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 : Rambus (RMBS) - Eagle or Penguin -- Ignore unavailable to you. Want to Upgrade?

To: Bilow who wrote (73546)	5/23/2001 8:17:21 PM
From: richard surckla	Read Replies (1) \| Respond to of 93625

A Massive Attack Of RDRAM From YAHOO by: kerry_lass 05/23/01 07:17 pm EDT Msg: 286852 of 286891 A Massive Attack of RDRAM [3:43 pm] Gavric The price cut for Pentium 4 CPUs has boosted the demand for RDRAM. Memory chip manufacturers have rushed to make more and more RDRAM modules. By the end of the year the production is expected to grow four times compared with today’s volumes. In the beginning of 2001 Samsung - RDRAM manufacturer #1 - produced 5 million of «equivalent 128Mbit» chips a month, but in March (with Intel’s assistance) it increased the volumes to 10 million. In June the company plans to double the output once again to make it 20 million chips a month (that will account for 30% of Samsung’s total production). Since February Toshiba manufactured 2 million of RDRAM chips a month, and by September it wishes to bring it up to 8 million a month (or to 60% of its total output). The third RDRAM manufacturers, Elpida Memory, is now releasing feeble 200,000 chips a month. In the second semester of 2001 it plans to reach the bar of 5 million chips a month (30% of the company’s output). The Taiwanese Winbond has scheduled to launch with Toshiba’s help a trial party of RDRAM, but the starting date of mass production and the expected ultimate output are not announced yet. Alongside with expanding the production of RDRAM for desktop PCs, Samsung and Toshiba are willing to manufacture more RDRAM chips for other sectors - net equipment and household appliances. Together they make about 20% of all the RDRAM market" xbitlabs.com

To: Bilow who wrote (73546)	5/23/2001 9:46:14 PM
From: gnuman	Read Replies (2) \| Respond to of 93625

You guys should hear what the Intel reps are saying about you in meetings under NDA: Intel rep's can only talk about us guys under NDA? What makes us guys so special, are we some kind of IP? LOL

To: Bilow who wrote (73546)	5/23/2001 11:19:04 PM
From: Jdaasoc	Read Replies (1) \| Respond to of 93625

carl: 478-pin socket will be used by the 0.13-micron Northwood Pentium 4 at year-end, supporting 512KB on-chip cache, up to 533MHz bus using dual PC1066 RDRAMs with dual PC2100 SDRAM buses in some chipsets, dual-CPU operation, computertimes.asia1.com.sg Please cite real technical white papers from leading manufacturers not $75 per 250 words hack articles. The article author could be a screwdriver shop owner from Singapore. (Nebojsa Novakovic is the managing director of a company that designs high-end computer systems.). This article you cite is pure speculation on future products. What you clip out totally contradicts what has been delivered from Intel in last few weeks. Why would Intel use a 478 pin socket for dual processor support when they already have technically superior solution for XEON 1.7 MHz with 600+ pin socket. You yourself stated that pins are cheap. Intel I guess is not wasting some of those 600 pins on DDR or dual DDR memory interface pins but for something more useful like state of the art performance. RE your statement Probably the thing to note is that: (1) Intel says PC266 may not work, but they said the same thing about PC133 before they chipsets for it. Please substitute 150 MHz FSB instead of 133 MHz which you can agree is in the works since the FSB have increased steadily from 4.33 MHz on the 8088 in 1981 to 133MHz now. In this situation RDRAM @ 600 MHz, which should not be technical difficult to produce at smaller DRAM geometries,(PC 1200) will produce all the bandwidth Intel will require; while DDR (PC2400) falls flat on it's ass due to much tighter timing requirements and lower percentage bus utilization due to dead bus cycles. Intel was found it's long term memory interface so expect fast turnaround on new P4 projects since they do not have to tweak DDR tight timing specifications. I was totally amazed dual XEON based solution came out so shortly after P4 release that puts 40% performance lead over best AMD Athlon solution which was the king of desktop CPU's. Keep the slop coming john