Blog

Almost every time I set up a Raspberry Pi these days, I use the 'Lite' version of Raspberry Pi OS. That version doesn't come with a GUI, it just boots to the console. It's much smaller in size and contains most things you'd need for a 'headless' Pi setup.

And if you know your way around the command line, it's not daunting to plug in a monitor, keyboard, and mouse, and explore via the shell if you need to.

But every so often, I've had a Lite install that I wanted to switch to GUI, but I'm too lazy to pull the Pi out of wherever it's installed, pull the microSD card, and re-flash it with the full OS, and then re-run my automation on it to set up whatever I had running before.

And that's why it's nice to be able to just install the GUI on top of an existing Lite install!

To do that (assuming you're running the latest Pi OS version, Bullseye as of this writing), just install Xorg and the Raspberry Pi 'PIXEL' environment:

In spite of Microsoft's cryptic announcement of Project Volterra, and Qualcomm's continuous lineup of 'flagship' ARM SoCs for Windows, Microsoft is still behind the 8-ball when it comes to ARM.

Apparently, in 2016, Microsoft entered into an exclusivity deal with Qualcomm. That's why all official 'Windows on ARM' devices use Qualcomm SoCs. At the time, Apple hadn't yet pulled off its third major architecture shift for macOS, from Intel X86 to ARM.

Looking back, products like the Surface Pro X and the myriad ARM for Windows laptops, were basically built to a budget and for portability above all else. They were never competitive with Intel/AMD-based computers. Microsoft seemed to think ARM would always remain in a niche, only used for light, mobility-first devices.

Like my last new shirt (It was DNS, I figured I'd mention my latest shirt design here on the blog in case you're interested:

As a parent of four, I thought it would be especially fitting on a onesie, therefore in addition to two 'normal' shirt options, there's a version for your baby:

This blog post is a lightly edited transcript from my most recent YouTube video, in which I explain some of Starlink's growing pains: slower speeds due to oversubscription, design challenges with their v2 hardware, and a major bet on much larger v2 sats and a rocket (Starship) that has yet to complete an orbital flight.

The video is embedded below, and the transcript follows:

I got Starlink during the Public beta, a little over a year ago.

I set up Dishy on my roof, I set up some advanced monitoring and tested it as a backup Internet connection, but ultimately passed it along to my cousin, who's using it on her farm.

I edit videos non-stop nowadays. In a former life, I had a 2 TB backup volume and that stored my entire digital life—all my photos, family video clips, and every bit of code and text I'd ever written.

Video is a different beast, entirely.

Every minute of 4K ProRes LT footage (which is a very lightweight format, compared to RAW) is 3 GB of space. A typical video I produce has between 30-60 minutes of raw footage (which brings the total project size up to around 100-200 GB).

I've been building a Mini ITX 'quiet-ish' server using a Supermicro motherboard and some Noctua fans.

I noticed sometimes the system would start 'revving' the fans up to max power. Then after a few seconds they would get quiet again. The CPU temps and other temps on the system were stable and not worrying, but popping off the server's cover, I noticed LED8 on the motherboard would blink red every time the fans would ramp up:

That LED indicates a 'fan failure' when blinking.