It Takes 400 MHz

Overcoming poor design with brute force power

By Steve Duff, Staff Writer, See staff list for e-mail address

I recently purchased a new PC with a Pentium II 400-MHz CPU, plus Windows 95 OSR2. It came well equipped with all the goodies -- an 11.5-GB UltraDMA drive, 64 MB of SDRAM, an Intel 740 2D/3D accelerator with 8 MB, a 32X CD-ROM and 56k modem. There was also the Corel WordPerfect Suite onboard and plenty of shovelware to ignore and maybe someday even uninstall. In the time I've had it, I've gone ahead and added a Diamond Monster 3D2 accelerator using the Voodoo2 chipset and 8 MB of RAM. That's 16 MB total graphics RAM.

The machine is a monster. It plays Quake2 and Jedi Knight at 800x600 with silky-smooth framerates. It raytraces rapidly as well. Using the 3x3 Oversampling setting in Cinema4D XL 5.1, and with every other option maxed out, it can raytrace the staircase test scene at 800x600x24 in 29 seconds! This, even though the algorithms used in the PC version of Cinema4D are slower than those used in the Amiga and PowerMac versions.

So why mention all this? Well, mainly because this is the first PC I've owned which offers an experience similar to what I get every day on my '060 Amiga. By this I mean that the new PC launches apps and opens/closes windows with the same blinding speed as my Amiga -- and I'm running my Amiga on a 16-bit Workbench rather than some super-fast 2-color screen.

The new PC multitasks smoothly and has even played two videos at once with no discernible loss in framerate. There are some rough spots -- it still takes a very long time to bring file download windows to the front, for example, while both Netscape Navigator and Internet Explorer remain stupendously slow at the 'Go Back' function, but overall, these new PCs offer a very nice and rapid computing experience and will certainly please most buyers -- until something goes wrong!

I have enough experience to realize that the heavy gaming use I put a PC through will sooner or later cause the grenading of Win95, and when that happens, fixing the mess won't be easy, or pretty. But even with this incredible drawback, the fact remains that these new PCs are a wake-up call. We have heard for a long time that we need more CPU power, but when you compare a 68060 Amiga to PowerMacs and PCs in the 200-MHz range, the Amiga simply feels faster. It is not, of course, categorically faster at everything; we all know that CPU-hungry apps will show the advantage of PowerMacs and Pentium PCs, but nonetheless, for most computing functions, an '060 @ 50MHz Amiga at least holds its own or flat-out blows away the Mac and PC.

But not anymore.

I say this because, having come to the Amiga after owning PowerMacs and Pentium PCs first, I have always been behind just about any scheme to get us a faster miggy. And that brings us to the Amiga, Inc. plan.

Clearly, not everyone at this magazine is happy with the AI plan. The newsgroups have also revealed many detractors. Typically, the anger is over whether or not the new machine will be a 'real' Amiga, or maybe because the current course has gummed up the PowerPC or DEC Alpha paths.

There's probably no easy answer that can satisfy everyone. I can say that I think AI is bending over backwards to offer legacy support through emulation. I can say that AI's announcement immediately improved the PPC outlook thanks to forcing phase 5 and Haage & Partner to become uneasy allies. I can say a lot of things, but in the end every Amigan will have to sit down and think long and hard over where they want to be.

In a previous article, I expressed my opinion that going head-to-head against Wintel in a desktop war, using a PPC-based Amiga, was a bad idea. At one time I thought it was a good idea, but now I have that loaded P-II 400. Sure, a 400-MHz PPC Amiga would be better than the PC. We know that, but Joe and Jane Public do not. The PPC Amiga would be priced comparably (we hope!) to a Wintel PC. Thus, anyone who appreciated the finer things in computing would choose the Amiga. Alas, most people have no idea, and educating them would require a stunningly expensive ad campaign. It just couldn't work.

The Amiga, Inc. plan is to compete in arenas that Wintel does not yet own or dominate. Thus it would seem that the main focus will be on a set-top game console for $500 or less. I personally cannot imagine a better strategy. The Wintel market has an installed base of over 100 million users. That's a lot. That's a big enemy. But there are still over 5 BILLION people in the world. The game is not yet over. If it was, Microsoft and Intel would not be so famously paranoid. They realize it can all end for them.

Remember that this century started with a much different power structure from what we see today. The supreme superpower was the British Empire. This was the largest empire in world history. Great Britain was so incredibly rich and powerful that they built their navy on the 'Two-Power Standard,' which meant it had to be as large as the next two largest navies in the world combined. Those navies belonged to France and Russia -- the US Navy and the German Kriegsmarine were quite far behind. Indeed, it hadn't been so long since the US Navy had been in peril of destruction at the hands of Chile over an international incident.

We obviously can't make direct analogies between global empires and giant corporations. Microsoft and Intel have in many ways a more insidious type of power than that of the great nations. But it's nonetheless true that the British Empire has faded, the USA has risen to world dominance, and that no empire of any kind lasts forever. This means the USA will inevitably decline and that Wintel will fall. It is the nature of things that giants have poodles nipping at their heels. Wolves today prowl the same territory once roamed by Tyrannosaurus rex. When T. rex ruled the forest, the ancestors of wolves were tiny, insignificant furballs. So given time, things change.

So I for one cast my vote with Amiga, Inc. There is no guarantee that their plan will achieve success, but it may be the only plan which does not guarantee failure. From all I have read, it seems they will take advantage of the AmigaOS's modularity to provide users with systems that can be scaled up or scaled down. We will be given wide latitude in how our GUI will look. We will have legacy support through emulation. We will even have parallel development by third parties on PPC and DEC Alpha machines. We will have a high degree of digital convergence and integration. In many ways it all sounds too good to be true, especially the very tight schedule being followed to give us all this by the end of '99.

We will all know whether the new machine is a true Amiga after we turn one on and have a little time to play around with it. Success in the marketplace will depend on the available software -- the games which must be in development right now, along with the word processors, web browsers and all the other applications that make up a full computing experience. AI did not take the 'safe' approach of simply going with a PPC port; they have instead gone for the gusto with a new and hopefully revolutionary hardware platform. Chances are that they will either succeed or fail in spectacular fashion -- I see no middle ground here. For all of us who value choice and quality, I hope it's the former. In any case, as a friend of mine once told me, "If you reach for the stars, at least you don't come up with a handful of mud.'



Return to the July 1998 (Volume 2 Issue 12) Main Index