As I’ve said in the past, a lot of people live in a sort of Eternal Now; no past, no future, just an elongated present.
It’s not a good way to approach life because life simply isn’t that way. The present comes from the past, and the future comes from both.
It’s the past that gives us perspective on current events. It gives us tools to determine what current events is important and what isn’t. It’s not perfect by any means, but it’s much better than being completely blind.
All Quiet On The Overclocking Front
We’re getting an increasing number of letters saying, “Why don’t you give us overclocking news?”
Before we answer that, let’s ask first, “What is “overclocking news?” I would call it news about overclocking that is new, different, significantly better and affordable. The ThoroughbredBs were overclocking news. The really cheap ThoroughbredBs were even bigger overclocking news.
The problem we face over the next twelve-eighteen months is that the equipment expected to come out during that time will fail to meet at least one of those four criteria.
To give a few current and near-future examples:
Via KT600 mobos are new, different and affordable, but not better than the competition.
Quad-channel RDRAM will be new, different, and a bit better, but probably not too affordable.
Athlon64s will be new and different, but it won’t initially be better, and it won’t be terribly affordable.
Prescott will be new and different, but the current heat problem makes “better” questionable (at least for overclockers), and it won’t initially be affordable.
We’re on the back end of .13 micron technology for both AMD and Intel. They’re both mature products. The capabilities of both products were figured out months ago. There’s been no significant jump in processor capability from either side in the last few months, and there isn’t going to be any.
This is neither surprising nor shocking to anyone who has been around a while. When you’re on the back end of a generation of processors, this is what heppans. It has happened before, it will happen again.
So the past tells us there would be a stall for some time even under the best of circumstances.
But the present tells us these aren’t the best of circumstances.
AMD hit a couple walls in the last couple years. They hit a wall first trying to make Thoroughbreds with the usual technology, then they hit a much different wall making Hammers with not-the-usual technology.
It’s beginning to look like Intel bought time and delayed hitting the wall at .13 micron due to the different design of the PIV, but they’re now faced with problems, perhaps a little sooner than they expected.
It’s not that there aren’t ways around the wall, but the ways around it are much more expensive, complicated and unproven than before. Delays and problems are to be expected as new technologies get mastered.
It’s a difficult and expensive biting-of-the-bullet which chip makers have ducked for a long time, but now they’ve run out of ducking room.
This will extend the drought of real “overclocking news,” and slow down the pace of improvement. A year from now, overclockers may be able to coax 20% more actual performance than they can do today. That’s not much.
To add to the whole situation, we’ll have new memory technologies, new video technologies, a crossover in hard drive technology. All of these will yield benefits in the future, but they won’t when they first come out. New technologies rarely do; they usually cost a lot and give relatively little more back.
Take all of these together, and you get a longer drought than usual. If you’ve never seen a drought before, this looks bad. It looks worse if you’ve never seen a drought before, or don’t even know what a drought is.
Tweaks and mods? Let’s face it, there’s just so much you can do with a motherboard, and that’s already been done with the current crop. Maybe someone will come up with something new out of the blue, but I wouldn’t count on it.
People will come up with things, but it’s hard to see a steady stream of major or new tweaks for old stuff, and most of you are going to have old stuff for quite a while.
So things will be quiet for quite a while.
But if you live in the Eternal Now, that answer doesn’t compute. You tell such a person these things, and since they don’t accept looking at past or future, they just tell you, “Give me overclocking news NOW or I’ll go somewhere else.”
And what will likely happen is that they’ll find someone who will tell them what they want to hear, and end up wasting a good deal of money.
We’re sorry, but in all good conscience, we’re not going to pimp people’s desire for something new. If we wouldn’t buy the stuff, how can we tell you to?
Time To Retrench and Consolidate
The answer to this is not to pretend the train is moving when it isn’t.
Nor do we think turning this into a P2P commentary site for a year is appropriate. We’re not going to stop covering it, but we’ll keep it within bounds.
We really don’t think posting a new article every three or five or seven or fourteen days is too good, either. We don’t think futile searches for new material should become a major source of revenue.
The irony is we’re in a golden age of overclocking now. Both Intel and AMD have inexpensive to very inexpensive systems that give you a lot for your money.
The smart thing to do is get up to speed (if you haven’t already), and wait out the drought until something significantly better comes along.
What we’ll probably do as a major part of saying something worthwhile in the upcoming months ahead is to go over all the information and best practices out there today, and put it into a format more easily referenced.
A big problem with the overclocking scene today is that it is catch-as-catch-can. There isn’t one place where you can get a solid grounding in what to do and how to go about it.
In essence, a sort of overclocking book.
Easier said than done, and something that will take a ton of time to complete, but we can put it out section by section over the course of time.
That looks better than the alternatives.