What is most fascinating (though often in a perverse way) about these debuts is not the technological but psychological displays.
It not only shows the computational limitations of the CPU, but also the computational limitations of the CPUers.
April 22 was the first day of Hammer, not the Last Day, as in Day of Final Judgment).
To me, looking at Opteron today tells me what to look for six months or a year from now. It identifies problem areas. That’s all.
For the current chips, at absolute worst, there’s nothing wrong with them that a lower price tag couldn’t fix.
Conversely, there’s nothing wrong with the Hammer that a higher clock speed couldn’t fix, either.
For those who think some motherboard magic or the like can do the trick, I’m somewhat skeptical about a major change, but since neither you nor I are going to be buying these things anytime soon, we can afford to wait and say “Show me.”
Even if the Hammer core at 130nm is etched in stone, there’s always 90nm migration and plenty of time to make changes at that point.
Given that one or more of these changes is at least possible, it’s hardly time to render final judgment. That doesn’t mean you can’t make any judgment for the moment; there are definitely areas of concern, but any and all judgments at this point are subject to change (or not) as the realities change (or not).
Outside of concerns about the financial shape of AMD, I’m not really all too concerned about how well or badly Athlon64 turns out in September or October, either.
That’s simply because almost all of you with socket A boards don’t want to change platforms at that point. You’re happy to play with cheap TBredBs, and you don’t see enough of a bang to justify the big bucks associated with a platform change.
It won’t be until 90nm chips come out (or don’t timely) that people will make their decision on the Hammer platform. They find no need to rush to final judgment, and neither do I.
The Other Road
To me and most of you, this way of thinking is hardly a major intellectual exercise. Yet it seems completely beyond some.
For those people, the mental process seems to be, “Choose sides, appoint yourself a mother hen, defend your adopted chicks against all perceived threats, and peck anyone else’s chicks whenever you can.”
For them, it doesn’t matter one little bit why you may think what you think, to what degree you may think it, or how permanent your thinking is. It does not register; there is no color in their universes, just black-and-white, yes-or-no, for-or-against. It’s “Are you with us or against us?” and that’s that.
In short, there’s an emotional attachment and involvement in a company or product or technology, and when that happens, much, most or all of the brain tissue doesn’t boot any more. You join the tribe and stop thinking treasonous types of thought.
Frankly, when it comes to getting emotionally attached to a company or product, neither Joe nor I can do that. To put it frankly and crudely, we just don’t give an emotional shit about this stuff. It’s just not something to get emotional about. We look at this, and like Yoda, we say, “Do or do not.”
We think that’s a healthy mental attitude to take about processed sand, and fortunately most of you share it.
For instance, you may prefer AMD or Intel, but it’s not a blind loyalty, but you prefer them because of what they do rather than what they are. You prefer them because they do what you want, and if they stop doing whatever that is, you’ll stop preferring them.
This is the way it should be, but try telling the self-appointed mother hen that there might be something wrong with her little chick.
There’s no rush on this. Sorry, but this is going to take a long time to play out. If you’re only capable of one immutable judgment on this, come back in July 2004. In the meantime, see if you can fix that rigidity.
If you need to join a tribe, try to find one with goals a little more worthy than worzhipping micrograven idols. Even ones that talk to you 64 rather than 32 bits at a time.
These are smart screwdrivers. They’re meant to be used, not loved.
And maybe chill a little, and try to see if you can upgrade your black-and-white set to color.