News and Analysis for the Strategic Decision Maker
Search tips Advanced search
_VIEWPOINT: January 10, 2000
A Look at DRAMs in 2000 (Part 1)
Jim Handy and George Iwanyc
After a long delay, Intel Corp. has finally started to ship its 820 chipset, the enabler for the use of Rambus DRAMs (RDRAMs) in the PC. Dataquest expects for this event to have a significant impact on the DRAM market over the next year.
Jim Handy and George Iwanyc are senior analysts with Dataquest/Gartner Group, a market research firm based in San Jose. The second part of their outlook for the DRAM market in 2000 will appear in the Jan. 17 issue.
PC-133 has been viewed by some as the interface that will make Intel rethink its RDRAM plans. When Intel announced during the last Intel Developer Forum (IDF) in September that a 133MHz DRAM would be supported by future Intel chipsets, some pointed to this as a turning point in the company's previously unrelenting push toward Rambus.
This appears not to be the case. 133MHz DRAMs are currently suffering from a significant price slide and it would have been ridiculous for Intel to have ignored the opportunity to scoop up some additional bandwidth for free. All those chipsets that Intel originally intended to ship using PC-100 are fine candidates for conversion to 133MHz. The only thing standing in the way is Intel's concern that systems using SDRAMs faster than 100MHz cannot be made to operate reliably with a wide array of DRAM module types plugged into the bus. The company now seems to have decided to stretch to meet this challenge. As a result, Dataquest expects to see 133MHz SDRAMs almost completely replace PC-100 parts over the course of the year.
Why has Intel chosen to ignore 133MHz technology until now? We do not believe that this has been solely the result of technical difficulties. Intel has been worried about a lack of focus at the DRAM manufacturers. Intel believed that it needed to push single-mindedly towards RDRAMs as the only solution in order to make DRAM manufacturers finally accept this as a possible outcome. Had Intel been seen to waver in this respect, they would not have been able to make DRAM manufacturers put in the time, effort, and expense of implementing viable RDRAM solutions. It would have been more than easy for DRAM manufacturers to justify putting off an investment in RDRAMs at a time when they were already losing significant sums due to history's most protracted DRAM oversupply.
There has been a lot of talk lately about double-data rate DRAMs, or DDR, and the possibility that this less-expensive technology could postpone the need for RDRAMs indefinitely. Dataquest does not buy into this theory for a number of reasons. First, the DDR interface, although easier to implement than RDRAMs (without the burdens of an increased die size and a royalty payment), is a solution only at the DRAM chip level. Many of the cookbook details that have been so neatly worked out by Rambus' engineers, details like signal paths, termination, and clocking, are left to the individual OEM in the case of DDR. This can significantly slow down time-to-market. Second, the high-signal frequencies used by DDR are alien to most circuit board designers and are likely to end up causing trouble when brought through one or more connectors to an indeterminate number of DRAM chips. Other reasons include a lack of the required support components and even a lack of rigid standardization that could impede the acceptance of this technology until the true winner is determined from an array of nearly compatible devices.
_
_
Copyright © 1999 Cahners Business Information, A Division of Reed Elsevier, Inc.
|
