Judging by the over-representation of the term “AI” in our news feeds these days, we’re clearly in the exponential phase of the artificial intelligence hype cycle, and very nearly at the dreaded “Peak of Inflated Expectations.” It seems like there’s nothing that AI can’t do, and nowhere that its principles can’t be applied to virtuous — and profitable — effect.
We don’t deny that AI has massive potential, but we strongly suspect that there will soon come a day when eyes will roll and stomachs will turn at yet another AI application that could have been addressed with something easier. An example of the simpler approach can be seen in this non-AI wildlife photo trap, cobbled together by [Sebastian] to capture pictures of some camera-shy squirrels. Rather than train an AI with gigabytes of squirrel images, he instead relies on his old Sony Alpha camera, which has a built-in WiFi. A Python script connects to the camera, which is trained on a feeder box and set to a very narrow depth of field. That makes a good percentage of the scene out of focus until a squirrel or other animal comes along looking for treats. The script detects the increased area of the scene that is now in-focus with a Laplace operator in OpenCV, and triggers the camera shutter. [Sebastian] ended up with some wonderful shots of the shy squirrels using this scheme; the video below describes the setup in more detail.
A dingy and cold early February in a small British town during a pandemic lockdown is not the nicest time and place to take your exercise, but for me it has revived a forgotten memory and an interesting tale of a technology that promised a lot but delivered little. Walking through an early-1990s housing development that sprawled across the side of a hill, I noticed a couple of houses with odd antennas. Alongside the usual UHF Yagis for TV reception were small encapsulated microwave arrays about the size of a biscuit tin. Any unusual antenna piques my interest but in this case, though they are certainly unusual, I knew immediately what they were. What’s more, a much younger me really wanted one, and only didn’t sign up because their service wasn’t available where I lived.
All The Promise…
Ionica was a product of Cambridge University’s enterprise incubator, formed at the start of the 1990s with the aim of being the first to provide an effective alternative to the monopolistic British Telecom in the local loop. Which is to say that in the UK at the time the only way to get a home telephone line was to go through BT because they owned all the telephone wires, and it was Ionica’s plan to change all that by supplying home telephone services via microwave links.
Their offering would be cheaper than BT’s at the socket because no cable infrastructure would be required, and they would aim to beat the monopoly on call costs too. For a few years in the mid 1990s they were the darling of the UK tech investment world, with a cutting edge prestige office building just outside Cambridge, and TV adverts to garner interest in their product. The service launched in a few British towns and cities, and then almost overnight they found themselves in financial trouble and were gone. After their demise at the end of 1998 the service was continued for a short while, but by the end of the decade it was all over. Just what exactly happened?
Those of us who work on the road have a constant dread of being stuck somewhere without power, facing a race between a publication deadline and a fast-failing laptop battery. We’re extremely fortunate then to live in an age in which a cheap, lightweight, and efficient solid-state switch-mode inverter can give us mains power from a car cigarette lighter socket and save the day. Before these inverters came much heavier devices whose transistors switched at the 50Hz line speed, and before them came electromechanical devices such as the rotary converter or the vibrating reed inverter. It’s this last type that [Robert Murray-Smith] has taken a look at, making what he positions as the simplest inverter that it’s possible.
If you’ve ever played with relays, you’ll probably be aware that a relay can be wired as a buzzer, and it’s this property that a vibrating reed inverter harnesses. He takes an octal relay and wires it up with a small mains transformer for an immediate and very cheap inverter. It’s not perfect, as he points out the frequency isn’t right. The relay will eventually wear out unless the arcing problem is improved with the addition of a capacitor. But it does make a rough and ready inverter if you find yourself in a MacGyver-style tight spot with only your junk box for salvation.
Since the introduction of the Raspberry Pi Compute Module 4, power users have wanted to use NVMe drives with the diminutive ARM board. While it was always possible to get one plugged in through an adapter on the IO Board, it was a bit too awkward for serious use. But as [Jeff Geerling] recently discussed on his blog, we’re not only starting to see CM4 carrier boards with full-size M.2 slots onboard, but the Raspberry Pi Foundation has unveiled beta support for booting from these speedy storage devices.
The MirkoPC board that [Jeff] looks at is certainly impressive on its own. Even if you don’t feel like jumping through the hoops necessary to actually boot to NVMe, the fact that you can simply plug in a standard drive and use it for mass storage is a big advantage. But the board also breaks out pretty much any I/O you could possibly want from the CM4, and even includes some of its own niceties like an RTC module and I2S DAC with a high-quality headphone amplifier.
Once the NVMe drive is safely nestled into position and you’ve updated to the beta bootloader, you can say goodbye to SD cards. But don’t get too excited just yet. Somewhat surprisingly, [Jeff] finds that booting from the NVMe drive is no faster than the SD card. That said, actually loading programs and other day-to-day tasks are far snappier once the system gets up and running. Perhaps the boot time can be improved with future tweaks, but honestly, the ~7 seconds it currently takes to start up the CM4 hardly seems excessive.