The Inquirer-Home

Intel to keep lead on notebook technology

But AMD will start to shiver its timbers
Tue Dec 28 2004, 10:58
2005 IN LAPTOPS WILL NOT be the steamroller of a year many predict for Intel, but it will be close. Intel has far and away the best laptop chip out there, the Pentium M. It also has some nifty marketing and branding campaigns surrounding it when you bundle it with the wireless chip. If I ever find anything out about it, I will write it up immediately. If Intel gives out M&Ms, I might even go to the press conference.

The Pentium M parts are in a class of their own, and have been since launch. In fact, there is nothing else in the same league. The 533FSB Dothan parts have a TDP of 27W, and it only goes down from there. Even at this low wattage, they perform quite well. Via has comparable power figures, but not for the same level of performance. AMD has better performance but 35W is a hefty 30% more, which translates into battery life of hours less. Not cool, pun intended.

So, what does Intel have on tap? The Dothan parts at 533FSB will be launched in mid-January, just weeks from now, as we predicted some weeks ago. The Alviso chipset will bring all the goodies to market, dual channel DDR2, PCIe, and power sipping chipsets. The integrated graphics on the chipsets won't wow any gamers, but they should keep their head up in the 5.5-6.5W parts crowd.

Overall, Alviso will be the chipset to own in the first half of 2005, and a 533 Dothan with be the chip to put into it. It will stay at the top of the heap until later in the year when the 2.13GHz model is replaced by the 2.26GHz version. That should rule the roost until Yonah pops up at the very end of the year. Current rumour rumblings have the 65nm process pushed out to really early 2006, and if that is the case, Yonah will obviously go with it.

Yonah, assuming it will be out in 2005, is one hell of a chip. It has everything you could want in a laptop chip, and some things that don't really matter, but brings some really impressive bragging rights to the microprocessor dunghill. The main one is dual cores - it will basically be two Dothans merged on a slab of silicon. Unlike the Smithfield cores, it won't be just two cores next to each other.

To impress your techie friends, point out to them that Yonah will have a shared L2 cache, and when it is running on battery power, one core will shut off, leaving the other core with the full 2MB cache. A 667FSB and the dual channel DDR2 should make this a potent chip.

Add in the massively improved SSE2 and new SSE3 units, and you have a multimedia monster. Other improvements, especially to IDIV will only add to this, you can probably encode a DVD on a flight, and still have battery life left over to watch the movie because this all will only consume 31W. Yonah, you go Grrl! Sorry about the lack of decorum, but it will be that good.

AMD is not without a response though. It currently has 2GHz mobile A64 parts, but 35 is a long way from 27, and even longer from the 21W of the 400FSB parts. The AMD notebook chips are a definite player in the desktop replacement laptop market where power is more important than battery life or weight, but in the thin and light or corporate markets, AMD is simply not a factor.

This will change sooner rather than later. The E0 step chips will have a reported 24% decrease in power, so you could put that same 2GHz A64 in a 28-29W envelope. Add in the north bridge features already in the CPU, the memory controller being the big one, and you are neck and neck.

This is not lost on AMD, and it has a Centrino killer coming very soon. No word on the name yet, but several people have them now, so it won't be long until you do also. I would expect it in late January, which is when things get interesting in the laptop arena.

On the down side though, there is TDP and there is TDP. If I had to guess, I would say Intel does not hit the TDP wattage on Dothan much, but AMD gets to its max numbers a lot more frequently. This means we will be in another round of benchmark games, with carefully constructed numbers telling anything but the real story. Either way, it won't be a bad chip - will hardly slay the Pentium M, but it will make Intel sweat.

The marketplace will look like this for 2005, Intel on everything that matters. AMD will make a valiant effort, but it will not dislodge Intel from prominence. It will bring the fight to Intel's doorstep, but without a bit more magic than I think they have up their sleeve, it won't be enough. AMD will hold ground and possibly gain a bit in the blue collar set's laptops, but the white collar guys will all have Pentium Ms.

When Yonah launches, unless AMD has real magic under its belt, it will be Intel leaping ahead once again, game over for now. Merom will be higher wattage initially, so it won't be a challenger for the crown, but an early appearing Napa64 might give AMD fits.

The only wild card to all of this is 64 bitness. AMD has it, Intel doesn't. Since the functional limit of 2GB is not a pressing limit among the notebook set, and high end gaming is lacking anyway, I don't think it will matter all that much. In fact, I think it will only matter in notebook advertising, which is more likely to sway blue than white collars.

I would think the market will end up with a meagre point or two above where it started the year at, mostly due to a halo effect from the desktop and server parts. On merit, it is all an Intel show, and if it does nothing to screw it up, and I don't think they will, it will laugh all the way to the bar at the Marriott, Santa Clara. ยต


Share this:

blog comments powered by Disqus
Subscribe to INQ newsletters

Sign up for INQbot โ€“ a weekly roundup of the best from the INQ

INQ Poll

Heartbleed bug discovered in OpenSSL

Have you reacted to Heartbleed?