A trick famously used by Nintendo to keep its slowly aging SNES console fresh against newer competition was to produce new games with extra support chips in the cartridge to push out hitherto-unthinkable performance. Chips such as the famous SuperFX gave us 3D polygonal graphics, but it would have been a few more years before even much faster platforms could achieve real-time ray-tracing. Nintendo may not have managed it, but here in 2020 [Ben Carter] has a SNES on his bench rendering a complex 3D ray-traced world.
Ray tracing refers to the practice of rendering a scene with accurate lighting by tracing the rays of light that go towards making each pixel. It can achieve results that even approach photorealism, but it remains an extremely computationally intensive job for any computer. To do this with a SNES he hasn’t resorted to a modern computer like the excellent Raspberry-Pi-based NES DOOM cartridge, instead he’s tried to create something that might have graced a Nintendo custom chip back in the 1990s. The tool may be a thoroughly modern DE10-Nano FPGA dev board, but what it implements could conceivably have been made as a 1990s-spec ASIC. In it are three ray tracing cores that do the work, but the final rendering is handled by the SNES itself. At 200 x 160 pixels and 256 colours it’s no graphical powerhouse, but the maximum frame rate of 30 fps makes it no slouch for the day. The video below the break supplies extra detail.
Perhaps an unexpected takeaway of the rendered scene lies in how of its era it seems. It comes from an age in which checker-board floors, mirrored balls, and azure blue skies looked so futuristic, and just before the likes of Toy Story redefined what the general public might expect from 3D rendering. If Nintendo had produced a ray-traced SNES game using a chip like this one, it would have certainly been a defining moment for gaming in that decade.
One of the most common complaints fielded by the agents of Big Candy regards the non-homogenous nature of their products. Skittles and M&Ms are two egregious offenders in this area, and it’s left up to the determined consumer to handle sorting these candies themselves. Of course, you can always do it by hand, but as we all know – machines will do the work.
This Skittle sorter is the creation of [Lewis] of [DIY Machines], and it’s a build targeted at the beginner level. Constructed out of cardboard, it uses a pair of servos to handle the transport of the candies into their requisite colored bins, via a rotating disc and chute. Skittles are scanned with a TCS34725 color scanner hooked up to an Arduino Nano, which changes the angle of the output chute to dump the candy in the proper location. The hopper is able to handle a standard 180 gram bag of Skittles without problems.
[Lewis] does a great job explaining each stage of the build, from the mechanical and electronic side of things, to the required calibrations to make everything play together nicely. The project teaches builders a multiude of useful lessons, like how to use limit switches and other concepts of automation. We’ve featured [Lewis] on these pages before, too; his stylish shelf clock is a particular delight. Video after the break.
It sounds like the start of a joke, but what’s the difference between taking Cornell’s CS6120 online and in-person? The instructor, [Adrian Samspon] notes that the real class has deadlines, an end-of-semester project, and a discussion board that is only open to real-life students. He also notes that you only earn “imagination credits.”
Still, this is a great opportunity to essentially audit a PhD-level computer science class on a fascinating topic. The course consists of videos, papers, and open source projects using LLVM and a custom internal representation based on JSON that is made for the class. It is all open source, too. You do however need access to the papers, some of which are behind paywalls. Your local library can help if you can’t otherwise find copies of the papers.
The topics include internal representation of programs, simple optimizations, data flow, global analysis and optimization, and practical topics about LLVM. With the basics out of the way, the class turns to some classic topics like loop optimization and alias analysis, along with some more modern topics like memory management and dynamic compilers.
The final part has some more cutting-edge reading about concurrency and parallelism. We suspect you should probably be pretty fluent in programming languages before attempting this course. Of course, not everyone needs to know how to write a compiler. But if you do, this is great stuff. Even if you don’t, it is a chance to expand your understanding of how things work under the hood.
You don’t have to wire this to the ignition or anything — all you have to do is power it with the cigarette lighter (or straight-up outlet, if you’re lucky). Every time you turn the key, this pump powers up and performs a little song to remind you to use it. Electronically speaking, it couldn’t be simpler — an Arduino UNO reads your hand from the distance sensor and activates a servo that dispenses three short pumps of isopropyl alcohol. Check it out in action after the break.
If you have a pile of old VHS tapes collecting dust in your attic or basement that you know you’ll never watch again, either because all of those movies are available on DVD or a streaming service, or because you haven’t had a working VCR since 2003, there might be a way of putting them to good use in another way. With the miles of tape available in just a few cassettes, [Brother] aka [Andrew] shows us how to use that tape as filament for a 3D printer. (Video, embedded below.)
The first step of the build is to actually create the filament. He uses a purpose-built homemade press to spin several tapes into one filament similar to how cotton or flax is spun into yarn. From there the filament is simply fed into the 3D printer and put to work. The tape filament needs to be heated higher than a standard 3D printer filament so he prints at a much slower rate, but the resulting product is indistinguishable from a normal print except for the color. It has some other interesting properties as well, such as retaining its magnetism from the magnetic tape, and being a little more brittle than PET plastic although it seems to be a little stronger.
While the VHS filament might not be a replacement for all plastic 3D prints, it’s still a great use for something that would likely otherwise head straight to the landfill. There are some other uses for this magnetic tape as well, like if you wanted to build a DIY particle accelerator.
To get started, [Daniel] fitted X-acto blades to a brushless outrunner motor, and tested their ability to cut grass. Satisfied with the performance, he built a trailer to tow behind an RC tank mounted to such a setup, with some success. With the concept beginning to bear fruit, he went with a clean sheet design for maximum performance.
The final build relies on an RC rock-crawler chassis, fitted with a brushless motor using field oriented control for maximum torque at low speeds. This allows the RC mower to slowly push through the grass without overwhelming the cutter heads. As for the cutter heads, the final rig has eight motors, each sporting two blades to chop down long field grasses with impunity.
The BBC has a long history of supporting technology education in schools. The BBC Micro introduced a whole generation of students to computers, and more recently the Micro:bit is teaching today’s children about embedded systems. [Michael Klements] happens to be a grown adult, but has whipped up a project using the little board to build an automatic plant watering system.
Rather than a simple timer-based system, [Michael’s] build measures soil moisture using a capacitive sensor. This has the benefit of not needing to be in direct contact with the soil as resistive sensors do, and thus the sensor can be built in a fashion that minimises corrosion. The Micro:bit reads this sensor using an analog input, and displays the moisture level using its inbuilt LED matrix as a graph. Once levels dip below a set threshold, a pump is activated to deliver water to the plant until the soil is suitably moist again.
It’s a simple project, but one that would be a great way to teach students about interfacing with pumps and sensors, as well as the basics of control systems. [Michael] also notes that further work could involve interfacing multiple Micro:bits using their onboard wireless hardware. We’ve thus far seen the Micro:bit used for everything from handheld gaming to gumball delivery. Video after the break.
Hovercraft never really caught on as regular transportation, but they are very cool. The Saunders-Roe SR.N1 was the very first practical example of the type, and served as a research vehicle to explore the dynamics of such vehicles. [mr_fid] was looking for a lockdown project, and set about crafting a radio controlled replica of his own.
The build is crafted out of a canny combination of plywood and balsa, the latter substituted in sections within the plywood hull to save weight. A pair of brushless outrunner motors are mounted in the central duct to provide lift, fitted with counter-rotating propellers in order to avoid torque effects on handling. Steering is via puff ports a la the original design, which allows the craft to spin very quickly in place to much amusement and no practical effect. The skirt is of a colorful design, carefully assembled out of polyurethane-coated nylon.
While it’s not the quickest way to build a hovercraft, it’s all the more beautiful for its attention to the details and function of the original prototype craft. We particularly like the sharp handling thanks to the puff port design. If you’re looking for a weirder design however, consider this Coanda Effect build. Video after the break.
At some point or another, many of us have tried to see how much of our digital lives could be accessed from the comfort of a terminal. We’ve tried Alpine for email, W3M for web browsing, and even watched Star Wars via telnet. But, in the increasingly socially-distant world we find ourselves in today, we find ourselves asking: what about video calling?
As you may have guessed, {Andy]’s solution replaces the conventional video stream we’re all used to with realtime animated ASCII art. The system works by capturing a video stream from a webcam, “compressing” each pixel by converting it into an ASCII character, and stuffing the entire frame into a TCP packet. Each client is connected to a server (meeting room?) which coordinates the packets, sending them back and forth appropriately.
As impressive as it is impractical, the only area in which the project lacks is in audio. [Andy] suggests using Discord to solve that, but here’s hoping we see subtitles in version 2! Will AsciiZOOM be replacing our favorite videoconferencing suite any time soon? No. Are we glad it exists? You betcha.
Modular synthesizers, with their profusion of knobs and switches and their seemingly insatiable appetite for patch cables, are wonderful examples of over-complexity — the best kind of complexity, in our view. Play with a synthesizer long enough and you start thinking that any kind of sound is possible, limited only by your imagination in hooking up the various oscillators, filters, and envelope generators. And the aforementioned patch cables, of course, which are always in short supply.
Luckily, though, patch cables and the modules they connect can be virtualized, and in his 2020 Remoticon workshop, Jonathan Foote showed us all the ways VCV Rack can emulate modular synthesizers right on your computer’s desktop. The workshop focused on VCV Rack, where Eurorack-style synthesizer modules are graphically presented in a configurable rack and patched together just like physical synth modules would be.
John started out with a simple example using the most basic of synth modules: a voltage-controlled oscillator (VCO), a voltage-controlled filter (VCF), and an envelope generator. Along with a few housekeeping modules to drive the computer’s soundcard and to use the keyboard as, well, a keyboard, participants were able to quickly assemble their virtual synthesizers. More complex instruments can be built up from literally thousands of available modules — some free, some open source, and some premium modules available for a reasonable price.
Hats off to Jonathan for a great workshop and handling my off-the-cuff question about reproducing the “Sound of the 80s” question with aplomb. It turned out not to be possible with the stock VCV Rack modules, but it still reminded us a bit of the gated reverb drum sound of “In the Air Tonight” and other classics. Make sure you check out not only the video of the workshop but the workshop page too, which has all the details you need to get started with your own virtual synthesizer.
We as humans are limited in the ways we can look at things ourselves, and rely on on the different perspectives and insights of others to help make sense of things. All it takes is one person to look at a data set and find something completely different that changes our fundamental perception of the universe.
Cecilia Payne-Gaposchkin discovered that stars are primarily made of hydrogen and helium, at a time when astronomers thought that the Sun and the Earth had no significant elemental differences. She proposed that hydrogen wasn’t only more common, but that it was a million times more common.
This outlandish conclusion was roundly dismissed at the time, and she aquiesced to tone down some of the conclusions in her thesis, until her findings were widely confirmed a few years later. Truly groundbreaking, the discovery of the prevalence of hydrogen in stars paved the way for our current understanding of their role as the furnaces for the heavier elements that we know and love, and indeed are composed of.
Meteorites, Comets, and Bee Orchids
Cecilia Helena Payne was born May 10th, 1900 in Wendover, Buckinghamshire, England. She was one of three children born to Emma and Edward, a lawyer, historian, and musician. Her father died with she was four years old, leaving her mother to raise the family alone.
The Great Daylight Comet of 1910 as seen by Lowell Observatory. Image via Wikipedia
When she was only a few years old, Cecilia saw a meteorite blazing across the sky and was completely fascinated. She continued to show interest in science after discovering a bee orchid growing in an orchard when she was eight. This was exciting because the bee orchid is native to the Mediterranean and not to England. A few years later, she witnessed both the Great Daylight Comet of 1910 and Halley’s Comet in the same year.
Cecilia’s mother moved them to London when she was twelve for the sake of her brother Humfry’s education. Cecilia’s education focused on languages, but she was more interested in botany and other sciences. She studied botany and math on her own, and one of her teachers took her to museums and lent her books to read.
At 17, Cecilia was finally able to study science. She transferred to St. Paul’s Girls’ School and spent two years there before earning a scholarship to Newnham College at Cambridge University. She majored in botany, physics, and chemistry, but her interest shifted solidly to astronomy after she attended a lecture about solar eclipses and relativity by Sir Arthur Eddington that transformed her world picture.
Cecilia works at the observatory. Harvard College Observatory via Brain Pickings
The Stuff of Stars
Cecilia’s options in the UK were limited to teaching, so she looked for grants that would get her to the States. After graduating Newnham in 1923, she won a National Research Fellowship which she used to pursue graduate studies at Radcliffe, a women’s college associated with Harvard, which was then restricted to men.
Her academic advisor persuaded her to write a doctoral thesis even though there was no astronomy program at Radcliffe or Harvard. The common belief at the time was that the Earth and the Sun were composed of the same elements. By studying tens of thousands of spectroscopic images, Cecilia determined instead that the Sun and other stars are mostly made of hydrogen and helium. She established a stellar temperature scale and accurately related the spectral classes of stars to their temperatures by applying astronomer Meghnad Saha’s ionization theory. Her thesis was published in 1925, and she was granted the first PhD in astronomy from both Harvard University and Radcliffe College.
Several faculty members and astronomers including Henry Norris Russell found her thesis problematic because it was so unorthodox. They went as far as forcing her to add a statement saying that the presence of hydrogen in the Sun was highly unlikely. Within four years, Cecilia’s findings were independently verified by others, including Henry Norris Russell. Ironically, Russell is often credited for this discovery, though he credited Cecilia in his own paper.
After receiving her doctorate, Cecilia studied high-luminosity stars in order to better understand the structure of the Milky Way. In 1933, she met astrophysicist Sergei I. Gaposchkin in Germany. They married the following year and settled in a small town near Harvard, eventually raising three children. Cecilia had become an American citizen in 1931 and helped Sergei obtain his citizenship a few years later.
Cecilia and Sergei worked together to survey all of the stars brighter than the tenth magnitude. Together with Sergei and her assistants, Cecilia made over 3,000,000 observations of variable stars. She published several books, including The Stars of High Luminosity (1930), Variable Stars (1938) and Variable Stars and Galactic Structure (1954). Cecilia retired from teaching in 1966, but continued doing research and editing books and journals that came out of Harvard Observatory. Various people did what they could over the years to elevate her standing at Harvard. She was the first woman promoted to full professor, and the first woman to head a department. She fought for equal pay, but spent much of her teaching career earning the same wage as an entry-level male professor.
Cecilia won plenty of awards in her lifetime, and was the first recipient of the Annie Jump Cannon award in astronomy. She died December 7th, 1979 in Cambridge, Massachusetts, but will always be remembered for her huge contribution to science.
It’s hardly a secret any more at this point that today’s game consoles from Microsoft and Sony are essentially AMD gaming rigs packed up into a custom package and with tweaked system software. So it’s not too surprising that enterprising hackers got the Playstation 2 emulator of RetroArch running on an Xbox Series X|S game console despite Microsoft’s attempts to stop them. (Video, embedded below.)
It’s possible to sneak the RetroArch app past Microsoft’s security checkpoints by shelling out $19 for a Microsoft Developer Account, setting up Developer Mode on the XBox console, and getting the Universal Windows Platform (UWP) port of RetroArch from the official website. This has the advantage of it being a blessed-by-the-Redmond-gods approach. But one cannot play retail games in Developer Mode and large games due to a 2 GB limit.
More recently, a hacker by the name of [tunip3] found a flaw in the Xbox app distribution system which allows one to download a ‘retail’ version of RetroArch. This involves marking the RetroArch app as ‘private’, allowing it to skip a review by Microsoft. People whose email address is on a whitelist are then granted download permission for that app on their Xbox console. The advantage of this ‘retail’ approach is that it does not feature the 2 GB filesize limits. The disadvantage is that Microsoft is free to take the app down and ban [tunip3]’s developer account.
My Way Versus the Highway
A lot about this comes down to a simple question of ‘why?’. Why even jump through these hoops to set up a limited, possibly ToS-breaking emulator on what is ultimately a gaming PC running Windows 10? Why not use that Raspberry Pi 4 or NUC system that’s been giving you sad eyes for the past months from where it’s been stuffed into a dusty corner?
The Playstation 5 should be more than capable of playing back any Playstation title from the original PSX up to the PS4 based on its hardware specifications, yet it only offers compatibility for the PS4. The XBox Series X|S on the other hand provides such backwards compatibility along the entire lineage of XBox games, though not encompassing every single title released for a Microsoft console.
Xbox One S playing PS2 version of the original God of War
Nintendo has had an on/off relationship with running their own blessed emulator solution, such as the Virtual Console (VC) for the Wii, Wii U and 3DS which even offered access to games for non-Nintendo games. In early 2019 Nintendo began to phase out VC, however. In many ways the VC service was the closest to what RetroArch offers today, even if one could argue about the limited number of games on the VC and the per-game cost to purchase the right to play it.
All of this makes one wonder what would happen if a multi-system emulation service like the VC but with a much larger game library and lower cost (e.g. full access as part of a PSN subscription) were to be made available. Would this be enough to make people stop trying to get RetroArch on their brand-new gaming console?
It’s Called a PC
The skeptic’s view on this matter would probably be that with the lack of truly exclusive titles for video game consoles these days one might as well stick an SFF rig of one’s choosing underneath the TV, running one’s favorite OS and controlled by a controller of one’s choice. Using Steam’s Big Picture feature or equivalent, it’d be about as easy to control with a controller as if it was a dedicated game console.
Installing RetroArch and similar on this gaming rig would be a snap too, and would likely work better for more emulators as it’d be a standard Windows or Linux system for which RetroArch is actually optimized. Bringing us back again to why people try to do things the hard way instead.
A large appeal of game consoles and other walled gardens has often been the ‘it just works’ selling point. Buy it, set it up, turn it on, start playing games, stop worrying. Not everyone is into debugging obscure compatibility and driver issues on Windows, or figuring out why launching a game makes the Xserver crash on Linux. From there it is appealing to still make it do a bit more, at least to those who even as a child found ourselves staring at devices and feeling that familiar itch in one’s fingers.
In the end it’s essentially just about hobbies and interests. Even if Microsoft et al. would vehemently disagree, nobody is harmed if someone hacks their PS5 or XSX|S to run additional software on it that brings the owner of said hardware more pleasure. That’s after all how most interesting hacks are born.
It’s hardly a secret any more at this point that today’s game consoles from Microsoft and Sony are essentially AMD gaming rigs packed up into a custom package and with tweaked system software. So it’s not too surprising that enterprising hackers got the Playstation 2 emulator of RetroArch running on an Xbox Series X|S game console despite Microsoft’s attempts to stop them. (Video, embedded below.)
It’s possible to sneak the RetroArch app past Microsoft’s security checkpoints by shelling out $19 for a Microsoft Developer Account, setting up Developer Mode on the XBox console, and getting the Universal Windows Platform (UWP) port of RetroArch from the official website. This has the advantage of it being a blessed-by-the-Redmond-gods approach. But one cannot play retail games in Developer Mode and large games due to a 2 GB limit.
More recently, a hacker by the name of [tunip3] found a flaw in the Xbox app distribution system which allows one to download a ‘retail’ version of RetroArch. This involves marking the RetroArch app as ‘private’, allowing it to skip a review by Microsoft. People whose email address is on a whitelist are then granted download permission for that app on their Xbox console. The advantage of this ‘retail’ approach is that it does not feature the 2 GB filesize limits. The disadvantage is that Microsoft is free to take the app down and ban [tunip3]’s developer account.
My Way Versus the Highway
A lot about this comes down to a simple question of ‘why?’. Why even jump through these hoops to set up a limited, possibly ToS-breaking emulator on what is ultimately a gaming PC running Windows 10? Why not use that Raspberry Pi 4 or NUC system that’s been giving you sad eyes for the past months from where it’s been stuffed into a dusty corner?
The Playstation 5 should be more than capable of playing back any Playstation title from the original PSX up to the PS4 based on its hardware specifications, yet it only offers compatibility for the PS4. The XBox Series X|S on the other hand provides such backwards compatibility along the entire lineage of XBox games, though not encompassing every single title released for a Microsoft console.
Xbox One S playing PS2 version of the original God of War
Nintendo has had an on/off relationship with running their own blessed emulator solution, such as the Virtual Console (VC) for the Wii, Wii U and 3DS which even offered access to games for non-Nintendo games. In early 2019 Nintendo began to phase out VC, however. In many ways the VC service was the closest to what RetroArch offers today, even if one could argue about the limited number of games on the VC and the per-game cost to purchase the right to play it.
All of this makes one wonder what would happen if a multi-system emulation service like the VC but with a much larger game library and lower cost (e.g. full access as part of a PSN subscription) were to be made available. Would this be enough to make people stop trying to get RetroArch on their brand-new gaming console?
It’s Called a PC
The skeptic’s view on this matter would probably be that with the lack of truly exclusive titles for video game consoles these days one might as well stick an SFF rig of one’s choosing underneath the TV, running one’s favorite OS and controlled by a controller of one’s choice. Using Steam’s Big Picture feature or equivalent, it’d be about as easy to control with a controller as if it was a dedicated game console.
Installing RetroArch and similar on this gaming rig would be a snap too, and would likely work better for more emulators as it’d be a standard Windows or Linux system for which RetroArch is actually optimized. Bringing us back again to why people try to do things the hard way instead.
A large appeal of game consoles and other walled gardens has often been the ‘it just works’ selling point. Buy it, set it up, turn it on, start playing games, stop worrying. Not everyone is into debugging obscure compatibility and driver issues on Windows, or figuring out why launching a game makes the Xserver crash on Linux. From there it is appealing to still make it do a bit more, at least to those who even as a child found ourselves staring at devices and feeling that familiar itch in one’s fingers.
In the end it’s essentially just about hobbies and interests. Even if Microsoft et al. would vehemently disagree, nobody is harmed if someone hacks their PS5 or XSX|S to run additional software on it that brings the owner of said hardware more pleasure. That’s after all how most interesting hacks are born.
It’s hardly a secret any more at this point that today’s game consoles from Microsoft and Sony are essentially AMD gaming rigs packed up into a custom package and with tweaked system software. So it’s not too surprising that enterprising hackers got the Playstation 2 emulator of RetroArch running on an Xbox Series X|S game console despite Microsoft’s attempts to stop them. (Video, embedded below.)
It’s possible to sneak the RetroArch app past Microsoft’s security checkpoints by shelling out $19 for a Microsoft Developer Account, setting up Developer Mode on the XBox console, and getting the Universal Windows Platform (UWP) port of RetroArch from the official website. This has the advantage of it being a blessed-by-the-Redmond-gods approach. But one cannot play retail games in Developer Mode and large games due to a 2 GB limit.
More recently, a hacker by the name of [tunip3] found a flaw in the Xbox app distribution system which allows one to download a ‘retail’ version of RetroArch. This involves marking the RetroArch app as ‘private’, allowing it to skip a review by Microsoft. People whose email address is on a whitelist are then granted download permission for that app on their Xbox console. The advantage of this ‘retail’ approach is that it does not feature the 2 GB filesize limits. The disadvantage is that Microsoft is free to take the app down and ban [tunip3]’s developer account.
My Way Versus the Highway
A lot about this comes down to a simple question of ‘why?’. Why even jump through these hoops to set up a limited, possibly ToS-breaking emulator on what is ultimately a gaming PC running Windows 10? Why not use that Raspberry Pi 4 or NUC system that’s been giving you sad eyes for the past months from where it’s been stuffed into a dusty corner?
The Playstation 5 should be more than capable of playing back any Playstation title from the original PSX up to the PS4 based on its hardware specifications, yet it only offers compatibility for the PS4. The XBox Series X|S on the other hand provides such backwards compatibility along the entire lineage of XBox games, though not encompassing every single title released for a Microsoft console.
Xbox One S playing PS2 version of the original God of War
Nintendo has had an on/off relationship with running their own blessed emulator solution, such as the Virtual Console (VC) for the Wii, Wii U and 3DS which even offered access to games for non-Nintendo games. In early 2019 Nintendo began to phase out VC, however. In many ways the VC service was the closest to what RetroArch offers today, even if one could argue about the limited number of games on the VC and the per-game cost to purchase the right to play it.
All of this makes one wonder what would happen if a multi-system emulation service like the VC but with a much larger game library and lower cost (e.g. full access as part of a PSN subscription) were to be made available. Would this be enough to make people stop trying to get RetroArch on their brand-new gaming console?
It’s Called a PC
The skeptic’s view on this matter would probably be that with the lack of truly exclusive titles for video game consoles these days one might as well stick an SFF rig of one’s choosing underneath the TV, running one’s favorite OS and controlled by a controller of one’s choice. Using Steam’s Big Picture feature or equivalent, it’d be about as easy to control with a controller as if it was a dedicated game console.
Installing RetroArch and similar on this gaming rig would be a snap too, and would likely work better for more emulators as it’d be a standard Windows or Linux system for which RetroArch is actually optimized. Bringing us back again to why people try to do things the hard way instead.
A large appeal of game consoles and other walled gardens has often been the ‘it just works’ selling point. Buy it, set it up, turn it on, start playing games, stop worrying. Not everyone is into debugging obscure compatibility and driver issues on Windows, or figuring out why launching a game makes the Xserver crash on Linux. From there it is appealing to still make it do a bit more, at least to those who even as a child found ourselves staring at devices and feeling that familiar itch in one’s fingers.
In the end it’s essentially just about hobbies and interests. Even if Microsoft et al. would vehemently disagree, nobody is harmed if someone hacks their PS5 or XSX|S to run additional software on it that brings the owner of said hardware more pleasure. That’s after all how most interesting hacks are born.
The first person shooter genre found its feet in the PC world, relying on the holy combination of the keyboard and mouse for input. Over time, consoles have refined their own version of the experience, and the gamepad has become familiar territory for many FPS fans. [Tech Yesterday] was a die hard controller player, but after trying out a mouse, didn’t want to go back. Instead, he built a truly impressive hybrid device.
The build begins with a standard Xbox 360 wired controller, somewhat of a defacto standard for PC gamepads. The left analog stick and triggers remain untouched, however the face buttons are all relocated using mechanical keyboard switches. The D-pad has been relocated to the left hand side with tactile switches, and the right analog stick removed entirely. In its place, a cut-down optical mouse is used on a flat 4″x4″ mousepad attached to the controller, strapped to the player’s thumb.
The resulting controller combines the benefit of analog stick movement and the precision aiming of a mouse. We’re amazed at how comfortable the controller looks to use, particularly in the improved second revision. While currently only used on PC, we can imagine such controllers shaking up the console FPS scene in a serious way.
SpaceX are perhaps most well known for their vertically-landing reusable rocket technology. The latest such effort is the Starship, which recently underwent a fiery test in the last month to attempt a bellyflop maneuver. [Nicholas Rehm] wanted to attempt a similar flight profile in the local park, and set to building an RC Starship of his own.
The build is like a few we’ve seen before, in which electric power is used to propel a rocket-like craft straight upwards using propellers and active stabilization. In this case, there’s a pair of twin motors with counter-rotating propellers which can pivot to direct their thrust, as well as four external control surfaces. These are all under the command of [Nicholas]’s custom flight controller.
Upon testing the rig, [Nicholas] was able to execute a smooth ascent, followed by a bellyflop, before a smooth return to vertical flight and descent. Landing vertically on the grass was out of the question, due to the rough surface, but we imagine it would be doable with the right landing gear attached.
While flight without wings in this manner isn’t particularly efficient, it’s great to see what can be achieved with smart use of control systems engineering to keep a craft stable. We look forward to seeing [Nicholas]’s next attempts, too. Video after the break.
The racing sim scene has always had a strong DIY subculture, as enthusiasts seeking the most realistic-feeling peripherals set out to modify off-the-shelf offerings for greater authenticity. Others go further and craft their own builds from the ground up. [ilge] has done just that, putting together his own set of pedals for sim racing.
The build relies primarily on 3D printed components, with a few springs and some nuts and bolts to hold everything together. Gear teeth on the pedal arms interface with matching gears mounted on potentiometers. These are then wired into an Arduino Pro Micro, which reads the individual pots via analog inputs and then acts as a USB Human Interface Device to the computer.
[ilge] tests the setup with a variety of games, including the popular Euro Truck Simulator and iRacing. It’s a great cheap way to get started with a pedal set for a sim rig. From here, the sky really is the limit; we’d love to see an upgraded version with a load-cell on the brake for better pedal feel. We’d be surprised if an H-shifter isn’t in the works, too. Video after the break.
This week retro-gadget collector and video blogger [Techmoan] featured perhaps the most delicious audio recording format that we know of — a chocolate gramophone record. Compared to his typical media format explorations, the chocolate record is of quite recent vintage. He first heard of them back in 2015 when Tasmanian artist [Julia Drouhin] offered chocolate recordings as part of her art project. The one that [Techmoan] finally obtained was from a UK chocolatier who offers them with custom labelling and your choice of two songs. There are some pointers in the video about how to playback your chocolate disk without ruining it (use the lightest stylus tracking force as possible). These disks are recorded at 45 RPM on one side only, and are about the same size as a standard single. But being about five times thicker, they pack a lot more calories than your typical phonograph disk.
No reflection on the Tewkesbury Town Band, but this is probably the lowest fidelity recording media ever, but at least you can eat it when you’re done listening — label and all. We hope the Mission Impossible movie producers are paying attention so we can see the secret audio briefing being eaten instead of going up in smoke next film.
If you’re not familiar with [Techmoan], he often dredges up and reviews old media formats. A while back we covered a review he did on a stainless steel wire recorder from the 1940s. If these things are of interest, you should definitely check out his channel.
“Time-domain reflectometry” sure sounds like something that needs racks of expensive equipment to accomplish. In reality, TDR is just measuring the time between injecting a pulse into a cable and receiving its echo, either from the other end of the cable or from some fault or defect along the way. It’s a useful technique, and as [Allen Wolke (W2AEW)] shows us, it can be accomplished with little more than a battery, a resistor, and an oscilloscope. And a little math, of course.
There are, of course, dedicated time-domain reflectometers, but all of them are really just elaborations of the basic principles [W2AEW] demonstrates with his simple setup. The oscilloscope is set up with a tee connector on one channel; one side of the tee is connected to the cable under test, while the shield conductor of the other side is connected to the negative terminal of a 9V battery. A resistor connected to the center conductor is used to complete the circuit, which sends a brief pulse down the test cable. The scope is set up to capture the outgoing pulse as well as the return pulse, allowing the time between the two to be measured. Some simple math gives the length of the cable, the distance to a fault, or with a little rearrangement, the velocity factor of the cable.
The video below shows the simple method at work on coax and Cat 5e Ethernet cable. It even worked on a roll of zip cable, which was a little surprising. If this technique is too simple, you can always elaborate a bit and roll your own TDR tester. Googly eyes optional, of course, but recommended.
There’s something satisfying about creating high voltages. Sure, there are practical uses like neon signs or doing certain experiments, but be honest — you really just want to see some giant arcs lighting up your dark mad scientist lair. [Mircemk] has just the prescription for what ails you. Using a two-stage approach, he shows a simple setup that generates about 110KV from a pretty tame 15V supply.
From the 15V, there is a stage that uses a flyback transformer and a switch to generate a reasonably high voltage. The final stage is a Cockroft-Walton voltage multiplier that can produce quite a bit of voltage. You can see the impressive arcs in the video below.
The multiplier circuit found fame with experiments by Cockroft and Walton, obviously, but was actually originated in the early 1900s with a physicist named Greinacher. The circuit uses diodes as switches and charges a bank of capacitors in parallel. The discharge, however, puts the capacitors in series. Neglecting losses and loads, the output voltage is equal to the peak-to-peak input voltage times the number of stages present. Real-world considerations mean you won’t quite get that voltage out of it, but it can still provide a potent punch. Click through the break for a video of the circuit in action!
A plane is a tool familiar to all woodworkers, used to shape a workpiece by hand by shaving away material. Regular planes are two-handed tools available at all good hardware stores. For finer work, a finger plane can be useful, though harder to find. Thankfully, [Daniel] put together a video showing how to make your own.
[Daniel]’s build relies on stabilized wood, useful for its density and consistent quality, though other woods work too. A 6″ pen blank is enough to make a pair of matching finger planes. A block and two side panels are cut out from the material, with attention paid to making sure everything remains square for easy assembly. The parts are glued together with a block set at the desired cutting angle for the plane. With the assembly then tidied up on the bandsaw and sander, [Daniel] installs the cutting blade. This can be made from a larger standard plane blade, or a cutdown chisel can be pressed into service. The blade is held in place with a wooden wedge beneath a metal pin. The pin itself is crafted from an old drill bit, cut down to size.
It’s a useful tool for doing fine plane work, for which a full-size tool would be ungainly. We can imagine it proving particularly useful in producing accurate scale models in smaller sizes. If you’re big into woodworking, consider giving your tools a good sharpen on the cheap, too. Video after the break.
In 1984 there weren’t many ways to listen to high-quality music, so an FM tuner was an essential part of any home hi-fi system. The Pioneer TX-950 picked up by [The Curious Lorenz] would have been someone’s pride and joy, with its then-cutting-edge microprocessor control, digital PLL tuning, and seven-segment displays. Astoundingly it doesn’t have an auto-tuning function though, so some work to implement the feature using an ATtiny85 was called for.
A modern FM tuner would be quite likely to use an all-in-one tuner chip using SDR technology under the hood, but this device from another era appears to be a very conventional analog tuner to which the PLL and microprocessor have been grafted. There are simple “Up” and “Down” buttons and a “Station tuned” light. One might imagine that given these the original processor could have done autotune. At least the original designers were kind enough to provide the ATtiny with the interfaces it needs. Pressing either button causes it to keep strobing its line until the “Station tuned” line goes high, at which point it stops. It’s an extremely simple yet effective upgrade, and since the ATtiny is so small it’s easily placed on top of the original PCB. The result is an ultra-modern tuner from 1984, that’s just that little bit more modern than it used to be.
Don’t you hate getting up in the middle of the night to use the bathroom? The worst part is not being able to see what you’re doing, but if you turn on a light, you’ll lose your night vision. Nightlights are supposed to be the best solution, but are usually too bright for 3 AM excursions and can end up leaking light into the bedroom. What the bathroom needs is a purpose-built nightlight that uses red light so you don’t lose your night vision.
This simple, wall-mounted night light is just the thing. All it takes is two AA batteries, a resistor, a red LED, and an SPST push button. [Vchaney] even made their own battery contacts. The genius part of this build is in the adjustable LED, which is fitted into a ball that moves around in a socket so you can aim it wherever you need to see. All the files are available if you want to print one for yourself.
For something that’s basically a coil of wire around some magnetic pole pieces, an electric guitar pickup is a complicated bit of tech. So much about the tone of the instrument is dictated by how the pickup is wound that controlling the winding process is something best accomplished with a machine. This automatic pickup winder isn’t exactly a high-end machine, but it’s enough for the job at hand, and has some interesting possibilities for refinements.
First off, as [The Mixed Signal] points out, his pickups aren’t intended for use on a guitar. As we’ve seen before, the musical projects he has tackled are somewhat offbeat, and this single-pole pickup is destined for another unusual instrument. That’s not to say a guitar pickup couldn’t be wound on this machine, of course, as could inductors, solenoids, or Tesla coils. The running gear is built around two NEMA-17 stepper motors, one for the coil spindle and one for the winding carriage. The carriage runs on a short Acme lead screw and linear bearings, moving back and forth to wind the coil more or less evenly. An Arduino topped with a CNC shield runs the show, allowing for walk-away coil winding.
We do notice that the coil wire seems to bunch up at the ends of the coil form. We wonder if that could be cured by speeding up the carriage motor as it nears the end of the spool to spread the wire spacing out a bit. The nice thing about builds like these is the ease with which changes can be made — at the end of the day, it’s just code.
There’s a good chance you already saw SpaceX’s towering Starship prototype make its impressive twelve kilometer test flight. While the attempt ended with a spectacular fireball, it was still a phenomenal success as it demonstrated a number of concepts that to this point had never been attempted in the real world. Most importantly, the “Belly Flop” maneuver which sees the 50 meter (160 foot) long rocket transition from vertical flight to a horizontal semi-glide using electrically actuated flight surfaces.
Finding himself inspired by this futuristic spacecraft, [Nicholas Rehm] has designed his own radio controlled Starship that’s capable of all the same aerobatic tricks as the real-thing. It swaps the rocket engines for a pair of electric brushless motors, but otherwise, it’s a fairly accurate recreation of SpaceX’s current test program vehicle. As you can see in the video after the break, it’s even able to stick the landing. Well, sometimes anyway.
Just like the real Starship, vectored thrust is used to both stabilize the vehicle during vertical ascent and help transition it into and out of horizontal flight. Of course, there are no rocket nozzles to slew around, so [Nicholas] is using servo-controlled vanes in the bottom of the rocket to divert the airflow from the motors. Servos are also used to control the external control surfaces, which provide stability and a bit of control authority as the vehicle is falling.
As an interesting aside, Internet sleuths looking through pictures of the Starship’s wreckage have noted that SpaceX appears to be actuating the flaps with gearboxes driven by Tesla motors. The vehicle is reportedly using Tesla battery packs as well. So while moving the control surfaces on model aircraft with battery-powered servos might historically have been a compromise to minimize internal complexity, here it’s actually quite close to the real thing.
Unfortunately, the RC Starship made a hard landing of its own on a recent test flight, so [Nicholas] currently has to rebuild the craft before he can continue with further development. We’re confident he’ll get it back in the air, though it will be interesting to see whether or not he’s flying before SpaceX fires off their next prototype.
