| Sunday, July 28, 2002 |
01:47 - ...But it's really not about speed
|
(top)  |
I've sneered a couple of times already at how when Microsoft showed off their Tablet PC to various journalists-- they first touted it as having "the best handwriting recognition in the industry", but then, when it became clear that under serious scrutiny it wouldn't hold up to such a claim, they held propaganda sessions to convince journalists and reviewers that "well, handwriting recognition doesn't actually matter". In other words, a particularly weaselly form of sour grapes. You can't prove superiority in some field, so you downplay the importance of that field.
It's at the risk of sounding as though I'm committing exactly that rhetorical crime by saying this; but while it can certainly be shown that the Mac's performance at various tasks is not only not "faster than any PC" (as the marketing PR would have it), but is not really in the same ballpark-- I must posit that the efficiency of computing does not, in fact, depend directly upon the speed of the CPU and chipset. It's not about speed. Not in the way we usually think of it.
A little background. When I first became interested in the Mac, it was not because I was under any impression that it was faster than Windows PCs. In fact, I was pretty well convinced that Macs were unquestionably slower at almost anything you might ask of them. This was the age of word processing and Photoshop and telnetting to UNIX servers to write rudimentary web pages without <BODY> tags; but even then, what attracted me to the Mac was not its speed. It was the user experience and the philosophy of design. It was a technology that was exciting. And that's the same feeling I have today.
I think it's great that Macs can be used to do high-end vector processing and rendering in workstation-class environments these days. I think it's great that the G4 has gotten to more than 1GHz, and that there are more improvements on the way. But raw speed isn't everything. No-- no way am I going to claim something stupid like "close the microprocessor R&D plants; there's no further use to be had from faster CPUs"; but I will say, however, that there's such a thing as greater efficiency in the use of the resources that are available to an operating system, in the way that that OS enables a user to do things.
The Mac is optimized, for instance, for tasks like video editing and image manipulation and audio mixing. Not optimized for raw CPU execution speed-- optimized for efficiency in the user experience, efficiency in getting things done. That's why pros use Macs for those things-- not because X processing task will complete 10% faster than on a different rig, but because of much more fundamental issues. Audio pros use Macs because the Mac is designed for low-latency, high-bandwidth audio throughput in the CoreAudio subsystem, something that Windows utterly lacks-- as well as built-in sophisticated MIDI support. Graphics and prepress pros use Macs because of ColorSync, a feature that Windows utterly lacks (and is incapable of incorporating because Microsoft can't guarantee standardization of its display hardware components). And casual consumers like me use Macs because of the way iTunes seamlessly integrates with the iPod, and the way iMovie plays with DV camcorders as though they were favorite puppies, or the way iPhoto knows how to do magic with images from just about any kind of digital camera without having to load drivers. Not because of how fast those things occur, but because of how well they work. It doesn't matter in the slightest to me how much slower or faster it is to rip a CD to MP3 in iTunes versus on a PC; what matters to me is that I can do it by putting in a CD and pressing a single obvious button.
We like how Apple has a Feedback page online where they solicit bug reports and new-feature suggestions for OS X, and that they listen to things we send them through it. We like how Apple actively wants us to rip/mix/burn our CDs and take our music wherever we go, with good ol' MP3 files instead of digital-rights-managed proprietary formats with RIAA-approved back-doors. We like how Apple thinks of little details like making it easy to install and deinstall applications by dragging-and-dropping them in any folder or the Trash, and how they've managed to take UNIX and make it into something so easy-to-use that one's grandma can mount someone's shared folder from across the Internet and file-share in a way that PC users have to use third-party P2P apps for. We like watching DVDs on wide-screen LCD monitors. We like being able to put custom icons on individual files and to specify per-file opener apps. We like the thought that our convenience and our creativity is the utmost goal of our computer company, rather than just more numbers inching upward.
Companies like Dell and Compaq don't have Macworld-style keynote events when they unveil each year's new product line. Why? Because it's boring. There's nothing new they ever have to show anybody that anyone wants to see in person or get a scoop on. It's just more of the same old same old, but <gasp> faster. Whereas Apple, two or more times a year, will announce a speed-bump in some machine or other-- but it's always in tandem with some major new thing they've just developed. "Look, it's got DVD burning now!" Or "Hey, an adjustable flat-panel all-in-one G4 machine!" or "Here's what OS X can do!" or "Hey, here's iPhoto!"
Journalists love Apple because they never have to muckrake in order to get a story out of them. Apple makes news.
More and more incremental speed in a rickety hunk of baling wire and bubble-gum doesn't excite me. But new technology that pushes the boundaries of what I can do, by putting entire new classes of abilities in my hands rather than just another boost to the speed of all the old ones, excites me. A 10% faster CPU doesn't make my heart beat 10% faster when I use it. But a feature like Sound Check or "Keep Music Folder Organized" or Smart Playlists makes it start to skip beats here and there.
When Jaguar gets here, it will bring speed improvements to the OS X user-interface. Welcome, indeed, it will be. But you know, I'm not looking forward to that speed increase anywhere near as much as I'm looking forward to Rendezvous, and sitting on the couch downstairs with my iBook on AirPort listening to Lance's Lord of the Rings recording streaming from the iTunes on his iMac upstairs. And when iSync is ready to go, and I can use it to convey all my schedules and contacts to my iPod with one button (or to a cellphone over Bluetooth), I won't care how fast it goes. This isn't so much the "it's amazing that the bear dances at all" argument as it is one of "my God, that bear's doing cold fusion".
Sure, frame rate is a fine goal for gamers. But that's probably why the Mac hasn't been a great gaming platform, historically: games, uniquely among software genres, benefit very concretely from the min-maxing of system speed; something that Apple has never considered to be of paramount importance, never more so than an efficient user experience and technology that's exciting.
I first got into Macs because Apple's technology excited me; I remain fascinated by Apple today for the same reason. The same goes for just about every engineer in the tech industry who has found himself magnetically drawn toward OS X lately. Speed has never been something that's inspired me. It's just another number-- and I suspect I'm not alone in thinking that. When a piece of revolutionary technology is dismissed because of the lack of luster on the speed statistic, such dismissal looks short-sighted and petty to us, and catches us by surprise. We're too busy being excited and inspired by the possibilities we see before us to be bogged down in such inane details. Having to wait an extra three or four seconds for a web page to render or to see the spinning cursor for a few moments while the OS resolves a blocking behavior is something we're willing to put up with, for the sake of the benefit of what that blocking behavior will reveal once it's resolved. And speed will always improve in the future. That's one thing we can always count on.
This may sound like lame rationalization for an undeniable handicap we can do nothing about (or maybe it sounds like Roger Meyers Jr. backpedaling in the courtroom, saying "Okay, so maybe my dad did steal Itchy-- but come on, the animation industry is built on plagiarism!"); but to call it that is to fail to "get" the Mac, as so frequently happens in the tech press and in society at large. Computing on the Mac really is a different thing. It operates by different rules. Different things are important. It's one of those things that one just has to experience, in a head-smacking moment of clarity, to fully appreciate.
A sufficiently souped-up Ford Escort, with all the ingenious tweaks and mods and decals in the world, can beat a Ferrari F355 down the track. But only one of those two cars excites me, and I'll tell you it ain't the Escort.
|
|
