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: Jacktoad who wrote (18808)	4/16/1999 5:49:00 PM
From: capt rocky	Respond to of 93625

unc. and all . jack toad too! vlsi is just the tip of the iceburg. rmbs is going into wireless com.! what great news. unc. is right, every device that now needs to communicate with memory at any speed at all, will be moving up to rmbs in its next upgrade. why not? the cost will not be that much more and they can brag rambus inside.it is happening as we sit here. rocky

To: Jacktoad who wrote (18808)	4/16/1999 6:09:00 PM
From: Zeev Hed	Read Replies (3) \| Respond to of 93625

A lurker E-mailed me the following, just another point of view. The critical part is this writer does not think that DRDRAM will capture more than 20% of the PC market. I think that author is dead wrong. I should add "it is the learning curve, stupid...". ELECTRONIC NEWS VIEWPOINT APRIL 12, 1999 The DRDRAM Mystery SHERRY L. GARBER IRLC1 RAMBBUS DRAM El (DRDRAM) is a mystery product - at the center of an incredible amount of debate. Controversy and problems are DRAM design change from page not unique to DRDRAM. Every mode to synchronous DRAM has introduced new challenges. Be cause progress is inevitable, the problems are solved. There- fore. the question is not whether it will be a product, but where it fits. Will it be the primary mem- ory product or merely fill a niche? Semico Research views DRDRAM as a niche product. Most DRAMs are consumed by four major industry segments; computer, communications, con- sumer and industrial. Histori- cally, the computer segment implements new higher perform- ance DRAMs which spread to the other applications as ASPs decline and supply increases. This trend is changing in 1999 and 2000. With promised changes in processors (Pentium 111), chipsets (Camino/Other) and DRAM types (DRDRAM, SDRAM 133MHz and Double Data Rate), the computer indus- try sub-segments will require dif- ferent types of DRAM. The server market will move to 128- megabit then 256Mb and evolve from EDO DRAM to SDRAM. The DRDRAM will debut in the high-end ($3500 and above price range) of the desktop PC market. Semico believes it will stay in this high-end niche and not migrate to the other computer price points. This is about 20 percent of the desktop units. For the next few years, SDRAM and other evolutionary DRAMs will corn- mand around 80 percent of the desktop market - plus the bal- ance of the computer and all of the communications, consumer and industrial markets. The desktop PC market itself has segmented into price points with distinct features. When the Pentium III 600MHZ processor was first imagined the desktop PC market was different. The $2000 to $2500 box was king and faster processors were anxiously anticipated. Then the $1000 PC came along. All of a sudden busi- nesses and consumers began to choose the inexpensive machines. Intel had already taught them that obsolescence was a fact in the computer industry. So com- puter buyers decided to buy cheaper machines. The PC is a true commodity product-just like DRAM. Cost is critical to the consumer and the PC OEM. In a commodity market cost of manufacture is key. The DRAM market has been evolu- tionary in design change. Since the cost of new DRAM facilities are expensive~ DRAM vendors need to reuse as much equipment for succeeding generations of product as possible. Part of the way to do this has been to make gradual design changes. This keeps the cost of DRAM produc- tion lower. Consider that it took six years to get the SDRAM into production. SDRAM was an evolutionary change. Issues in the DRAM market always relates back to manufac- turing. How fast can product ramp, how fast can it be cost- reduced? How cheaply can it be made because price competition is always present? Rambus has done a great mar- keting job. By having DRAM vendors sign an agreement to make no negative comments about the product, the status of DRDRAM has always been in question. This air of uncertainty has kept it on the front page of the industry periodicals for two years. That is some kind of record. How- ever, the down side is that because the problems were not openly dis- cussed between vendors and OEMs, the product is not ready to ship in volume this year as forecast by Rambus and other industry analysts. This is an industry that can solve technical problems, but it takes the informed cooperation of all the participants. By creating an atmosphere where anyone asking questions is deemed an enemy and anyone who asks no questions is a friend, the Rambus group has kept the industry from concentrating on the issues that need to be resolved. It is unfortunate, but Semico believes when the Ram- bus war is over, there will be more losers than winners. One example is the obsoles- cence of the 64Mb DRDRAM before it even began shipping. How can DRAM managers jus- tify to their management the cost of developing a generation of DRAM that is not needed at a time when development money is at a premium and means sur- vival in the industry? So DRDRAM will happen and it will be in the high-end desktop PC. Product availabil- ity is expected by September. With the push-out of the Camino chipset, DRAM ven- dors will have a little more time to ramp production. Semico believes demand will be about 20 million 64Mb-equivalent DRAMs in 1999. Sherry L. GARBER is senior vice president of Semico Research Corp, Scottsdale, Ariz. E-mail her atsherryg@semico. com