When choosing between what computer to buy, many people prefer a tower case, because it’s “upgradeable” (that is, components can be replaced with better alternatives in the future) and “customizable” (that is, components can be removed, or different ones added).
Common pieces of hardware for prospective upgrades are the CPU, the amount of RAM and hard drive space and the graphics card. To a lesser extent, replacing the optical drive or adding features such as TV tuner or optical audio outputs are also frequently cited as reasons on how and why this is useful.
I, too, subscribed to this idea for a long time — and I convinced others of the same. It was clear to me that it’s in a customer’s best interest to have as much flexibility as humanly possible.
Now I believe it’s not a priority, for two reasons.
First, I have observed, with me as well as with many, many others, that the potential is not being taken advantage of: years later, the computer still, for the most part, carries the exact same hardware. Even if components were added, they are often external; examples of that include a webcam (perhaps for distant friends or loved ones), an external hard drive (perhaps for backups), a media card reader (perhaps for one’s newly-purchased digital camera), or even more obvious choices such as a new printer. As external interfaces have improved (with the move away from traditional serial and parallel ports and external SCSI, towards USB as well as FireWire), this possibility has become more and more appealing. Why bother opening your computer’s case when you can simply hook the device up externally, without much sacrifice (aside from a usually insignificantly higher price)?
But secondly, I believe it’s not truly relevant (for most people) because, rather than (or perhaps in addition to) truly solve a problem, it adds adds another. Upgrades have a tendency to create a bottleneck. Further updates have a tendency to merely move that bottleneck elsewhere.
Consider a media card reader. Let’s say you bought your computer a few years ago, and it came with USB 1.1. And the digital camera you have also only transfers data with USB 1.1. Finally, it came with a CompactFlash card, and rather slow one, too. And now you’re angered by the slowness, because somehow, the fast and snappy of yesteryear is suddenly the slow and sluggish of today. What are you going to do?
Well, you would have to start by actually adding a USB 2.0 port to your computer. Fortunately, you have a computer that’s internally expandable, so all you have to do is get a $15 PCI USB 2.0 card. There, done.
Or not: the system that came with your computer (you never bothered to upgrade) doesn’t actually support USB 2.0. And the third-party driver that comes on a CD with the card causes crashes — blue screens, kernel panics, whatever.
Ah, you convince yourself: time to upgrade the OS.
$150 later, you get your USB 2.0 card running. And? It doesn’t crash. Satisfied, you try and connect your camera.
A warning along the lines of “your device may not be able to use USB 2.0 High Speed mode” puts you off. Didn’t you just insert a faster card? And get a newer, allegedly “better” OS? And for the nearly two hundred bucks, plus time and effort, now you don’t have any real benefit whatsoever? Ah, that’s right: the camera isn’t USB 2.0-capable.
Oh, but you could always get a new camera. It’s kind of outdated anyway, right? But that wouldn’t quite fit the whole idea of upgrades; it’s a complete replacement. “Never!”, you proclaim: you buy a media card reader ($15) instead. Universal as they tend to be, the one you bought of course has the right slot for your cards. Next to half a dozen other ones, and several dozen claimed different formats you’ve never even heard of.
So, you take the card out of your camera and into the reader, and you connect the reader (USB 2.0!) to your computer (USB 2.0!) and wait. Da-ding, it works. Your system isn’t quite smart enough to figure out that all those other ports, and the fact that there’s nothing in them, don’t really have to be displayed, but right now, you don’t care. You finally have a fast solution to reading your freshly-shot photos off your camera, right?
Right?!
No, you don’t. You see, the card you’re using is nowhere near fast enough to make much use of USB 2.0. And the card format you’re using couldn’t possibly benefit that much from USB 2.0 High Speed either. You could get a faster card, but it wouldn’t be much use — and then you run risk of the card not being backwards-compatible with your camera.
You comfort yourself: you still have a working camera, and now you have a slightly faster workflow. And, on top of that, you have a completely new OS — imagine the possibilities!
But you know that you’re lying. You expected a much more, and you won’t ever possibly use the new features in the OS, seeing as you barely used those in the old one either.
And then it dawns on you: sure, you would have invested a lot more, but if you had simply waited a few more years until both your computer and your camera became outdated, and then purchased new ones, you would have been much, much more satisfied.
The one single time you actually made use of your computer’s ability of internal upgrades and customizations, you didn’t end up enjoying the results. Why would you have? They weren’t noteworthy, involved a lot of money, time, effort and frustration, and didn’t in any way meet your expectations.
Maybe next time you buy a computer, you won’t be fooled again.
Others' Thoughts
Comment on May 31st, 2006 at 11:53 pm
Well, that the whole computer industrie is focused so much on upgradeability is rather odd. Most other electronic products are following a more or less revolutionary technical progress instead of an evolutionary. If you buy a TV or a game console it will be at least a few years state of the art. If you buy a new computer it will be obsolete the moment you walk out of the store.
BTW: MenuTemperature doesn’t work with my new MacBook!
Comment on May 31st, 2006 at 11:59 pm
Exactly! And it’s easy to get aggravated over. In a few months, with Merom, there will be MacBook Pros with significantly faster, yet cooler chips. There will be features, such as dual-layer burning, that I’m missing out on. The only way not to get angry is to simply buy when you know you’re getting exactly what you want — last year I planned on not getting a PowerBook until they hit 2 GHz (which they never did); this year, the MacBook Pro, which is pretty much just a rebadged PowerBook anyway, has not one, but two cores at 2 GHz each, so I had no reason to wait any longer — I got what I had wanted.
Indeed. Check the attached beta 6.
A final version of 1.0.1 is due out tomorrow.
Comment on June 1st, 2006 at 8:51 pm
I think the biggest confusion is the idea of interchangable parts vs. upgradability. When you just think of parts as interchangable, you don’t get the big expectations that an “upgrade” gives - you expect roughly the same functionality.
Also, there’s the “sloppy mechanic” concept again. The less work you put into your machine, the easier it will be to replace, simply because you have less investment in it.
And I’m waiting for a stable OS X 10.5, personally.
-Sud.
Comment on June 1st, 2006 at 9:18 pm
I can’t say I’m experiencing an unusual amount of stability problems, and I’ve used 10.4 extensively for just about as long as it’s been out.
Comment on June 2nd, 2006 at 2:35 am
Yeah… I agree for the most part. However, there are certain features of my laptop that I simply would love to upgrade slightly… Upgrades that were offered to me on the purchasing of the laptop that i didn’t take such as an ATI mobile graphics chipset rather than the built in intel chipset… There’s very little about my computers that change… but the graphics upgrade would be a very nice feature.
Your Own Thoughts
I'd love to hear your input. Just try to stick to a few rules:
Before you comment for the first time (or, after you have deleted cookies), you will have to answer a little challenge to prove that you are not a spammer.
Comments are written in Markdown.