HOW TO USE THE FLOOMPTAGORPT RISING/FALLING TONE GENERATOR


INTRODUCTION:

This program generates a tone that seems to rise or fall forever.

It's sort of like the tone described in

  http://en.wikipedia.org/wiki/Shepard_tone

except that this application allows you to adjust various parameters.

It's a Java application that can be downloaded and run on your local 
system.  It requires Java (not the same as JavaScript) to run.  Setting 
it up and using it may get a little technical.


NOTES and CAUTIONS:

Some Java security options may or may not interfere with this program's 
operation.  If something doesn't work and there's no obvious reason, you 
may want to check the security settings.

It mainly needs to be able to play audio into a mixer channel.  In 
addition, the drop-down menu in one of the parameter-setting boxes needs 
to be able to set itself for "Always on top."  Also, if you want to 
paste the session log into a text file to save it you'll need to allow 
the program to access the system clipboard.

The defaults on at least some systems allow these.

It does NOT need to write disk files or send email.

Do NOT run it as root or with administrator privileges, especially if 
you don't know me or you got it from a third party.


GETTING STARTED:

Download the file

  Floomptagorpt.jar

into any convenient directory (You may wish to create a brand new 
directory to keep this project separate from everything else).  Open a 
command shell, go to that directory, and do

  java -jar Floomptagorpt.jar

which should cause the application's window to appear.  (Commands are 
given in Unix/Linux sytax.  Windows or Mac may differ.)

Once it's there, open your mixer, turn the master volume down to avoid 
unpleasantly loud surprises, click the application's Run button, and 
then turn the volume back up until you hear the tone at a comfortable 
level.

If you need to stop the tone quickly (for example, if it's too loud and 
you can't turn the volume down) you can click the Stop or Pause button, 
or drag the output level slider (the one on the left) down to zero.

Although many different settings of the various parameters are possible, 
the defaults will work reasonably well as a starting point.

If something seems not to work, or only partially work, check the system 
security settings.  Also, if you get no audio, exit the program and 
verify that your system's audio works with other software.  


SIMPLE CONTROLS:

The Run button starts the tone playing, the slider at the right edge of 
the application window changes how fast the tone goes up or down, the 
slider at the left edge varies the output level, and the Pause and Stop 
buttons silence the tone.

There may be a slight delay between moving a slider and hearing the 
change.  This is due to the time required for the data to work its way 
through the buffers.  (Possible future change:  Smaller buffers?)

The difference between Pause and Stop is that Pause just pauses 
everything so that when you hit Run it resumes where it left off, while 
Stop assumes you are going to be changing settings and thus resets a 
bunch of internal stuff.

You will also notice that Stop enables the Advanced Controls (see 
below), while Run disables them and Pause has no effect.

The sweep rate slider allows adjusting the frequency sweep rate.  For 
any given setting the frequencies change at an exponential rate 
(straight lines on a semi-log scale).  This can be expressed in units 
such as octaves per second.

Positive numbers mean the tone's frequency is rising, while negative 
numbers mean it is falling.  At the exact zero point the frequency 
remains constant.

The relationship between the numbers on the slider and the actual sweep 
rate is non-linear.  The actual rate in octaves per second is shown in 
the session log (see below).  It can range from several seconds for a 
change of one semitone up to ten octaves per second in either direction.

The output level slider adjusts the output level on a linear scale.  It 
is calibrated in percent of maximum output level.  Due to the nonlinear 
nature of human hearing, changes in the upper portion of the range may 
appear to have little effect.  They can, however, be seen on an 
oscilloscope.  You may prefer to use the level control in your system's 
audio mixer for most adjustments. (Possible future change:  Audio 
taper?)

There is also a Session Log that lists settings and such.  If your 
system's security permissions permit access to the Clipboard you can 
paste the contents of the log into a text file for future reference.


ADVANCED CONTROLS:

WAVEFORM:

The Waveform radio buttons allow a choice of four waveforms: Sine, 
Sawtooth, Square, and Triangle.  They all have the same peak amplitude 
(before any phase shifts and such between the tone generator and the 
speakers).  Thus their RMS values will differ.

The amplitude is calculated so as to peak just below the clipping level 
in the event that all components ("voices") happen to add in phase with 
the output level control at 100%.  Therefore the RMS levels 
will decrease as the number of voices is increased.


FREQUENCY PARAMETERS:

For a given waveform and sweep rate, operation is defined by four 
parameters: Number of voices (individual tone components), frequency 
separation (ratio between any two voices), minimum frequency, and 
maximum frequency.

If these are all constant then given any three of them it is possible to 
calculate the fourth.

Assume the frequencies are rising.  Eventually one of the individual 
components ("voices") will reach the maximum frequency limit.  At that 
time that generator will be reset to the minimum frequency limit and its 
frequency will begin rising from there.  Thus all the individual 
frequencies will be between the lower and upper limits.  Since they are 
all rising exponentially their spacing will remain constant in 
logarithmic frequency space.

Given the number of voices and their frequency ratio, you can calculate 
the ratio between the minimum and maximum frequencies.  It is simply the 
ratio expressed in logarithmic units such as semitones or octaves times 
the number of voices (An octave is a two-to-one frequency ratio.  A 
semitone is a twelfth of an octave, or a ratio of the twelfth root of 
two.).

The number of voices and their separation (in semitones) are two of the 
items that can be set from the Frequency Settings box.

We choose to start the sweep in the middle of its range, which is the 
geometric mean of the minimum and maximum frequencies.  Thus we set that 
value as our third parameter and use it to calculate the frequency 
limits.  The results of these calculations are shown in the Session Log.


NOTE:

The program does not check the settings to verify that any particular 
combination is reasonable.  Subsonic components, frequencies (or 
harmonics thereof) high enough to violate Nyquist sampling criteria, and 
so on are possible.  Also, ringing in analog low-pass filters in the 
output hardware may result in peaks greater than the theoretical maximum 
level.

The frequency ratio between voices is normally calibrated in 
equal-temper musical semitones (powers of the twelfth root of two).  If 
you prefer to work with integer ratios (3/2, 5/4, etc.) use the 
"rational" option when invoking it:

  java -jar Floomptagorpt.jar rational

The difference is rather subtle and may not be audible, but can be 
easily seen on an oscilloscope, especially with a small number of 
voices.


FADE-IN AND OUT:

For the effect of a continuous rising or falling tone to really work 
well, components that reach the upper or lower frequency limit and reset 
to the opposite limit should fade and out back in gradually so as not to 
be noticed.  This is the default of the "Fade In/Out" check box in the 
"Transitions" box.

If you turn this off the individual voices will remain at a constant 
amplitude, switching frequency suddenly at the limits.  This may be 
useful for certain tests and demonstrations, espcially when viewing 
waveforms on an oscilloscope.


CANCEL/DEFAULT/APPLY:

These buttons control what happens to the settings you make in the 
Waveform, Frequency Parameters, and Transitions areas.

"Cancel" returns the controls to the settings that were in effect when 
the tone was last playing.

"Defaults" returns the controls to their factory defaults.

"Apply" causes the control settings to take effect.

These actions are noted in the Session Log.

The frequency parameter settings have no effect on the Sweep Rate and 
Output Level sliders.  Thus you can compare different settings at the 
same sweep rate.


                        -- END --

2011-07-14 23:41:23