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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Tenchusatsu who wrote (239514)	8/29/2007 3:17:56 AM
From: graphicsguru	Read Replies (2) \| Respond to of 275872

Now we know why CSI is taking so long. This unwinding mechanism sounds scary to me in its complexity. It's a big risk. On the other hand, Intel has done more complex things in the past, and gotten them to work. The P4 replay mechanism, for example. I find it impressive that there were no fdiv-like recalls on it. And maybe the CSI unwind complexity is revealing relative to the question about why the MP version is a year later. I can easily imagine that the unwinding mechanism becomes a whole lot more complicated for MP systems than for two socket systems. In a system with more than 2 sockets, there are many more opportunities for race conditions that need to be detected and corrected. It really sounds like a Q/A nightmare to me. But if Intel can make it work robustly, it's going to kick butt.