Physical Computing, Tangible Interface, TUI/NUI, Interactive Scultpure, Music Control, OSC
6.29.2007
ftir pre-alpha - display test
This is my el-cheapo DIY multi-touch interface in progress. So far, it's just the display half of the project. I've taken an old overhead projector and laid it on it's back pointing up through a piece of glass onto some typing paper. There is an lcd overhead panel displaying a crude smiley made with a monome and this is rear-projected onto the paper. It's high noon with the shades drawn.
What's kind of amazing is that I took no measurements and just sort of laid the stuff out to see what sort of structure I'll have to build, and yet it already almost works well enough. Though blurry, this sort of resolution (big blobs of color) is plenty for most preliminary audio/display apps I can think of. Adjusting (raising) the table height will get the diffusor into a better plane for this projector and result in a much crisper image. It'll help when the sun goes down, too.
Next I'm going to try the infrared blast approach to generating the finger touch blobs, using a budget night-vision security camera.
more pics
touch ribbon test
If I put 5v thru the outer leads it sends a range of 0.1-5 volts from
the center lead. It is very sensitive to touch in this mode.
If I remove power, resistance at the center lead ranges
from 0 to 10.5 K ohms - it requires slightly more force to get
a steady reading.
More pics here
More Thinking
And thinking
sketches 1
sketches 2
ribbon tests
shebang
6.28.2007
translucent touch pots
...well, sort of. Here are shots of two of the more promising types of pots that spectrasymbol sent, showing the relative translucence.
This shows the yellowish type that I think has the best tactile feel.
This shows it sticky-side-up
This is a wider type with a sort of grid of wires inside, and on ly 2 leads. The surface of this one is kind of like mylar. I tested it for capactiance and resitance. The pad appears to be divided into 4 distinct horizontal areas, giving values of 6, 5, 4, and 3 near the bottom. I'm not sure if you can see but it is sort of split down the middle but both halves give the same readings. On a capacitance setting it looksed like the values were greater in range but it was hard to get steady readings. Maybe I just need a data sheet on this thing.
More Thinking
And thinking
sketches 1
sketches 2
ribbon tests
shebang
This shows the yellowish type that I think has the best tactile feel.
This shows it sticky-side-up
This is a wider type with a sort of grid of wires inside, and on ly 2 leads. The surface of this one is kind of like mylar. I tested it for capactiance and resitance. The pad appears to be divided into 4 distinct horizontal areas, giving values of 6, 5, 4, and 3 near the bottom. I'm not sure if you can see but it is sort of split down the middle but both halves give the same readings. On a capacitance setting it looksed like the values were greater in range but it was hard to get steady readings. Maybe I just need a data sheet on this thing.
More Thinking
And thinking
sketches 1
sketches 2
ribbon tests
shebang
LED touch sensing
image from Jeff Han
Jeff Han's LED touch-sensing video can be seen here:
http://cs.nyu.edu/~jhan/ledtouch/index.html
Here is an article worth reading: Dietz et al, MERL
And this was a frustratingly dead link which purported to show a schematic, then I found an archived version here: ...motestruments.com/led-touch-sensor-circuit/
How to Build a 2-Note Ribbon Controller
an article from paia's arkive by John Simonton:
http://www.paia.com/ProdArticles/dual-ribbon-howto.htm
6.27.2007
phineus - riptide
7 new tracks from phineus:
meoomp.mp3 2:55
atunal.mp3 1:46
riptide.mp3 6:54
wuhwuhwuhw.mp3 2:08
wind-upwind.mp3 1:14
wavehello.mp3 1:32
longwalker.mp3 2:20
These were made during a weeklong stay in a cape cod beach house for a long-needed off-grid rest. I recorded the sounds of the house and the wind and the surf with my Edirol R1. Some tracks are made primarily from these sounds, while others include samples from the freesound project, such as plucked fiddle and glasses dinging. Sounds were played using the monome 40h and mlr running on a Mac Powerbook and recorded into the Edirol. The whole rig including the laptop, monome and Edirol weighs about 8.5 lbs.
meoomp.mp3 2:55
atunal.mp3 1:46
riptide.mp3 6:54
wuhwuhwuhw.mp3 2:08
wind-upwind.mp3 1:14
wavehello.mp3 1:32
longwalker.mp3 2:20
These were made during a weeklong stay in a cape cod beach house for a long-needed off-grid rest. I recorded the sounds of the house and the wind and the surf with my Edirol R1. Some tracks are made primarily from these sounds, while others include samples from the freesound project, such as plucked fiddle and glasses dinging. Sounds were played using the monome 40h and mlr running on a Mac Powerbook and recorded into the Edirol. The whole rig including the laptop, monome and Edirol weighs about 8.5 lbs.
6.26.2007
FLATHAND
In 2003 LEIF KRINKLE & ZACH LAYTON of ITP built the FLATHAND, using strip potentiometers from spectrasymbol.
6.25.2007
thinking out loud
So I'm thinking about the monome kit a lot since it just became available for pre-order. Both the logic board and the buttons are a great thing to be able to get for diy projects of all types, especially those involving microcontrollers and music and robots and stuff. Anyways, I was thinking about the challenge of making a faceplate. I looked today at sites about pouring resin into molds, laser-cutting plexiglass to order, and thought of other interesting approaches including simply hand-sawing it slowly out of wood or plexi with a small coping saw.
Meanwhile, I received a number of samples a few weeks back from a company that makes those blister buttons for xerox machines and they also sent me a couple of touch-strips. I was thinking about making a row of vertical touchstrips to act as faders in a monome-like instrument, and the samples they sent were basically translucent stickers with adhesive backs, but touch-sensistive. I assume they change resistance as you move your finger up and down. Since the strips are translucent, you could put a row of colored leds behind them to act as indicators. So the "xenome" could be used in conjunction with your monome to give you fader-like control of 8 channels, as well as animated output of sound levels, or whatever... cause it's all programmable.
So I figure if I'm going to all the trouble to order or build a custom faceplate for a monome 40h kit, why not do something special? I'm imagining one faceplate that could accomodate a monome kit and a xenome kit into one unit. (An additional feature of the xenome kit could be a built-in MIDI converterfor those who are not ready to OSC).
More Thinking
And thinking
sketches 1
sketches 2
ribbon tests
shebang
Meanwhile, I received a number of samples a few weeks back from a company that makes those blister buttons for xerox machines and they also sent me a couple of touch-strips. I was thinking about making a row of vertical touchstrips to act as faders in a monome-like instrument, and the samples they sent were basically translucent stickers with adhesive backs, but touch-sensistive. I assume they change resistance as you move your finger up and down. Since the strips are translucent, you could put a row of colored leds behind them to act as indicators. So the "xenome" could be used in conjunction with your monome to give you fader-like control of 8 channels, as well as animated output of sound levels, or whatever... cause it's all programmable.
So I figure if I'm going to all the trouble to order or build a custom faceplate for a monome 40h kit, why not do something special? I'm imagining one faceplate that could accomodate a monome kit and a xenome kit into one unit. (An additional feature of the xenome kit could be a built-in MIDI converterfor those who are not ready to OSC).
More Thinking
And thinking
sketches 1
sketches 2
ribbon tests
shebang
6.18.2007
monome 40h kit pre-orders
Warm up the irons, break out the Mouser catalog and choose your favorite flavor led:
begin: http://monome.org/articles/2007/06/18/kit-pre-orders/
begin: http://monome.org/articles/2007/06/18/kit-pre-orders/
6.14.2007
GATTACA Music
On this website you can enter a genetic sequence and have it turned into a MIDI file. If you don't happen to know your own genetic sequence they have some example sequences you can try out: http://www.doe-mbi.ucla.edu/cgi/pettit/gene2musicweb
The results are surprisingly musical. Here's a repository of known gene sequences: NCBI
The results are surprisingly musical. Here's a repository of known gene sequences: NCBI
6.10.2007
DIY FTIR interface UPDATE
(What's an FTIR interface? read this)
I've been collecting parts to build a budget FTIR multi-touch interface:
$25.00 Picture frame from IKEA
$50.00 vintage briefcase-style overhead projector
$50.00 Polaroid PolaView 3000 lcd overhead panel
free misc wire, power supply, and electronic bits
$25.00 19x19 plexiglass sheet, 1/4" thick w/ edges polished
-------
$150.00
Need to get:
- 20 infrared LEDs
- led pcbs (or just hand-wire them?)
- Fire-i™ Digital Board Camera firewire w/ IR and changeable lenses
- 4 telescoping legs
- diffuser layer / compliant surface (tbd)
So I'm thrilled with my eBay purchases. I skipped the modern projectors (LCD projector = too expensive, new o-head projectors = too flimsy) and found an old brief-case style by Buhl that looks like it was made in the 60's or 70's. $30 plus $20 shipping - such a deal. It folds up cleverly into a textured red and black tolex case, has dual bulbs, temp sensitive fan, everything is made of metal and it works beautifully. I found an inexpensive lcd overhead panel display that sits on top, making a very nice big monitor image on the wall. For the FTIR I'll need to get the reflector close enough to project a small (19" wide) image that will fit inside the frame yet still be in focus. It might require lengthening the head adjuster track but not by much.
The plan is to add 4 adjustable legs to the picture frame to make it into a tall tabletop. The projector and lcd will lie "on it's back" pointing up through the picture frame and plexi, at a distance of around 3 feet (tbd). The plexiglass will be covered by a piece of translucent material (tbd) to catch the image. Around the edges of the plexiglass will run 4 rows of infrared leds, shielded so as to only allow IR (infrared) to enter the edge of the plexi. Next to the projector lens, also pointing up, will be a firewire camera that has been modified to see only IR light. Fingers or objects placed against the tabletop, by "frustrating" the internal reflective properties of the plexiglass, will reflect blobs of IR downwards towards the camera. The camera will be hooked up to blob-tracking software (lots of openSource code/tools available on the web). The blobs will control a music interface (tbd). The resulting image is projected up through the plexi in normal (non-IR) light to display against the diffusor.
Right now it seems like the biggest challenge will be sourcing/making the correct material to use as a diffusor, see: How to create a Zero-Force Compliant Surface for FTIR Based Multi-touch on a Back Projected Table
Updates:
1) It seems someone on the NUIgroup forum has discovered that the bottom of a stock Dance Dance Revolution pad makes a good translucent/compliant surface. Hm, we have one of those in the closet!
2) I am also exploring the possibility of using the IR projection method instead of ftir. MS Surface uses this method. Basically, instead of using internal reflection of the plexi, they simply shoot ir up from underneath and catch the reflections of things that are touching the table. Not as precise as ftir but it seems to avoid a lot of the trickier requirements, especially the compliant surface issue. Also you can use regular glass, and off-the-shelf IR beamers are readily available (in fact they sell several "night vision" cameras that have built in IR beamers).
Lots o' links on the subject here: My Delicious FTIR DIY
I've been collecting parts to build a budget FTIR multi-touch interface:
$25.00 Picture frame from IKEA
$50.00 vintage briefcase-style overhead projector
$50.00 Polaroid PolaView 3000 lcd overhead panel
free misc wire, power supply, and electronic bits
$25.00 19x19 plexiglass sheet, 1/4" thick w/ edges polished
-------
$150.00
Need to get:
- 20 infrared LEDs
- led pcbs (or just hand-wire them?)
- Fire-i™ Digital Board Camera firewire w/ IR and changeable lenses
- 4 telescoping legs
- diffuser layer / compliant surface (tbd)
So I'm thrilled with my eBay purchases. I skipped the modern projectors (LCD projector = too expensive, new o-head projectors = too flimsy) and found an old brief-case style by Buhl that looks like it was made in the 60's or 70's. $30 plus $20 shipping - such a deal. It folds up cleverly into a textured red and black tolex case, has dual bulbs, temp sensitive fan, everything is made of metal and it works beautifully. I found an inexpensive lcd overhead panel display that sits on top, making a very nice big monitor image on the wall. For the FTIR I'll need to get the reflector close enough to project a small (19" wide) image that will fit inside the frame yet still be in focus. It might require lengthening the head adjuster track but not by much.
The plan is to add 4 adjustable legs to the picture frame to make it into a tall tabletop. The projector and lcd will lie "on it's back" pointing up through the picture frame and plexi, at a distance of around 3 feet (tbd). The plexiglass will be covered by a piece of translucent material (tbd) to catch the image. Around the edges of the plexiglass will run 4 rows of infrared leds, shielded so as to only allow IR (infrared) to enter the edge of the plexi. Next to the projector lens, also pointing up, will be a firewire camera that has been modified to see only IR light. Fingers or objects placed against the tabletop, by "frustrating" the internal reflective properties of the plexiglass, will reflect blobs of IR downwards towards the camera. The camera will be hooked up to blob-tracking software (lots of openSource code/tools available on the web). The blobs will control a music interface (tbd). The resulting image is projected up through the plexi in normal (non-IR) light to display against the diffusor.
Right now it seems like the biggest challenge will be sourcing/making the correct material to use as a diffusor, see: How to create a Zero-Force Compliant Surface for FTIR Based Multi-touch on a Back Projected Table
Updates:
1) It seems someone on the NUIgroup forum has discovered that the bottom of a stock Dance Dance Revolution pad makes a good translucent/compliant surface. Hm, we have one of those in the closet!
2) I am also exploring the possibility of using the IR projection method instead of ftir. MS Surface uses this method. Basically, instead of using internal reflection of the plexi, they simply shoot ir up from underneath and catch the reflections of things that are touching the table. Not as precise as ftir but it seems to avoid a lot of the trickier requirements, especially the compliant surface issue. Also you can use regular glass, and off-the-shelf IR beamers are readily available (in fact they sell several "night vision" cameras that have built in IR beamers).
Lots o' links on the subject here: My Delicious FTIR DIY
6.08.2007
tilt-ometer-ized monome 40h/se
How cool is this:
http://monome.org/media/se_tilt.mov
I believe they've just made a run of 16 of these:
http://monome.org/articles/2007/06/05/40hse/
6.07.2007
monome + flash
Check out this blog where a guy is using his monome to interface w/ Flash to do cool video drawing stuff: http://www.ilikegravity.com/real/archives/monome/index.php
6.06.2007
more monome models!
monome just announced they are going to produce multiple models of their button pads based on the 100h platform. There will be a 64-button model that is smaller than the 40h, a 128-button wide version, and the already-planned 256-button unit (100h). Also, they are renaming the models to "the 256", "the 128", "the 64". I prefer the hex designations.
Subscribe to:
Posts (Atom)