I completed building the Pip-Boy Glove a while ago now so I’m writing this as a bit of a retrospective, but I figured rather than write about how I eventually built the thing, there would be more value going over the path that got me to the final prototype.

For some reason I’ve always held myself to a high standard and with that comes the expectation that I should be able to do things perfectly first time, but that isn’t reality and actually it’s pretty normal to make a bunch of mistakes along the way, so I’d rather talk about that and normalise the process.

At the end of Part 2 I had put out call for help for anyone with a working device that was capable of pulling a full disk copy to see if that could get me out of a hole, and much to my surprise, someone did! So shout out to Jim D who went out of his way to buy the required hardware, get inside this thing and take a full disk image, unbelievably helpful and shed new light on this whole project.

In part 1 we identified all the debug connections we could use to try and get to the bottom of why this thing isn’t working, so in this part we’re going to make use of those to dive a little bit deeper. To make things easier I removed everything from the board, mounted it to my test stand and connected up each of the debug connections, UART, JTAG and I also made an adaptor cable to extend the USB Mini-B out to a regular USB Type A socket. The UART console will give us an interactive serial console to the device and some useful debug messages during boot, for deeper memory analysis or debugging we can use the JTAG.

When I first picked up one of the many broken ThinkGeek Pip-Boy’s back in 2024, I had fairly early on made the decision that I wouldn’t bother dealing with the broken board and instead replace all of the electronics and code with more modern and custom parts. Since then the broken board has been sat on my workbench and recently caught my attention again, and I’m just not the kind of person to write something off as being mysteriously broken, so I decided it was time to do a deep dive into it and see what’s really going on.

Previously in Bringing new life to an old CRT I had managed to get a decent enough video output signal to my old Amstrad CRT by putting together a small VSync splitting board and with some hackery got a decent enough monochromatic image and was fairly happy with it. Some time has passed since then and as discussed in What is a terminal anyway? I had a bit of a change of heart with how I wanted to approach this project, and I ditched the Raspberry Pi and decided to build something with a microcontroller.