An anonymous tipster pointed us to the forums of Finnish Mac-modding site Mak Tek where we found the image above. Yes, that really is Leopard running on a Tomtom satnav. (We think it’s a Go but it’s a little hard to tell from the photo.)

From what we can tell from the original post and the poster’s comments in the thread which follows (all gleaned via Google translate — any of our readers speak Finnish?), this feat was achieved using a version of Apple’s opensource Darwin compiled for the ARM CPU inside the Tomtom. This is all getting a little over my head so it’s over to Æ’s resident code monkey Harry Tannenbaum for the expert analysis:

“Yes, this is really stupid. I mean, sure, it’s possible to compile XNU for ARM — that’s what Apple does for the iPhone and iPod Touch and all the necessary defines and processor specific stuff is in the open source releases — but are we meant to believe that they (whoever “they” are — looks like the site’s down right now) also managed to get the whole GUI stack running on a CPU it’s never been compiled for? Give me a break.

“If you’re going to claim you got any OS X running on a Tomtom you should go for the iPhone OS. I mean, it even has a touch screen like the iPhone and with the GPS and everything you might even have gotten a couple of people to believe you.”

In unrelated news, Æ will no longer be paying a no-questions-asked flat fee for un-confirmed rumours and sneak peek photos.

• MakTek Forums

I can probably claim to have used just about ever computing platform under the sun with only the smallest probability of some harder-core geek tripping me up by name-checking a system so obscure that I’ve never given it a go. I tell you this so that when I say that the Mac continues to provide me with the best computing experience you understand in which context I speak. But that’s not to say it’s perfect.

As a geek I occasionally find myself with an itch which the Mac just won’t let me scratch. I speak, of course, of the joys of self-build. In the same way that you can’t really call yourself a true petrol head unless you’ve owned at least one Alfa Romeo, you haven’t truly earned your geek spurs until you can point to a humming tower case, fully booted and tweaked to within an inch of its life, and say, “I built that.”

I shall attempt to labour the automotive analogy a little longer (or at least until the wheels fall off). The Mac is a prestige brand, an Audi or Mercedes or BMW. When you buy one you buy the complete package. You wouldn’t expect to strip down and rebuild your Mac any more than you’d expect to stick a body kit and ridiculous rear spoiler on your A6. (Hmm… is this working? Time to change gear.)

Yesterday’s article about the mysterious “Brick” touched upon the subject of the xMac, a fabled beast which it seems at time exists only in the imagination of readers of ArsTechnica. It is the mythical mini-tower Mac: unlike the iMac it is upgradable, and unlike the Mac Pro it is affordable to mere mortals.

So what I’m currently wondering is: has the xMac’s time finally come. Let’s take a look at the signs. Apple’s move to Intel was successfully completed some time ago. Since Mac innards are now basically the same as PC innards many of the barriers to a box which can make use of standard PC components are gone. Drivers are always a problem, but as the Mac’s market-share continues to expand it doesn’t seem unreasonable for PC component manufacturers to invest the little time needed to produce them, in the same way they have started to for Linux, and in the way printer and other accessory manufacturers have for years. (Given what joy Apple’s IOKit driver development system is to work with I suspect that once word gets out that one vendor has tried it, found it not at all painful and begun to reap the reward, others should quickly follow.)

Those of us who have dabbled in the Hackintosh scene can experience a little of the possible xMac thrill, running (totally illegally, so don’t try it at home) OS X on standard PC hardware. However, the emphasis has always been more on finding compatible kit, rather than allowing your tweaking and crazy-powerful-system-building talents run wild. (The Hackintosh scene is also a source of drivers for various bits of non-Apple kit, which gives those component vendors at least a toehold on which to build.)

Of course, the key to all of this is a willingness on Apple’s part to create the machine in the first place, sacrificing in one stroke both repeat sales (as non-upgradable Macs reach the ends of their working lives) and its own position as vendor of the majority of Mac-specific add-ons and upgrades (although we’ve all smart to their RAM prices these days, aren’t we?) Although the chances of them letting us paint go-faster stripes down the side of their perfect bodywork seem slim, it can’t hurt if we all only cross our fingers and hope they see the light.

• image by 37prime

While before last week we might not have been expected to know who Mr Wei-han Lien was and what he did (what? you still don’t know? Here, remind yourself), the fact that Apple has an ARM CPU Architecture Team shouldn’t come as a surprise to anyone.

The ARM family is pretty-much ubiquitous nowadays. There are probably about two dozen ARM chips within five meters of you at this very moment. Unless you’re out in the middle of the desert, in which case there will only be several. Like sand they get everywhere: your phone, iPod, router, simply everywhere.

A couple of factors have been key to this success. The ARM’s low power drain is certainly one of them, but equally so is its modular design. ARM itself doesn’t get its hands dirty with the mucky business of actually fabricating the chips. Instead they licence their designs for others to make. These Original Design Manufacturers chose what extras they’d like on the same piece of silicon, which in turn means fewer components to achieve the same goals.

The ARM chip came out of the British home computing boom of the early Eighties. In many ways its all that remains. Originally called the “Acorn RISC Machine”, it was designed by Acorn Electronics to power the successor to their classic BBC Micro. A generation of British school kids will remember computing labs full of Acorn Archimedes computers. They may have lacked the glamour of the Atari STs and Commodore Amigas in their bedrooms at home, but this really was an amazing machine. Despite what Apple would later claim, it was the Archimedes and not the PowerMac which was the first desktop RISC computer.

You wouldn’t think Apple would make a mistake like that, given how involved in the history of ARM they were. They began working with Acorn in the late Eighties, and when the renamed “Advance RISC Machines” was created in 1990 they held a significant share in the company. The ARM CPU would go on to power first the ill-starred Newton and then the moderately more successful iPod in all its incarnations. Twenty years later and it looks like the Apple-ARM relationship is still as close as it has ever been.

• Image by ‘esquareda‘