Insights on Intel

[anhline]

In case you haven't heard, Apple announced last week that it will begin a phased transition to Intel processors in 2006. I've heard and read a ridiculous amount of misinformation and general retardedness in the last 7 days, as is expected w/ this kind of news. Most of this retardedness involves people assuming that Apple will be using Intel chips

Comments

[brentturbeaux]

The only few places where Apple's going to have a hard time in the transition, as far as I can guess, would be:

1) Sales on current towers and notebooks will take a dive because of people waiting for the next big thang. I'm in this boat, actually, because I was thinking about G5-ing myself in the next few months--prior to hearing this news. How can a ~3% market share possibly withstand a damn good reason NOT to buy a Mac?

2) Apple will have an incredibly hard time convincing non-Macvangelists that an Intel Mac should be more expensive than an indentically equipped Intel PC (you know it's inevitable). Macvangelists, of course, can be convinced of anything.

3) Intel releases new chips with breakneck frequency, and publishes their roadmap well ahead of time (Apple couldn't possibly like that). To keep up with PCs, they'll have to dump the latest Pentium in there every few weeks or else look like slackers.

[metaclops]

Your first point has a name: The Osborne Effect.

The move to x86 was unavoidable, IBM just could not deliver. Kudos to Apple for seeing this and maintaining x86 OSX builds.

[brentturbeaux]

IBM could deliver, but not to a customer who purchases such a staggeringly low number of chips. I can see the board meeting now. "IBM, we want you to make the fastest desktop CPU in the world. And we want you to make.... ten thousand of them."

[anhline]

Having a nice performance roadmap is definitely important but I'm sure when it came down to the bottom line, it was all about the money, from both sides of the story. With more dough flowing in from Apple to IBM, I'm sure they could have resolved their 90nm issues in a sufficient time frame. From Apple's perspective, why pay IBM more money for more development when they could probably get the same if not better from Intel for less money?

Hey, ten thousand chips though? A million Macs every 3 months is a little more than that. ;-)
But yeah, not nearly the same numbers as all three of the gaming consoles would get for IBM.

[metaclops]

Thank god I'll never have to hear about the megahertz myth again.

;-)

[anhline]

Well, unfortunately, with Intel's processor roadmap (and AMD's for that matter), you will probably hear about the Mhz myth for quite some time to come. Their next gen CPUs (and most of the current Pentium-M stuff) will clock significantly lower than a P4. Centrinos weigh in at what, like in the 1.6-2Ghz range? Intel's dual-core/64 bit stuff looks like it'll come out somewhere around 2.5Ghz or possibly lower. More and more, the focus is switching from frequency to bandwidth. Intel has publicly stated that they will probably not be able to hit 4Ghz w/ their current line and are developing more efficient architectures for their future offerings.

[fetalpig2]

Jargon Jargon Jargon, what this comes down to is a soultion for the hot ass G5 powerbook, i can finally trade in my 17" powerbook for a slick ass G5... so i can more effectivly write my papers for school, this G4 is just not cutting it, WHO WANTS SOME MONEY?

I, Cringely

[everydaylloyd]

I'm sure you've read this article?

Re: I, Cringely

[anhline]

Yeah, that guy is a fucking ignoramous.

"moving to the Cell Processor would have made much more sense than going to Intel or AMD"

The other stuff is so bad it's not even worth mentioning...

Re: I, Cringely

[brentturbeaux]

Yeah, moving from one hard-to-produce boutique processor to another. SMART.