Awakening A Dragon From Its Slumber

January 12, 2020 by Jenny List 18 Comments

For all the retrocomputing fun and games we encounter in our community, there are a few classic microcomputers that rarely receive any attention. Usually this is because they didn’t sell well and not many have survived, or were simply underwhelming machines that haven’t gathered a huge following today. One that arguably falls within both camps is the Dragon 32, a machine best known in those pre-Raspberry Pi days for being the only home computer manufactured in Wales, and for being nearly compatible with the Tandy Color Computer due to both machines’ designs coming from the same Motorola data sheet. Repeat restorer of retrocomputers, [Drygol], has given a Dragon 32 the full restoration and upgrade treatment, offering us a rare chance to take a look at this computer.

The Dragon arrived with a pile of contemporary books and software, but no power supply. A significant modification was made to the internal PSU board then to allow it to work with an Amiga unit, and the black-on-green Dragon text came up on the TV screen. Recapping and a replacement for a faulty op-amp fixed poor video quality, then it was time for a 64K memory upgrade with some neatly done bodge-wiring. Finally there’s a repair to the very period-looking analogue joystick, and a home-made interface for the more common Atari/Amiga style sticks.

The Dragon may be only a footnote in the history of 8-bit home computing, but with its good expandability and decent quality keyboard it perhaps deserved to reach more homes than it did. This appears to be the first time a Dragon has featured here, though its Tandy CoCo cousin has made it into a few stories.

Maze Solving Via Text Editing

January 12, 2020 by Al Williams 18 Comments

Linux scripters usually know about sed — the stream editor. It has a simple job: transform text as it whizzes from input to output. So if you wanted to solve a maze, this wouldn’t be the tool you’d think to use, right? Well, if you were [xsot], you’d disagree.

You build a maze using spaces for empty space and # for walls. There’s an S to mark the start position and an E to mark the end. Of course, the maze can also contain newlines. The sed script does an amazing job of solving the problem.

Continue reading “Maze Solving Via Text Editing” →

Generating Beetles From Public Domain Images

January 12, 2020 by Sharon Lin 11 Comments

Ever since [Ian Goodfellow] and his colleagues invented the generative adversarial network (GAN) in 2014, hundreds of projects, from style transfers to poetry generators, have been produced using the concept of contesting neural networks. Unlike traditional neural networks, GANs can generate new data that fits statistically within the same set as the training set.

[Bernat Cuni], the one-man design team behind [cunicode] came up with the idea to generate beetles using this technique. Inspired by material published on Machine Learning for Artists, he decided to deploy some visual experiments with zoological illustrations. The training data was found from a public domain book hosted at archive.org, found through the Biodiversity Heritage Library. A combination of OpenCV and ImageMagick helped with individually extracting illustrations to squared images.

[Cuni] then ran a DCGAN with the data set, generating the first set of quasi-beetles after some tinkering with epochs and settings. After the failed first experiment, he went with StyleGAN, setting up a machine at PaperSpace with 1 GPU and running the training for >3 days on 128 px images. The results were much better, but fairly small and the cost of running the machine was quite expensive (>€125).

Given the success of the previous experiment, he decided to transfer over to Google CoLab, using their 12 hours of K80 GPU per run for free to generate some more beetles. With the intent on producing more HD beetles, he used Runway trained on 1024 px beetles, discovering much better results after 3000 steps. The model was moved over to Google CoLab to produce HD outputs.

He has since continued to experiment with the beetles, producing some confusing generated images and fun collectibles.

Continue reading “Generating Beetles From Public Domain Images” →

A Supercapacitor Might Just Light Your Way One Day

January 12, 2020 by Jenny List 32 Comments

Sometimes the simplest hacks are the most useful ones, and they don’t come much simpler than the little supercapacitor LED flashlight from serial maker of cool stuff [Jeremy S. Cook]. Little more than an LED, a supercapacitor, USB plug, and couple of resistors, it makes a neat little flashlight that charges from any USB A power socket and delivers usable light for over half an hour.

It’s neat, but on its own there’s not much to detain the reader until it is revealed as a “Hello World” supercapacitor project from an article in which he delves into the possibilities of these still rather exotic components. Its point is to explore their different properties when compared to a battery, for example a linear voltage drop in contrast to the sharp drop-off of a chemical cell. In the video below the break we see him try a little boost regulator to deliver a constant voltage, with consequent severe loss of lighting time for the LED. It’s by this type of experimentation that we learn our way around a component unfamiliar to us, and the article and video are certainly worth a look if you’ve never used a supercapacitor before.

Continue reading “A Supercapacitor Might Just Light Your Way One Day” →

NanoVNA Tests Antenna Pattern

January 11, 2020 by Al Williams 6 Comments

When [Jephthai] wanted to build his own Yagi antenna, he turned to MMANA software for antenna modeling. This is an antenna analysis program that uses the moment method to calculate parameters for different antenna geometries. After building the Yagi, the predicted tuning and impedance matched the real antenna nicely. But what about the radiation pattern? To test that, he used a NanoVNA and a clever test setup.

He needed a test spot out of the antenna’s near field so he set up his workstation 18 feet away from the test antenna which was on a mount that could rotate. On the edge of the workstation table — affixed with painter’s tape — is a NanoVNA connected to a laptop.

Continue reading “NanoVNA Tests Antenna Pattern” →

This Artist Drags His Feet Across Sand And Snow

January 11, 2020 by Sharon Lin 10 Comments

You may have seen Simon Beck’s work a few years back. The snow artist, known for creating large-scale works of art with nothing but snowshoes, has been creating geometrically inspired fractals and mathematical forms for years. An orienteer and map-maker by day, he typically plans out his works in advance and chooses sites based on their flat terrain. The lack of slopes prevents skiers from traversing the area beforehand and helps with measuring the lines needed to create the drawing.

He starts off by measuring the distance he has to be from the center by using a compass and walking in a straight line towards a point in the distance, making curves based on relative position to other lines. Once the primary lines are made, he measures points along the way using pace counting and joins secondary lines by connecting the points. The lines are generally walked three times to solidify them before filling in the shaded areas. The results are mesmerizing.

He has since expanded to sand art, using the same techniques that gained him fame in ski resorts and national parks on the sandy shores. Unfortunately, tidal patterns, seaweed, and beach debris make it slightly harder to achieve pristine conditions, but he has managed to create some impressive works of art nonetheless.

Continue reading “This Artist Drags His Feet Across Sand And Snow” →

Plucky Kalimba Plays Itself

January 11, 2020 by Kristina Panos 23 Comments

[Gurpreet] fell in love with the peaceful, floaty theme from the Avatar series and bought a kalimba so he could hear it resonate through his fingertips. He soon realized that although it’s nice to play the kalimba, it would be a lot cooler if it played itself. Then he could relax and enjoy the music without wearing out his thumbs.

After doing a bit of experimentation with printing tine-plucking extensions for the servo horns, [Gurpreet] decided to start the design process by mounting the servos on a printed base. The servos are slotted into place by their mounting tabs and secured with hot glue. We think this was a good choice — it’s functional and it looks cool, like a heat sink.

[Gurpreet]’s future plans include more servos to pluck the rest of the tines, and figuring out how feed it MIDI and play it real time. For the demo after the break, [Gurpreet] says he lapel mic’d the kalimba from the back and cut out the servo noise with Audacity, but ultimately wants to figure out how to quiet them directly. He’s going to try lubing the gears and making a sound-dampening enclosure with foam, but if you have any other ideas, let him know down below.

We don’t see too many kalimba projects around here, but here’s one connected to a Teensy-based looper.

Continue reading “Plucky Kalimba Plays Itself” →