noisefloor.org

People have been experimenting with computers as creators of music for, what, half a century now? Even so, I think the field is still in its infancy. Music combines all of what makes us human, and computers, so far, are not that great at being human.

About three years ago, I found an application called Pure Data. From that site: it "is a real-time graphical programming environment for audio, video, and graphical processing." And ever since, exploring it has been on the bottom of  my to-do list. Well TODAY I finally started in on it. I got as far as exploring sine wave generators and random numbers. As it happens, this is a fairly sophisticated application. I quickly got some random notes out of the sine wave generator, but decided that it wouldn't be very hard to expand on that. Doing a lot of cut and paste allowed me to get pretty serious with randomization. I was able to randomize the way in which it randomizes, if that makes any sense.

Here's the Pure Data patch file I created, if you have Pure Data, or want to get it and play with this. Have at it!

Here's a sample MP3 of the patch in action:

sample.mp3 Stereo, 3.3MB, MP3, 320kb/s

And here's a screenshot of the Patch. You'll need some of the values, because when you open the patch, the initial max-values don't get stored in the patch, and it doesn't work without them.

To give you an idea of how much I like this patch, I've been listening to it for about 6 hours now. Hmmm. Perhaps I'm just a little off-center? heh.  Thanks for reading!

I never know when or where I'll be surprised by some new sound, yet it almost always is a surprise. It happened again while attending to some seasonal errands at the nearby craft store. This time it was a package of 50 or so medium size jingle bells. They were packed pretty tightly, and didn't make much of the usual bell-like sound. But the plastic container did make for a nice 'amplifier' of the sound of tiny metal balls rolling against the insides of the bells.

Of course I brought them home. I dug through a few boxes of wires and stuff left over from when I moved two years ago, and found a piezo transducer. I stuck it to the package of bells and started recording. I very slowly rotated the package of jingle bells. The noise is produced by the balls in the bells rolling around. The loud clunks are whole bells knocking against each other. I added some reverb, and here it is:

Merry_Christmas.mp3, Mono  928kB, MP3.

This one really pleases me.

When I went on vacation in 2009, I took with me not only recording gear, but lots of radios, too. I enjoy vacationing in places that are far from any sizable human habitation. This is due partially to my tendency to be a hermit. It is also because I really enjoy short wave listening.

And so I did a great deal of that this year. I heard something that really got me curious. On the Long Wave (LW) band, I heard what at first sounded like  a pure carrier. This would have been the exact opposite of rare, except that I suddenly began hearing an extremely small changes in frequency.I realized that what I was hearing had to be data. Of course, I recorded it. When I got back home, I opened the recording in Audacity.  Sure enough, I could see that my ears weren't deceiving me. I wasn't able to discover anthing about this data, but I did stick it in the back of my mind.

Now it has been four months since then. One of the blogs I read every day is Make Magazine online. I saw someone had built a really neat High Quality Sine Wave Generator (which is very cool all by itself). But what caught me like a bear trap was the reason they built this device: to produce Four Frequency Shift Keying (4-FSK) data signals using the MEPT-JT protocol. It was developed by Nobel Prize laureate and astrophysicist Joe Taylor, who was Professor of Physics at Princeton University. This protocol is nearly unbelieveable. People have documented reception of digital signals transmitted at a power of one milliwatt at a distance of over ten thousand miles! Have human beings ever accomplished that? The receiving decoder can decode data that is 25dB below the noise floor¹. Not above the noise, below it - so deep inside the noise that humans can often not hear anything at all, yet there is the received data on the screen.

One of the reasons for this  is because the system relies on an extremely narrow passband: about Six hertz. Second, it relies on an "out-of-band" sync signal: A GPS provided clock.

So, I've downloaded the software and my little Acer Aspire One netbook is dutifully decoding signals and logging them to this web page. You can find sample wave files included in the software.

1 - Oh yes, I've been wanting to use that phrase in a post here for a long time.