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.

Non-Tech : The Critical Investing Workshop -- Ignore unavailable to you. Want to Upgrade?

To: Voltaire who wrote (31591)	9/1/2000 9:08:14 PM
From: Voltaire	Respond to of 35685

An interesting Micron situation analysis from an anonymous poster: Well as usual, these guys have it mostly wrong. This lawsuit is potentially a company killer for Micron (and now for Hyundai as well), because if the suit is not settled and Rambus wins then not only does Micron owe royalties, plus interest, plus costs, but Rambus is not legally required to license them to build ANY DRAM. If Rambus wins then it clearly and decisively owns ALL the memory market for the foreseeable future. Micron is out of business as SDRAM is 90% of their revenue. I guess they could become a fab house and sell capacity to companies who have legal product to make. If Rambus loses, on the other hand, it still owns RDRAM and DRDRAM and QRDRAM plus the communications and other applications for the technologies they have patented. The suits only challenge 8 of Rambus's 100 patents and 100 patents pending. In fact we are in the strange position of watching Micron begin to ramp up production of DRDRAM under a license from Rambus while suing them to over turn patents which affect SDRAM and DDR RAM. Micron and Hyundai have both asked for Summary Judgment against Rambus. This is done when a company feels an immediate and real threat to their business. In this case the threat, as I see it, is that Micron, Hyundai and others have an market initiative to sell DDR plus the AMD Athlon plus the Via chip set as a real, viable competitor to RDRAM plus P4 and the 840 chipset from Intel. The P4 launch is in October 16, 2000. If DDR and the Athlon don't hit the market in essentially the same time P4 will be left without a viable competitor and they will run away with the market. So time is of the essence. On the other hand, DDR will not be accepted by the oems under a clouded title. The oems have already watched RAMBUS shut down Hitachi and their customers who designed in unlicensed Hitachi chips. They will not risk developing and launching a new product based on AMD Athlon and DDR RAM sans a license from Rambus or a ruling that RAMBUS patents are voided. It would be irresponsible to do so. They will continue to ship existing products with SDRAM and (graphics cards with) DDR RAM because they are in production and they have no choice but to continue to ship existing products and hope for a settlement soon, or negotiate a separate license wit Rambus if semi / Rambus negotiations break down. This may be why Micron asked for a Summary Judgment.Their DDR initiative is at risk in the very short term future. They believe (correctly or not, it is unproven)they have a competitive solution. They have invested a lot of money to develop DDR. At present it goes only into graphics boards but they believe it will be a strong competitor to the Intel P3 and P4 alternatives, costing less than RDRAM while providing comparable speed, at least given current PC CPU and memory performance demands and system level (software, firmware and component) bottlenecks. DDR has been under development since 1996. A lot of money has been spent to make it a viable competitor in the PC business. So far, no one has fielded a DDR based PC, due to problems and delays in producing workable components that will allow DDR to work on a high speed PC motherboard. Fortunately for DDR, RDRAM and Intel have experienced their own setbacks, including bottlenecks in the Intel 820, recalls of the P3 based motherboards with 3 RIMM sockets due to noise problems, a slower that desired (by Intel) production ramp resulting in higher than expected costs, a somewhat larger than expected chip size affecting yields and resulting in higher than expected costs,etc. Micron and others have exploited the delays and problems with RDRAM to push their DDR initiative. These delays appear to be ending. Several vendors have now signed on to RDRAM in volume and several more as set to do so to take advantage of the RDRAM only P4 launch on October. Volume production, coupled with competition of the sockets means that RDRAM will now drop in price rapidly, unless Intel totally screws the pooch. We already know that the Play Station 2 is a giant success with demand ramping to 1 million PS2s per month in September. The volume is there today and it will double again in October. Five vendors have started to compete for this business. Already the oem prices for RDRAM are dropping rapidly. Aftermarket prices are still high because the new production is just ramping and has not reached the after market outlets yet. But they are dropping as well, down in the past two weeks to $229 from a high of more than $800 six months ago and $329 last month. What this all means: Micron and others must successfully launch DDR in the 4th Quarter of face a much leaner, less problematic, lower cost RDRAM based solution with all system components optimized to use the RDRAM. They will also face much more aggressive pricing from RDRAM vendors (possibly including themselves). Customers will demand it. PC performance will require it. Competition will insure it as they all scramble for the biggest share of a rapidly expanding market. Launching DDR and Athlon against a now much more price competitive DRD RAM based P4 competition will be a much tougher job. It will be an impossible job if DDR is not a licensed product. Why would oems buy trouble with a new, unproven set of components (DDR, Via, Athlon, Mobo) versus the competition (RDRAM and P4) for no significant ( and certainly no proven) price or performance benefits in the face of a potential ITC order to stop shipments coming from Rambus. This is the environment in which we find ourselves. In summary, Rambus' price/volume problems are ending. They now have 75% of the WS market, they are exclusive with P4 for at least the next three quarters, they are the fastest PCs today and as prices drop they are taking increasing PC share. They have Sony PS2 in volume and ramping. They have five MM licensees ramping volume and beginning to compete on price for the rapidly growing RDRAM/P4 business. The Memory Manufacturers have to get in the P4 game now or leave it. They have all licensed RDRAM and they have all announced that they want a piece of the P4. They have all put the test equipment and other volume production facilities in place and they have announced publicly their intentions to be players. They will go after it. The competition is on for all the RDRAM sockets here now and coming October 16th. My predictions: 1. Micron and others will settle the lawsuits and ship licensed DDR. Otherwise DDR will die and go away within the next two quarters except for Graphics applications and possibly some Server applications next year. After the 4th Quarter it will be too late to overtake RDRAM. 2. RDRAM and Intel will have a successful launch and oem RDRAM prices will come down dramatically. RDRAM based systems will take increasing market share as prices come down over the next two quarters. 3. No one will want a high speed processor with SDRAM by Q2 2001. They will insist on high speed components matching the higher speeds of Athlon and P3 and P4 from Intel. 4. Value PCs based on AMD and Intel parts will use low cost SDRAM, but DDR will miss its window and will not be needed as an interim solution. By the time Micron settles and launches, (if it ever does) DDR will be a say late and a dollar short.