Computation08

Last Prototype

Here is my last prototype:

Abstract Video of D.C. Madam from dj pimptress on Vimeo.

"Red Sky at Night, Shepherd's Delight. Red Sky at Morning, Shepherd's Warning."

Red Sky at Morning (Shepherd's Warning) is an exploration of new media, fine art, and surprises. The title implies that there is something unexpected coming, and relates to the rhyme, "Red Sky at Night, Shepherd's Delight. Red Sky at Morning, Shepherd's Warning."

My intention was to surprise anyone looking at the projection with the actual content of the image. Would the sheep copulating get their attention? Would the party hat make them look twice? I ended up getting the same amount of attention when I projected still sheep than when I switched the images every five minutes. Turns out that simply projecting the picture onto canvas is enough to get attention. The surprise came when the image did not appear to react to motion or any other outside stimulus.

I did another test run in the lobby at 2. west 13th, and got similar reactions to the 10th floor lab. I'd still like to take the picture outside to test it in public, outside of the school environment. Anyway, here is the movie from yesterday's test.

Experiements with the game of life

Last week I created a variety of rule sets for the game of life.  I began this weeks work with the aim to combine several rule sets into one program.  I began by attempting to create my own rule set from scratch. I was getting a better handle on how the code worked and was having more success with controlling the programs outcome.  This was due in large part to the use of the 'print' statement.  The rules are dictated by a series of if statements: if there are 3 neighbors than die etc.  I put a print statement in each if statement and had it spit out a number (the first rule spit out one the second spit out two etc.).  By doing this I was able to see which rules where firing and how often.  In the process I found some interesting results. 

Traditionally, Cellular Automata programs work by looking at each pixel individually every frame and comparing them with the surrounding pixels.  I was working with a canvas size of 400 by 400 pixels and the program was set to run  at 8 frames a second.  That's 1,280,000 calculations per second.  This is nearing the limit of Processing's power.  When trying to understand code I like to use print statements to get a handle on what and when things are happening.  With this program many calculations happen every frame Processing was not able to run with both the pixels being drawn and the print statements being displayed.  I could do one or the other but not both.  This made it very difficult to get a handle on what was happening.  However the print results were interesting.

Click to view the printout as HTML.  Your browser can zoom the text in and out. on mac: use "cmd & +/-"  on pc "cntrl & +/-".  Try it out.
  Note: the white space is the number 1.  At the zoom level the images was captured at the 1s don't show up.

This is the program used to make the printout. 

These print outs are interesting to look at, but they also could act as blueprints for the game environment  if it were to ever be developed further. 

I made print outs of some of my other life programs to see the results.  I continued tinkering with my code and gaining a better understanding of how the CAs worked.  It became clear that I would need some kind of system to keep track of which rule set was being applied.  The different rule sets would also need to be recognized by each other in order for the system to act as an integrated ecosystem.  There are many ways to do this, however because of the sheer number of calculations being preformed each frame already I needed a way that was fast.  Integers are the fastest data type so I decided to base the system around them.  The way game of life program that I am working with works, is by looking at each pixel and adding up the value of it's neighbors.  This total represents the number of neighboring pixels that are turned on.  If you have multiple rules working together and iterating with one another this won't work because the system will have no idea which rule is on.  I developed a system that would encapsulate all of the rules into one three digit integer. The "blue" count would be held in the 'ones' place the "red" count would be held in the 'tens' place and the green count would be held in the 'hundreds' place.  I than wrote a function that could parse the three digit number and return which ever place was being checked.  The function follows:

int nb(int N, int num) {
  for(int i =100; i >= 1; i=i/10) {
    int x;
    x = num/i;
    num = num - i*x;
    if(N==100) {
      if( i == 100) { //green
        return x;
      }
    }
    if(N==10) {
      if( i == 10) {//red
        return x;
      }
    }
    if(N==1) {
      if( i == 1) { //blue
        return x;
      }
    }
  }
  return 0;
}

The function returns an integer.  The first parameter is place that you want returned ('1', '10', '100') and the second is the number of neighboring pixels that are turned on (1-8 for blue, 10-80 for red, 100-800 for green).  so num might be 232.  That is 2 green 3 reds and two blues.  The function uses a for loop.  Each loop variable x is set to num and divided by i than it checks what value was entered and which step the loop is on.  If they match it spits out the number of neighboring pixels of the specified rule set.

With this function I set out to combine the three rule sets starting with ones that I had conceived of in my note book.

visual sketches of rules:

I tried to implement these rules.  The results were mixed, I was able to get interesting results but as the program runs the output gets more and more chaotic. 

rule combination 1

I made several iterations of this rule set and had trouble gaining control.  I than had the idea of applying my multiple rules method to some of the life rule sets I had made previously and was successful.  Below are the results.

 

Ruleset 1

and
Ruleset 2

The result:

Combination of life 2 and life 3.

3 rule set combination

The atmosphere is the last component to add.  Below is a composite of how the gas might look once it is added.

I found several cellular automata machines here is a good one if you want to experiment with them http://www.fourmilab.ch/cellab/manual/chap5.html

Using techniques of mathematical logic, von Neumann was then able to deduce that such self-reproduction should in fact be possible. His proof hinged on the idea that an automaton could have a blueprint for building itself, and that in self reproduction, two steps would be necessary: 1) make an exact copy of the blueprint, and 2) use the blueprint as instructions for making a copy of the automaton. The role of the blueprint is entirely analogous to the way DNA is used in biological self-reproduction, for here the DNA is both copied and used as instructions for building new proteins.

The complexity of a reservoir full of floating machine parts hindered von Neumann from making his proof convincing. The next step came from Stanislaw Ulam, who was working with von Neumann at Los Alamos during those years. Ulam's suggestion was that instead of talking about machine parts in a reservoir, von Neumann should think in terms of an idealized space of cells that could hold finite state-numbers representing different sorts of parts.

Below are von Neumann's theories put into practice as software:

Wikipedia article here

Day 7: On-route speculation and prototypes

A re-visit  to motivations and  what became of them:

1. I wanted to visualize data - any kind
2. Tried at coming up with a correlation to determine/ some question
3. Finally dwelt on: architecture and its influence on tourist/ general experience of city space
    Hoped to it to generate visualization for a travel book.

4. Parameters dwelt with age/style of architecture and demographics of the space
5. Process became a little confused conceptually as mapping data was not looking at style but at age
    Demographic data was more of use to perhaps historical conservation agencies than a tourist.
6. A future comparison of architecture in various cities may yield information of similar spatial experiences.

An illustration of the process:

The Present form (?):
Mapping New York's architecture by age

Who can use the visualization?
1. A Tourist:
>>Guage style and density of the style of architecture by area
>>Plan tour accordingly
>>Get a sense of scale of architecture to visualize how it would feel being in that space

2. A conservationist/ urban planning department/ historian
>>Guage style and density of the style of architecture by area
>>Guage age of buildings and spread of NY buildings over time
>>Guage relationships between architecture vs rent
>>Guage relationships between architecture vs ethnicity/ownership
>>Guage relationships between architecture vs land use
etc.

Download nyBuildings.html
Download nyBuildings.jar
Download nyBuildings.pde

Pros and Cons + Where from here?

1. Cons: The 'need' for the visualization has to be more fleshed out to determine what data needs to be correlated. Perhaps requires interviews with conservationists and tourists to determine that.
2. Pros: Got a hand at data collection techniques.
3. Where from here?: (a)Refine ways in which the visualization of building and styles can be shown (b)use the prototype to query conservationists and tourists to figure whether this tool is of any help.

venn revisited

Mini Thesis: part 6

First "Tourina Box" user test:

My first user is 28, female and has lived in NYC for 8 years.  She used it to create a tour for her Brother, who has never visited her in NY.  It tours the places she has lived around NYC since she moved here in 1999.   The box includes artifacts, voice recording and maps. Here is her interpretation of  "Tourina Box":

-

-

-

Additional Charts/diagrams:

-

 

-

 

Diagram 2

Initial content

The interest in the project started with the relationship between horror movies and war. is there a corrolation? a delayed reaction? a cause effect relationship? a inverse corrolation? or anything?

Movies fall in to the cultural artifacts catagory which may be a product of many factors, can we find a relationship between them and one of these large subjects such as war?

How do comparison relationships contextualize components being compared.

Economist: China's GDP

In visualizing the relationships:

What is the best way to explore possible relationships between two sets of data?

animation?
bar graph?
line graph?

What can be added to better visualize these types of data relationships?

visual exploration

"macro/micro" INSTALLATION

THURSDAY MAY 8th 4:30-8 pm

25 east 13th street ((FINE ART/architecture building))

COME SEE!  video documentation ... 1st large scale prototype ... more to come
((apply for grant for future installations with higher tech materials and more space))

 

            

   

Prototypes, 4

Venn Diagrams, re-mapping:

Venn Diagram, cleaned up:

DB Schema, revised: Download dbschema2.pdf

User Scenario Pamphlets: