Okay, far be it from us to pass judgment on any of you, but, well, we're doing it anyway: some of you need serious help. A handful of faithful viewers (who shall remain nameless) have indicated that they're jonesing for dirt on the filthy secret lives of interconnected nodes in massively-parallel clustered computing environments. Not to blow our PG rating or anything, but apparently their misguided assumption is that when you cram several hundred wickedly fast systems together in a series of custom-designed racks and hook 'em up with a ton of internode bandwidth, it's a good bet that as soon as the cleaning staff leaves for the night the whole place turns into the computational/virtual equivalent of a scene from Caligula. That's as far as we're walking that scary path, other than to reiterate our recommendation for intense psychiatric care-- and to acknowledge that while the 1100 G5s destined for the Virginia Tech "Terascale Computing Facility" may not necessarily be promiscuous, they are undoubtedly cheap.

See, a wealth of details about that formerly hush-hush Virginia Tech G5 supercomputer have finally been made public; Think Secret was right last week when it claimed there was a campus informational meeting about the project slated for yesterday, and Mac Rumors points to detailed notes from the presentation available at Chaosmint. Sadly for some depraved viewers, there's precious little about the licentious activities those G5s get up to behind closed doors, but Virginia Tech has revealed that it went with Power Macs because they were the least expensive available option; they provide the most (pardon the expression) "bang for the buck."

Strange but true! Virginia Tech considered Sun's SPARC systems, but not for long; a SPARC-based cluster with enough oomph would have "required too many processors" and been "too expensive." How about IBM systems running AMD's Opteron, or HP units based on the Itanium? Nope; for the necessary performance, both options would have required double the number of processors and cost twice as much. Even Dell was rejected as "too expensive," which is a little alarming given how far they've been willing to slash their prices to win contracts in the past. And don't believe for a second that Dell didn't try; reportedly the reason this whole project was top secret in the first place was because "Dell was exploring pricing options during bidding" ever since February. In the end, evidently they just couldn't get prices down far enough to close the deal.

Once the dust had settled, only one vendor was still standing: Apple, whose Power Mac G5 represented the "lowest price" available from a pure "cost vs. performance" standpoint. Even if Virginia Tech paid full educational price for those 1,100 Power Macs (and you just know they didn't), the bill would "only" come to a bit under $3 million; the school is paying $5.2 million for the entire project. Did you ever think for even a split second that you'd live long enough to see Apple win a contract based on price? Because we sure didn't, and to us, that's more exciting than the thought of 1,100 aluminum G5 towers getting up to virtual shenanigans when no one's looking. Marginally.

Sick of wondering just when the heck new PowerBooks are going to show up? Of course you are-- so is everybody else on the planet with a pulse. Don't worry, we're not even bringing it up. At least, no more so than to say we're not bringing it up. And, um, also to tell you that we only brought it up to say we're not bringing it up. The way people have been harping on the issue, you'd think Apple didn't make anything except PowerBooks, when, as everyone knows, they also make a killer slab of fruit leather. Oh, and lots of other computer-type thingies. So instead of waiting for pro portables that probably won't show up until President Kutcher throws out the first pitch of the Hover Baseball season, let's talk about the other products in the on-deck circle at One Infinite Loop and when we might finally see that stuff.

First up: Mac OS X 10.3, known in more dramatic circles as "Panther." This has been an issue for a while, as you well know; news reports prior to the operating system's official introduction pegged the product for a mid-September release, but in June His Steveness stated only that it'd be available before the end of the year. After that, though, rumors swirled that Panther was indeed destined for a mid-September launch after all, specifically at Apple Expo in Paris on the 16th; for a while it really looked possible, given how quickly development had been progressing, but now with less than two weeks until the Paris gig and Panther still in beta, mid-October is looking more and more likely.

Or is that early October? Now AppleInsider is claiming that Apple's official target date for Panther's unleashing is October 3rd. Get it? 10/3? 10.3? Ooooh, it's all clever and stuff! Apparently we're supposed to expect all sorts of cute in-store celebratory tie-ins come the Big Day, so mark your calendars now just in case. Dare we suggest live panthers in every store, ready to pounce on and maul anybody who says that Exposé "isn't actually all that great"?

Meanwhile, what about that wireless mouse 'n' keyboard set whose existence has been confirmed in so many different ways that Paul Simon's about to record a song called "Fifty Ways To Leak Your Input Devices"? Well, despite earlier reports that Apple had hoped to intro them alongside the G5 back in June, Mac OS Rumors now reports that the company will be sitting on them until January-- but hey, at least they'll be all warm and toasty when we finally get them in the dead of winter, right? (Then again, a January unveiling certainly means Macworld Expo, and the dead of winter in San Francisco isn't all that dead after all.)

Lastly, on the "recklessly unconfirmed" front, some guy wearing a powder-blue tux and carrying a pineapple ran past us at the bus station while muttering something about storage bumps to the current iPod line (with the top-of-the-line model hitting the Big 4-0) coming as early as Monday, but as everyone knows, you just can't trust a man with a pineapple. Or can you? Well, no. You can't. Forget we said anything.

Okay, we've been staring at this thing for two days, now, and we're no closer to cracking the mystery; indeed, if anything, dwelling on it has only raised more questions. Ugly questions. Questions like, "why the letter 'U'?" And "why is only the capital one a problem?" And of course the biggie, "just when exactly did Apple start paying Microsoft to come up with utterly baffling and seemingly insane bugs for its products?"

At first we thought it, like the evil monkey that lives in our closet, was just the fevered hallucination of a sleep-deprived mind, but then faithful viewer EelBait insisted that he saw it, too, and other viewers followed: an official support document from Apple's own servers confirming that you can't use a capital "U" in your Open Firmware password, or else said password "will not be recognized during the startup process." It's real. This is a genuine, honest-to-weirdness bug involving, of all things, a restriction on using a specific letter of the alphabet-- uppercase only, of course. Duh.

So now we're starting to wonder about that "hallucinational" monkey, too. This could be bad. (Anyone seen Outbreak?)

You know about Open Firmware passwords, right? If you're the paranoid type and you think evil monkeys are creeping out of your closet in the middle of the night to boot your Mac and mess with all your files (replacing the body of your résumé with an excerpt from Monkey Hamlet, for instance), you can actually set a password which your Mac will require before it tries to boot a disk other than your designated startup volume-- so people (or simians) can't just boot off a CD-ROM or external hard drive and then mount your hard disk and wreak havoc. It's a nice feature if you're the crates-of-Beefaroni-in-the-bomb-shelter type-- as long as you don't insist on using a password with a capital "U" in it.

So why no "U"? Clipboard-wielding denizens of the AtAT laboratory have as yet been unable to find any technical reason for the restriction, which leads us to believe that it's somehow intentional. So far the most plausible explanation we've come up with is that someone at Apple really doesn't want people using the titles of Prince songs as their passwords. Something to ponder over the weekend while you're fending off the monkeys.