When we look at how everyone’s favorite flying dinosaurs get around, we can see that although they use their wings a lot too, their legs are at least as important. Even waddling or hopping about somewhat ungainly on legs is more energy efficient than short flights, and taking off from the ground is helped by jumping into the air with a powerful leap from one’s legs. Based on this reasoning, a team of researchers set out to give flying drones their own bird-inspired legs, with their findings published in Nature (preprint on ArXiv).

The prototype RAVEN (Robotic Avian-inspired Vehicle for multiple ENvironments) drone is capable of hopping, walking, jumping onto an obstacle and jumping for take-off. This allows the drone to get into the optimal position for take-off and store energy in its legs to give it a boost when it takes to the skies. As it turned out, having passive & flexible toes here was essential for stability when waddling around, while jumping tests showed that the RAVEN’s legs provided well over 90% of the required take-off speed.

During take-off experiments the drone was able to jump to an altitude of about 0.4 meters, which allows it to clear ground-based obstacles and makes any kind of ‘runway’ unnecessary. Much like with our avian dinosaur friends the laws of physics dictate that there are strong scaling limits, which is why a raven can use this technique, but a swan or similar still requires a bit of runway instead of jumping elegantly into the air for near-vertical take-off. For smaller flying drones this approach would however absolutely seem to have legs.

Fresh off a world record for the fastest quadcopter, [Luke Bell] decided to try his luck with something more own to earth, namely trying to tackle the world record for the fastest RC car, with the current record set at 360 km/h. Starting off with a first attempt in what will be a video series, the obvious approach seems to be to get some really powerful electric motors, a streamlined body and a disused runway to send said RC car hurtling along towards that golden medal. Of course, if it was that easy, others would have done it already.

With the quadcopter record of nearly 500 km/h which we covered previously, the challenge was in a way easier, as other than air resistance and accidental lithobraking there are no worries about ground texture, tire wear or boundary layer aerodynamics. In comparison, the RC car has to contend with all of these, with the runway’s rough tarmac surface being just one of the issues, along with making sure that the wheels would hold up to the required rotation speed. For the wheels you got options like foam, hard rubber, etc., all with their own advantages and disadvantages, mostly in terms of grip and reliability.

So far speeds of over 200 km/h are easy enough to do, with foam wheels being the preferred option. To push the RC car to 300 km/h and beyond, a lot more experimentation and trial runs will have to be performed. Pending are changes to the aerodynamic design with features also commonly seen in F1 race cars such as downforce spoilers, diffusers and other tricks which should prevent the RC car from (briefly) becoming an RC airplane.

Raspberry Pi just dropped the new Raspberry Pi 500, which like its predecessor puts the similarly named SBC into a keyboard. In a detailed review and teardown video, [Jeff Geerling] goes over all the details, and what there is to like and not like about this new product.

Most of the changes relative to the RP400 are as expected, with the change to the same BCM2712 SoC as on the Raspberry Pi 5, while doubling the RAM to 8 GB and of course you get the soft power button. As [Jeff] discovers with the teardown, the odd thing is that the RP500 PCB has the footprints for an M.2 slot, as seen on the above image, but none of the components are populated.

Naturally, [Jeff] ordered up some parts off Digikey to populate these footprints, but without luck. After asking Raspberry Pi, he was told that these footprints as well as those for a PoE feature are there for ‘flexibility to reuse the PCB in other contexts’. Sadly, it seems that these unpopulated parts of the board will have to remain just that, with no M.2 NVMe slot option built-in. With the price bump to $90 from the RP400’s $70 you’ll have to do your own math on whether the better SoC and more RAM is worth it.

In addition to the RP500 itself, [Jeff] also looks at the newly launched Raspberry Pi Monitor, a 15.6″ IPS display for $100. This unit comes with built-in speakers and VESA mount, but as [Jeff] notes in his review, using this VESA mount also means that you’re blocking all the ports, so you have to take the monitor off said VESA mount if you want to plug in or out any cables.

After recently putting together the paper tape reader for his custom tube-based UE1 computer, [David Lovett] did get squiggles on the outputs, but not quite the right ones. In the most recent video, these issues are addressed one by one, so that this part of the UE1 1-bit computer can be called ‘done’. Starting off the list of issues were the odd readings from the photodiodes, which turned out to be due to the diodes being misaligned and a dodgy solder joint. This allowed [David] to move on to building the (obviously 6AU6 tube-based) amplifier for the photodiode output signals.

Much like the Bendix G-15’s tape reader which served as inspiration, this also meant adding potentiometers to adjust the gain. For the clock signal on the tape, a clock recovery PCB was needed, which should provide the UE1 computer system with both the clocks and the input data.

Using the potentiometers on the amplification board, the output signals can be adjusted at will to give the cleanest possible signal to the rest of the system, which theoretically means that as soon as [David] adds the permanent wiring and a few utility boards to allow the code to manipulate the tape reader (e.g. halt) as well as manual inputs. The UE1 computer system is thus being pretty close to running off tape by itself for the first time and with it being ‘complete’.

Since the construction of the first commercial light water nuclear power plants (LWR) the design of their fuel rods hasn’t changed significantly. Mechanically robust and corrosion-resistant zirconium alloy (zircalloy) tubes are filled with ceramic fuel pellets, which get assembled into fuel assemblies for loading into the reactor.

Now it seems that silicon carbide (SiC) may soon replace the traditional zirconium alloy with General Atomics’ SiGa fuel cladding, which has been tested over the past 120 days in the Advanced Test Reactor at Idaho National Laboratory (INL). This completes the first of a series of tests before SiGa is approved for commercial use.

One of the main advantages of SiC over zircalloy is better resistance to high temperatures — during testing with temperatures well above those experienced with normal operating conditions, the zircalloy rods would burst while the SiC ones remained intact (as in the embedded video). Although normally SiC is quite brittle and unsuitable for such structures, SiGa uses a SiC fiber composite, which allows it to be used in this structural fashion.

Although this development is primarily part of the Department of Energy’s Accident Tolerant Fuel Program and its focus on melt-down proof fuel, the switch to SiC could also solve a major issue with zirconium, being its use as a catalyst with hydrogen formation when exposed to steam. Although with e.g. Fukushima Daiichi’s triple meltdown the zircalloy fuel rods were partially destroyed, it was the formation of hydrogen gas inside the reactor vessels and the hydrogen explosions during venting which worsened what should have been a simple meltdown into something significantly worse.

Recently a company called Z-Polymers introduced its new Tullomer FDM filament that comes with a lofty bullet list of purported properties that should give materials like steel, aluminium, and various polymers a run for their money. Even better is that it is compatible with far lower specification FDM printers than e.g. PEEK. Intrigued, the folks over at All3DP figured that they should get some hands-on information on this filament and what’s it like to print with in one of the officially sanctioned Bambu Lab printers: these being the X1C & X1CE with manufacturer-provided profiles.

The world of engineering-grade FDM filaments has existed for decades, with for example PEEK (polyether ether ketone) having been around since the early 1980s, but these require much higher temperatures for the extruder (360+℃) and chamber (~90℃) than Tullomer, which is much closer (300℃, 50℃) to a typical high-performance filament like ABS, while also omitting the typical post-process annealing of PEEK. This assumes that Tullomer can match those claimed specifications, of course.

One of the current users of Tullomer is Erdos Miller, an engineering firm with a focus on the gas and oil industry. They’re using it for printing parts (calibration tooling) that used to be printed in filaments like carbon fiber-reinforced nylon (CF-PA) or PEEK, but they’re now looking at using Tullomer for replacing CF-PA and machined PEEK parts elsewhere too.

It’s still early days for this new polymer, of course, and we don’t have a lot of information beyond the rather sparse datasheet, but if you already have a capable printer, a single 1 kg spool of Tullomer is a mere $500, which is often much less or about the same as PEEK spools, without the requirement for a rather beefy industrial-strength FDM printer.

We were recently notified by a reader that [Tom Evans] had filed a copyright claim against [Mark]’s repair video on his Mend it Mark YouTube channel, taking down said repair video as well as [Mark]’s delightful commentary. In a new video, [Mark] comments on this takedown and the implications. The biggest question is what exactly was copyrighted in the original video, which was tough because YouTube refused to pass on [Mark]’s questions or provide further details.

In this new video the entire repair is summarized once again using props instead of the actual pre-amp, which you can still catch a glimpse of in our earlier coverage of the repair. To summarize, there was one bad tantalum capacitor that caused issues for one channel, and the insides of this twenty-five thousand quid pre-amp looks like an artistic interpretation of a Jenga tower using PCBs. We hope that this new video does stay safe from further copyright strikes from an oddly vengeful manufacturer after said manufacturer event sent the defective unit to [Mark] for a repair challenge.

Since this purportedly ‘audiophile-level’ pre-amplifier uses no special circuits or filtering – just carefully matched opamps – this is one of those copyright strike cases that leave you scratching your head.

The Apple FlatMac was one of those 1980s concepts by designer [Hartmut Esslingers] that remained just a concept with no more than some physical prototypes created. That is, until [Kevin Noki] came across it in an Apple design book and contacted [Hartmut] to ask whether he would be okay with providing detailed measurements so that he could create his own.

Inside the 3D printed enclosure is a Raspberry Pi 4 running an appropriately emulated Macintosh, with a few modern features on the I/O side, including HDMI and USB. Ironically, the screen is from a 3rd generation iPad, which [Kevin] bought broken on EBay. There’s also an internal floppy drive that’s had its eject mechanism cleverly motorized, along with a modified USB battery bank that should keep the whole show running for about an hour. The enclosure itself is carefully glued, painted and sculpted to make it look as close to the original design as possible, which includes custom keycaps for the mechanical switches.

As far as DIY projects go, this one is definitely not for the faint of heart, but it’s fascinating to contrast this kind of project that’s possible for any determined hobbyist with the effort it would have taken forty years ago. The only question that’s left is whether or not the FlatMac would have actually been a practical system if it had made it to production. Although the keyboard seems decent, the ergonomics feel somewhat questionable compared to something more laptop-like.

Thanks to [Daniel Doran] for the tip.

When thinking of home computers and their portable kin it’s easy to assume that all of them provided BASIC as their interpreter, but for a while APL also played a role. The most quaint APL portable system here might be the Ampere WS-1, called the BIG.APL. Released in Japan in November of 1985, it was a very modern Motorola M68000-based portable with fascinating styling and many expansion options. Yet amidst an onslaught of BASIC-based microcomputers and IBM’s slow retreat out of the APL-based luggables market with its IBM 5110, an APL-only portable in 1985 was a daring choice.

Rather than offering both APL and BASIC as IBM’s offerings had, the WS-1 offered only APL, with a custom operating system (called Big.DOS) which also provided a limited a form of multi-tasking involving a back- and foreground task. Running off rechargeable NiCd batteries it could power the system for eight hours, including the 25 x 80 character LCD screen and the built-in microcassette storage.

Although never released in the US, it was sold in Japan, Australia and the UK, as can be seen from the advertisements on the above linked Computer Ads from the Past article. Clearly the WS-1 never made that much of a splash, but its manufacturer seems to be still around today, which implies that it wasn’t a total bust. You also got to admit that the design is very unique, which is one of the reasons why this system has become a collector’s item today.

If a tree falls in a forest with nobody around, does it make a noise? In the case of the rainforests equipped with the Rainforest Connection’s Guardian system someone most assuredly will.

Originally created by the people behind the US nonprofit Rainforest Connection (RFCx) using upcycled smartphones to detect the sounds of illegal logging, their project now has grown into something much larger, keeping not only tabs on sounds of illegal activity, but also performing bioacoustic monitoring for scientific purposes.

Currently active in ten countries, the so-called Guardian Platform no longer features smartphones, but custom hardware inside an IP66 weatherproof enclosure and a whole range of communication options, ranging from cellular bands to satellite communications. The petal-shaped solar panels provide the module with up to 30 watts of power, and double as a shield to help protect it from the elements.

Not only is the real-time microphone data incredibly useful for rangers trying to stop illegal logging, it also provides researchers access to countless hours of audio data, which will require detailed, automated analysis. Even better is that if the audio data is available to the general public as well, via their Android & iOS apps (bottom of page), just in case you wanted to hear what that sneaky wildlife in the jungle of Peru is up to right now.

Back in the day the Apple iPod was the personal music player (PMP) to get even if mostly because everyone and their dogs had one. These days most people just use their smartphone as a PMP, but what if you were to take, say, a 5th generation iPod and modernized it? That was the basic idea that [Zac Builds] picked up and ran with, with the results as shown in the video he made about it.

The 5th gen iPod was the first one capable of playing video, and was released in October of 2005. Powering it is a Broadcom BCM2722 for video playback, and came with a 30 or 60 GB HDD. First thing that [Zac] tosses is the old (3.7V, 650 mAh) battery, which appears to be already a replacement for the original, followed by the 60 GB 1.8″ HDD. Next tossed is the 2.5″ 320×240 QVGA screen, which gets replaced by a compatible modern LCD. The case is replaced with a transparent case, along with a transparent touch wheel, and the HDD is replaced with a 256 GB SD card in an iFlash Solo SD card adapter for iPods.

Next up was the installation of more off-the-shelf mods, such as a ‘taptic mod’ – which adds a rumble motor – and replacing the iPod’s 30-pin connector with a USB-C connector, requiring some fiddly soldering and desoldering. Following this a Bluetooth audio transmitter was added, extreme PCB mods performed with a cut-off wheel to make everything fit with a custom midframe and rear case.

Ultimately, the parts left of the original iPod were most of the mainboard and some flex cable, which raises the question of whether it might not have been faster and easier to start off with designing a custom PCB. Perhaps the true value is in the modding journey and not the destination?

Thanks to [Keith Olson] for the tip.

Phosphor-based displays like CRTs rely on the phosphor to emit light for a set amount of time after being activated, allowing them to display a seemingly persistent image with one drawing beam per color. Translated to UV-sensitive PLA filament, this means that you can totally use a printed sheet of this material in combination with a 405 nm laser diode to create a display that doesn’t look dissimilar to an early CRT. This is exactly what [bitluni] did in a recent video, meshing together said laser diode, UV-sensitive PLA, stepper motors and two mirrors with an Arduino-based controller to create a rather interesting vector display.

In the video, [bitluni] goes over the development steps, including a range of improvements like being able to turn off the laser when moving between the end of a line and the beginning of a new one. While the Arduino Nano board does the driving of the stepper motor controllers, an ESP32 provides the drawing instructions. The STL and other project files including Nano & ESP32 firmware can be found on the GitHub project page.

While far from being a practical display with a single-digit Hz refresh rate, it does provide an interesting demonstration of these types of persistence of vision based displays, and without the use of exotic MEMS mirror modules or the like.

Magnetic skyrmions are an interesting example of solitons that occurs in ferromagnetic materials with conceivable solutions in electronics, assuming they can be created and moved at will. The creation and moving of such skyrmions has now been demonstrated by [Yubin Ji] et al. with a research article in Advanced Materials. This first ever achievement by these researchers of the Korea Research Institute of Standards and Science (KRISS) was more power efficient than previously demonstrated manipulation of magnetic skyrmions in thicker (3D) materials.

Magnetic skyrmions are sometimes described as ‘magnetic vortices’, forming statically stable solitons. In a broader sense skyrmions are a topologically stable field configuration in particle physics where they form a crucial part of the emerging field of spintronics. For magnetic skyrmions their stability comes from the topological stability, as changing the atomic spin of the atoms inside the skyrmion would require overcoming a significant energy barrier.

In the case of the KRISS researchers, electrical pulses together with a  magnetic field were used to create magnetic skyrmions in the ferromagnetic  (Fe₃GaTe₂, or FGaT) film, after which a brief (50 µs) electric current pulse was applied. This demonstrated that the magnetic skyrmions can be moved this way, with the solitons moving parallel to the electron flow injection, making them quite steerable.

While practical applications of magnetic skyrmions are likely to be many years off, it is this kind of fundamental research that will enable future magnetic storage and spintronics-related devices.

Featured image: Direct imaging of the magnetic skyrmions. The scale bars represent 300 nm. (Credit:Yubin Ji et al., Adv. Mat. 2024)

Recently China’s new CHIEF hypergravity facility came online to begin research projects after beginning construction in 2018. Standing for Centrifugal Hypergravity and Interdisciplinary Experiment Facility the name covers basically what it is about: using centrifuges immense acceleration can be generated. With gravity defined as an acceleration on Earth of 1 g, hypergravity is thus a force of gravity >1 g. This is distinct from simple pressure as in e.g. a hydraulic press, as gravitational acceleration directly affects the object and defines characteristics such as its effective mass. This is highly relevant for many disciplines, including space flight, deep ocean exploration, materials science and aeronautics.

While humans can take a g-force (g₀) of about 9 g₀ (88 m/s²) sustained in the case of trained fighter pilots, the acceleration generated by CHIEF’s two centrifuges is significantly above that, able to reach hundreds of g. For details of these centrifuges, this preprint article by [Jianyong Liu] et al. from April 2024 shows the construction of these centrifuges and the engineering that goes into their operation, especially the aerodynamic characteristics. Both air pressure (30 – 101 kPa) and arm velocity (200 – 1000 g) are considered, with the risks being overpressure and resonance, which if not designed for can obliterate such a centrifuge.

The acceleration of CHIEF is said to max out at 1,900 gravity tons (gt, weight of one ton due to gravity), which is significantly more than the 1,200 gt of the US Army Corps of Engineers’ hypergravity facility.

Who doesn’t like dial-up internet? Even if those who survived the dial-up years are happy to be on broadband, and those who are still on dial-up wish that they weren’t, there’s definitely a nostalgic factor to the experience. Yet recreating the experience can be a hassle, with signing up for a dial-up ISP or jumping through many (POTS) hoops to get a dial-up server up and running. An easier way is demonstrated by [Minh Danh] with a Viking DLE-200B telephone line simulator in a recent blog post.

This little device does all the work of making two telephones (or modems) think that they’re communicating via a regular old POTS network. After picking up one of these puppies for a mere $5 at a flea market, [Minh Danh] tested it first with two landline phones to confirm that yes, you can call one phone from the other and hold a conversation. The next step was thus to connect two PCs via their modems, with the other side of the line receiving the ‘call’. In this case a Windows XP system was configured to be the dial-up server, passing through its internet connection via the modem.

With this done, a 33.6 kbps dial-up connection was successfully established on the client Windows XP system, with a blistering 3.8 kB/s download speed. The reason for 33.6 kbps is because the DLE-200B does not support 56K, and according to the manual doesn’t even support higher than 28.8 kbps, so even reaching these speeds was lucky.

As impractical as most overclocking of computers is these days, there is still a lot of fun to be had along the way. Case in point being [Pieter-Jan Plaisier]’s recent liquid nitrogen-aided overclocking of an unsuspecting Raspberry Pi 5 and its BCM2712 SoC. Previous OCing attempts with air cooling by [Pieter] had left things off at a paltry 3 GHz from the default 2.4 GHz, with the power management IC (PMIC) circuitry on the SBC turning out to be the main limiting factor.

The main change here was thus to go for liquid nitrogen (LN₂) cooling, with a small chipset LN₂ pot to fit on the SBC. Another improvement was the application of a NUMA (non-uniform memory addressing) patch to force the BCM2712’s memory controller to utilize better RAM chip parallelism.

With these changes, the OC could now hit 3.6 GHz, but at 3.7 GHz, the system would always crash. It was time to further investigate the PMIC issues.

The PMIC imposes voltage configuration limitations and turns the system off at high power consumption levels. A solution there was to replace said circuitry with an ElmorLabs AMPLE-X1 power supply and definitively void the SBC’s warranty. This involves removing inductors and removing solder mask to attach the external power wires. Yet even with these changes, the SoC frequency had trouble scaling, which is why an external clock board was used to replace the 54 MHz oscillator on the PCB. Unfortunately, this also failed to improve the final overclock.

We covered the ease of OCing to 3 GHz previously, and no doubt some of us are wondering whether the new SoC stepping may OC better. Regardless, if you want to get a faster small system without jumping through all those hoops, there are definitely better (and cheaper) options. But you do miss out on the fun of refilling the LN₂ pot every couple of minutes.

Thanks to [Stephen Walters] for the tip.

On today’s installment of UE1 vacuum tube computer construction, we join [David Lovett] once more on the Usagi Electric farm, as he determines just how much work remains before the project can be called done. When we last left off, the paper tape reader had been motorized, with the paper tape being pulled through smoothly in front of the photodiodes. This left [David] with the task to create a PCB to wire up these photodiodes, put an amplification circuit together (with tubes, of course) to amplify the signal from said photodiodes, and add some lighting (two 1-watt incandescents) to shine through the paper tape holes. All of this is now in place, but does it work?

The answer here is a definite kinda, as although there are definitely lovely squiggles on the oscilloscope, bit 0 turns out to be missing in action. This shouldn’t have come as a major surprise, as one of the problems that Bendix engineers dealt with back in the 1950s was effectively the same one: they, too, use the 9th hole on the 8-bit tape as a clock signal, but with this whole being much smaller than the other holes, this means not enough light passes through to activate the photodiode.

Here, the Bendix engineers opted to solve this by biasing the photodiode to be significantly more sensitive. This seems to be the ready-made solution for the UE1’s tape reader, too. After all, if it worked for Bendix for decades, surely it’ll work in 2024.

Beyond this curveball, the rest of the challenges involve getting a tape punched with known data on it so that the tape reader’s output can actually be validated beyond acknowledging the presence of squiggles on the scope display. Although the tape guiding mechanism seems more stable now, it also needs to be guided around in an endless loop due to the way that the UE1 computer will use the tape. Much like delay line memory, the paper tape will run in an endless loop, and the processor will simply skip over sections until it hits the next code it needs as part of a loop or jump.

With semi-modern components, paper tape is easy to handle. Automatic tape feed only adds a little complexity.

Formerly known as Unidentified Flying Objects, Unidentified Anomalous Phenomena (UAP) is a category of observations that are exactly what the UAP label suggests. This topic concerns the US military very much, as a big part of national security involves knowing everything that appears in the skies. This is the reason for the development of a new sensor suite by the Pentagon called GREMLIN. Recently, a new report has provided more details about what this system actually does.

Managed by the All-domain Anomaly Resolution Office (AARO) within the DoD, GREMLIN blends many different sensors, ranging from radar to ADS-B and RF monitors, together to establish a baseline and capture any anomalies within the 90-day monitoring period to characterize them.

UAPs were a popular topic even before the 1950s when people began to see them everywhere. Usually taking the form of lights or fast-moving objects in the sky, most UAP reports can be readily classified as weather balloons, satellites like Starlink, airplanes, the Northern Lights, the ISS, or planets like Mars and Venus. There are also curious phenomena such as the Hessdalen lights, which appear to be a geological, piezoelectric phenomenon, though our understanding of such natural lighting phenomena remains limited.

But it is never aliens, that’s one thing we know for sure. Not that UFO’s don’t exist. Really.

At first glance, trying to play chess against a large language model (LLM) seems like a daft idea, as its weighted nodes have, at most, been trained on some chess-adjacent texts. It has no concept of board state, stratagems, or even whatever a ‘rook’ or ‘knight’ piece is. This daftness is indeed demonstrated by [Dynomight] in a recent blog post (Substack version), where the Stockfish chess AI is pitted against a range of LLMs, from a small Llama model to GPT-3.5. Although the outcomes (see featured image) are largely as you’d expect, there is one surprise: the gpt-3.5-turbo-instruct model, which seems quite capable of giving Stockfish a run for its money, albeit on Stockfish’s lower settings.

Each model was given the same query, telling it to be a chess grandmaster, to use standard notation, and to choose its next move. The stark difference between the instruct model and the others calls investigation. OpenAI describes the instruct model as an ‘InstructGPT 3.5 class model’, which leads us to this page on OpenAI’s site and an associated 2022 paper that describes how InstructGPT is effectively the standard GPT LLM model heavily fine-tuned using human feedback.

Ultimately, it seems that instruct models do better with instruction-based queries because they have been programmed that way using extensive tuning. A [Hacker News] thread from last year discusses the Turbo vs Instruct version of GPT 3.5. That thread also uses chess as a comparison point. Meanwhile, ChatGPT is a sibling of InstructGPT, per OpenAI, using Reinforcement Learning from Human Feedback (RLHF), with presumably ChatGPT users now mostly providing said feedback.

OpenAI notes repeatedly that InstructGPT nor ChatGPT provide correct responses all the time. However, within the limited problem space of chess, it would seem that it’s good enough not to bore a dedicated chess AI into digital oblivion.

If you want a digital chess partner, try your Postscript printer. Chess software doesn’t have to be as large as an AI model.