On Sun, 30 Apr 2000, you wrote:
>
>
> What did he(Assassin message at bottom) just say to me Doug? I don't
> think I quite understand what I'm supposed to try.
First of all, he's talking about the same approach that I described,
using a microcontroller to control the LEDs. He further talks about
matrices (ie grids) of LEDs, parts you can buy that have many LEDs
in a single little "brick" of plastic. This brick contains both the
LEDs and circuits to drive them. Finally he refers to the signals
(voltages) being sent from the microcontroller to the LED unit. It
works like this: wires connect between pins on the microcontroller
and pins on the LED controller. A program applies volatges high or
low to the pins and sends these signals to the LED unit. There are
documented descriptions of what signals to send and how that come
with the LED drivers.
Note that although we are talking about the same problem here (how
to control LEDs), his suggestion was different than mine. I suggested
that it might be interesting to weave tiny individual LEDs into
the fabric, whereas he is speaking of modules of many LEDs. Both
can be useful depending on what you want to do. I think it would
be kinda cool to weave the individual LEDs in, because then it
doesn't look like a computer, it looks like fabric that lights up.
> brain... about a week after the iswc conference last year...I was really
> inspired by the speaker that closed the conference. He was from DARPA and
> he had talked about clothing that changes as you move... camouflage that
> changes with the background...
That sound cool!
> My first thought was that there would be little output devices on the
> outside of the garment (attached like sequins or fish scales) that
> could be like teeny tiny sony camcorder screens (but hopefully much
> lighter {even a sequin gown is a good ten pounds}) that could have
> radio connections to known points in an area that could send signals
> from a fixed cameras (probably with location from some type of local
> triangulation process GPSish thing)
If you're gonna dream, dream big! <grin> What you describe would be
very cool indeed, but for now it is stuff off science fiction. Some
day soon though these ideas will come a lot closer to reality. You
could easily do triangulation with beacons, for example the fabric
might have infrared receivers all over the place (cheap and small)
while fixed beacons have infrared transmitters. Another thing that
this reminded me of is the game boy cameras. Inside of these are
tiny mitsubishi digital cameras. Apparently these are low cost
devices, and many can be networked together. In theory you could
for example put them on all sides of a vehicle, or perhaps fabric
too. There are some fairly detailed hacker pages on the web about
these gameboy cameras. But don't get too excited yet because it's
not a trivial exercise and the cameras by today's standards are
not cheap in low volume.
> Also I considered fish and birds natural coloration one color
> form one angle an other color on an other so camouflage works
> from top and bottom to make a body less noticeable.
This is a VERY cool idea. I hope you implement it some day.
> So then I thought having small video cameras all over the body
> wired to the opposite side of the body on all sides of the body
Aha!
> might provide all around changing of the clothing coloration that would
> mute to tones similar to the background regardless of what angle(within
> reason) the garment was viewed from. (in my opinion a much better solution)
Cameras are going to be an expensive proposition right now, but
there are many other sensors that you could consider. Light sensors
would be cool, they could make the clothing look different depending
on whether you are in dark or light. I have been researching sensors
of all kinds and it is quite fascinating.
> By this time I was thinking of LEDs because I thought a hundred tiny screens
> would be really neat but cost a ton 'o' cash bee totally heavy and might
> not even exist.
The tiny screens are hideously expensive. For example I am about to
shell out a whopping $3100 for a wearable display. Inside the eye
peice is a tiny half inch screen that does 800x600 resolution. Most
of the cost is the driver electronics, but an extra display alone
is $600. Youch.
> I narrowed my scope to a simple one or two color project maybe a tie that
> could change patterns with different scanned images or something of similar
> size a panel on the shirt that shows whats going on on the other side of
> the body as if we were looking through a hole. Think of henri Magreet
> paintings of an easel painted exactly like the background.
That's a good idea for starters. Start small and work your way up
to the more complex versions.
> later we got one of those plasma gas screens in at work and looking at it
> closer showed tiny little specks of colors close together so each tiny
> little speck is made of three little parts one red, green, and blue.
Yes, most displays use pixels of RGB. Unfortunately they tend to be
either big heavy cathode ray tubes (CRT) or flat square liquid crystal
displays (LCDs). I would bet good money though that one day these
things will flex somehow. Then perhaps your clothing could be a
display. That would be bizarre. Hee hee, you could change clothes
dynamically with software!
> So I thought easy. I'll just put little clusters of these LEDs together and
> if I can figure out bright/contrast I could just call on different lights to
> light up more or less depending on the pattern I wanted.
Okay the fundamental starting points are inputs, processing, and
outputs. Inputs are the sensors themselves. You need to pick some
simple ones to start with like light sensors. The "processing"
can be done with hardware (ie fixed circuit) or with software
(dynamic control). You probably want to end up with some low
cost fixed circuitry eventually. But in order to experiment with
many different possibilities, I recommend using a microcontroller.
You attach sensors and LEDs to pins on the microcontroller. Then
you use basic programming to read the sensors, and turn on the
LEDs. You can start with just a couple of LEDs, then when you
get to the point of wanting to control an array of them, you
will need to add some other parts that can "multiplex" or
switch between the LEDs. So I'll go back to what I said before,
to get started, and easy way is to buy the BasicX BX-24 starter
kit from www.basicx.com. All you have to do is connect the
micrconroller to a PC with a serial cable and load the software
that comes with the kit. There is already example programs for
switching LEDs on and off. To run them you simply open up the
program files and select a menu item called "compile and run".
There are other example programs for reading light sensors,
infrared remote control receivers, reading potentiometers,
driving relays etc. There is also some good example code at
www.phanderson.com.
> I HAVE NO IDEA HOW TO ACCOMPLISH THIS.
There are two fundamentally different approaches. The first is
pure electronics. You would have to start reading some books
on electronics, and perhaps get some educational kits. This
is recommended anyways regardless of which way you choose to
do it, because you'll end up needing at least some minimal
electronics knowledge. There are some little books called
"engineer's mini notebooks" at radio shack for $2 each.
Take a look at those. In particular there is one for sensor
project ideas, I think this might give you some really good
ideas on how to drive the inputs.
The second approach is a software one, using a microcontroller
to monitor the sensors and drive the LEDs. Nice thing about
this is that you can try out many different sensors and
LED combinations, without hard-wiring anything (no soldering).
I suggest that you get the basic-x kit, the sensor book from
radio shack, some LEDs, and a few simple sensors like light
sensors, buttons, potentiometers and such. Also get a
"bread board" which will allow you to temprarily plug in
parts and jumper (wire) them together. It may sound quite
intimidating at the moment, but really it's not that hard.
If you get the basicx kit and get stuck, I would be happy
to give you some tips along the way.
> dress that would blend into whatever wallpaper I was leaning
> against
You seriously should pursue stuff like this as a long term
goal. I think it would make waves. I see this as artwork
that could end up in galleries and such.
> I really wonder if I should start some place a little more rudimentary
Start with simple sensors connected to batteries and switching
LEDs on and off. Move on to using the basicx kit. I think you
will enjoy it once you get rolling.
> I'm thinking I should work on something more usefull and less fantastic eg.
> something a little more like a wearable computer lizzie type thing.
In terms of practicality, those "pretty little lights" can tell
us lots of things. I use them to tell me if I am connected to
the network, if I am able to reach machines across the network,
battery status, that sort of thing. With some more creativity
it could be useful and interesting. It would certainly be
easy to make graphs of information for example. On the more
complex side you could draw pictures or render letters and
numbers.
> I realise this is a bit out in left field.
Nothing wrong with left field. And your questions are directly
related to wearables, so there is no problem discussing them
here.
-- Doug
--
Subcription/unsubscription/info requests: send e-mail with subject of
"subscribe", "unsubscribe", or "info" to 
Wear-Hard Mailing List Archive (searchable): http://wearables.blu.org
From Wear-Hard Mailing list Archive (WH)
Maintained by R. Paul McCarty
Archive created with babymail