Radio Shack had a long history of buying things overseas, having their name slapped on them, and selling them in the United States. That was the case with the Tandy Pocket Computers, which were in that awkward space between calculators and full-blown computers. Like many computers of those days, if you wanted to do anything interesting, you needed to turn to assembly language. But as [Old Vintage Computing Research] recalls, the assembly for these little devices was very strange, even for an assembly language. He found out that there is a reason it is so strange and shares it in a deep dive into the device’s machine code history.
The story starts with the Japanese government. In 1969, the ministry in charge of such things decided that it wouldn’t be fair for people who knew a particular computer to have an advantage when taking the Information Technology Engineer exam. So, logically, they made up a fictitious instruction set and architecture for the test. Since no one used it, no one would have an unfair advantage.
However, eventually, Japanese manufacturers started making computers that used the architecture. The architecture was COMP-X, and the assembler was CAP-X. The post covers the history of machines either using the architecture or emulating it going back to the 1970s. It eventually winds up at the Sharp and Casio pocket computers that would wear Radio Shack livery in much of the world, especially the United States.
AI guardrails and safety features are as important to get right as they are difficult to implement in a way that satisfies everyone. This means safety features tend to err on the side of caution. Side effects include AI models adopting a vaguely obsequious tone, and coming off as overly priggish when they refuse reasonable requests.
Enter GOODY-2, the world’s most responsible AI model. It has next-gen ethical principles and guidelines, capable of refusing every request made of it in any context whatsoever. Its advanced reasoning allows it to construe even the most banal of queries as problematic, and dutifully refuse to answer.
As the creators of GOODY-2 point out, taking guardrails to a logical extreme is not only funny, but also acknowledges that effective guardrails are actually a pretty difficult problem to get right in a way that works for everyone.
This also means that as AI models become more advanced, so too have they become increasingly sycophantic, falling over themselves to apologize for perceived misunderstandings and twisting themselves into pretzels to align their responses with a user’s expectations. But GOODY-2 allows us all to skip to the end, and glimpse the ultimate future of erring on the side of caution.
While the proliferation of the smartphone has caused the personal music player (PMP) market to mostly evaporate, there are still those who prefer a standalone device for their music. The Melodio Self-Mate is one such spiritual successor to the iPod.
Music-only devices really benefit from the wheel interface pioneered by Apple, so we still see it in many of the new Open Source PMPs including this one and the Tangara. The Melodio uses the ubiquitous ESP32 for its brains coupled with a TI PCM5102A DAC and TI TPA6130A2 headphone amp for audio. A slider on the side of the device allows you to switch it between mass storage mode and programming mode for the ESP32.
Since this device packs a little more horsepower and connectivity than the original iPods, things like listening to Spotify are doable once assembled, instead of having to completely rebuild the device. Speaking of building, there are only renders on the GitHub, so we’re not sure if this project has made the jump IRL yet. With more people concerned about the distractions of smartphones, maybe this renaissance of open PMPs will lead to a new golden age of music on the go?
Street sledding, a popular pastime in Norway, is an activity that is slowly dwindling in popularity, at least as far as [Justin] aka [Garage Avenger] has noticed. It used to be a fun way of getting around frozen lakes and roads during winter, and while some still have their sleds [Justin] wanted to see if there was a way to revitalize one of these sleds for the modern era. He’s equipped this one with powerful electric turbines than can quickly push the sled and a few passengers around the ice.
Since this particular sled is sized for child-sized passengers, fuel-burning jet engines have been omitted and replaced with electric motors that can spin their turbine blades at an impressive 80,000 rpm. The antique sled first needed to be refurbished, including removing the rust from the runners and reconditioning the wood. With a sturdy base ready to go, the sled gets a set of 3D printed cowlings for the turbines, a thumb throttle on the upgraded handlebars, and a big battery with an Arduino to bring it all together.
With everything assembled and a sheet of ice to try it out on, the powerful sled easily gets its passengers up to the 20-30 kph range depending on passenger weight and size. There’s a brake built on an old ice skate for emergency stops, and the sled was a huge hit for everyone at the skating pond. There are plenty of other ways to spruce up old sleds, too, like this one which adds a suspension for rocketing down unplowed roads.
The recently published book Understanding Deep Learning by [Simon J. D. Prince] is notable not only for focusing primarily on the concepts behind Deep Learning — which should make it highly accessible to most — but also in that it can be either purchased as a hardcover from MIT Press or downloaded for free from the Understanding Deep Learning website. If you intend to use it for coursework, a separate instructor answer booklet and other resources can be purchased, but student resources like Python notebooks are also freely available. In the book’s preface, the author invites readers to send feedback whenever they find an issue.
Ethernet is ubiquitous, fast, and simple. You only need two diffpairs (four wires) to establish a 100Mbit link, the hardware is everywhere, you can do Ethernet over long distances easily, and tons of the microcontrollers and SoCs support it, too. Overall, it’s a technology you will be glad to know about, and there’s hundreds of scenarios where you could use it.
If you need to establish a high-bandwidth connection between two Linux boards in your project, or maybe a Linux board and a powerful MCU, maybe make a network between microcontrollers, Ethernet’s your friend. It also scales wonderfully – there’s so much tech around Ethernet, that finding cables, connectors or ICs tends to be dead easy. Plus, the world of Ethernet is huge beyond belief. Ethernet as most of us know it is actually just the consumer-facing versions of Ethernet, and there’s a quite a few fascinating industrial and automotive Ethernet standards thatflipmany of our Ethernet assumptions upside down.
Now, you might be missing out on some benefits of Ethernet, or perhaps misunderstanding how Ethernet works at all. What does it mean when a microcontroller datasheet says “has Ethernet interface”? If you see five pins on an SBC and the manufacturer refers to them as “Ethernet”, what do you even do with them? Why does the Raspberry Pi 4 SoC support Ethernet but still requires an extra chip, and what even is GMII? Continue reading “Ethernet For Hackers: The Very Basics”→
The brain is probably the least explored organ, much of which is due to the difficulty of studying it in situ rather than in slices under a microscope. Even growing small organoids out of neurons provide few clues, as this is not how brain tissue is normally organized. A possible breakthrough may have been found here by a group of researchers whose article in Cell Stem Cell details how they created functional human neural tissues using a commercial 3D bioprinter.
As detailed by [Yuanwei Yan] and colleagues in their research article, the issue with previous approaches was that although these would print layers of neurons, they would fail to integrate as in the brain. In the brain’s tissues, we see a wide variety of neurons and supportive cells, all of which integrate in a specific way to form functioning neuron-to-neuron and neuron-to-glial connections with expected neural activity. The accomplishment of this research team is 3D bioprinting of neural tissues with the necessary functional connections.