The Inquirer-Home

Intel pulls 45 nanometre chips into 1H 07

So says the rumour mill
Sat Feb 10 2007, 21:54
THE HOT RUMOURR of the moment is that Intel is about to pull its 45 nanometre shrink into the first half of this year and so basically six months early.

It may want to, but is it technically feasible? It isn't like it is simply buying off the shelf parts, because this is bleeding edge tech.

There are two parts to the process, the design and the manufacturing. Think of this as the pretty patterns you draw and the things you draw them with. Both need to be on time and with the desired characteristics or you get nowhere fast. Intel has both planned forthe fourth quarter and to pull Penryn in, both are needed.

The design side is the more probable candidate. The core is, by all accounts, in great shape, better than Core Number Numeral was at this stage of the development.

But when we heard the rumour of a massive 45 nanometre breakthrough, we dismissed it because of the manufacturing process. Pulling in the tools, ramping the process, and just physically moving all that stuff into the fabs was going to be somewhere between damn hard and impossible.

In talking with manufacturing people, we've learned this might not be as impossible as it seems. The 45 nanometre process was basically done and set in mid-2006. With the recipe set, it was just a matter of equipping the fabs and getting the tools in place. Shake them down, and hopefully you are in business.

The tools are the critical components, and they may or may not be done in time for Intel to pull the schedule in. The key to this is the critical lithography tools. At 45 nanometres, Intel is not changing litho tools to a large degree, it is going for two pass dry lithography using many of the same parts as it did at 65 nanometres.

In both cases, this boils down to it being technically possible, barely. Assuming perfect execution, no stumbles along the way by Intel or its partners, and the stars aligning correctly, means it might happen.

That brings up the question of whether it make financial sense? You are pulling in 45 nanometres with the attendant expenses and losing quarter of the life of your 65 nanometre process. These are billion dollar footnotes, not trivial little things. Even if it can, Intel may choose not to for financial reasons.

Basically what you end up with is a rumour that is on the fringes of possibility, but does hit it. Intel could pull 45nm in to 1H/07 but it would be a long shot. Then again, stranger things have happened in the industry lately. µ


Share this:

blog comments powered by Disqus
Subscribe to INQ newsletters

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

INQ Poll

Heartbleed bug discovered in OpenSSL

Have you reacted to Heartbleed?