Wavebird Controller Soars Once More with Open Source Adapter

After scouring the second-hand shops and the endless pages of eBay for original video game hardware, a pattern emerges. The size of the accessory matters. If a relatively big controller originally came with a tiny wireless dongle, after twenty years, only the controller will survive. It’s almost as if these game controllers used to be owned by a bunch of irresponsible children who lose things (wink). Such is the case today when searching for a Nintendo Wavebird controller, and [James] published a wireless receiver design to make sure that the original hardware can be resurrected.

The project bears the name Wave Phoenix. The goal was to bring new life to a legendary controller by utilizing inexpensive, readily available parts. Central to the design is the RF-BM-BG22C3 Bluetooth module. Its low power draw and diminutive footprint made it a great fit for the limited controller port space of a Nintendo GameCube. The module itself is smaller than the GameCube’s proprietary controller connector. Luckily for projects like this, there are plenty of third-party connector options available.

When it comes to assembly, [James] insists it is possible to wire everything up by hand. He included an optional custom PCB design for those of us who aren’t point-to-point soldering masters. The PCB nestles cleanly into the 3D-printed outer casing seen in the image above in the iconic GameCube purple. Once the custom firmware for the Bluetooth module is flashed, pairing is as simple as pressing the Wave Phoenix adapter pairing button, followed by pressing X and Y simultaneously on the Wavebird controller. The two devices should stay paired as long as the controller’s wireless channel dial remains on the same channel. Better yet, any future firmware updates can be transferred wirelessly over Bluetooth.

Those who have chosen to build their own Wave Phoenix adapter have been pleased with the performance. The video below from Retrostalgia on YouTube shows that input responsiveness seems to be on par with the original Nintendo adapter. Mix in a variety of 3D printed shell color options, and this project goes a long way to upcycle Wavebird controllers that may have been doomed to end up in a dumpster. So it might be time to fire up a round of Kirby Air Ride and mash the A button unencumbered by a ten-foot cord.

youtube.com/embed/uDWsT5ocKWY?…

There are even more open source video game controller designs out there like this previous post about the Alpakka controller by Dave.

hackaday.com/2025/12/07/wavebi…

Hackaday Links: December 7, 2025

We stumbled upon a story this week that really raised our eyebrows and made us wonder if we were missing something. The gist of the story is that U.S. Secretary of Energy Chris Wright, who has degrees in both electrical and mechanical engineering, has floated the idea of using the nation’s fleet of emergency backup generators to reduce the need to build the dozens of new power plants needed to fuel the AI data center building binge. The full story looks to be a Bloomberg exclusive and thus behind a paywall — hey, you don’t get to be a centibillionaire by giving stuff away, you know — so we might be missing some vital details, but this sounds pretty stupid to us.

First of all, saying that 35 gigawatts of generation capacity sits behind the big diesel and natural gas-powered generators tucked behind every Home Depot and Walmart in the land might be technically true, but it seems to ignore the fact that backup generators aren’t engineered to run continuously. In our experience, even the best backup generators are only good for a week or two of continuous operation before something — usually the brushes — gives up the ghost. That’s perfectly acceptable for something that is designed to be operated only a few times a year, and maybe for three or four days tops before grid power is restored. Asking these units to run continuously to provide the base load needed to run a data center is a recipe for rapid failure. And even if these generators could be operated continuously, there’s still the issue of commandeering private property for common use, as well as the fact that you’d be depriving vital facilities like hospitals and fire stations of their backup power. But at least we’d have chatbots.

Well, that won’t buff right out. Roscosmos, the Russian space agency, suffered a serious setback last week when it damaged the launchpad at Site 31/6 during a Soyuz launch. This is bad news because that facility is currently the only one in the world capable of launching Soyuz and Progress, both crucial launch vehicles for the continued operation of the International Space Station. As usual, the best coverage of the accident comes from Scott Manley, who has all the gory details. His sources inform him that the “service cabin,” a 20-ton platform that slides into position under the rocket once it has been erected, is currently situated inside the flame trench rather than being safely tucked into a niche in the wall. He conjectures that the service cabin somehow got sucked into the flame trench during launch, presumably by the negative pressure zone created by the passage of all that high-velocity rocket exhaust. Whatever the cause of the accident, it causes some problems for the Russians and the broader international space community. An uncrewed Progress launch to resupply the ISS was scheduled for December 20, and a crewed Soyuz mission is scheduled for July 2026. But without that service cabin, neither mission seems likely. Hopefully, the Russians will be able to get things tidied up quickly, but it might not matter anyway since there’s currently a bit of a traffic jam at the ISS.

We saw a really nice write-up over at Make: Magazine by Dom Dominici about his impressions from his first Supercon visit. Spoiler alert: he really liked it! He describes it as “an intimate, hands-on gathering that feels more like a hacker summer camp than a tech expo,” and that’s about the best summary of the experience that we’ve seen yet. His reaction to trying to find what he assumed would be a large convention center, but only finding a little hole-in-the-wall behind a pizza place off the main drag in Pasadena, is priceless; yes, that mystery elevator actually goes somewhere. For those of you who still haven’t made the pilgrimage to Pasadena, the article is a great look at what you’re missing.

And finally, we know we were a little rough on the Russians a couple of weeks back for their drunk-walking robot demo hell, but it really served to demonstrate just how hard it is to mimic human walking with a mechanical system. After all, it takes the better part of two years for a new human to even get the basics, and a hell of a lot longer than that to get past the random face-plant stage. But still, some humanoid robots are better than others, to the point that there’s now a Guinness Book of World Records category for longest walk by a humanoid robot. The current record was set last August, with a robot from Shanghai-based Agibot Innovations going on a 106-km walkabout without falling or (apparently) recharging. The journey took place in temperatures approaching 40°C and took 24 hours to complete, which means the robot kept up a pretty brisk walking pace over the course, which we suppose didn’t have any of the usual obstacles.

hackaday.com/2025/12/07/hackad…

I wonder what D&D would look like without ability modifiers?

AD&D didn't really have them, they were scoped so differently. I've seen a lot of OSR stuff use them these days, so much OSR pulls from OD&D and also 3rd edition for some reason. I feel like the scoping of them as linear is weird, especially with the standardization of the d20, so commonly people are getting +3 to +4 at level 1, and then need a +1 from its abilities to represent skill. So 1/5th to 1/4 of the die's spread is covered. So a 10 is "easy" and a 15 is "normal". And then it makes this giant chasm between "skilled" and "unskilled". So a level 8 wizard with 8 strength might have a 30% chance of failing to climb a wall a level 1 fighter with 18 strength would have a 50% chance of scaling. But to keep pace it means your characters are quickly getting +10s or more, or +0s, so your DCs kinda fall into "impossible or trivial" with no other options.

Similarily there's a bunch of design space you could in theory touch with the 3-18 scale using separate dice or using some tricks around the weighting at the top or bottom of the scale. I dunno, it's a thing I noodle over a lot.

USB Video Capture Devices: Wow! They’re All Bad!!

[VWestlife] purchased all kinds of USB video capture devices — many of them from the early 2000s — and put them through their paces in trying to digitize VHS classics like Instant Fireplace and Buying an Auxiliary Sailboat. The results were actually quite varied, but almost universally bad. They all worked, but they also brought unpleasant artifacts and side effects when it came to the final results. Sure, the analog source isn’t always the highest quality, but could it really be this hard to digitize a VHS tape?

The best results for digitizing VHS came from an old Sony device that was remarkably easy to use on a more modern machine.
It turns out there’s an exception to all the disappointment: the Sony Digital Video Media Converter (DVMC) is a piece of vintage hardware released in 1998 that completely outperformed the other devices [VWestlife] tested. There is a catch, but it’s a small one. More on that in a moment.

Unlike many other capture methods, the DVMC has a built-in time base corrector that stabilizes analog video signals by buffering them and correcting any timing errors that would cause problems like jitter or drift. This is a feature one wouldn’t normally find on budget capture devices, but [VWestlife] says the Sony DVMC can be found floating around on eBay for as low as 20 USD. It even has composite and S-Video inputs.

For an old device, [VWestlife] says using the DVMC was remarkably smooth. It needed no special drivers, defaults to analog input mode, and can be powered over USB. That last one may sound trivial, but it means there’s no worry about lacking some proprietary wall adapter with an oddball output voltage.

The catch? It isn’t really a USB device, and requires a FireWire (IEEE-1394) port in order to work. But if that’s not a deal-breaker, it does a fantastic job.

So if you’re looking to digitize older analog media, [VWestlife] says it might be worth heading to eBay and digging up a used Sony DVMC. But if one wants to get really serious about archiving analog media, capturing RF signals direct from the tape head is where it’s at.

Thanks to [Keith Olson] for the tip!

youtube.com/embed/OTOChbbTRgs?…

hackaday.com/2025/12/07/usb-vi…