Not On The Internet

Whenever you need to know something, you just look it up on the Internet, right? Using the search engine of your choice, you type in a couple keywords, hit enter, and you’re set. Any datasheet, any protocol specification, any obscure runtime error, any time. Heck, you can most often find some sample code implementing whatever it is you’re looking for. In a minute or so.

It is so truly easy to find everything technical that I take it entirely for granted. In fact, I had entirely forgotten that we live in a hacker’s utopia until a couple nights ago, when it happened again: I wanted to find something that isn’t on the Internet. Now, to be fair, it’s probably out there and I just need to dig a little deeper, but the shock of not instantly finding the answer to a random esoteric question reminded me how lucky we actually are 99.99% of the time when we do find the answer straight away.

So great job, global hive-mind of über-nerds! This was one of the founding dreams of the Internet, that all information would be available to everyone anywhere, and it’s essentially working. Never mind that we can stream movies or have telcos with people on the other side of the globe – when I want a Python library for decoding Kansas City Standard audio data, it’s at my fingertips. Detailed SCSI specifications? Check.

But what was my search, you ask? Kristina and I were talking about Teddy Ruxpin, and I thought that the specification for the servo track on the tape would certainly have been reverse engineered and well documented. And I’m still sure it is – I was just shocked that I couldn’t instantly find it. The last time this happened to me, it was the datasheet for the chips that make up a Speak & Spell, and it turned out that I just needed to dig a lot harder. So I haven’t given up hope yet.

And deep down, I’m a little bit happy that I found a hole in the Internet. It gives Kristina and me an excuse to reverse engineer the format ourselves. Sometimes ignorance is bliss. But for the rest of those times, when I really want the answer to a niche tech question, thanks everyone!

Dual Power Supply In A Pinch

Recently I needed a dual voltage power supply to test a newly-arrived PCB, but my usual beast of a lab power supply was temporarily at a client’s site. I had a FNIRSI programmable power supply which would have been perfect, but alas, I had only one. While digging around in my junk box I found several USB-C power-delivery “trigger” boards which I bought for an upcoming project. These seemed almost too small for the task at hand, but after a little research I realized they would work quite well.

The ones I had used the Injoinic IP2721 USB-C power delivery chip, commonly used in many of these boards. Mine had been sold pre-configured for certain output voltages, but they were easy to re-jumper to the voltages I needed, +5 VDC  and +20 VDC. The most challenging aspect was physically using them — they are the size of a fingernail. This version had through-hole output pads on 0.1″ centers, so I decided to solder them to the base of a standard MTA pin header. A few crimps later and I was up and running, along with the requisite pair of USB-C cables and power adapters.

For just a few dollars each, these trigger boards are useful to have in your toolbox, both for individual projects and for use in a pinch. We reviewed these modules a couple of years ago, and check out the far more flexible PD Micro that we covered last year.

Singing Fish Nails Sea Shanty Audition

The Big Mouth Billy Bass and other singing fish were a scourge first delivered to us in the late 90s. [Kevin Heckart] has been teaching them to sing new songs without the tinny sound quality and hokey folk tunes. For this, he must be applauded.

A Teensy 4.1 or Teensy 3.2 is used to power [Kevin]’s various singing fish builds. There are two motors inside a singing fish, typically — one motor to pivot the fish’s body, and one to open and close the mouth. Hook these up to a motor driver, and command that with the Teensy, and you’re up and running. To sync the fish with the music, MIDI data is sent to the Teensy over USB. The Teensy takes in note data and uses this to command the motors to make the fish appear to sing along.

The tutorial linked above is a great way to learn how the hack was achieved. However, the real money is in the performance. A video of [Kevin]’s fishy chorus performing the famous Wellerman sea shanty has over 50 million views on YouTube and he’s collected over 26 million likes on Tiktok.

Sometimes the simple hacks are the ones that bring the most joy. Video after the break.

Continue reading “Singing Fish Nails Sea Shanty Audition”

Kotonki: Agricultural Vehicle Built For Customization

Agriculture on any scale involves many tasks that require lifting, hauling, pushing, and pulling. On many modern farms, these tasks are often done using an array of specialized (and expensive) equipment. This puts many small-scale farmers, especially those in developing countries, under significant financial pressure. These challenges led a South African engineering firm to develop the Kotonki, a low-cost hydraulically powered utility vehicle that can be customized for a wide variety of use cases. Video after the break.

The name Kotonki is derived from the Setswana phrase for a donkey kart. It is in essence a self-propelled hydraulic power pack, capable of hauling 1 ton of anything that can fit on its load bed. It comes in front-wheel drive or four-wheel drive versions, with each wheel individually driven by a hydraulic motor. The simple welded steel frame articulates around a double pivot, which allows it to keep all 4 wheels on the ground over any terrain. At a max speed of 10 km/h it won’t win any races, but neither would most other agricultural vehicles. The Kotonki is built mostly using off-the-shelf components and is powered by a common 12HP Honda engine. In the world of DRM agricultural equipment, this makes for simple repairs, low running costs, and easy customization for the task at hand. This can include mounting log splitters, water pumps, lifting beds, or anything else that can be driven by its hydraulic and rotary PTOs (Power Take-Off).

Continue reading “Kotonki: Agricultural Vehicle Built For Customization”

Computer Vision Extracts Lightning From Footage

Lightning is one of the more mysterious and fascinating phenomenon on the planet. Extremely powerful, but each strike on average only has enough energy to power an incandescent bulb for an hour. The exact mechanism that starts a lightning strike is still not well understood. Yet it happens 45 times per second somewhere on the planet. While we may not gain a deeper scientific appreciation of lightning anytime soon, but we can capture it in various photography thanks to this project which leverages computer vision machine learning to pull out the best frames of lightning.

The project’s creator, [Liam], built this as a tool for stormchasers and photographers so that they can film large amounts of time and not have to go back through their footage manually to pull out the frames with lightning strikes. The project borrows from a similar project, but this one adds Python 3 capabilities and runs on a tiny netbook for more easy field deployment. It uses OpenCV for object recognition, using video files as the source data, and features different modes to recognize different types of lightning.

The software is free and open source, and releases are supported for both Windows and Linux. So far, [Liam] has been able to capture all kinds of electrical atmospheric phenomenon with it including lightning, red sprites, and elves. We don’t see too many projects involving lightning around here, partly because humans can only generate a fraction of the voltage potential needed for the average lightning strike.

A 3D-printed device labelled "BlixTerm" plugged into the back of a Commodore PET

BlixTerm Brings Full-Speed YouTube Video To The Commodore PET

If you’ve ever used a home computer from the late 1970s or early 1980s, you’ll no doubt be familiar with the slow speed of their user interfaces. Even listing the contents of a BASIC program from RAM could take several seconds, with the screen updating one line at a time. Video games were completely optimized for speed, but could still handle just a few slowly-moving objects at the same time. Clearly, playing anything resembling full-motion video on hardware from that era would be absolutely impossible – or so you might think.

In fact, [Thorbjörn Jemander] has managed to persuade a Commodore PET to play YouTube videos at a completely reasonable 30 frames per second. He describes the process of designing the “BlixTerm” hardware and software in his video (embedded below), along with lots of useful information on how to push digital systems to their absolute limits.

A video of a drifting car, as rendered by a Commodore PET displayNaturally, the PET needs a bit of assistance from modern hardware, in this case a Raspberry Pi Zero 2 W hooked up to the “User” expansion port. The Pi connects to YouTube through WiFi and loads the requested video, then downconverts it to a 640×200 grayscale stream and transforms each frame to an 80×25 grid of characters, using those from the PET’s ROM that most closely resemble the pattern needed.

While it took quite some effort to squeeze enough performance out of the Pi to do all of this in real time, the trickiest bit was getting the resulting character stream into the PET’s video memory fast enough. To do this, [Thorbjörn] designed a special interface card with 2 KB of dual-port SRAM, which enabled the Pi to store its video frames as soon as they were ready on one side, and the PET to load them at its own pace from the other side. With just sixteen microseconds available to process each byte, the PET’s CPU can execute only four or five machine code instructions; barely enough to load and store a single character and jump to the next memory address.

The end result, as you can see in the video, is really impressive. Even within the constraints of the Commodore character set, the resulting image is clearly recognizable, while the frame rate seems to defy the hardware’s limitations.

If you’re a Commodore aficionado and wondering what the hell that weird PET 600 model is all about, [Thorbjörn] made a video about that too; it’s a rebadged 8296 aimed at the Swedish market. We’ve actually seen a project to generate live video on the PET before, although at a much lower frame rate. Thanks for the tip, [Keith Olson]!

Continue reading “BlixTerm Brings Full-Speed YouTube Video To The Commodore PET”

Converting A Sigma Lens To Canon, Digital Functionality Included

These days, camera lenses aren’t just simple bits of glass in sliding metal or plastic housings. They’ve often got a whole bunch of electronics built in as well. [Dan K] had just such a lens from Sigma, but wanted to get it working fully with a camera using the Canon EF lens fitting. Hacking ensued.

The lens in question was a Sigma 15-30mm f/3.5-4.5 EX DG, built to work with a Sigma camera using the SA mount. As it turns out, the SA mount is actually based on the Canon EF mount, using the same communications methods and having a similar contact block. However, it uses a mechanically different mounting bayonet, making the two incompatible.

[Dan] sourced a damaged EF lens to provide its mount, and modified it on a lathe to suit the Sigma lens. A short length of ribbon cable was then used to connect the lens’s PCB to the EF mount’s contacts. When carefully put back together, the lens worked perfectly, with functional auto-focus and all.

It goes to show that a little research can reveal possibilities for hacking that we might otherwise have missed. [Dan] was able to get his lens up and running on a new camera, and has taken many wonderful pictures with it since.

We’ve seen some great lens hacks over the years, from 3D printed adapters to anamorphic adapters that create beautiful results. If you’ve got your own mad camera hacks brewing up, drop us a line!