Netbooks . . .

Their day will come, just not quite yet.

As yesterday’s article mentioned, some PC manufacturers are very concerned about tinyPCs.

A few days ago, we got a “Man Bites Dog” moment when the head of Intel told us how much one of his products sucks.

You know, this is sort of less than a third of the performance of our Centrino, so you are dealing with something which is not — you know, most of us wouldn’t use and you are dealing with something that is principally designed for net access, for web access, not really to run robust applications and I think it’s likely to stay that way for quite some time. As a comparative, you would not want to run YouTube video on these things all day long. You would not want to do any photo editing on these all day long, or even for a little bit of time.

So I really think it’s a first-time buyer kind of thing and it will sort itself out, just like there is a different market for lower priced, lower performing cars than higher end cars.

(To be fair, he said that only after relentless haranguing about Atom by the analysts. Earlier on, someone had let slip that Atoms had a lower gross profit margin than the Intel average (more on that later), and the analysts reacted like Otellini had just declared Intel bankruptcy.)

We’ve pointed out repeatedly that eventually, tinyPCs are going to wreck the PC market as we’ve known it and no doubt more than a few PC giants are going to turn into dinosaurs and go extinct.

However, “eventually” is not now, or even very soon. We are at the dawn of tinyPCs, not noon, not even early morning.

The big argument for tinyPCs is that most people don’t need much computing power to do what they normally do with a computer, and will gladly swap that for other advantages (i.e. cheaper, smaller, lighter, more mobile, easier-to-use).

There’s no doubt that this will eventually become the norm for the average computer user. What there’s complete doubt about now is what the norm will look like, in the home, in the office, on the road, walking around.

What we are seeing now is the first generation of “netbooks,” and it shows. These range from pretty to very compromised machines, and they’re not usually all that cheap, either, especially compared to low-end “full” notebooks.

However, they have one huge advantage; they’re light, around two pounds/a kilogram, as opposed to around six pounds or more (2.5+ kilograms) for a typical “full” notebook.

Yes, there are very light “ultraportable” notebooks around, but they are much more expensive than regular notebooks, much less these netbooks.

However, that alone is not a good reason to buy one of these things. It’s a false economy to save some money buying a cheap box, then pay it back and more in lost productivity waiting for the computer to do whatever, or spending your time incessantly scrolling around to see what you need to see.

No, a good candidate for these initial netbooks (and those that follow) is someone who has to bring one around for a few tasks which include the need to do some real typing, but those things aren’t computational squat.

Otellini was quite correct in suggesting that these devices aren’t powerhouses. I would suggest that many people buying now will be kicking themselves two years from now, because these devices will probably evolve rapidly and be much better by then (or sooner).

But wasn’t it interesting that Otellini mentioned YouTube as an excellent reason not to buy Atom-based systems? Doesn’t that sound like, “Oh my God, Centrino can’t lose the back-to-school crew this year!” in disguise?

The problem is, Centrino is going to lose the non-gaming back-to-school crew pretty soon once performance moves up a bit and somebody sticks a decent enough video playback chip in there, because if Intel doesn’t, nVidia will.

Once that happens, I think a pretty big proportion of full notebookers will follow. Personally, I can’t recommend that anyone in my current stable of notebook users dump what they have now in favor of one of these, but when it comes time to replace what they now have; it’s hard for me to imagine suggesting anything else.

One Last Point

Everyone noticed Otellini trashing Atom, but they didn’t notice this item from another exec:

I expected the Atom platform to be about 10 product margin points lower than the mainstream of the product line, which was dual-core, and quad-core to be about 10 product margin points higher than that. So my view hasn’t changed and I’ll update that again in Q1 next year when we have the investor meeting.

Hmmm, so quadcores have close to a 70% profit margin? Quadcores are presented as a new cash cow, a hyperprofitable offset to Atom?

I’m sure that figure includes quad-Xeons, but even given that, given Intel’s promise of higher gross margins the next few quarters, and so long as Atoms are suspected of dragging that number down, Intel is going to be under a lot of pressure to keep those quad prices high.

Maybe I’m crazy, but between quotes like these and the Nehalem luxury socket where you’ll pay $300 more for an extra 266MHz and comments like “We could sell quads for $100, but why?” I sense the old Intel arrogance coming back.

Just as lack of money is AMD’s Achilles’ heel, arrogance is Intel’s. Yes, they have much to be arrogant about at the moment, but let’s face it, Intel’s current domination is as much due to AMD’s failings as Intel’s achievements.

I suspect Intel thinks it can make a lot more money and kill AMD. I don’t think they can do both. I think it’s more likely (assuming that AMD can get its act together and shift production to 45nm quickly with little problem) that trying to do both will give AMD enough breathing room to stop the death spiral and survive.

Ed