It’s Now Or Later, Much Later
After all the developments and conferences over the last six months, we now have a decent idea what to expect over the next two years.
First, there is now. “Now” isn’t going to change very much the next nine months, a few price slips here and there, maybe a 1.8GHz dual core A64 for two-something around the turn of the year, but no dramatic changes worth waiting another nine months for if you’re hurting now.
Second, there is the second-half of next year. AMD won’t get 65nm chips out until then, and odds are they won’t be much faster than the current 90nm. What is more likely is that they’ll be relatively cheaper on the dual-core side than they are today, so those who like everything about dual-core except the price will likely get some relative bargains by waiting. Then again, that “bargain” will include the price of DDR2 RAM and a PCI Express video card.
Intel? Well, I just don’t see the point in buying a missing link CPU, and anybody with a late-model Northwood or better system isn’t exactly hurting, so an extra six months will kill few. How will they compare to AMD’s chips? I would suspect they’ll be more-or-less on par, simply because I think AMD is sitting on its laurels at the moment; we haven’t heard squat about any Hammer successor, and any improvements we’ve heard about are basically “me toos” that Intel will also provide.
This will likely be a good time for Intel loyalists to step up, but again, nobody is going to get their doors blown off.
Finally, sometime in 2007 edging into 2008, there will be the Vista era. On the CPU side, we’ll see 45nm chips, but they’ll be relatively minor compared to what will be happening on the video side.
It all boils down to high-definition becoming mainstream. HD video on the PC is going to mean some big changes.
The first will be hardware-based DRM. Expect to see the necessary chips to show up at any and all motherboards near you fairly shortly, without much if any publicity.
Right now, a lot of people think that’s a joke. Well, if it turns out that way, expect people like MS and Intel to be sued big-time by Hollywood, and Hollywood trying to block sales of PC-based HD recorders. That’s why I think it isn’t going to be a joke.
The real issue will not be “Can pirate versions be made by anyone?” but rather “Can the average person make a pirate version?” This is one area where the Internet won’t foment piracy; it’s one thing to download a 5Mb MP3; it’s quite another to download a 20Gb movie stripped of all the DRM.
In any event, HD DRM will at least complicate the purchase of video cards and monitors from now until the Vista era. For many, it will be “damned if you do, damned if you don’t.” If you buy a “Vista-compatible” video card/HD recorder/monitor, you likely will make any video piracy very unlikely. If you buy DRM-free equipment, you may well find yourself not being able to play HD-material at all.
In the middle of this and of course the usual multiple standards for HD material, we’re also faced with a change in display standards. DisplayPort is a big unknown which will chill both short-term video card and monitor sales among the aware. I know I don’t want to buy a new monitor or suggest one to anyone who’ll expect their next one to last five years to buy one until I get a clearer idea about this.
The Dark Horse
It seems to be dawning on Sony, very, very slowly, that people don’t terribly want a $500 gaming console. So they’ve been mumbling, more and more, about the PC possibilities of the PS3.
Can Sony make the PS3 a reasonable alternative to the PC, something the kids can play games and do their homework on? If I had to bet, I’d bet “No,” I don’t think Sony is bold enough or will put up enough bucks to make that bet, but if I’m wrong, all bets are off in the PC industry.
The PS3 could be disruptive technology. When overclockers write me to say that they’d go to a PC PS3 because they don’t want to pay the prices they’re seeing, you know there’s going to be a lot of Joe Sixpacks who’d think the same given the chance.
Different people have different circumstances, but if we had to give generic advice, if you’re hurting now, you’re probably better off buying sooner rather than later, get a few year’s use out of it, then buying new stuff once these new standards become mainstream in, say, 2008. Machines just aren’t going to get a whole lot faster any time soon.
More importantly, with these standard changes, the upgrade decision has added a new dimension. It’s no longer just a matter of how much better a buy later will be than a buy now. It will also be whether something you buy now will even do what you’ll want to do later.