All Terrain Thinking

A Compendium of things I think are Important

Generally Speaking, Think on this...

Why Computers Will Never Be Fast Enough

By Lucas Roebuck

The geometric increase in the ability of computers to perform increasingly complex operations in shorter amounts of time is unlike any other advance in the history of technology.

For example, a mere 35 years ago, man made it to the moon with less computing power than what I carry in my cell phone today. In the annals of world history, 35 years is rather insignificant. In the world of computing technology, 35 years is a score of lifetimes.

Intel founder Gordon Moore most famously suggested the geometric increase in computing power, arguing at different times over the past few decades that the number of transistors on a chip and/or raw power would double every 18 months to two years. The ability to count on this perpetual upswing of computer force has become known as Moore’s Law.

Still the Same

True, the rise of computing power has enabled man to innovate in many ways. The advent of digital media and new-era communication technology (Internet, cell phone) in the past 10 years shows how much computing power has the ability to change things.

Unfortunately, while computing hardware—specifically the micro-processor has become astronomically more powerful, the daily application of what is done with all that power hasn’t really evolved in the past 10 to 20 years.

The applications for which most people use computers are essentially the same as they were 10 years ago. For example, I first used computers for building newspaper pages in 1989 —15 years ago. I used a computer with an 8MHz processor, a 40MB external hard drive, an 8-bit flatbed scanner, and a laser printer. Today, I build newspapers with a 2,500MHz dual-processor computer with a 160,000MB hard drive, a 32-bit flatbed scanner, and a laser printer.

For all the thousands of dollars that have been dumped into numerous upgrade cycles since the days that newspaper editors were using 8MHz machines, we still essentially create the same product in the same amount of time. The only real innovation has been the introduction of high-resolution digital photography and the Internet.

Even the Internet isn’t a new innovation. Scientists and university profs have been sending e-mail over TCP/IP networks for more than 20 years.

It seems that, even though we have computers that are thousands of times faster than those of even 10 years ago, we are essentially doing the same things with our computers. Here’s a theory that is based on the truism that necessity is the mother of invention: As programmers have had a nearly unlimited toolbox of processor power and memory banks to develop useful applications for us, they have grown slothful and sloppy, chasing slow feature evolution instead of true application innovation.

Has human ingenuity ultimately been supplemented or harmed by the meteoric increase in computing power?

Let’s pick on two prominent computer companies: Microsoft and Apple Computer.

Microsoft and Apple ’s Lack of Innovation

I have been using Microsoft Word since version 4—I’m writing this article in version “X”—and the only real advancement I use is the automatic underline feature of misspelled words. Even with the myriad menus and features, including Max (the Macintosh answer to Clippy), I am basically creating the same documents I created on my 1986-era computer using Word 4.

Apple, which used to lead in operating system innovation, has taken all those millions and millions of processor cycles at its disposal and given us not a natural-input operating sys-tem driven by speech and handwriting, but rather just a lush, über-graphical version of the original Mac operating system of 1984, using the same desktop metaphor.

Don’t even get me started on the “innovation” of Windows XP. It’s easier on the eyes than Windows 2000 (and even that’s debatable), but basically adds more feature molasses to an overly complex system, when all people want to do is write in Word and send e-mail. One of my PC-loving friends tells me that the XP stands for “eXtra Pain.”

Not that I don’t enjoy Mac OS X’s “genie” effect, or Microsoft Word’s “correct-as-you-type” features, but somehow I expected much more out going from 8MHz to 5,000. Unfortunately, instead of using the new sys-tem resources to innovate the ways we use computers, programmers find trivial ways to eat up those extra processing cycles. Or just do more of the same by supporting more colors at a higher resolution.

I was very excited in 1993 when I purchased an “AV” computer that could not only capture video, but also interpret spoken commands. The soft-ware command system had problems, however, because of office noise and other logistical/usage issues—and the fact that the 20MHz chip couldn’t really provide the raw power needed to interpret audio. In fact, Apple Computer still ships the same little-used voice-recognition software on every Mac today, 10 years later.

The software doesn’t have processing power problems anymore, thanks to the super-clocked G5 chip from IBM. But for some reason, Apple decided that it wasn’t worth it to make speech recognition work well—to really integrate voice commands into the operating system the same way it integrated the mouse back in 1984. Even though my computer can process literally 1,000 times the instructions in one second that it could in 1994, in 2004 I am still talking to my computer using lame 1994 technology.

How many more buttons will Microsoft add to the next version of Excel that we will never use? How many more polygons will Doom 4 have to render the same basic first-person-shooter game? How many more different typefaces will I load into my system that will never grace a page or screen?

Rising to the Challenge

Historically, when you limit the amount of resources engineers and programmers have, great things hap-pen— people rise to a challenge. Consider the first Macintosh, the last great innovation in personal computing.

A handful of people took four years to build that computer from the ground up. They didn’t design it to use the fastest processor and the largest RAM space; instead, the project was specifically designed to make a computer that was simple and affordable. Designers were required to use minimal resources. They had to write efficient code and make use of every processor cycle.

Other cases of designers creating truly remarkable products working with a limited platform include the original Palm and the iPod—both examples of how engineers working with strict computing resource options were able to create great products. The Palm PDA has already “suffered” too many resources, becoming more complex and less useful. So far, the iPod has avoided that fate.

Today, there is no efficiency in the bloatware churned out from Cupertinono and Redmond. Which is why, no matter how fast IBM, AMD, Free-scale and Intel make the processors, they will never be fast enough to make programmers find truly innovative ways to use them.

Prior | Tell us what you think | Next

Add to Your Social Bookmarks: Del.icio.us Digg Furl Reddit Simpy Spurl Y! MyWeb - StumbleUpon