FUTURE OF OS/2 REPORT

Booting It Up with Salaryman

Salaryman set his coffee mug on his desk. On a coaster, of course. Salaryman respected his banking firm too much to stain their solid oak desks with his fresh-ground Brazilian coffee.

Smacking his lips, Salaryman hit the space bar on his black keyboard with his index finger and sat back. There was a quick electronic beep and some fans under his gargantuan desk whirled to life. The LCD lit up and Salaryman typed his BIOS username and password.

Username: salaryman
Password: •••••••••

Unseen PowerPC: The Cores That Didn't Make It

Ever since Apple, IBM, and Motorola jumped into bed together twelve years ago, tech media—and Apple watchdogs especially—have had a field day with speculation, rumors, and actual news regarding new PowerPC projects. Apple gave a familiar face and a flair of iconoclast to the affair while IBM lent a grave sobriety. This thing could really happen, then, someone to challenge the Microsoft/Intel duopoly. And the stories just kept coming, well into the next decade. But not all of them were so real.

So what were these projects that came and went faster than a Quad Xeon Mac? Handily enough, they're right below, broken down by project. Read on to find out what Apple, IBM, and Motorola had in store for us throughout the Nineties and what didn't make the cut. Through all the rumors, one thing was certain: AIM never had a lack of imagination, even if it didn't always end up in silicon.

Why Apple Really Ditched PowerPC

Apple wants to make their switch to Intel chips seem like a no-brainer, but the reality of it was a lot more complicated than just faster chips for Macs. Apple's claims of their Intel systems being "4-5x faster" than their PowerPC systems is a little much to swallow, especially with Intel Macs landing in users' hands and failing to live up to the hype. So if these Intel chips aren't really that much faster than the G5, why did Apple make the switch? The answer to this question is a lot more interesting than what Apple's telling you.

The Ninety Nanonmeter Speed-Bump

Last Summer, Apple launched a serious upgrade with the Power Mac G5 and for the first time in a decade used a totally new PowerPC core not based on the PowerPC 603. With IBM's PowerPC 970, Apple is using a mainframe-level chip capable of massive parallel computing, access to hyper bus speed, and huge volumes of cache. Since the industry hit the wall shrinking to 90nm, however, the Mac community has expressed unrest at the clock-starvation: Memories of Motorola's 500 MHz Fiasco five years ago bubble up to the surface.

Kill your worry processes, Mac users. There's no clock-stall in the PowerPC's future any time soon. Motorola's failure to achieve speeds above 500 MHz was a result of it recycling the PowerPC 603 core far too many times, something that IBM avoids in using its Power series core. The speed-bump at 2.5 GHz is the result of a one-shot problem involved in shrinking the die. But let me explain in more detail.

Power Mac G5³

Recently, many in the Mac community have been discussing the possibility of a G5-based Cube design, similar to Apple's Power Mac G4 Cube. I don't think this will happen, as the thing that killed the original G4 Cube, and that would damn the G5 cube to the same fate eventually, is the lack of market for the thing. Yes, I would like to have one, but being a Mac geek is not a characteristic most Mac users share. Let me explain the lack of market for the Cube and why it's destined to fail.

Sovereign Semiconductor

In the best decision out of Motorola in years — now that Chris Galvin has resigned — the Motorola Semiconductor Product Sector will be spun off into its own independent corporation. After years of mismanagement and dwindling mindshare, setting SPS free could spark the rebirth of the sleepy chipzilla, but sadly for Apple and Mac users the move has come too late to benefit Macintosh.

Motorola Semi's Failure to Innovate

Motorola struggling isn't news to anyone who watches Apple, nor is it news to anyone else with a vested interested in Motorola's semiconductor branch. But it's not like the company is beleagured. It's not like it takes Motorola's last gasping breath to release a new G4. The problem with Motorola's struggle is Motorola's attitude toward innovation.

Trading Megahurtz For Megahertz

For the last few years Motorola has been the sole supplier of Apple's high-end chips, all from the G4 family. And for the last few years, Mac fans and industry pundits alike have expressed grief over the speed — or lack thereof — Motorola has reached with these processors. While Intel and AMD reach speeds nearing 3 GHz, or 3,000 MHz, the Motorola/Apple camp have slowly crawled to 1.25 GHz.

A cacophony of possible solutions to the Megahurtz problem have been heard from within the Mac community, and finally an end is in sight. The light at the end of the PowerPC tunnel is shining, Mac faithful, and we can all breathe a sigh of relief.

Well, not quite yet.

The PowerPC Conspiracy

Since the late 90s, when Apple introduced the PowerPC G4 and Intel introduced the Pentium III, there has been a severe performance gap between the venerable Macintosh and the ubiquitous PC. Not only did the Pentium III have a higher clock speed than Apple's G3 or G4, but its performance per clock also increased. The days of Apple/IBM/Motorola (henceforth AIM) triarchy in the microprocessor business were at an end, its pinnacle reached with the Mach V, a PowerPC 604 variant that outperformed the Classic Pentium, MMX Pentium, and Pentium Pro clock-for-clock. With plans for the Mach VI and Mach X (the PowerPC 604r and 605, respectively) canned, Intel took sweeping strides toward the throne of CPU superiority and has held on with an iron grip ever since.

What Motorola doesn't want you to know is that this obsolescence was planned from the beginning of its involvement in the PowerPC fiasco. Owing to spite and jealousy over Apple's choice of IBM's PowerPC architecture over its own Ripfire 88k series, Motorola decided to trumpet Apple's decision despite the fact that it had something better Apple desperately needed at that point in time: the M68060, the sixth CPU in the hot 68k family. Keeping its hands quietly in its pockets and staring at the floor in silence, Motorola remained mum on the new specs for the '060 and let Apple purchase the PowerPC 601 from IBM, which would act as a time bomb set to destroy Power Macintosh performance.

The PowerPC 601 was the Piltdown Man of its family: it bridged the gap between the older POWER instruction set architecture (ISA) and its direct successor, the PowerPC ISA; in fact, the 601 actually included some POWER instructions and emulated others by stringing together PowerPC instructions—it could run binaries compiled for POWER chips unmodified. This little dynamo also out shined the Pentiums of the day, which ran at 50, 66, and 75 MHz. The 601 was generally 1.2-1.5 times as fast as a Classic Pentium at the same clock speed. Apple and IBM were pleased while Motorola snickered silently in the shadows: their M68060 was running circles around both the Pentium and the PowerPC 601 in a beast called the Amiga. To put it in perspective, the '040 and '060 Amiga lines actually used PowerPC chips as text co-processors!

As Intel glacially moved from the Classic Pentium to the MMX-enhanced Pentium (and also down-clocking the new line), AIM released the PowerPC 602, 603, 604, and 620 parts, all designed to conquer niches in the market that the stop-gap 601 could never be specialized for. The 602 was used in stadium scoreboards, remote-controlled Transformers, and the popular Nintendo64. The 603 was a very power-friendly chip and could turn its various subsystems off and so was used in low-end desktops, laptops, and network computers. Its older brother, the 604, was created to compete with the secret P6 project that would eventually produce the monstrous Pentium Pro. Finally, the 620 was a 64-bit überzilla with support for up to 128 MB of L2 cache and top clock speeds exceeding 200 MHz.

Years later, in 1997, the PowerPC 602, 603, 604, and 620 chips had either been canned or were severely losing steam. In a seemingly clever move, AIM took the 603 and augmented it. The result was the PowerPC 750, which came to be known as the G3. Motorola's plan to sabotage Apple's desktop performance had just been realized. By the end of the year, Apple was selling Power Macs that ran at 233 and 266 MHz while the last of its PowerPC 604-based line had been running at 350 MHz! Aside from the obvious fabrication limitations of the G3, it also ran slower per clock than the 604 had. Of course, with Apple's interim CEO Steve Jobs at the helm, Mac users did not question this blatant paradox. G3 systems were snatched up by the thousands.

Presently, it is clear that Motorola's nefarious plot has almost come to its fruition. Apple just recently reached 1,000 MHz while the Intel (and now also AMD) attained clock rates of 2,400 MHz in the same time frame. Motorola injected additional roadblocks into Apple's plans by not letting it move away from the abysmally performing 603 core—the G4 was just a hacked-up G3 with AltiVec and an FPU borrowed from the outdated 604!

It is clear that for Apple to survive, it must look beyond Motorola's PowerPC: IBM's own processors would be a good choice, and a logical one at that. One thing remains plain: Motorola is slowly killing Apple—from the inside.

The PowerPC 7455 Is Too Little, Too Late

Let's face it: Apple's new offerings in the Power Mac line are too little, too late. The PowerPC 7455 is a cop-out for the real deals (the 7460, 7500, and the 8500). If Apple doesn't get its act together soon on the high end, it'll be relegated to a consumer-only nitch and dwindle until it's bought out for its brand name.

Apple's been severely slacking when it comes to keeping its high end customers appeased. Just how long did the Power Mac stay at 500 MHz? 18 months? Well, you do have to hand it to Apple: they broke Moore's law. But don't drone on about Motorola's bug in the 7400 that kept it at 500 MHz and no higher. IBM has a 1 GHz G3 out now (the PowerPC 750FX) and could have easily provided Apple with the firepower it needed then. No AltiVec you say? Motorola's a greedy miser. They could easily release an AltiVec-only co-processor, but they want to keep it tied to PowerPC so they're guaranteed business. Business from a company too stupid to drop deadbeat technology, Apple.

Motorola's PowerPC 7455 is a compromise. It's basically a rehash of heretofore substandard processor technology with a few new fabrication features added to let it crawl towards the 1 GHz mark. There's nothing new on the table with it, and that's what makes Apple look even stupider. How long have rumor sites been predicting Apollo (the 7460)? And it's a well-known fact that the 8500 has been in testing for over a year. Yet Apple finally breaks the gigahertz barrier with something that barely is capable of doing so, a silly token upgrade to the 7450.

Why do we Mac users put up with Apple delivering slop from a bleeding company that can't keep a schedule? The only thing that makes these systems "fast" compared to the new 2.2 GHz 786/Pentium4 or AMD's XP is Steve Jobs's Reality Distortion Field. That in itself is amazing, but not perpetually sustaining. Eventually (hopefully) Mac users will smarten up to this kind of marketechnology.

It's really kind of funny. Apple has awesome machines and stays ahead of the competition hardware-wise but runs Mac OS 8 and 9, which aren't at all native to PowerPC and can't do SMP, then it gets an OS that has memory protection, SMP support, full native PowerPC code, preemption, etc. (Mac OS X) and lets its hardware fall behind by a year.

I remember back when when Apple was encroaching on SGI's low and mid end systems; now you need a PowerPC 74xx to run Mac OS X because the 750 is under-powered but still shows up in the iMac and iBook lines. That's called selling snake-oil.

If I were you I'd consider the above and think about jumping ship. I didn't like to admit it but once I was honest with myself I felt like technology was going somewhere besides the Barbie aisle.

The PowerPC G4 Is a Lie

At the heart of the current high-end Macs, routers, and switches is the PowerPC G4, which is what Apple and Motorola claim to be their fourth generation CPU that is the result of the three-way Apple/IBM/Motorola alliance, which has been designing and fabbing various PowerPC chips since 1991.

I contend that “G4” is a blatant misnomer by Apple and Motorola to spur sales and compete with Intel's Pentium 4 product and nomenclature. Below I'll give some historical background, technical information, and plain facts that support my claim that the PowerPC G4 is really a second-generation processor, and the broader notion that the PowerPC family has not evolved significantly since 1995, something Apple and Motorola propaganda has repeatedly accused the competition of in recent years. But first, the background.