Warble tones - and a request for new forum

[Dave Jones]

[XPosted to Software Tools and Everything Else]

What exactly is a warble-tone? Is there a standard definition?

I wrote a little program that generates a WAV file of what I imagined a warble-tone to be. I frequency-modulated the center tone. It seems to work, which is to say, it has the correct center frequency, it warbles at the specified rate, and the RMS amplitude is .707 times peak. However, it sounds different from the warble-tone produced by Speaker Workshop given the same parameters. Mine has a higher perceived variation in frequency. It sounds as though it "warbles wider." One of us is wrong, and I suspect it is I.

I frequently have theoretical questions of the newbie variety. Often I figure the question has been asked and answered, but searching proves time-consuming futile. There is no good place for my questions on the diyaudio forums. I am posting this to "Everything Else," where it will be seen by few and lost among, well, everything else. I would like to see a forum or forums for theoretical, best-practices, and newbie questions. The good answers could be made sticky and locked.

Added in proof: Maybe Speaker Workshop's spread-parameter is total. I figured that a warble of 1/3 octave means it varies in instantaneous frequency from 3/4 to 1 1/3 times the center frequency. Which brings up a third question. Speaker Workshop is supposedly open-source. Does anyone know a painless way to snarf the source code? I tried to install Turtle CVS, but no joy.

[mickeymoose]

A warble tone is basically a tremolo, an amplitude modulated tone.
In aoustical tests it is used with a slowly swept sine frequency, warbeled at around 4x per second, the am is about +/- 10% of the main signal. The reason for the warbel tone is to reduce the possibility of standing waves in the room measured.
A frequenzy modulated note is a vibrato.
E

[sreten]

Hi,

For a 1/3 octave warble tone it needs to vary +/- 1/6 of an octave around
the centre frequency, i.e x the sixth root of two and 1 / sixth root of two.
i.e. 1.225 and 0.891, that is very different to the range you are using.

Each 1/3 octave is spaced by a constant multiple of third root of 2, 1.260.

I assume the modulation should be triangular and frequency logarithmic,
not linear, not sure how you would go about making the latter true.
Use a linear VCO and bend the driving signal a little (lot) to suit ?

rgds, sreten.

Think about a 2 octave warble tone, half f to twice f,
the modulation cannot be linear, must be logarithimic.

[mickeymoose]

OK, I looked up "warble" in Webster's Encyclopedic Unabridged Dictionary (any spelling errors?): it says: 2, lump in the skin - ups, sorry, wrong one! 3, (of electronic equipment) to produce a continuous sound varying regularly in pitch and frequency.
E

[Dave Jones]

Originally Posted by sreten

Hi,

I assume the modulation should be triangular and frequency logarithmic,
not linear, not sure how you would go about making the latter true.
Use a linear VCO and bend the driving signal a little (lot) to suit ?

I used sinusoidal modulation on a log scale. Maybe if I changed the sine wave to a saw-tooth, it would sound exactly like the warbling that Speaker Workshop does. That had already occurred to me. What kind of signals do analog warble-machines warble? Anyone know?

[Dave Jones]

Originally Posted by mickeymoose

A warble tone is basically a tremolo, an amplitude modulated tone.
In aoustical tests it is used with a slowly swept sine frequency, warbeled at around 4x per second, the am is about +/- 10% of the main signal. The reason for the warbel tone is to reduce the possibility of standing waves in the room measured.
A frequenzy modulated note is a vibrato.
E

That's a new one on me. I'll have to give that a ponder.

[mickeymoose]

Dave:
My reference is the Audio Cyclopedia by Howard M. Tremaine (4th. printing). Although the dictionary says one thing, ie general language use, the same word in tech speak may meen something quite different