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: hmaly who wrote (79697)	5/8/2002 9:51:16 AM
From: Joe NYC	Read Replies (1) \| Respond to of 275872

Harry, As you can conclude on your own by looking at the necessary math, there isn't a single Pentium 4 DDR solution available today that can offer the amount of bandwidth necessary to feed a 4.26GB/s 533MHz FSB. True, not today because all that's available are single DDR channel chipsets. This will change, and between Q4 2002 and Q2 2003, Sis, Via and Intel will all have dual channel DDR chipsets. Joe

To: hmaly who wrote (79697)	5/8/2002 3:56:19 PM
From: Ali Chen	Read Replies (3) \| Respond to of 275872

"According to who? Not to Anand. Read this part." Why should anyone read this blurb? There is no such thing as "necessary bandwidth it feed a xxxGHz" processors, any bandwidth is not enough since times when CPU core speeds started to exceed the multi-cycle memory access time. What is more important wrt Rambus is that the 533/PC1066 has not delivered anything close to what was expected. For truly scalable memory the SPEC scores for a 2533 NW processor must be at 930/950 level, yet Intel reported only 830-860 range in their PR. Also, if you look at www6.tomshardware.com the system with PC800 in several cases OUTPERFORMS the PC1066 system: MPEG4: 2400/133/400 = 35.55 fps, 2400/133/533 = 35.18 fps ( www6.tomshardware.com ) MPEG2 with the same result. In 3D Studio Max 4.3, www6.tomshardware.com 2400/133/PC800 and 2400/133/PC1066 are equal, while 2533/133/PC800 (143s) outperforms the 2533/133/PC1066 (146s) (if I got his notations correctly. There is some discrepancy in the section that lists the memory used). In any case, the REVERSED gain in performance should trigger some questions. The answer could be as simple as BIOS misconfiguration, but I think it is a manifestation of a general scalability problem with Rambus. I think the latency of PC1066 remains largely the same, and this causes those weird discrepancies in benchmarks scalability. Looks more and more like a dead end, pawing the road to DDR+. - Ali