discrimination threshold
and we have the precedence effect helping as well. In the
second case the low brass have a problem tied to the
physiology and psychophysics of hearing,(14) especially
when playing in the lower register.
Aural discrimination II
When a brass instrument begins a notes, time is required
to reach the peak volume of the notes (this is called
'onset'). Stimulation of an auditory nerve fibre as a
physiological response to the presence of a sound takes a
minimum of time and energy.(15) A vibration must reach a
minimum set of values in pitch, sound pressure and length
before the neurons are fired and there can be any neural
stimulus that is perceived as sound.(16) This component is
very small, but it is there and it is affected by factors
such as pitch and spectral color.
Playing into the stand
Even though the sound of the trombone is essentially
omnidirectional through most of its range, placing a music
stand between the bell and the front of the orchestra causes
a reduction in the amount of the initial wave front that
arrives at the front of the orchestra. Since the stand is so
close to the sound source it is possible that a significant
sound shadow can be produced, especially in the high
register. Much of the sound is bounced off of the floor and
other nearby surfaces and arrives fairly directly, but there
is a dual additional effect. First, the component of the
sound that is reflected from the stand adds to the
elongation of the sound path and second, there is the
possibility that the sound spectra might be affected by the
repeated absorption and reflection. It would be easy to say
that the biggest detriment to playing into a stand is heard
when the members of a trombone section cannot agree on
whether they should or should not all play into the stand
for the sake of consistency. The truth is that it is
slightly less important whether the bass trombone plays into
the stand. This is because the bass trombone sound is almost
always omnidirectional in nature due to its tessitura and
because the wavelengths of pitches in the general bass
trombone register are much greater than the physical
dimensions of the stand.(17) As a result less 'shadow
effect' is present.(18) The greater effects of a bass
trombone playing into a stand might be those that alter the
spectral color of the bass trombone sound and the visual
impact on the audience/conductor. It should be notesd that
playing into the stand would theoretically be an even
greater problem for trumpets except that they tend to have
their stands lower since they need not make room beneath the
stand for the slide. They tend to play over, or around, the
stand and avoid this problem.
There is also another problem associated with playing
into the music stand. When the sound of the instrument is
instantly reflected to the player it can be argued that the
player is reacting to strong sonic cues not related to the
acoustic space when attempting to adjust for 'lateness' and
for balance within the orchestra. Adding this factor to the
false sense of volume and spectral color that is produced by
such a close reflective surface it seems obvious that not
playing into the stand at all, or at the very least moving
it as far away as is practical, is preferred for all members
of the trombone section.
A minimum time is necessary to generate a standing wave in
the instrument
The mass of measurements and theory on this subject is
far too complex for the purpose of this article so I will
just say that the mass of air, pliability of the lip, method
and energy with which air is supplied to the lip reed, the
overall length of the instrument, and the efficiency of the
instrument design are all factors that affect the time it
takes for a notes to be generated. No matter how hard we try,
it does take time to start a notes, but once again this
effect does not have to be exaggerated by the poor, slow, or
stuttered beginning of a notes due to inattention or poor
function.
We hear it late in our minds and accept it
Most important is whether or not we clearly imagine
what we want to sound like. This goes for brass sections
just as well as for any individual player. We have grown to
accept lateness as a norm, but the amount of lateness that
is heard as acceptable is much greater than it needs to be.
As the charts show the first sound can be heard no more than
31 milliseconds late if we play by sight and without other
compensating anticipation. That is far less delay than that
found in many orchestras and is at least partly compensated
for by the precedence effect. Although there are many other
physical adjustments that can be made to the whole process
of playing it is the willingness to allow ourselves to hear
the truth of our performance and an honest willingness to
change that will make the most difference.
If you have an 'on time' sound in your imagination you will
unconsciously make the adjustments that are necessary to
make the imagined sound real. This principle is true when
applied to the problem of lateness and true when applied to
the work of producing a singing tone or fine accurate
technique. It is endless dedication to musical idea(l)s
which makes our art satisfying and engrossing. It is the
music which we must ultimately serve.
It is fortunate that many of these problems are common to
all the players in the orchestra. For example; the problem
posed by simple distance is blurred by the reflections of
sound in the shell and the fact that instruments closer to
the podium than the trombones are also late, but by a
smaller amount. Sound arrives as a part of an envelope
beginning with the violin and viola desks under the
conductor's nose and ending with the tuba, percussion and
finally the reflected sound in the concert hall itself. The
sound of instruments in the middle of the orchestra arrive
between these extremes and this helps blur perception of
lateness.
The delay problem can actually be worse for some players of
instruments other than low brass within the orchestra.
Although the horns have a longer sound path than any of the
other brass because of the backward pointing bells and the
distance between the basses and the percussion section can
be as much as 80 feet in some orchestras! Add the other
factors and it is amazing that we sound like we are together
at all. In fact, it is clear that we already effectively
employ a variety of both conscious and unconscious remedies
for lateness.
One interesting attempt to resolve this problem has been to
have the entire orchestra play well behind the beat of the
conductor (I am personally reminded of Guy Frasier Harrison
in Oklahoma City and Eugene Ormandy in Philadelphia). In
this way it is possible to gauge the interval of response
(after much practice) and for the front and rear of the
orchestra to respond at different times as necessary for all
the sound to arrive at once. There are many disadvantages to
this method and it is quite an understatement to say that
this solution is not recommended.
Experience shows us that we can anticipate by an appropriate
amount if the tempo is steady and we make a concerted effort
to stay on the 'front of the beat'. Of course this is much
more difficult to do when the tempo is changing. Clear,
pointed attacks help a great deal as does a near rear wall,
excellent conducting technique, minimal distance from front
to back in the orchestra, absence of sound shields, playing
exclusively by sight, playing to the side of the stand, and
nearly instantaneous physical response. It takes a great
deal of effort to move a brass section toward this way of
playing and it requires consistent effort both on the part
of the players and the conductor, but the rewards are
certainly worth it.
At least we can understand most of the main components of
this problem. It would be nice to be able to plead that
'it's the distance' and be done with it. Unfortunately, by
itself, the calculated delay is not nearly as much as we
might have hoped! We must work to compensate for the
physical, acoustic and psychoacoustic elements of lateness
so that the sound of the orchestra as a whole can have
maximum clarity, beauty, and impact.
endnotess:
1 With 12 ft. of sonic no-man's-land in front of
the low brass before seating for violas and celli begins.
2 Chart B
3 Arthur H. Benade, Fundamentals of Musical Acoustics
New York: Oxford University Press, 1976), 204.
4 It is fascinating to speculate as to why the brain
retains the ability to collect data on sounds over periods
ranging from 1 to 70 milliseconds depending on a variety
of conditions associated with the sonic events.
5 Jurgen Meyer, Acoustics and the Performance of Music
(Frankfort am Main: Verlag Das Musikinstrument, 1978), 89.
6 The inverse square law states that as the distance
increases the energy decreases at an accellerated rate.
For example, for a given unit of energy on 1/4th of the
energy is present at 2 units of distance and only 1/9th
of the energy is present at 3 units of distance and
1/16th at 4 units of distance. { E=1/(D^2) }. The design
of acoustic spaces is an effort in large part to diminsh
the effect of the inverse square law at the distance of
the audience by controlling the reflective and absorbtive
qualities of the acoustic space.
7 Juan G. Roederer, Introduction to the Physics and
Psychophysics of Music 2nd ed. (New York: Springer-Verlag,
1975), 146.
8 (The angle of incidence equals the angle of reflection.)
The sound reflected by large flat surfaces of sound shields
produces the most obvious and disruptive effects.
9 It is an interesting to notes that the use of 'click track'
in the recording studio is nearly an absolute necessity for
accurate rhythm in a recording. The use of a 'click track'
is an art in itself, and is only effective if it is
provided to the players at a rather high volume since it
must compete with their own direct sound (provided them
via headphones) and their own sound via bone conduction
(present because they are wearing the headphones).
10 Barry Leshowitz, Measurement of the Two-Click Threshold,
(The Journal of the Acoustical Society of America: Vol. 49,
No. 2 (part 2), February 1971), 462.
11 Jens Blauert, Spatial Hearing: The Psychophysics of Human
Sound Localization, (Cambridge, Massachusetts: The MIT Press,
983), 230-231.
12 Charles K. Friedberg, Diseases of the Heart, Vol. I, 3rd
ed. (Philadelphia and London: W. B. Saunders Company, 1966), 77.
13 Murray Campbell and Clive Greated, The Musician's Guide
to Acoustics (New York: Schirmer Books, 1987), 112.
14 A wide array of interesting topics is involved. The masking
effect of complex tones pitched at intervals above the general
tessitura of the trombone section greatly affects the volume
at which we must play some passages in order to be in balance
with the rest of the orchestra.
15 It is interesting to know just how sensitive the ear is.
Under laboratory conditions Leshowitz demonstrated in 1971
that we can discriminate clicks which are separated by as little
as 10 microseconds. In 1976 Green demonstrated that we can hear
sounds which displace the eardrum by as little as 10-11 meters,
one hundredth of the diameter of a hydrogen atom!
16 Juan G. Roederer, Introduction to the Physics and
Psychophysics of Music 2nd ed. (New York: Springer-Verlag,
1975), 48-53.
Name : Chris de Ruiter
System Manager
Department : Epidemiology & Biostatistics
Erasmus University Medical School
Internet adres : deRuiter@xxxxxxxxxxxxxxx
Phone : +31 10 4088252
Fax : +31 10 4365933
Adres : Erasmus Universiteit
Dr. Molenwaterplein 50
3015 GE Rotterdam
Netherlands
P.O. Box : Postbus 1738
3000 DR Rotterdam
Netherlands- Prev by Date: Re: Combinations
- Next by Date: cornet question
- Previous by thread: Repiano Pronunciation
- Next by thread: Re: your mail
- Index(es):