Shared on Thingiverse

I decided to share the how to of the device in the last post with the world via Thingiverse.

Check it out.

Cordislumen (Tentative Name) progress

Wow... last little while has been piss poor blogging performance for me, hasn't it? I apologize, many things have been happening in my professional life, the majority of which I do not feel comfortable sharing online.

Anyway, onto the cool stuff.

Cordislumen (Tentative Name) progress part 1 from Andrew Parnell on Vimeo.

I just finished building one of the ... I believe three ... components of my next piece. (Tentatively known as Cordislumen, hence forth not referred to by name). In function it's like the system of pulleys in my last piece. But practically speaking...it works so much nicer. In part this is because I'm not having to deal with co-opting a piece of building infrastructure (though, I did really like that element of Tangólumen) and also because I learned a lot from Tango's design, and was able to better design this one.

This portion of the project contains an outlet, a dimmer switch, and a servo (and for now: a dirty rubber band). The servo operates the dimmer, which in turn controls the amount of current going through one of the outlet's two sockets. (The other one delivers consistent power to the micro-controller / servo via a wall wart.) This is a really nice way to control an outlet...because in all truth, full source voltage is scary, scary stuff.

Some more photos...right here:

Tweenbots

Since I first saw them on Saturday, Tweenbots seem to be making a serious buzz. This isn't a surprise really, I mean who can resist a cute cardboard robot that navigates urban landscapes through human intervention?

This is exactly the type of project I want to create. It rehuminizes this inhuman technology. There's something really sweet about it.

<3

[Tweenbots] [via NOTCOT]

Tangolumen Completed / Post Review Reflections

After living in my studio for the better part of a week, reviews finally came, and Tangólumen was up and doing things. It wasn't doing exactly what it was supposed to...but it was doing things, and that's something.


Tangólumen In Action from Andrew Parnell on Vimeo.

People seemed to enjoy it, and I got a lot of really useful feedback ranging from what to do next, tweaks to make, things to consider, possible redirection, and so forth. Now that I've gone back and adjusted the lights and programming so that it's actually to spec, I really need to start inviting people into the space and use data I gather from that to further development.

I am the for loop.

There's one thing that seems nearly thematic in my current project: repetition.

This is a side effect of the modular design I'm really proud of across the board in this project, but at the same time it's making the assembly portion of fabrication really repetitive.

Some examples:



There are 64 pieces of chipboard in this project, each with a 17" long strip of aluminum tape running up their center. Each piece is then taped into a grid, with cat 5 cables shared among every eight panels.



When I designed all the acrylic pieces for the pulley system, I designed it around an interchangeable system of pieces. Six pieces combine to make 1 part, there are 32 parts in this system, meaning that I cut 192 pieces (not including extras, or the pulleys). Cutting took ~30 minutes (well, an hour, but that's because I had to do it twice), assembling them was another story all together.


Review prep checklist:
[19/64] Strip Cat 5 cables + Glue them to the panels with conductive epoxy
[ ] Assemble flooring
[ ] Attach Vc to flooring
[ ] Recut 1 servo-box
[ ] Cut 2x 1' of threaded pipe
[ ] Attach pulley rig to dimmers
[ ] Cut and crimp steel cable.

i++;

Some Assembly Required

The downside to laser cutting is that if you want undercuts, you basically have to create them as separate pieces and then laminate them together. The upside is that they look, and are, awesome.



Last leg of cutting is scheduled for tomorrow. Then I just have to cut the threaded rods down length and assemble.

Prototype

Untitled from Andrew Parnell on Vimeo.

Fun With Lasers

It's been a while since I've given a project update here. In part because I've been really busy (1 job, 2 internships, and school) but mostly because I'm trying to find a way to tell you all about my progress / what I'm doing without giving away the ending.

I don't think I'm risking too much with this update, so here it goes.

The project I've been working on involves a lot of both large and small scale work. The small scale stuff requires a lot of parts to be fabricated. This gave me an excuse to design stuff to be made on the laser cutter we have in the building here.

Last week I cut some test pieces:


Which I then assembled into a 3d form so they can serve their purpose of joining threaded rod at perfect 90 degree angles.



Eventually I'm going to cut a few more of these (with modifications made, this is a learning experience.) and eventually wind up with this:


For now though I've gotta say, lasers are just awesome.

Xbox Live Vision

The Xbox Live Vision is an excellent webcam. You can find it for about $15 on ebay and it uses a universal driver, making it compatible with pretty much any computer without hassling with drivers. This makes it an ideal choice for interactive video installations. I used two of them in The Eye. But if regular visual data isn't enough? Modify it to see infrared!

I really need some exposed film.

My first contribution to the Arduino Playground

The Arduino: Playground has given me so much that I felt it was time to contribute back.

So I wrote this article: Arduino + PERL

Blogger Upgrade

Apparently blogger upgraded one of the backend pieces of their atom / rss feeds, migrating from OpenSearch 1.1 to OpenSearch RSS 1.0.

There aren't a lot of people this really effects, but since this blog is driven by a blogger feed, it screwed with me for a second.

Note to self: always double check ratios when ording gears.

I'm a little irritated with myself. I ordered some servos, servo gears, shafting, and plain bore gears last night, and while doodling up schematics today, I realized that I reversed the ratio in my order. Seriously lame.

I went back and ordered the propper ratio (4:1 not 1:4). What irritates me is that if I'd gotten it right the first time, I would have saved a buck, this way I spent an extra twenty-three.

Lesson learned.

Remember an Article by Veerle

Thinking about flash based sites I remembered an article by Veerle.

Read it.

Thinking about flash

I'm part of the team that's building the website for the 2008 digital arts bfa class @ UO - Portland.

Today we all got into a good, hour long...maybe less, but I doubt it...discussion/fight centered on the idea of having a mute button for sound effects on the site. I am, and always will be, of the opinion that if you're going to have sound on your website you must have a mute button. The easiest example at to why is that many (if not most) people browse the web while listening to their music library, or to a lesser extent watching tv (or some sort of video). Not giving the option to mute, you're sending a message to those people saying quite clearly "look, if you can't dedicate all of your senses to this website, leave." That's totally unnecessary, if not a little offensive.

Another reasoning I have behind always being able to mute specific items is that it creates a barrier to those who need to use a screen reader. This doesn't really apply in this case though because we're developing the site in flash.

The more I think about it, the less fond of this second fact I am. Infact, I'm not fond of this fact at all, and really never was, but failed to bring this up. (Why, I don't know.)

Flash is a means of web development. Yes, it's great for interactives, but it's awful for basic tasks. For starters: it completely excludes anyone with ANY accessibility concerns. Secondly, flash is proprietary software meaning that if for some reason you can't get a player from Adobe, you can't view the page.

Alaternatively a standards based website can have the text scaled, have the colors changed if need be, and information can be ported to other devices (such as a screen reader) so that the same site can be used by anyone. Also the only requirements for access are to have a web browser and an internet connection.

For many the solution here is to create a second standards based version of the site so that those who for whatever reason can't/won't visit the flash site can still access the site. This is a good first step. But often these non-flash sites get neglected in terms of design. Here you're creating a seperate but equal senario. (And we all know how that winds up)

The better solution is to ONLY make a standards based website. "But Andrew!" You might exclaim, "if I do that I'll have to sacrafice all the nice fluid elements of my interface." And to that I would say you're mistaken. The solution here is to incorperate ECMAScript, otherwise known as JavaScript (no relation to Java). Why is this better? It's standards based, allows you to all those web interface things you might wish to do with flash, but without telling those with accessibility concerns that they belong at the back of the bus.

Tumblr

Hate Tumblr. Soooo much. Mostly their rss feed. Far as I can tell there's no way to get it to give you more than the latest 20 posts. Their api has a read function that's more usable...but it's a really ugly xml file.

Little blog recode.

I just altered that reads the blog on AndrewParnell.com

I'd noticed that the default feed from blogger (which powers this blog) only shows the 25 most recent entries, which really isn't good enough for me.

However, after some searching I found that if you append the query string ?start-index=# to the feed path, you can get older entries. Better yet, throw on &max-results=5 and rather than having to parse the entire feed and then have php display only 5 entries, I can just show the entire feed.

It's pretty sweet.

Why.

So, thanks to mem on the arduino forum, I managed to figure out the damage to the code:

Now it works!

Fixed! from Andrew Parnell on Vimeo.

What was wrong?
Short version: I was being really wasteful.

Long version: the way the code was written, each LED took a whole byte to assign. With the fixed code, each row takes a byte. (Taking it from 64 bytes per pattern, to 8. Serious savings!)

Code in the comments.

Why!!!!

Ok, this is my personal project update/post for the evening. (Also explains why I'm up at 4 am.)

I've found that adding 15 or more "characters" for the code to go through, totally breaks the matrix. 14 or less, fine. 15 or more, mess. Also 15 screws it up differently than 16+ (15 just freezes the loop, 16+ start creating garbage) What's weird is that there is nothing changed in the code except the length of the array, and the adding of a few more array entries. So I can't figure out why it's breaking like this.

Wanna see what I'm talking about? Sure.

Good

Working from Andrew Parnell on Vimeo.

Bad/Broken

Broken from Andrew Parnell on Vimeo.


And if you care, like last time, code is in the comments.

Cork Board as Protoboard

I'm prepping to do a 1024 LED sculpture, and the suggestion came up of scaling it from 2' x 2' to 4' x 4'. I was concerned that this new size without an equivalent increase in LEDs (which I really am not prepared to do) so I built a sample square foot panel to try the spacing. Of course, I don't have a square foot protoboard (does anyone?) so instead I built it on a piece of cork board. What I really liked about this method was that it allowed me to prototype it as a 2 layer board.





The larger scale actually works a lot better than I anticipated.