The Inquirer-Home

Hammer the Memory

Will it Hammer the competition or itself into nothing?
Sat Nov 23 2002, 00:38
AMD ISN'T doing so well. It's losing money and having to lay off staff. One of the questions that keeps cropping up is, "why don't AMD release the Hammer now in the hope of pushing up sales?"

The problem might well be one of memory. Originally the Clawhammer was going to use DDR266 (PC2100) but that's obviously ridiculous now that DDR333 is the high-end standard. We've also got DDR400 starting to impact the market and nVidia are going to be using DDR-2 500 in the GeForce FX. The memory market is in such a rapidly changing state that producing a Northbridge - the traditional memory controller chip - is a nightmare at the moment. And the Hammer has its Northbridge built in.

Why risk introducing the Hammer while memory marchitecture is in such a state of flux? There is no good reason. Rumours are flying that the VIA KT400A will use dual channel DDR400, a sensible move to keep up with Intel who are moving to dual channel systems themselves. A Hammer with single channel DDR might well perform quite badly in comparison.

So, will we see Hammers being rapidly upgraded to take in advances that competing systems can introduce? Is the single channel memory controller inside the Hammer going to be replaced? The fact that the Hammer chips aren't due for another four months could be a good thing. It might mean we'll see a better chip. On the other hand, all of the motherboards and chipsets being touted by the big players are single channel. AMD may have made a stunningly huge mistake in integrating the Northbridge. In Hammer, we might well see a chip that's dead on arrival. Time will tell. µ

Arron Rouse is a freelance Technical Author and Business Analyst The copyright is his, get your thieving hands off it..

L'Inq
This one

 

Share this:

blog comments powered by Disqus
Advertisement
Subscribe to INQ newsletters

Sign up for INQbot – a weekly roundup of the best from the INQ

Advertisement
INQ Poll

Heartbleed bug discovered in OpenSSL

Have you reacted to Heartbleed?