Revisiting Using AI Coding Assistants: You’re Holding It Wrong Edition

June 8, 2026 by Maya Posch 100 Comments

After scathing accusations of skimping on due diligence, as well as other feedback to my article on trying to use an ‘AI coding assistant’ for the first time, the only rational, academic response is to lick one’s wounds following a particularly bruising peer review and try to address the raised issues. Reality after all does not care about one’s feelings, and there may be more to this AI assistant technology that can be coaxed out with a more in-depth look.

To this end I’ll do my best to try and work through each raised point, criticism and accusation, to see what I – and perhaps others – can learn of this endeavor. Said points include the use of the wrong frontend – i.e. Copilot – and the wrong model – being Claude Haiku 4.5 – as well as the egregious flaw on my end of ‘prompting wrong’.

For the sake of due diligence the best frontend and models will be investigated for particular tasks, with finally the verbal minefield of ‘prompt engineering’ examined for industry-standard approaches.

Continue reading “Revisiting Using AI Coding Assistants: You’re Holding It Wrong Edition” →

Repairing A Apple PowerBook 150 With Serious Issues

June 7, 2026 by Maya Posch 2 Comments

Sometimes you purchase an old device that is very cheap for a good reason. So too with the 1995 Apple PowerBook 150 that [Hugh Jeffreys] purchased for a single Aussie buck back in 2018. After finally taking it off the shelf recently, the issues are very apparent. Without even trying to turn it on, the visible damage ranges from the display that’s gone full vinegar with wolverine scratches, to the snapped hinge. Naturally the HDD also turned out to be dead.

Without a functioning display there was little point in continuing, so the disassembly started there, revealing many broken plastic clips. The cause of the vinegar symptom is the degrading polarizer, which with some finesse can be removed like a thick screen protector. Fortunately, here it’s put on top of the glass layer of the display, so after peeling it off the remaining glue can be safely dissolved and scraped away.

Inside the case the RTC battery was found to have started leaking, causing corrosion and damaging a variety of important traces for the keyboard and display. All of this damage seemed fixable, but after a while the damage was just too severe. Fortunately he was able to obtain a replacement for the affected daughter PCB, which allowed the display to come back to life, so that a new polarizer could be installed after cutting a large sheet down to size.

A replacement hinge was then printed in PETG and glued to the part of the lid where it had broken off, while snapped plastic clips were reinforced with glue where they had hung on. Finally, the IDE HDD was replaced with a CF card via an IDE adapter and the entire system reassembled.

Unfortunately [Hugh] wasn’t able to immediately source or create MacOS floppies with a version that the laptop wanted to install from, so that part couldn’t be tested yet, but there’s a good chance that this old PowerBook 150 has finally been cured of at least its biggest ills, without spending much more than the original asking price.

Continue reading “Repairing A Apple PowerBook 150 With Serious Issues” →

Desalinating Seawater With Solar And No Brine

June 7, 2026 by Maya Posch 39 Comments

Although desalination is very commonly used these days to convert seawater into fresh water, one of the major disadvantages of current approaches is that commercial desalination plants produce a lot of brine, which has to be dumped somewhere ideally without causing major environmental issues. A new solar-thermal method as demonstrated by [Luheng Tang] et al. was published in Light: Science and Applications, with accompanying PR article.

This method is claimed to require no pre-treatment or leave brine, using special panels that wick water across their surface and then use solar radiation to distill this water. This differs from previous similar methods through a special surface treatment that prevents build-up of salts which would require cleaning or replacement.

The salts and other contaminants that would normally end up in the brine slough off these cells and can then be further processed to recover everything from plain table salt to lithium as well as gold, uranium and other substances of interest that are prevalent in seawater.

So far these self-cleaning cells have been tested with water from a number of oceans with a claimed 74% solar-to-vapor conversion efficiency and nearly 100% salt extraction. As always the challenge will be in scaling this up to industrial levels, but so far it looks promising.

Building A Gifford-McMahon Cryocooler With 3D-Printed Parts

June 6, 2026 by Maya Posch 12 Comments

Although cryocoolers are capable of pretty impressive cooling, for many of them the underlying working principle is simple enough that you do not need any special skills or a big budget to make your own version. Take the Gifford-McMahon cryocooler for example, which works using nothing more than some kind of coolant gas and a piston in a cylinder that you can even 3D print, as demonstrated by [Hyperspace Pirate] in a recent video.

The lowest temperature reached across the two prototypes was only -84°C, but this was mostly due to some sub-optimal design choices, such as the use of regular air and a clear acrylic tube to get a good glimpse at the inner workings. The trickiest part of this type of cryocooler is probably that you need to move the piston containing the regenerator between both ends of the cylinder to get a cool and a hot side.

That particular problem was solved by using magnets to move the piston externally, which worked beautifully until the problem of using regular compressed air from the shop compressor caused massive ice formation that jammed up the piston. Obviously this was not an unexpected issue, and for the next step the coolant gas will be replaced by helium, as making that gas freeze up requires quite a bit more effort.

Continue reading “Building A Gifford-McMahon Cryocooler With 3D-Printed Parts” →

Stratasys Vs Bambu Lab: Industrial Vs Consumer ABS Showdown

June 6, 2026 by Maya Posch 15 Comments

The test parts being printed on the Stratasys Fortus 450mc. (Credit: My Tech Fun, YouTube)

Professional Stratasys FDM printers demand a pretty hefty price premium over your typical hobbyist-level machine, with the gold-plating continuing even with the special filament cartridges that you buy for some of their printers.

This raises the question of in how far this eye-watering price tag is justified, and how much is just you paying for support and the brand name. After acquiring a spool of Stratasys ABS filament via a US viewer, [Dr. Igor Gaspar] set to work to try and answer this question.

The viewer had already liberated the spool of ABS+ P430 filament from its cartridge, making it easy to use that directly with the Bambu Lab FDM printer.

To make it a fair comparison, [Igor] also needed to have a sample printed on a real Stratasys printer, for which he used a local company’s services. An interesting sidenote here is that the US viewer’s company moved away from Stratasys to Bambu Lab printers.

[Igor] was able to see his test parts being printed on the Stratasys printer, as said company is in the same city. This showed him that it took 14 hours to print the parts versus 3.5 hours on the Bambu Lab printer, suggesting that his worries about the right printing parameters for the Stratasys filament were warranted. Sussing those out was thus paramount for a fair comparison and warranted some test prints.

From a sheer aesthetic point of view the Stratasys-printed parts looked much cleaner, and their dimensional accuracy was also significantly better due to the slicer adjusting for this. Between the used Stratasys M30 and Bambu Lab ABS filaments there’s no clear winner, with both trading blows. Amusingly enough, the older Stratasys ABS type in the form of the ABS+ P430 filament performed the best of all when printed on the Bambu Lab printer at its preferred temperature setting.

Moral of the story is thus that – unless you really want to pay for that service contract – to loot old Stratasys ABS spool cartridges and use them in your hobbyist FDM printer. As [Igor] says in the conclusion, the nicer looks is probably due to them printing very thin layers, much finer than the 0.2 mm layers he used. This would also match the much longer print time and is thus something we can replicate on any FDM printer with a temperature-controlled printing environment.

Continue reading “Stratasys Vs Bambu Lab: Industrial Vs Consumer ABS Showdown” →

Using Brand New NiMH Cells After Sitting 12 Years Unused

June 5, 2026 by Maya Posch 25 Comments

You know your batteries are old when their labels have faded. (Credit: DiodeGoneWild, YouTube)

After finding a pack of NiMH rechargeable cells that had never been used since buying them in 2014, [DiodeGoneWild] decided to test whether they could be tossed or not. After previously testing different brand cells that had gone high internal resistance after only about five years, he wasn’t expecting much. Amazingly, the batteries not only recovered, but seems to be not that much worse off for wear.

Three of the four precharged cells still held some voltage and happily charged back up to their rated 2,000 mAh capacity basically with the first cycle. One of them read 0V initially, but was revived using the typical manual charging approach involving a bench power supply. After a few charge-discharge cycles only the deep discharged cell showed some noticeable degradation with slightly reduced capacity, but all of them read healthy internal resistance values.

What this mostly shows is that not all NiMH cells are made the same, with the Tronic ones that previously failed after a few years doing much worse than these Activ Energy cells which are apparently sold primarily at Aldi stores. Overall NiMH is a pretty robust battery chemistry, so it’s always worth it to try reviving a cell before tossing it.

Continue reading “Using Brand New NiMH Cells After Sitting 12 Years Unused” →

Investigating The S3 Virge’s Reputation As A 3D Decelerator Card

June 5, 2026 by Maya Posch 15 Comments

The special 512x384 mode with S3 card installed. (Credit: Bits und Bolts, YouTube) — The special 512×384 mode with S3 card installed. (Credit: Bits und Bolts, YouTube)

Back in 1996 the 3D gaming market on PC was beginning to heat up, with hot new titles like Tomb Raider coming out that year and requiring much more graphics power than what was needed for old titles like Doom and Duke Nukem 3D to experience good graphics. Thus you had to pick some kind of 3D accelerator card to buy. Here a common joke was that of the available options, the S3 Virge GPU was so bad that it was actually worse than running in software rendering, but was this true? Cue [Bits und Bolts]’s investigation to finally put this myth to rest.

On software rendering mode a zippy Pentium 166 would struggle to render at 640×480 resolution, so if you wanted more than 320×240, or really knock down graphical fidelity, you had to get that 3D accelerator card. After combining a P166 with an S3 Virge/DX – a minor update to the original Virge – the Tomb Raider game was first compared while running in 512×384 resolution, which the game offers you with an S3 card installed along with bilinear filtering.

After hitting a capped 30 FPS on that first test, 640×480 was tried and hit a solid 15 FPS with bilinear filtering enabled, but the conclusion is basically that the special 512×384 resolution mode is pretty good. Perhaps the main causes of the myth was the wide variability in quality of the various GPUs using the S3 Virge chip, as well as trying to run at anything other than this special resolution which appears to target the card’s strengths.

Continue reading “Investigating The S3 Virge’s Reputation As A 3D Decelerator Card” →