Mysterious Files PH

OLED displays solve many of the problems suffered by LC displays, including color fidelity, dynamic range and power usage. That said, especially in the early days OLED gained a reputation for dim screens, short lifespans and burn-in. Over time better organic dyes were developed, along with burn-in prevention methods that have made OLEDs much closer to LCDs in terms of longevity. In a recent comparison between OLED TVs by RTings it’s however clear that between 2017 and 2023 there haven’t been any major advances beyond bumps in brightness.

The relatively dim screens were a major problem, as they prevented OLEDs from displaying HDR content. This issue has been well and truly addressed, as confirmed by RTings’ testing, but after an over 10,000 hours stress test that simulates about 10 years of regular use at maximum SDR brightness, especially static elements like the CNN TV banner happily burned in even on the newest models with all burn-in prevention measures enabled.

Here the biggest take-away is probably that even if the expected panel lifespan at full brightness is still the same, this higher brightness budget means that you can gain some lifespan by cranking the brightness way down. It’s also essential to keep features like pixel refresh cycles enabled, as demonstrated by [Hardware Unboxed] and their abuse of a QD-OLED monitor where a worst-case 6,000 hour stress-test managed to create some impressive levels of burn-in from uneven subpixel wear.

Aside from global access to cat videos, the presence of thousands of Starlink broadband internet access satellites in LEO has a very pleasant side effect for atmospheric researchers. Starlink publicly publishes near-real-time ephemeris data on its individual satellites. From this data you can deduce many details about the atmosphere at that altitude, including its density at specific altitudes at specific times, information which otherwise would be very hard to gather. Recently, this allowed [Mamoru Yamamoto] to determine the density of the thermosphere using tomography.

In a similar 2025 paper by [Zhuoliang Ou] et al. as published in Remote Sensing, this same data source was used to investigate details of the thermosphere. With Starlink publishing this data since 2021, this provides an invaluable dataset for studying this outermost part of the atmosphere.

Commencing just before the generally recognized transition into ‘space’ at 100 km altitude and below the Earth’s exosphere, the thermosphere‘s thickness fluctuates due to factors like solar irradiation and, with it, the exact altitude at which the exosphere begins. Generally, though, it is well above 600 km altitude. This places Starlink satellites as well as both active space stations (ISS and Tiangong) in the thermosphere.

[Zuholiang Ou] et al. established that the Starlink data matches well with that from a dedicated research satellite like SWARM-B, thus making it a useful source of scientific data.

The innovation in [Yamamoto-san]’s paper is that instead of using the typical two-line element (TLE) set, a more comprehensive tomographic approach was used, which essentially uses more data for a larger reconstruction, with the resolution claimed to be about on par with that of the SWARM satellites. This implies that although these Starlink satellites were never designed to be more than data relays, they may have accidentally become the biggest development in thermospheric research in a long time.

Of course, you can’t please everyone.

FTP isn’t exactly cutting-edge technology. These days, if you control both ends of a connection, you’re probably using scp, SFTP, rsync, or something even fancier. But FTP refuses to die, especially if you are perusing old public FTP servers or talking to retrocomputers. Every now and then, you still need an FTP client. Naturally, there are plenty of graphical clients. But some of us would rather stay at the command line. You could just type ftp, of course. It works, and if you haven’t used it lately, it is probably better than you remember. However, I’ve long been a fan of NcFTP. While some other FTP clients have caught up, it still has unique features that make FTP a lot more productive.

Not Your Father’s FTP

Before maligning the standard ftp command, though, we should point out that it probably isn’t the FTP client you remember from 30 years ago. For example, on openSUSE Tumbleweed, /usr/bin/ftp is really tnftp, a portable version of NetBSD’s enhanced FTP client. Debian uses it too; the ftp package in both Bookworm and Trixie leads you to tnftp. Since current Raspberry Pi OS is based on Debian Trixie, you’ll encounter tnftp there, too. That’s significant because tnftp has already fixed many of the irritations you might associate with old-fashioned FTP.

You get command-line editing, history, and filename completion, things that are also in ncftp. Both understand passive FTP and IPv6. The tnftp client can also retrieve HTTP, HTTPS, and file: URLs, so commands such as:

ftp https://example.com/something.tar.gz

aren’t necessarily typos, although ncftp lacks this ability. But ncftp does have some killer features.

Remember Me?

One of NcFTP’s nicest creature comforts is bookmarks. Connect to a machine, move to a useful directory, and save it:

ncftp /pub/micros> bookmark oldstuff

Then later you can simply type:

ncftp oldstuff

The bookmark can remember more than just the hostname, making frequently used FTP sites feel much more like named resources than anonymous servers you repeatedly have to navigate.

NcFTP also maintains a cache of remote directory listings. If you’ve ever used FTP over a slow link, you know how annoying it is to ask for the same directory listing over and over. NcFTP can often work from what it already knows instead. Neither feature sounds earth-shattering, but together they make an interactive FTP session considerably more pleasant.

Get All The Things

Another difference becomes obvious when you want an entire directory. NcFTP supports recursive transfers:

get -R foo

or:

put -R foo

That seems obvious if you’re accustomed to modern tools, but traditional FTP is fundamentally organized around transferring individual files. NcFTP does the tedious directory walking for you. It also handles resuming interrupted transfers more naturally, something particularly welcome when the file in question is a multi-gigabyte disk image rather than README.TXT. With tnftp, you have to explicitly ask to resume an interrupted file. NcFTP will detect it and, depending on configuration, either resume or, at least, offer to resume the transfer.

Go Away, I’m Busy

NcFTP also has a clever background-transfer system. Commands such as:

bgget giant-file.iso

Hand a transfer to NcFTP’s spooler rather than tying up your interactive session. There are corresponding facilities for uploads. That’s an interesting distinction from simply detaching a shell command. NcFTP knows that this is a transfer job and maintains a queue of FTP work that can be retried and processed independently.

Shell Games

But perhaps the biggest reason to know about NcFTP is that NcFTP isn’t just one program. The package includes commands such as ncftpget, ncftpput, and ncftpls. These perform FTP operations directly from the Unix shell without starting an interactive FTP command interpreter. For example:

ncftpget ftp.example.com /tmp /pub/widget.bin

or:

ncftpput ftp.example.com /incoming widget.bin

This is much nicer in a script than sending commands to ftp using, for example, a here document and automating login with .netrc. For example:

ftp <<EOF
open ftp.example.com
cd incoming
put widget.bin
quit
EOF

Sure, it works, but any time you send input to an interactive program it is, at best, messy. The ncftpput program expresses what you actually wanted to do in the first place: put this file there. That’s much more Unix-like.

Don’t Do This At Home

None of these conveniences change FTP’s fundamental problem: ordinary FTP is not secure. Usernames, passwords, and data can travel without encryption. If you’re designing a new system and control both ends, you usually have much better choices. But sometimes you don’t control both ends. If FTP is something you run into, ncftp is worth knowing about. Bookmarks, cached directories, recursive and background transfers, and especially the script-friendly companion commands turn an antique protocol into something that feels surprisingly at home on a modern Unix command line.

Of course, just as you can use FUSE to mount an ssh server, you can use ftpfs, to make a remote server look like part of your file system. You never know when FTP is going to crop up.

In the 1960s, the home computer was barely a twinkle in anyone’s eye. This was the decade in which computers were used by a handful of companies and organizations and took up entire rooms. It was the decade that saw the birth of the first integrated-circuit computer, the Apollo Guidance Computer, and, rather strangely, one of the first home computers of sorts: the Digi-Comp 1.

In 1963, E.S.R. introduced this little programmable educational toy computer. It relies on simple mechanical logic gates, programmed by positioning tubes on sliders. Their original intention was to create actual computers, but one thing led to the next, and E.S.R. became the premier educational computer toy company of the 1960s. The Digi-comp is a rather simple device. A program is entered by soda-straw-like tubes placed on sliding rails. These tubes interact with spring-loaded levers which move in accordance with the “clock” switch. The result is a simple binary computer you can program.

[Chris Staecker] got a remarkably well-preserved Digi-Comp I to demonstrate its unique functioning. With some rubber bands and a touch of modern technology in the form of 3D printing, [Staecker] got the toy working like new. The programming process is quite manual, and its functioning is not obvious. But it can compute binary outputs of given functions. Unfortunately, this is not a Turing-complete process, but rather a state machine with 8⁸ (or 16,777,216) possible states. The toy, with all its limitations, is highly reflective of the state of computing at the time, and a rather neat look at what the future was, back in 1963.

Make sure to take a look at E.S.R.’s successor to the Digi-Comp I, the creatively named and quite different Digi-Comp II. We’ve looked at other old “sort of” computers like the Geniac. You can even simulate that one in your browser.

Usually, the name ‘Famicom’ and the associated Nintendo Entertainment System (NES) don’t exactly bring to mind downloadable content and online services. However, at least in Japan, a modem add-on for the Famicom offered those features. Released in 1988, the Family Computer Network System (FCNS) offered all that using its modem to dial into Nintendo’s servers. Recently, [Throaty Mumbo] got his mitts on one of these units to try and see what you can still do with it in 2026.

Although only mildly successful in the Japanese market, Nintendo still sold 130,000 of these modem add-ons, with tens of thousands of users for the stock brokering, banking, and Super Mario Club services offered. The experiences with the FCNS would form the foundation for Nintendo’s Satellaview for the Super Famicom and from there into today’s Nintendo Online offerings.

Of course, the main challenge with reviving the FCNS in 2026 is that the remote servers were turned off decades ago, along with the minor detail that to do anything with it you need to insert a special card. This was missing in the used version that [Throaty Mumbo] bought, but fortunately he got a workable Super Mario Club card. The next struggle? He doesn’t speak Japanese, so it took a while to eke out the details.

With a phone line simulator, it’s at least possible to try to have the FCNS try to dial, but that only throws another error. Ultimately it turned out that the FCNS uses NTT’s proprietary DDX-TP (in Japanese) protocol, which isn’t really documented outside of some Japanese sources. This is where the NES Dev site and its forum thread full of FCNS enthusiasts proved essential, as they had already done all the reverse-engineering.

NES Dev user [Ben Boldt], for example, made a DIY version of the FCNS access (tsuushin) card that lets you configure its details. For development, the PLCC-based CPLD was a bit of an annoyance, so the FCNS got somewhat modified along the way for more direct access.

After five months of work, he was finally able to connect the Famicom via the FCNS and a telephone line simulator for some Super Mario Club action. This connects to the SMC Server simulator on the PC, with the detailed manual that comes with the access card helping a lot.

After all this, the Super Mario Brothers game was graced with a game patch that allows for network-based SMB two-player action. This adds a ‘dial’ and ‘answer’ option so that the two Famicoms can connect and commence the two-player session. Although one can question the latency this would entail without a direct POTS connection as here, it is definitely a nifty demonstration of what ultimately would become the multiplayer gaming on Nintendo consoles today.

You can still do Famicom programming. There were even clones.

We’re no strangers to unusual hardware failures around these parts, but even so, a swarm of jellyfish clogging up the works is a new one to us. That’s exactly what happened to the cooling system at the Gravelines nuclear power plant in France earlier this week, resulting in three reactors having to be shut down. This isn’t even an isolated incident, as POLITICO points out that the same thing actually happened last year around this same time.

So why are jellyfish staging an annual protest against nuclear power? A warming of the North Sea has extended breeding seasons and produced larger populations of the plankton that the jellyfish feed on. This is great for the gelatinous sea creatures, but not exactly ideal if you’re trying to pump the water they live in through a complex cooling system. The French installed filter screens and monitoring systems after this happened in 2025, but clearly they’re going to need to keep working on the problem.

From a new problem to a very old one, a BBC investigation has revealed that the UK’s National Health Service (NHS) Blood and Transplant department was inadvertently leaking private patient data by sending it in the clear over the pager network. Sniffing pager messages was fairly trivial decades ago, and is even easier these days thanks to the proliferation of cheap software-defined radios and open source decoding software. Given the fragility of the underlying communication network they use, the NHS announced they were going to stop using pagers by 2021, but clearly not everyone got the memo.

As an aside, we were amused to see that the BBC had to devote a section of the article to explaining what a pager is and why they became popular in the medical field. Of course, it makes perfect sense. Many — perhaps even most — of the people reading the article would never have seen one of the devices in the real world, let alone actually used one. Just a sign of the times, and a great reminder of how old some of us are getting.

Speaking of getting old technology, we came across this web-based emulator for the Flight Data Subsystem used by the Voyager spacecraft. The website lets you write your own assembly code or load up some example programs and watch it execute at a blistering 806.4 kHz. The site also features a real-time display of how far Voyager 1 is from Earth, the current communications delay, and a trove of original Jet Propulsion Laboratory documentation, so even if you’re not interested in writing assembly code for a 1970s computer, it’s worth checking out.

On the subject of running old code in the modern web, you can now play DONKEY.BAS in the browser. The game, now 45 years old, was written by Bill Gates and Neil Konzen to demonstrate the graphics and sound capabilities of BASIC on DOS. For the uninitiated, despite what you might think from the name, the game actually tasks the player with controlling a race car. The donkeys are standing in the road for some reason, and you need to avoid them by switching lanes. With the way the industry is now, we wouldn’t be surprised to hear somebody is working on a gritty origin story film to expand the lore.

Finally, turning our attention to slightly more recent history, [Marcin Wichary] has written a fascinating retrospective on Apple’s Touch Bar. The thin touch screen was introduced in 2016 as a replacement for the top row of the keyboard on the MacBook Pro, and was the subject of some impressive reverse engineering efforts by hackers that wanted to use the unique high-resolution display in their own projects. Apple quietly abandoned the feature in 2023, and [Marcin] makes the case that the problem was never the hardware itself but rather how the feature was implemented in software.

The article includes numerous examples of the redundant and at times downright bizarre user interface elements that would pop up on the Touch Bar, made all the more perplexing given Apple’s usual panache for such things. Although we’re not necessarily convinced the feature ever made a whole lot of sense, given the examples [Marcin] has collected here, it certainly seems like the poorly implemented software side of the equation wasn’t doing it any favors.

See something interesting that you think would be a good fit for our weekly Links column? Drop us a line; we’d love to hear about it.

It used to be a rite of passage to build some sort of AM radio receiver. Many people started with a crystal radio, of course, but you’d graduate to maybe a TRF or superhet design. FM radio projects were not as common, because demodulating an FM signal was harder, as was making things work in the 100 MHz range. These days, though, you can get everything on a chip like the TEA5767, and [turtushig22-blip] has an open design that uses that device on a breakout board, a display, a knob, and an external amplifier. An ESP-32 pulls it together. You can see two videos about the project below.

Unlike older FM radio designs, the TEA5767 doesn’t require any adjustments. RF goes in, and stereo audio goes out. You can control the device through I2C or a 3-wire serial interface.

Software-wise, you send a five-byte control word and the chip sends back five status bytes. You don’t set the frequency directly, but provide a PLL divisor. For example, with a 32 kHz reference crystal, setting 100.1 MHz would be: 4*(100.1 MHz + 0.225 MHz)/0.032768 MHz = (about) 12247.

The radio uses a 225 kHz IF and can autonomously seek stations. You can select whether the search is up or down, and the minimum acceptable signal strength. One unusual feature: the TEA5767 can select the positive or negative mixer output, which might be useful for image rejection in some cases.

The output is purely analog, and there’s no RDS/RBDS decoder. So don’t expect station information on the OLED display. Still, a nice, easy FM radio build. Naturally, not the first TEA5767 build that we’ve seen. Or even the second.