S&B

AMD released its tricore processors to generally lackluster reviews, but in the middle of all that, this little item got slipped in.

“AMD doesn’t have the resources to spin a dual-core Phenom die . . . .”

Oh. They just figured this out? Then why did they promise these chips, even gave them names and model numbers and speeds? Didn’t they waste all the development time and money dedicated to these chips? Or does AMD just make these things up even before doing any work on them?

Well, we figured that out four months ago.

There aren’t going to be any “native” tricores, either:

AMD has no plans to make a separate triple-core die, simply because it would require quite a bit of engineering resources and the need for triple-core CPUs diminishes over time as quad-core adoption increases. Right now AMD is focused on bringing its 45nm Phenom processors to market and those are occupying all of AMD’s availability engineering resources.

Oh. It’s better to break quads than to make tris and chew up 285mm of die space to compete against Intel chips which chew up 143mm (at 65mm) and 107mm (at 45nm). It’s even better not to make anything for the 90%+ of the market that remains dual-core.

This is, of course, ridiculous and insane unless you never planned on making very many to begin with.

Do you really think AMD plans on selling many tricores? Look at the retail pricing! What’s the slogan, “Get 75% quad performance at 90% of the price?”

But 45nm is coming, you say. OK, what does AMD plan on doing with those? As of right now, we’ll see some quads showing up 4Q 2008. No duallies, no tris. The 45nm duallies are supposed to show up last, just like the aborted 65nm chips. At the moment, it’s alleged that 45nm duallies will show up 1H 2009, just like the now extinct 65nm duallies were supposed to show up 1H 2008.

But just when you might expect AMD to be able to make enough 45nm chips to actually make duallies, then they “plan” to start converting over to high-K 45nm. They’re also supposed to make “samples” of Bulldozers around then, too, but again, no promises of actual production.

I don’t know about you, but I’m seeing a recurring pattern here. As I said two months ago.

AMD can’t make these chips in volume, and because they can’t, they’re stunting with low-volume quads to make themselves look more advanced than they really are, and hide the fact that they can’t ramp these new generations to serious levels of production.

This company ought to change its name. They certainly don’t make advanced micro devices any more. How about S&B “Stunts and BS?” That seems to be their major products these days.

If you think I’m being a wee bit too pessimistic, there’s a very simple question some analyst, anybody can ask the head honchos to settle this: When do you expect 45nm crossover, in units sold?

If they give a definitive date, any definitive date, that’s a sign of seriousness, a time by which the stunting will have stopped.

But I would bet they wouldn’t answer, and that’s proof of it.

Ed


Be the first to comment

Leave a Reply