Kill -9

Fuzebox kit procured

killminus9 — Sat, 02 May 2009 03:33:39 +0000

I’ve procured a kit for a Fuzebox, from http://www.ladyada.net/make/fuzebox/

It’s an 8-bit, open source video game console. Mono sound output. RCA composite output at 240 x 224 pixels, with a total of 256 unique colors

The really interesting part is the microcontroller. It’s not even a microcontroller, it’s simply a tiny computer on a single chip. It’s an Atmela Atmega644 chip. It has 64k of flash rom for instructions storage, and 4k of SRAM for a run time memory space. It gets up to 20 MIPS, and there are tools to let you program it in C. It’s running the Atmel AVR Core, whose machine language has only 131 verbs, and its registers only 8 bits of width. So it’s a very intelligible system. Learning how to use one of these chips would open up many doors that remain closed to the BS2…with only 2K of EEPROM that is shared between both instruction storage and runtime memory.

Here’s the project board…time to get started soldering

Under the microscope

Remote control

killminus9 — Wed, 22 Apr 2009 07:58:14 +0000

pk1 with his new master, the BB 8700c

So, I have an old blackberry 8700c. I pulled it out of the closet and dusted it off. I read up on it a little bit. Apparently it has 64mb of Flash memory, a 300 mhz ARM (RISC) processor. Hell, this little phone is a faster machine than half the computers I’ve owned in my life. The thing that really interested me about it is that Blackberries run J2ME, and they have Bluetooth capabilities, internet access, and a few other cool features.

So, I thought I should try to use Bluetooth to enable my Blackberry to be a remote control for Pk1. and after about 3 days of work, I’ve got it up and running. It’s J2ME utilizing RIM’s API for bluetooth access…the bluetooth connection goes straight to the EB500 Embedded Blue module on Pk1, and…Pk1 just waits for commands from the phone.

Future interesting applications would be to attach the Blackberry to the base of Pk1, so that Pk1 can offload data storage and processing to its code written in J2ME. This expands upon the ideas I have below, except this would allow it to be more portable and self contained. This way, after writing a better brain for Pk1, I wouldn’t have to follow him around with my laptop, I could just DL the code to the BB, and attach it to Pk1, and he’d be free to roam….and kill small rodents for energy.

New brain implementation

killminus9 — Wed, 14 Jan 2009 23:37:15 +0000

I tried implementing a robot brain in Java, but I couldn’t get that working due to problems using the Java Native Interface to play nicely with the native serial port communication code. I tried C#, which worked…marginally. After using MyEclipse, I really don’t like Visual Studio as an IDE.

So I’m back where I started…using Python 2.5 to write the Brain for PK-1. The reason for this is because I wanted to take advantage of Python’s awesome 3D library VPython. So I spent a little more time on it, and was able to get the two way RS232 conversation happening between a Python program and PK-1′s RS232-emulating Bluetooth tranceiver.

So the code is being written now… PK-1 will turn on, explore his environment, and the Python Brain will construct a 3D model of the world as PK-1 sees it. It should be pretty cool. I’ll post screenshots when it’s ready for testing.

killminus9 — Thu, 08 Jan 2009 03:13:52 +0000

So, I’m building an army of killer robots. So far, I’ve developed Prototype K-1. He’s a Parallax BoeBot with the Ping sensor and a couple IR sensors.

Here’s PK1 looking sharp.

: Prototype K1 with new sonar sensor

I set up another BS2 stamp microcontroller to try to use PK1′s Ping sensor to do some sonography…sonar imaging…scanning of just any room I was in, or of any objects…but the resolution was so low that the image was unrecognizable.

Now I’ve acquired a Compass sensor and a Bluetooth transceiver. The compass sensor is going to be used to assist with locomotion and navigation…so that it will be easier to execute commands such as “turn 30 degrees left” or “turn 90 degrees right”, and to be able to execute them accurately.

The purpose of the Bluetooth transceiver is to allow Prototype K-1 to wirelessly communicate with a laptop or any other computer device running a brain program in a higher level language than what PK1 runs on his BS2 chip. My plan is to write a program using C# .NET that will collect sensory data from PK1′s components and organize the data into a higher level schema, which can be used to make more intelligent, higher level decisions. It will be much easier to work in C# to create a system of higher complexity, and leave the programming on the BS2 chip (which is done in a language called PBasic) to only concern itself with manipulating hardware and communicating with the brain.

I’ve got the basic “hello world” functionality for the bluetooth tranceiver, meaning that the C# program and microcontroller can enjoy a two way conversation. the bluetooth tranceiver emulates standard RS232 serial port communication.

so now that I’ve got all the hardware and hello world up and running…I just have to program it to actually….”do” something….

killminus9 — Sat, 19 Apr 2008 09:01:33 +0000

penaldecree: I know what you mean. most people haven”t stuck around for long. the few that I have known for like half my life, I’ve never even met irl.
Fluid06: it’s really a tragedy
Fluid06: most people can go visit their childhood playgrounds even after they are adults
Fluid06: every now again for a little nostalgia
Fluid06: but when your childhood playground has since been deleted or erased
Fluid06: it’s just a piece of you that’s gone
penaldecree: you’re so emotional brandon

lucid virtual

killminus9 — Wed, 09 Apr 2008 05:36:36 +0000

were Nietzsche alive today, and had he a blog, it would probably contain the post:

“a large part of the discontent and anxiety we have for the new digital era is not due to the feeling we have knowing that everything being made today is now virtual, but due to the realization that all along everything always was virtual.”