Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Charles Gryba who wrote (149403)	11/25/2001 11:20:28 PM
From: wanna_bmw	Read Replies (2) \| Respond to of 186894

Constantine, Re: "r u the same guy who a month agi was saying how impressed you were with hammer? What changed?" First, can we try to spell out our words, rather than looking like we just graduated from middle-school? No offense, but I guarantee that anyone that says "r u" instead of "are you" in an argument will automatically be placed at a disadvantage. To answer your question, though, I was an still am impressed with AMD's implementation, but like I said then, what's on paper doesn't necessarily translate to a very good product. Most of Hammers improvements are not revolutionary. The largest change has to do with the coherent Hypertransport controllers for multiprocessor systems, which I believe are only going to be a unit in the Sledgehammer variation of the design. I imagine that in order to reduce cost and validation time, Clawhammer will only have one Hypertransport connection. Additionally, the Clawhammer chip will probably not at first support all the functionality of the "Sledge" chip, which may include 16 byte memory support, scrubbing, or chipkill. Anything to reduce validation time will probably be thrown away in order to get Hammer to market faster. The integrated memory controller will benefit a lot on some applications, and the improved TLBs and BPUs will probably add more to IPC. I highly doubt, however, that these small changes can in any way account for the 50% in IPC that Milo predicted (actually, Milo said there would be at least at 50% increase in IPC). You might be able to see why I'd be skeptical. Now, I am certainly not willing to call Hammer a failure until after it hits the market. For all I know, it may offer an appreciable increase in performance, and Intel may stumble along the way. In all likelihood, I disagree with the theory that Intel will stumble or that AMD will have a much better product than anyone can imagine, but I do give them credit for reaching another generation of their micro-architecture, which they will need to continue their competitive climb. Anything short of current expectations, though, will probably not save them from future Intel initiatives. wbmw