Economy Size Geek - A Pico-Sized Platform with Potential
The important thing to understand about the SD card you're using is that it needs to be partitioned and formatted in a very specific way. Using fdisk or gparted, remove all existing partitions on the card. Then, create two partitions. The first one needs to be 50MB and formatted to fat32. The second partition can use the rest of the space on the card and should be formatted to ext3.
For the OS installs, you will be putting a uImage file on the fat32 partition. The rest of your files will go on the Linux ext3 partition. As you go through the process of building your install image, make sure you keep your uImage and Linux root filesystem in sync with the same kernel versions.
The fine people at Ångström have worked hard on a distribution that is easy and stable to run on embedded devices. This is a good place to start if you just want to boot your Beagle Board into a Linux environment quickly.
I even found a handy tool to build your own custom Ångström image at amethyst.openembedded.net/~koen/narcissus. This is still experimental, but it was nice to be able to choose what I wanted in my image quickly.
If you want something even easier, Ångström also offers a tarball containing a working system that demonstrates the capabilities of the Beagle Board. Simply download the latest tarball from www.angstrom-distribution.org/demo/beagleboard, and get the uImage that has the same timestamp. Untar the tarball directly onto the ext3 partition of your SD card, and put the uImage (make sure you rename it to uImage) on the fat32 partition.
The nice thing about Ångström is how easy it is to get up and running. The downside is I am not very familiar with that distribution as an environment. I ran into some problems configuring the Wi-Fi for the board, so I decided to switch to Ubuntu. Installing Ubuntu is more complicated upfront, but once installed, it was a lot easier for me to work with.
The other nice thing about Ubuntu is that Canonical announced late last year that it would officially support the ARM architecture (www.ubuntu.com/news/arm-linux). This means Canonical would be working on making sure that the packages I know and love on my x86 are all available on the ARM. Because the OMAP processor in the Beagle Board is an ARM chip, it will benefit from all this support.
A step-by-step guide for installing Ubuntu can be found at elinux.org/BeagleBoardUbuntu. By the time this article is published, Karmic Koala should be officially released, so instead of giving you out-of-date instructions, I recommend you follow the link to get the latest information.
I actually installed Jaunty, as that was the stable version of Ubuntu at the time of this writing. Once it was up and running, any software I needed was simply an apt-get away (in most cases). In this case, that meant adding wicd (light network manager), Firefox and VLC.
It would be easy enough to use the Beagle Board as a simple workstation. You can browse, look at documents, listen to music and even watch video. Those tasks demonstrate the board's flexibility combined with the wealth of open-source software. On one hand, it is really amazing to have so much processing power packed into such a tiny form factor. On the other hand, turning the Beagle Board into a desktop seems like a waste of that form factor (unless you mount it inside a keyboard).
My initial goal was to use it as a set-top box. I was going to use it to stream Hulu to my living-room television. I did some research and found that you can build your own IR USB receiver (USB-IR-Boy). That would make it easy for me to interface a standard remote to the Beagle Board. Before I started ordering parts, I decided it would be better to see how good the streaming capabilities were. This walked me straight into a problem with no clear workaround. In order to stream Hulu, your browser must support Flash. At this time, the only Flash available for the ARM processor family is Flash Lite. Flash Lite seems to be licensed in a way that makes it impossible to distribute to a single end user. It is targeted more at handset vendors who will be shipping millions of units. I installed Gnash just to see if it was compatible, and it didn't work either. But, Adobe has announced that it is working on Flash 10 for ARM-powered devices (www.adobe.com/aboutadobe/pressroom/pressreleases/200811/111708ARMAdobeFlash.html).
So all is not lost. On the open standards front, I can hope that the progress being made on HTML 5 and video will remove the need for a proprietary solution to see video on the Web.
My first plan dashed, I realized there was another project that I've been meaning to play with—namely building a MAME cabinet. In honor of the Beagle Board's tiny size, I started looking at building a Mini-MAME cabinet. This basically is a smaller version of an arcade cabinet, and it has some advantages—namely that it will not take up too much space, and it will be usable by my one-year-old son sooner. Once again, before I ordered any parts, I installed all the MAME packages just to test things out. Bam! (That's the sound of me hitting another wall.) In this case, the MAME packages have been compiled and packaged, but they are not set correctly to run on an ARM processor. I made some attempts at hacking the Makefile to fix the problem, but I was unsuccessful. There are videos of the Beagle Board running MAME, but I was not able to duplicate their success.
|Non-Linux FOSS: libnotify, OS X Style||Jun 18, 2013|
|Containers—Not Virtual Machines—Are the Future Cloud||Jun 17, 2013|
|Lock-Free Multi-Producer Multi-Consumer Queue on Ring Buffer||Jun 12, 2013|
|Weechat, Irssi's Little Brother||Jun 11, 2013|
|One Tail Just Isn't Enough||Jun 07, 2013|
|Introduction to MapReduce with Hadoop on Linux||Jun 05, 2013|
- Containers—Not Virtual Machines—Are the Future Cloud
- Non-Linux FOSS: libnotify, OS X Style
- Linux Systems Administrator
- Lock-Free Multi-Producer Multi-Consumer Queue on Ring Buffer
- Validate an E-Mail Address with PHP, the Right Way
- Technical Support Rep
- Senior Perl Developer
- UX Designer
- Introduction to MapReduce with Hadoop on Linux