Two Too Many . . .

Add Your Comments

We will see 4-core processors sooner rather than later, and early next year, Intel will come out with a second one that won’t cost a thousand dollars.

Should you care? No.

Why shouldn’t you care? Take a look at the numbers (as opposed to the hype) in this article. All hollering aside, it is just so obvious that a quad-core does no more than a dual-core in just about every real-life application, and any significant “wins” in the benchmarking is simply due to the 4Core being overclocked to a frequency higher than those it is being compared to.

The reason for that is simple, typical desktop software isn’t designed to use four cores, and until it is, 4Core isn’t going to do squat for you.

And let’s face it, software developers as a whole aren’t exactly the biggest early adopters in the world, as you can see from all the optimized dual core or even 64-bit software floating around.

Yes, eventually software will start to take advantage of these standards. We’ll start seeing a serious migration to 64-bit a year to two years after Vista introduction, probably pretty much the same timeframe for dual-core.

But it’s going to be a very long time before a lot of software optimized for 4Core will become mainstream in the desktop market; five years or longer.

There’s a two-step approach one ought to take when judging whether or not you need some new development that won’t have an immediate impact on current performance.

The first step is to ask, “How much extra will it cost me?” If the extra cost is at or near zero, well, who cares?

If the extra cost isn’t nominal, then there’s a second set of questions, “How much more will it cost me to get it now as opposed to later (and how much hassle will retrofitting “later” mean)? When will the benefits start to kick in? How long do I expect to use this?”

People will of course place different values on the answers, even in the same situation. An extra $100 is a dealbreaker for some, a trifle for others. Something with an expected life of a year or two will be looked upon quite differently than something with a expected lifespan of five years plus. A person who leaves a door off his case so he can make hardware changes on the fly will have a much different attitude about replacing parts later on than someone who would open up his computer as readily as he would open up his abdomen. Even the tinkerer just mentioned might think twice about doing a brain transplant for his notebook.

Nonetheless, for the vast majority of people, really everyone other than those who can take immediate or short-term advantage of it, 4Core makes no sense for at least the next couple years. It will cost a lot more than duals for at least a couple years, it will provide almost no one with any real advantages immediately or anytime soon, and by the time it’s likely to do the average person some real good, you’ll be better off buying a new system then.

Ed


Leave a Reply

Your email address will not be published. Required fields are marked *