To: LindyBill who wrote (18135 ) 2/16/2000 9:16:00 PM From: Mihaela Respond to of 54805
RE: RAMBUS People, I hate to be a spoil sport, and I really hope you all get rich on Rambus.But remember this:If Intel makes any kind of statement that they are in a hold position of any kind on Rambus, the stock will drop 50% upon opening the next day! We have been there people! Why in God's name do you want to take this kind of risk? I have got 4 or 5 stocks in my portfolio that have gone up 35% or better in the last two weeks. None of them have this type of risk involved. I have read the posts on this Board about Rambus for the last 24 hours, and it sounds like the posts for Qualcomm the last week of December, '99. Sure, it is a great product, and has a great future, but not yet! And if Rambus forges ahead, and you have no problems, don't come to me with "You were wrong, Lindy!" I will not have been wrong, you will just have escaped the Guillotine! LindyBill LindyBill, I believe Intel is architecting their systems so that the processor, memory and I/O can be scaleable to the needs of their computer systems for years to come. They (Intel) take architecture choices very seriously, so that the value-chain can be ready for these changes. A successful 15 year parallel ATA arch. to a serial ATA arch. for the next 10 years is made over many years and re-thought over the development lifecyle. A 133 Mhz bus (Carmel) to a 400Mhz bus (Tehema) is planned well in advance. Problems do/can come up (Camino) for any technology development project. If a major problem comes up, Intel can/will put a project on hold. There is always that big risk. I believe Intel Timna should use RDRAM for performance & cost reasons. Intel now says they will first bring out Timna this fall at 667Mhz with SDRAM instead of RDRAM. This will slow the ramp of RDRAM for low cost appliances & PC's by ~6 months. All this noise (to a LTB&H person) just changes your expectations for growth. Will Intel drop RDRAM on Timna? Here is an article about serial ATA. Chip-to-chip fast I/O will be very important for future needs, just like fast busses and fast memory. The server farm vendor (& Intel) want faster I/O right now, so Intel is busy getting these specifications done. Just discussing this Watch & Wait candidate. Here are snips from the article below: "The parallel storage interface we use today was designed more than 15 years ago and is a potential bottleneck for tomorrow's more powerful computing platform," said Pat Gelsinger, vice president and general manager of Intel's desktop products group. "By working together, this group of industry leaders will ensure that the transition to serial ATA happens as smoothly and as quickly as possible, ultimately resulting in enhanced system performance and flexibility." ... The serial ATA specification is expected to have at least a 10-year life span, reaching a target of 1.5 Gbits/s when devices roll out about a year from now. The specification should be finalized this fall, with additional 2X and 4X serial ATA extensions planned to double and quadruple bandwidth in the future. techweb.com