Hardware

Last week's EmTech 09 meeting played host to a panel discussion on the future of data storage. All three of the panelists were from companies that have a poorly known product on the market, and each of them discussed improvements that are in the pipeline, which we'll cover towards the end of this article. But they also provided a more general overview of the challenges facing storage technology at a time when data production is beginning to outstrip our ability to cope with it.

This past Wednesday, September 25, we hosted our second Ars Premier Chat. We'd like to thank again all those who participated and submitted questions. We got so many great questions and comments that Jon didn't have enough time to answer them all in our allotted hour of time.

So we asked Jon to tackle the best four remaining questions. If you're simply interested in reading a transcript of the live event, scroll down or click here.

If you've been putting off buying a new PC because you're waiting for Windows 7 to ship, you're in luck. At least one manufacturer plans to make Windows 7 PCs available to customers early, nine days to be exact, and Redmond isn't going to get in its way. In fact, Microsoft is perfectly fine with the idea. "October 12 is the date that Microsoft enables our Authorized Replicators to begin shipping Windows 7 to Microsoft OEM Authorized Distributors," a Microsoft spokesperson confirmed with Ars.

"As the fulfillment process through our supply chain takes time, we begin initial product distribution in advance of General Availability. Therefore, this represents our Release To Manufacturing (RTM) process for OEMs who purchase indirectly through Distribution. So, while in theory a System Builder could get Windows 7 product pre-GA, based on supply chain analysis, the intended timing for customers to receive Windows 7 PCs from System Builders should be close to GA, October 22nd. This represents the same process Microsoft managed with past Windows releases."

For PC gaming, this week was the best of times and the worst of times—new graphics hardware on display from AMD and Intel, but aimed at an ailing PC gaming market that's in its worst shape ever.

At AMD's big VISION event earlier this month, the company gave a technical deep dive on its next-gen GPU family, codenamed Evergreen. The embargoes are now up on the technical details and reviews of the first product in this family, the Radeon 5870, an embargo lift that was no doubt timed to take a little of the wind out of Intel's long-awaited Larrabee reveal yesterday.

Recently, we launched Ars Premier 2.0, our new and improved subscriber program that offers some fantastic new benefits for subscribers. Tonight sees our second insider-only feature: a live, moderated webchat with Ars co-founder and Deputy Editor, Jon Stokes.

The chat will start at 8pm EDT/5pm PDT tonight, so return to this page at that time to participate. Jon will be answering premier subscribers' questions on all things CISC vs. RISC, multicore, SoCs, and computer architecture. All you need is a Web browser for this and all of our other Moderated Live Discussions. If you're not a subscriber yet, and want in on this and our other Premier offerings, there's still plenty of time to sign up.

Ten years ago, I waded into the then-raging "Mac vs. PC" wars with a lengthy treatise on "RISC vs. CISC: the Post-RISC Era." In the conclusion to that article, I declared the "RISC vs. CISC" debate over—the terms were irrelevant, I argued, because they were artifacts of a bygone era that had now been robbed of any once-useful technical meaning by Apple's marketing department.

In 2009, Apple has abandoned PowerPC for Intel, RISC designs have been squeezed into the very highest and lowest ends of the processor market (Big Iron and embedded systems), and the only place to buy a "RISC workstation" is on eBay. Meanwhile, x86 continues to press upwards into high-performance computing and to squeeze downwards into the mobile market. 

It appears, then, that to the extent that it's still worthwhile to think about the technology rivalries of the past decade in terms of "RISC vs. CISC," CISC has clearly won. And if CISC won, then RISC was a useful idea whose time came and went, and some of its better ideas live on because Intel has adopted them for its x86 family. 

Except that it's not quite fair to either history or the present moment to speak of a victory for one side. The real story is much more complex, and more interesting than a simple victory.

SAN FRANCISCO—CEO Paul Otellini kicked off the 2009 Intel Developer Forum with a keynote that quite literally laid out Intel's plan for world domination. The keynote's theme was "a continuum of computing," and Otellini was keen to stress the 32nm process node as the point at which x86—or "Intel Architecture" (IA), as the chipmaker prefers to call it—can claim a place inside everything from smartphones to servers.

One of the features of Ars Premier 2.0 is subscriber-only content such as live chats and articles. Tonight, we're announcing our next Premier 2.0 feature, a retrospective on the past ten years of RISC vs. CISC since the publication of RISC vs. CISC: the Post-RISC Era in 1999 by Ars cofounder Jon Stokes followed by a live chat.

The article will be published Tuesday evening, September 22 and will be followed by a moderated live chat to begin at 8pm EDT/5pm PDT on Wednesday, September 23. Jon will be answering premier subscribers' questions on RISC vs. CISC, CPU design, and other hardware topics. As was the case with the last week's chat with John Siracusa, all you'll need is a Web browser.