SI
SI
discoversearch

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) News Only -- Ignore unavailable to you. Want to Upgrade?


To: Thomas C. Donald who wrote (86)12/6/1998 6:10:00 PM
From: Thomas C. Donald  Respond to of 236
 
DRAM suppliers support new memory standard
Jo Pettitt

11/30/98
Newswire (VNU)
Copyright (C) 1998 VNU Business Publications Ltd.; Source: World Reporter (TM)

A number of DRAM chip suppliers have today announced their support for Double Data - Rate Memory(DDR) The group, which includes IBM, NEC and Fujitsu and eight other companies, are hoping that by announcing their support for the development and manufacture of DDR SDRAM memory components and modules, they will increase industry awareness of the benefits of the technology and push its adoption. DDR SDRAM is an open standard developed and approved by the Joint Engineering Device Engineering Council(JEDEC). The specification addresses the high data throughput requirements demanded by dramatic increases in microprocessor speeds. The standard is particularly beneficial to servers, workstations, PCs, data communications and consumer products in which performance parity with the microprocessor is critical. Jim Handy, director and principal analyst at Dataquest commented: An industry-wide push for DDR is critical to its adoption. This demonstrates the industry's seriousness about this technology. DDR SDRAM offers immediate advantages in terms of performance and cost for applications that require large memory configurations or that require an evolutionary range of performance from 66MHz to 200MHz. He added: This criteria applies to servers workstations and data communications but many other applications can just as easily benefit from this continuity and flexibility.





To: Thomas C. Donald who wrote (86)12/6/1998 6:15:00 PM
From: Thomas C. Donald  Respond to of 236
 
DRAM options forcing hard choices
Mark Ellsberry

11/23/98
Electronic Buyers' News
Page 24
Copyright 1998 CMP Publications Inc.

As far back as anyone can remember, DRAMs have been a commodity business. But if you look closely at the parts coming out today, it's pretty easy to figure out that we're on the threshold of market specialization.

PC-100 SDRAM, the main memory used in today's PCs, is giving way as the industry prepares for a rapid shift to Direct Rambus DRAM. But even as this transition begins, emerging memory types, like double - data - rate (DDR) SDRAM and SLDRAM, threaten to fragment the computer market.

Because of this, DRAM manufacturers and PC makers are being confronted by a series of tough choices.

Suppliers are bombarding OEMs with proposals for improving DRAM performance, leaving their customers two options: pick one memory type to create a single, new standard, or allow the technology path to fracture into as many as three new directions.

Indeed, if everyone had their way, vendors next year would be dealing with EDO, page mode, PC-66, and PC-100 SDRAM; Direct RDRAM; DDR; and SLDRAM parts, all at the same time.

In addition to the wider field of DRAM architectures, manufacturers must also make choices regarding new memory densities. Historically, the industry has expanded from one generation to the next by a factor of four, i.e. 4-, to 16-, to 64-Mbit parts.

But the move to 256-Mbit DRAM presents suppliers with a two-fold technology challenge: To ensure that parts are cost-effective and available in the right package, vendors must migrate to a design rule no larger than 0.18 microns, while at the same time converting to larger, 12-in. wafers. Both challenges represent significant technical hurdles, and are very expensive transitions at a time when the industry is not well set up to make these kinds of investments.

In fact, the hurdles are so large that they will force the industry from an historic trend. For the past 20 years, DRAM densities have generally followed three-year cycles. From 1994 to 1997, for example, 16-Mbit DRAM was the most cost-effective part. If history repeated itself, then the 64-bit would be the most cost-effective part from 1998 through 2000.

However, Hyundai and other companies have elected to stick a half-generation, 128-Mbit device in this time slot. The result is that customers will get the denser memories they really need at better cost.

To do this, however, the industry will have to borrow from the 64- and 256-Mbit generations. Instead of 64-Mbit appearing as the most cost-effective part until 2000, it will extend only to 1999. In 2000 and 2001, 128-Mbit will be the most cost-effective part, with 256-Mbit taking on that distinction in 2002.

Memory configurations present manufacturers with other issues. In the past, the DRAM industry has improved chip performance by going to a wider part. A x1 device, for instance, took about 100 ns to decode one bit of data, while a x16 part takes about the same time to decode 16 bits of data. This technique worked well, but its cost-effectiveness is exhausted at x32.

Switching to Rambus may correct this issue. Direct RDRAMs transfer data at either 600 or 800 Mbytes/s-depending on the system design-while using an 8-bit-wide bus. But the performance improvement will come at the cost of device complexity.

These factors are forcing DRAM suppliers to make some hard decisions. In the past, the premise was to build DRAM to be the smallest, most cost-effective memory.

However, OEM performance requirements are starting to splinter that low-cost, small-die model. That's fine, as long as OEMs recognize that DRAMs may not achieve the same economies of scale as before. The historic learning curve, which has enabled suppliers to lower costs by an average of 30% annually over the past 25 years, may be affected.

The bottom line with this new, expanded DRAM menu is that we're on the verge of losing standard-part pricing. The traditional method of comparing different costs by factoring a premium into the price of a commodity architecture may become meaningless if there is no common denominator available for comparison.

-Mark Ellsberry is vice president of marketing for the Semiconductor Division of Hyundai Electronics America, San Jose.

November 23, 1998



To: Thomas C. Donald who wrote (86)12/6/1998 6:20:00 PM
From: Thomas C. Donald  Respond to of 236
 
Predict The Future? Forget About It!

11/16/98
Electronic News
Page 8(1)
COPYRIGHT 1998 Cahners Publishing Company Copyright 1998 Information Access Company. All rights reserved.

I have learned not to predict how the winds will blow in this crazy business. As cliche as it may sound, the only constant in this business is change. Some might wonder why there would be so much change associated with a product as simple as memory. I mean, let's face the fact that memory is not brain surgery. It all comes down to how fast it can go, how much of it is available in the marketplace, and how much it costs. In those three simple issues lie the basis for the changes that seem to come at us faster than we wish they would. Let's take a quick look at each of these issues.

There are so many competing memory technologies on the horizon that the question of how fast memory can go has become very confusing. We have PC133 around the corner as well as Rambus, Sync Link, Double Data Rate and other technologies all ready to take us beyond the current PC66 and PC100 speeds. Which will be the standard of choice? Rambus certainly looks strong, but beyond that, all I am willing to say is that speeds will increase - and dramatically!

As far as the worldwide memory supply, it seems like just yesterday that there was more than enough memory out there to satisfy all takers and then some, but now we are hearing words like, "allocation" and "shortages" again. Of course, that affects pricing. Just when we were all ready to give up as prices continued to slip to obscene levels, the shortages are creating pricing spikes that don't appear to be leveling off anytime soon. These are all developments that only few could have predicted and only then with some serious luck!

So, let me conclude by saying that I am sorry to not offer something more concrete. I guess the best I can do is offer the advice: Hang on and enjoy the ride!

Craig Yerkes is managing director for Memory Card Technology in the U.S. More information about MCT is available at www.memory-card- technology.com.



To: Thomas C. Donald who wrote (86)12/6/1998 6:26:00 PM
From: Thomas C. Donald  Respond to of 236
 
Servers And Workstations Are Likely Niches -- DDR quietly making inroads
Amber Howle

11/02/98
Computer Reseller News
Page 155
Copyright 1998 CMP Publications Inc.

Irvine, Calif. -- While industry support swells for the Direct Rambus memory systems expected to be rolled out next year, a competing technology has already made its quiet debut.

Some vendors currently are shipping samples of Double Data Rate (DDR) SDRAM to PC OEMs, with volume shipments expected to reach the VAR channel by the end of next year.

DDR SDRAM essentially doubles the data-transfer rate from a DRAM chip to the main processor. While Direct RDRAM, developed by Rambus Inc., Mountain View, Calif., and Intel Corp., Santa Clara, Calif., requires a whole new architecture, DDR SDRAM is scalable with the current generation of SDRAM.

Both technologies offer comparable performance, but DDR SDRAM most likely will find a niche in the server and workstation segments over the next two years, said analysts and vendors.

"DDR [SDRAM] is a pretty good fit for large servers and workstations that need a large number of DRAMs and can support wider data buses," said Steve Cullen, analyst at Cahners In-Stat Group, Scottsdale, Ariz.

Direct Rambus technology supports a 16-bit-wide bus transferring data at 800 bits per second (bps) per pin. Although DDR will only operate at 200 bps per pin, it will support a 64-bit bus, resulting in speed comparable to that of Direct Rambus. Any data path under 32 bits will be a good candidate for Direct RDRAM, while DDR SDRAM is suitable for data paths over 64 bits, Cullen said.

The competing technologies are pretty close in terms of performance, said Sherry Garber, senior vice president at Semico Research Corp., Phoenix. "The most attractive thing about DDR [SDRAM] is that it's a more evolutionary solution . . . and can go out there at prices comparable to current SDRAM," she said.

Samsung Semiconductor Inc., San Jose, Calif., was the first to ship DDR SDRAM samples to OEMs, doing so in August. It was followed by, in order of appearance, Mitsubishi Electronics America Inc., Micron Technology Inc., Hitachi Semiconductor America Inc. and Fujitsu Microelectronics Inc. All of these vendors have signed up for Direct Rambus support as well.

DDR SDRAM will find its place in the memory industry, said Keith Horn, director of product marketing at Fujitsu, San Jose, Calif. "I think Direct RDRAM and DDR SDRAM can coexist, and our [OEM] customers are telling us that by requesting both technologies for different segments of their markets."

Intel's involvement enhances Direct Rambus' potential of surpassing DDR SDRAM in market share because more than 70 percent of the DRAM industry supports Intel products, Garber said. "But despite all the gearing up for Direct RDRAM, we believe there will be a tremendous shortfall in production," she said. That might help DDR SDRAM.

Cost issues with Direct Rambus and a likely DRAM shortage next year will open the door for competing technologies such as DDR and a third memory technology, SLDRAM, said Cecil Conkle, assistant vice president of DRAM marketing at Mitsubishi, Sunnyvale, Calif. Customers will need something available to secure bids quickly, he said.

VARs next year are likely to stick with what they know: SDRAM. But by 2001, industry observers said, DDR SDRAM and Direct RDRAM will have significant shares of the market.

November 02, 1998





To: Thomas C. Donald who wrote (86)12/6/1998 6:31:00 PM
From: Thomas C. Donald  Read Replies (1) | Respond to of 236
 
Japanese chip makers spin first DDR DRAMs
Anthony Cataldo

10/12/98
Electronic Engineering Times
Page 37
Copyright 1998 CMP Publications Inc.

Tokyo - As if awaiting the same cue, three of Japan's memory suppliers have just announced 64-Mbit double - data - rate DRAMs to add to the mix of high-speed DRAM types that will start to ramp up late this year.

Fujitsu Ltd., Hitachi Ltd. and Mitsubishi Electric are trying to push the new open-standard DRAM technology into a range of applications ranging from PCs to high-end workstations and servers, where the clock frequencies of CPUs have started to outstrip DRAM speeds. While many vendors expressed optimism that DDR will be adopted in high-end systems, questions remain as to whether they will make a good fit at the midrange to low end of the PC market.

DDR devices, which have been made into a Jedec standard, achieve high speeds by including DLL circuitry so that the device reads data on both the rising and falling clock edges of each clock cycle, effectively doubling the bandwidth. The concept is not unlike that used in Direct Rambus DRAMs, which will also soon start rolling out.

The fact that Direct Rambus is a protocol-based interface allows it to use a narrow 16-bit channel to achieve high speeds with a low pin count. DDR is a more con- ventional approach based on the low-voltage-swing 2.5-V Stub Series Terminated Logic-2 (SS-TL-2) interface for synchronous devices. Both claim to provide up to 1.6-Gbyte/ second peak bandwidth.

Because the DDR devices conform to a Jedec specification, they use a bidirectional strobe signal, a 2.5-V SSTL-2 interface and 66-pin, 400-mm thin small-outline packaging. New184-pin dual in-line memory modules are also required.

"This new technology is an open, non-proprietary evolution of standard SDRAM which doubles the peak bandwidth and is much easier to use for error correction than other advanced DRAM types," said Cecil Con-kle, assistant vice president of marketing at Mitsubishi Electronics America (Sunnyvale, Calif.). "That's why we expect DDR SDRAM to be used widely in high-end servers and workstations for scientific, financial, networking and communica- tions applications. Depending on how development efforts progress for related chip sets, these cost-effective advanced DRAMs may also be popular in high-performance PCs."

Along with needing the support of off-the-shelf chip sets, DDR devices will need new high-speed clock drivers and new Jedec-standard modules to be used in high-end PCs, Conkle said. Because of these infrastructure requirements, chip- set suppliers, motherboard makers and PC companies have been reluctant to adopt DDR for midrange to low-end PCs because of the additional costs they entail, said Misao Higuchi, manager of memory engineering of NEC Corp. For the middle to low end of the PC market, NEC is pushing its Virtual Channel Memory architecture, which has garnered support from three of Taiwan's chip-set manufacturers but still lacks widespread support from memory vendors.

While DDR DRAMs have so far failed to gather backing from chip-set suppliers-most notably Intel Corp., which has chosen Direct Rambus for future chip sets-a number of manufacturers have announced they will produce the memory devices and are optimistic that high-end system manufacturers will develop their own memory controllers with hooks to DDR.

Hitachi will soon offer DDR devices in by-4, by-8 and by-16 configurations, each having 83-, 100- , 125- and 133-MHz speed grades. Measured on a per-pin basis, data-transfer rates are 166, 200, 250 and 266 Mbits/s, respectively. They will have sequential and interleaved burst sequence options, with burst lengths of 2, 4 or 8 per pin. CAS latency options are 2 and 2.5.

Made on a 0.25-micron process, the DDR devices will be available from Hitachi in November in Japan. The company will make 10,000 of the DRAMs per month starting in January and will increase volume production to 800,000 units/month by the end of next year, according to a Hitachi spokesman. The price will be about $14, except for the 133-MHz parts, which will cost $22.

Fujitsu, meanwhile, will also start selling samples of 64-Mbit DDR chips with by-4, by-8 and by-16 I/O widths in November for about $14. Made on Fujitsu's 0.32-micron process, the devices will come in 83- and 100-MHz speeds.

Fujitsu will also sell 184-pin DIMM modules with DDR SDRAMs. Four configurations will be offered: 8M x 64 and 16M x 64 without ECC, and 8M x 72 and 16M x 72 for ECC types.

For its part, Mitsubishi is offering 64-Mbit SDRAMs in by-4, by-8 and by-16 configurations, with speeds as high as 133 MHz. There are three CAS latency options available: 1.5, 2 and 2.5, with burst lengths of 2, 4 and 8 for each data pin. Mitsubishi said its "-7.5" specification for the memory offers a first access time of 37 ns with a 133-MHz bus at CAS latency 2.5.

October 12, 1998