Author: Tyler August

437 Articles

Variable-Pitch Propellers For More Efficient Quadcopter

April 8, 2026 by 20 Comments

Quadcopters tend to have very poor efficency because of their high disk loading. High disk loading– that is, how much weight each square meter of area swept by the propellers must carry–is almost unavoidable with conventinal quadcopters, which are controlled by throttling the four props. Make the propellers too big, and their inertia slows down that control loop, leading to stability problems. [rctestflight] had an idea to solve this, by borrowing a technology from the world of fixed-wing aviation: variable-pitch propellers.

In aircraft use, they are not new, dating back to the end of the first world war. They’re made for everything from the largest turboprops to the  75 kW(100 HP) Rotax 912. By varying the propeller pitch, you can keep the engine turning in its ideal RPM range but still vary thrust by taking a larger or shallower ‘bite’ out of the air with each sweep of the prop. You can probably see how this applies to the quadcopter: a well-designed pitch-change mechanism is going to be much quicker than throttling a big prop with lots of rotational inertia. That’s the theory.

To test it, [rctestflight] builds some large 3D-printed variable pitch props, hooks them up to regular drone motors via a belt drive, before going on–you guessed it–an RC test flight. To make that work, he’s got the pitch servo being driven from what should be the flight controller’s thrust output to each motor. Aside from the vibrations from imperfect balance on the 3D-printed props, it flies quite well– and much better with pitch control than trying to vary the RPMs of those heavy props. He’s even able to reverse the propeller pitch, making this perhaps the first quadcopter capable of autorotation. Well, almost, given that it lost control and came apart when he cut the throttle.

Continue reading “Variable-Pitch Propellers For More Efficient Quadcopter” →

an animated gif of the eye in motion.

Bending Faux-Neon LEDs Make For Animations Glass Tubes Can’t Match

April 8, 2026 by 10 Comments

Odds are, if you like neon lights, you’re not thrilled with the LED faux-“neon” strips that are supposed to replace them. They’ve got their advantages, but the light quality of RGB LEDs lacks something compared to the emission spectrum of nobel gas, at least to purists. On the other hand, you cannot create an animation by bending glass tubes, like [David Hamp-Gonsalves] has demonstrated with his Neon Animated Eye.

Back in the day, you’d have needed dozens of tubes for a flickery animation, but [David] figured that since these LED strips are flexible, why not flex them? He’s using addressable LEDs — WS2812s, specifically — so activating and deactivating the pupil of the eye is easy-peasy. Opening and closing the lid is accomplished with a geared motor driven by a TB6612 driver turning a barrel cam. The ends of the stiff LED strip being brought together and pulled apart result in the blinking effect here, but as [David] points out you’re hardly limited that specific motion. There’s a whole world of Tron-like glowing animatronics that can be created with this technique. Code and STLs are available on GitHub, though, if you want to replicate the eye exactly.

[David] says he’d like to see this in a storefront someday, but given that fatigue life is a thing, it might be something to keep in your back pocket for seasonal displays like Christmas and Halloween rather than something that’s going to run 24/7. On the other hand, if you’re careful about limiting flexion and which faux-neon strip you buy, you might be able to create an animation that can last for years.

This is hardly the first time we’ve seen these faux-neon strips , but it is the first time we’ve seen them animated. We can’t help but think the Hauntimator software we featured before would be a good paring with this hack.

Are We Surrendering Our Thinking To Machines?

April 8, 2026 by 82 Comments

“Once, men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them.” — so said [Frank Herbert] in his magnum opus, Dune, or rather in the OC Bible that made up part of the book’s rich worldbuilding. A recent study demonstrating “cognitive surrender” in large language model (LLM) users, as reported in Ars Technica, is going to add more fuel to that Butlerian fire.

Cognitive surrender is, in short, exactly what [Herbert] was warning of: giving over your thinking to machines. In the study, people were asked a series of questions, and — except for the necessary “brain-only” control group — given access to a rigged LLM to help them answer. It was rigged in that it would give wrong answers 50% of the time, which while higher than most LLMs, only a difference in degree, not in kind. Hallucination is unavoidable; here it was just made controllably frequent for the sake of the study.

The hallucinations in the study were errors that the participants should have been able to see through, if they’d thought about the answers. Eighty percent of the time, they did not. That is to say: presented with an obviously wrong answer from the machine, only in 20% of cases did the participants bother to question it. The remainder were experiencing what the researchers dubbed “cognitive surrender”: they turned their thinking over to the machines. There’s a lot more meat to this than we can summarize here, of course, but the whole paper is available free for your perusal.

Giving over thinking to machines is nothing new, of course; it’s probably been a couple decades since the first person drove into a lake on faulty GPS directions, for example. One might even argue that since LLMs are correct much more than 50% of the time, it is statistically wise to listen to them. In that case, however, one might be encouraged to read Dune.

Thanks to [Monika] for the tip!

TinyGo Boldly Goes Where No Go Ever Did Go Before

April 7, 2026 by 26 Comments

When you’re programming microcontrollers, you’re likely to think in C if you’re old-school, Rust if you’re trendy, or Python if you want it done quick and have resources to spare. What about Go? The programming language, not the game. That’s an option, too, with TinyGo now supporting over 100 different dev boards, along with webASM.

We covered TinyGo back in 2019, but they were just getting started at that point, targeting the Arduino and BBC:micro boards. They’ve grown that list to include everything from most of Adafruit’s fruitful suite of offerings, ESP32s, and even the Nintendo Game Boy Advance. So now you can go program go in Go so you can play go on the go.

The biggest drawback–which is going to be an absolute dealkiller for a lot of applications–is a lack of wireless connectivity support. Claiming to support the ESP8266 while not allowing one to use wifi is a bit of a stretch, considering that’s the whole raison d’être of that particular chip, but it’s usable as a regular microcontroller at least.

They’ve now implemented garbage collection, a selling point for those who like Go, but admit it’s slower in TinyGo compared to its larger cousin and won’t work on AVR chips or in WebAssembly. It’s still not complete Go, however, so just as we reported in 2019, you won’t be able to compile all the standard library packages you might be used to. There are more of them than there were, so progress has been made!

Still, knowing how people get about programming languages, this will please the Go fanatics out there. Others might prefer to go FORTH and program their Arduinos, or to wear out their parentheses keys with LISP. The more the merrier, we say!

The Heat Island Effect Is Warming Up The AI Data Center Controversy

April 7, 2026 by 58 Comments

There’s been a lot of virtual ink spilled in environmental circles about the cooling water requirements of data centers, but less consideration of what happens with all the heat coming out of these buildings. Naturally, it’s going to warm the surrounding environment, but how much? Around 2 C (3.6 F) on average, and potentially much more than that, according to a recent study on the data heat island effect.

It’s common sense, of course: heat removed from the data center doesn’t go away. That heat might go into a body of water if one is available, but otherwise it’s out into the atmosphere to warm up everybody else’s day. In some places — like a Canadian winter — that might not be so bad. In others, where climate change and urban heat islands are cranking up the summertime temperatures, it very much could be. Especially if you’re in the worst-case scenario micro-climate described by the paper, which saw a predicted increase of 9.1 C (16 F).

Now, these results are theoretical and need to be ground-truthed, but anyone who has huddled next to the air-exchange unit of a large building for warmth knows there’s something to them. Unfortunately there don’t seem to be before-and-after measurements available for existing data-centers — AI or otherwise — to show exactly what their heat output is doing in the real world, but the urban heat island effect from all the dark asphalt in our cities is well known. Cooling paint and green roofs can help with that, but they won’t do much for the megawatts being pumped out to keep your cousin’s AI girlfriend online.

Some would argue that all this heat wouldn’t be a problem if we could launch the data centers outside the environment — just have a care the front doesn’t fall off.

Image of data center cooling by Анна from Pixabay

Flying Cell Towers For Lower-Latency

April 6, 2026 by 25 Comments

When the inevitable Kessler Syndrome cascade sweeps Starlink and its competitors from Low Earth Orbit in what will doubtless be a spectacular meteor shower of debris, the people behind Sceye and its competitors are going to be laughing to the bank. That’s because they’re putting their connectivity rather lower than orbit — in the stratosphere, with high-altitude dirigibles.

The advantages are pretty obvious: for one, the dirigible isn’t disposable in the way the very-low-orbit satellites Starlink and its planned imitators use. For another, the time-of-flight for a signal to get to a dirigible 20 km up is less than a tenth of the time it takes to get 480 km up — and that affects latency. Thirdly, the High Altitude Platform System (HAPS) concept won’t require any special transmitters. Regular cellular modems using ordinary 4G and 5G bands and speeds are usable, which eliminates a big barrier to rollout.

Continue reading “Flying Cell Towers For Lower-Latency” →

The Smallest Dialup ISP Is A Raspberry Pi And A Prison Phone

April 4, 2026 by 29 Comments

There were a plethora of tiny, local ISPs in the days of dial-up internet. Along with the big providers, many cities would have more than one. Some of those have survived broadband, but none of them were as small as [Jeff Geerling]’s Pi ISP — a tiny dialup ISP built so his Aunt’s old G3 MacBook can get online at 36kbps, as God and [Robert Kahn] intended.

Hardware-wise, the Raspberry Pi is at one end of the chain, and your retrocomputer at another. In between, you’ll have a USB modem plugged into the Pi, and a device called a “two-way line simulator” to create a dial tone for that plain-old-telephone goodness. [Jeff] notes that these were commonly used in prisons for the phones that visitors use to talk to inmates.

Of course, since these devices are designed strictly for voice transmissions, which POTS was built for, you’re not going to get over 36 kbps, and that’s even with high-quality gear. The cheaper options might drop you down to 28k… just like with an ISP back in the day. ‘You get what you pay for’ is very rarely false.

Now, you can use this technology to just connect two computers together — as we’ve featured previously — but [Jeff] has gone the extra mile to put together, via Ansible, an easy-to-install software package that will let the Raspberry Pi act just like your ISP’s servers once did, and connect you to that series of tubes once called the World Wide Web. Of course, the World Wide Web isn’t built for dial-up anymore, so you’re going to be waiting… a while. Hackaday’s front page isn’t especially heavy, weighing about 4MB at the time of this writing, but that’s 15 minutes of load time, and you still aren’t reading the articles.

You also won’t be able to access much on old machines that can’t do HTTPS, but [Jeff] thought of that and bundles [rdmark]’s MacProxyClassic to translate the modern web into HTML tags that Netscape can understand and serve them over HTTP. You’ll still be waiting for our modern bloat, but perhaps not quite so long.

If you want the “authentic” dial-up experience, you’ll need to see the lightweight webpages of Yesteryear, and MacProxyClassic contains a Wayback Machine extension for that purpose. We featured a similar project a while back that did that, but without all the joys of dial-up. Now get off the computer, we’re expecting a call!

Continue reading “The Smallest Dialup ISP Is A Raspberry Pi And A Prison Phone” →