The Megahertz Can Of Worms September 12th, 2000 There are moments when you have to face serious problems and you wonder if there is a simple, or at least viable, way to save the day. Apple and its processor issue is one of these cases. How does one get out of it? For those who are not too sure of what I mean, here is a quick background. At the introduction of the Power Macintosh G3 in 1997, Apple trumpeted the "twice as fast" claim to show how the new PowerPC 750 is simply on a different planet, leaving Intel and other processor manufacturers behind to eat dust. With its architecture and L2 cache, this baby smoked. Remember the snail ad? More or less at the same time, other PowerPC developments were going well enough for the industry to hear about plans for the future. Copper chips would replace aluminum chips to break barriers. There was speculation about even crossing the GHz hurdle. As if this was not enough, we all knew that Apple and Motorola were working on the G4 processor, a new generation of chips that would take Macs to the next step. We wondered what PC manufacturers could do about all of this. Happy happy, joy joy! Not for long, though. Things got more complicated, especially when the G4 plans emerged from Motorola. IBM and Motorola had different plans and parted ways, shaking the sacred Apple-IBM-Motorola alliance soon after frictions related to the end of clones had also weakened relations. Then, Intel and AMD broke the gigahertz barrier. After dominating the industry for sheer speed, Apple had to face something new. Faster (or apparently faster) processors, in speed or number of MHz, competed with its best offer available. Add to this the fact that Motorola fumbled G4 production, being unable to produce it in great numbers at fast speeds, forcing Apple at one point to revise its G4 line downward to 350-450 MHz for a while. Today, Apple does not offer any Mac with anything faster than 500 MHz processors. When you talk about processors in the industry, you open a can of worms. Here are the standard arguments used: Megahertz mean nothing; we know better and plain performance should make the difference. Consumers care about megahertz as a reference and industry insiders underestimate the impact of specifications on the common folks. A PowerPC at 500 MHz is much faster than Pentium, Celeron or Athlon at the same clock speed. That is cute, but PC manufacturers sell computers with 1 GHz chips that are fast enough to beat the current PowerPC processors. Apple answered by sticking a second chip in its G4-based Macs. Sure, but to smoke Pentium and Athlon chips, you need to use symmetric multiprocessing as well as Velocity Engine and very few software titles are ready for that. You get the point. Argument after argument, pundits and armchair experts will trade until they run out of gas and everybody realizes that Apple has a tricky situation on its hands. Will Motorola bounce back or fumble again? What should Apple do? Is IBM ready to manufacture faster chips in great quantities to serve Apple? Should Apple dump Motorola? Everybody has a viewpoint on the issue and it is interesting to watch the "ditch Motorola, go IBM" slogans flying around. Some industry insiders pretend to know that Motorola has a 600-800 MHz G4 chip in the ranks and that by January, it will be able to produce them in numbers large enough for use in G4-based Macs. Another industry insider, a friend of mine, tells me that IBM has a 1.2 gigahertz G4 processor (with Velocity Engine technology!) ready and that all that Apple has to do is to switch PowerPC manufacturers. I advise that all of us take these two assertions with a grain of salt, especially the one about IBM. Nevertheless, one has the right to dream... I would be careful about dumping Motorola. As you know, Apple and Motorola have dealt with each other for years. You should not dump a business partner with a decision from the top of the head and according to the situation of the moment. It is a double-edged sword and it will take you down just when you need help. Payback is a bitch, or so they say. If IBM is able to manufacture faster G4 processors in numbers and unless Apple has a restrictive exclusivity contract with Motorola, maybe Apple could explore the possibility to use chips from both companies. Would that be possible? They say that nothing is impossible. Having faster G4s from IBM and slower ones from Motorola could even open an opportunity in Apple's product line. Some G4s could be low-end and others could be high-end. Who knows? I am no Apple executive or engineer to know all the dirt about this. The bottom line is that if megahertz hurt Apple's image for speed, something has to happen. Education is usually a prime way to get almost anything done. If you educate customers and explain in simple terms why a 500 MHz PowerPC can beat a 700 MHz Pentium, they have a better chance to understand. When I mean education, I mean more than developer notes and papers addressing the issue for industry insiders. I really mean broader audiences to make it clear that there is more to computers than megahertz. Mostly because benchmarks and Photoshop tests do not do the trick, no matter how credible they are. Something has to prove it to everybody for the last time. One can only hope that Motorola will pull a magic trick out of the hat, but nobody would bet on it. Meanwhile, many of us wonder what will happen and if this whole situation will affect Apple's product strategy and sales in the long term. Bold moves could be risky for now, but Apple will not be able to wait anymore to act if Motorola fumbles again in the next 6 months. Your comments are welcomed.