CHAMP LIBRE VOLUNTEER HELP

Good morning,

Thanks to the volunteers who came out to help this past week and weekend! We could use some more help today, tomorrow and Wednesday. Here is the information:

Monday:

- at the TML  - completing the cells and lighting from now until 6:00 pm

Tuesday:

- at the Monopoli Gallery from 11 am onwards - hanging the cells, setting up the network of electronics
- at the TML from 10 am onwards - electronics, helping to bring the work down to the Monopoli Gallery (would anyone be able to help drive a shipment from the TML to the gallery?)

Wednesday:

- at the Monopoli Gallery from 9 am to 12 pm - completing the installation, hanging.

Please let me know what works for you. - by email: gregorybeckrubin@gmail.com or by phone 514.848.2424 ext. 4351 (the TML lab) or mobile 514.502.0995.

Thanks for your attention and looking forward to hearing from you.

Regards,

Gregory

cell production, volunteers schedule

 

Starting now we would like your help putting together the

various parts of the Champ Libre installation. Following are some important

times to look out for volunteering:

 

Thursday, September 18: Electronics making/assembling

Location: meet @ ESKI Studio <http://eskistudio.com/contact/> (171 Rue Richardson, 514-510-5777)

close to Metro Charlevoix by foot. (it’s a HUGE

building…)

Time: 9:00 am to 1:30 pm

We’re aiming to start at 9 sharp; let’s first meet in the lobby of the

building at 8:50 am. Please confirm your attendance by e-mailing me

directly.

 

Friday, September 19: Cell production

Location: TML, EV-7.725

Time: 2:00 pm and onwards.

 

Saturday, September 20: Cell production/Various

Location(s): TML & Monopoli Gallery <http://maps.google.com/>

Time(s): 12:00 pm at the TML, 3:00 pm at Monopoli

 

Sunday, September 21: Cell production/Various

Same details as Saturday.

 

Monday, September 22: Various work, inc. electronics & cells production

Location: Monopoli Gallery<http://maps.google.com/maps?f=q&hl=en&geocode=&q=monopoli+gallery+montreal&ie=UTF8&ll=45.502106,-73.55823&spn=0.02292,0.051155&z=15>,

close to Metro Place d’Armers, 181 Rue Sainte-Antoine Ouest

Time: starting at 11:00 am

 

Tuesday, September 23: Same as above.

 

 

Just to clarify: experience with electronics is not a prerequisite! The work

is hands-on, accessible, and fun to learn.

 

 

The show opens on Wednesday, the 24th. By Saturday, we’ll be adding

    volunteer schedule info for the days following Sunday. Please keep your eyes

    peeled for these times/dates as well!

 

Thanks for your attention. Please e-mail <gregorybeckrubin@gmail.com> to confirm. Looking forward to

hearing from you.

 

Gregory and the rest of the crew.

 

Drawings from Evan

Evan and Patrick’s drawings for the publicity poster and postcard for Champ Libre.

Plan_Section_CHAMPLIBRE_WEB champlibre_FACADE_WEB

80 Nodes Produced

eski 238

led api 0.03

Initial attempt at defining a high level api for led control from Max:

Pneuma_API_0.03

software+hardware system diagram Champ Libre

pneuma_system_diagram04

pneuma_system_diagram-shanghai05

coding dummy sensor munge - state brain - led out

Navid and Morgan and JC
It would be great if

(1) you can parse the data into Max from at least 2, ideally closer to 6 photocells in cardboard tubes,
fanned out in an arc of directions,
PLUS an ultrasound.

TML can buy this one from Navid since it is expensive.

(2) make a dumb “state engine” patch that maps from sensor vector
to a list of values packaged for output to serial (USB) in the format that I wrote up on blog:
http://topologicalmedialab.com/blogs/pneuma/2008/09/06/code-questions/

(3) play, put some time-morph, delay, compress (speed up), and shaping (eg crude wavetable).

remember the photocell data will be from sun’s SLOW passage,
but the ultrasound will be from jittery or slow people.
We’ll need to be ready to interpret them

Xin Wei

sensor-dock code email

Hi Vincent,

Yes, please give us polling, thanks.

Beyond-Pneuma Q1. Would it be possible to also make a push model where one sensor channel’s data is emitted only when there is delta above a user-settable noise threshold? This has been really useful for us where we want to distinguish “constant” gesture from

Beyond-Pneuma Q2. is there an easy way in this current design to accommodate more that 8-bit resolution, say a 12-bit photo sensor?

On the LED “output API”, what’s reasonable to assume for starters ? I’m just spit-balling, presumably something minimal like:

(1) the LED network itself has an autonomous behavior of wave-like attenuation;

(2) a format like the following:
node_id (0..255), command (0..127), value (-127..127)

Command : { absolute_level | relative_increment | multiply_by… }

if the command =
absolute_level, then intensity of that node is set to absolute_level;
relative_increment, then increase or decrease the intensity of the node by relative_increment;
multiply_by, then multiply the intensity of the node by value (rescaled to (-1.0..+1.0)

I won’t presume to ask for anything more than a bare minimum till we get a chance to play with your version 0 of the LED controller API.

Later, it may be useful to have parameters like
ramp time,
spatial distance measure
spatial attenuation factor (each cycle multiply cell by factor),
modes, such as setting some cells to act as “reflectors” (if that makes sense given a topology),
or “pinned-value” (acting like a source or sink). (we can talk and see if it _may_ be trivial, or not :)
I don’t know. I’ll learn with you what’s feasible given current constraints.

Thanks, compliments,
Xin Wei

On 5-Sep-08, at 10:26 AM, Vincent Leclerc wrote:

Thanks JS.

It might be more robust/stable for your machines if instead of
streaming data constantly, I sent the reading of all 32 sensors each
time you “bang” the serial object. This way the sensor dock will only
be sending data when you ask for it. What do you think?

Vincent

On Thu, Sep 4, 2008 at 10:31 PM, Jean-Sébastien Rousseau
<jsrousseau@gmail.com> wrote:
Hi guys,

Here are some Max5 patches, in order to play/work with the SensorDock that
Vincent brought to TML today.

I tested them on my MacBookPro, everything worked fine, except for the two
times when I closed the help patch and the system hanged, and needed a
reboot. Probably need to tell Max to let go the usbserial connection before
to close the patch.

Note that the SensorDock needs to be connected to your USB port before you
start Max, in order to be seen by the ’serial’ object used by the patches.

Plus, you need to have installed the VCP driver (virtual COM port) from
FTDI. find it at: http://www.ftdichip.com/Drivers/VCP.htm

Have fun

JS

On 4-Sep-08, at 3:31 PM, Vincent Leclerc wrote:
The sensor dock is fully functional…. and extremely fast!
I need the contact info of whoever will design the MAX-MSP patch to
obtain data from the dock.

Or you can forward him/her the following info:

———

Sensor Dock Connection Info

If you don’t have the Arduino IDE installed on the machine, you need
the VCP drivers from FTDI:
http://www.ftdichip.com/Drivers/VCP.htm

Connect to the dock using a patch similar to this one:
http://www.tigoe.net/pcomp/code/category/arduinowiring/108

The Baud Rate for the Sensor Dock is 115200.

Each message is of the following format:

SENSOR NUMBER [0:31]
COMA [,]
SENSOR VALUE [0:255]
NEW LINE [\n]
CARRIAGE RETURN [\r]

Vincent

code questions

What’s reasonable to assume for starters ? I’m just spit-balling, but I suppose minimally something like:

(1) the LED network itself has an autonomous behavior of wave-like attenuation;

(2) we can send something like the following:
node_id (0..255), command (0..127), value (-127..127)

Command : { absolute_level | relative_increment | … }

if the command = absolute_level
then intensity of that node is set to absolute_level

if the command = relative_increment
then increase or decrease the intensity of the node by relative_increment

Later, it may be useful to have parameters like
ramp time,
spatial distance measure
and a spatial attenuation factor (each cycle multiply cell by factor)
but I don’t want to ask till we get a chance to play with version 0 of the LED controller.

shipping, programming

Shipping
JC has researched process — now needs to talk with Patrick and me about CARNET ATA
bc we need to decide this weekend what pieces to ship and what we shall carry in hand.
JC and I think we should crate and ship the electronics (maybe we should ask Vincent if it
would be possible to print a unique ID on EACH board, numbered serially. I will ask.)
The considerations are:
(1) radically reduce weight and volume of shipped crate;

(2) we can much more easily travel with plastic sheet, than electronics;
(3) we can pack the electronics carefully in pre-prepared crate;
(4) we can pre-id the electronics and list them (are serial numbers better or would they generate needless work?);

Cut File
Peter, Pat seemed to say he was beginning to do drawings, to be converted to cut files. Patrick talked with me a bit about the 3D volumetrics. Let’s schedule an ichat / skype this weekend bc I feel we need to sync up now that we have
electronic prototypes from Vincent, and are about to launch into integrating prototype cell + electronics.

Max on Mac Programming
This Saturday Sep 6, Navid Morgan and I will work in the TML from 2 PM onward on the sensor-dock —> Max with sample sensors that Navid has
(TML will reimburse or replace : please add to a parts wish list which is ftp’d onto pzp@raid.arch.umanitoba.ca/tml-pneuma/common-budget-schedule/

I would like to play with the complete sequence form sensor to LED this weekend of possible.
At least we’ll simulate the light using the halogens, via the DMX patch (thx JS, Harry)

Next Page »