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: cordob who wrote (81318)	3/2/2002 12:48:34 PM
From: NightOwl	Read Replies (1) \| Respond to of 93625

Cor, Obviously I am not a qualified counter, so please excuse my obstinance, but what difference does "price" make if the cost of (a) producing the chips; and (b) building the platform for it are always higher for DRDRAM than the alternative competition? Am I missing some old "new math" here? Were I a Bus riding Thread Moron, instead of a Hard "Core" Pedestrian, I would be praying for some high priced, money is no object, value added, niche design wins for this IP that would allow manufacturers to project some profits if they produce it. As best I can recall, RMBS IP has three design wins which have or *may* lead to a product that serves that kind of market. When PS/2 goes to industry standard they'll be down to two "maybe's." This is one of the principle reasons I believe that CB is correct about the Bus being Dead Dead Dead. Their entire operation is driven by the expectation of delivering more DRAM for less money but they missed that brass ring. They need some "special thing" that nobody else has: This is not hard. Everybody with a patent has a "special thing" that nobody else has. RMBS has the multiplexed bus IP. It tried to get some other "special things," but got caught using public domain, cut rate methods to get them. They still have the good old Bus. They need to find people who can be persuaded to buy their "special thing:" This is tough when most people who buy "special things" think you stink. One can usually over come such perceptions by the simple application of a good deodorant soap and lots of hot water. In the case of RMBS, I suspect a full colonic will be required, i.e., Tate and the "so called" Board of Directors must go, assuming you can pay someone enough to step in and take the bullet. Of course, the "cleansing treatment" will also require an end to all the smelly lawsuits. Unfortunately, the only way of getting that done this side of a Bankruptcy Court would be by means of a capitulation and change in control. But so far RMBS has shown such little interest in coming clean, that one can only assume that they still believe their manure is odorless. I suspect the "problem" will begin to register with their olfactory senses after they've spent the 3rd of INTC's 4 annual license payments. ...Mom & Pop could probably improve the current performance of Tate's nose by dumping all their shares and refusing to ride the Bus until the interior as well as the driver have been completely replaced. But Mom & Pop will probably still be hanging around the Net posting questions like: "How much do you think those 200 patents will bring if they do file bankruptcy?" long after that massive cash kitty has been drained. They need to sell enough of their "special thing" to keep the doors open: Without INTC forcing the Bus out of the garage and down the street every morning, this is going to be tough. They need an end user with deep pockets and a box in need of a multiplexed Bus or Samsung will soon tire of its two headed DRAM line. So far their best idea seems to be to "hype" 4200 RIMMS and that will not get it done. That leaves them competing against the same wide body bus platform that just forced INTC to run away screaming "Give me OUM!" Moreover, none of the mechanics over at the COMM Bus garage appear interested in taking INTC's place as Official RMBS Bus Pusher. The fact that we have PR touting the 4200 RIMMS and Sis driven motherboards is a fair indication that RMBS IP will not be a player on the high volume COMM memory map. They need to be working on a "next best thing" that people will pay for with or without persuasion: What happens 15 years from now is clearly a minor problem for the Bus, but it is a problem none the less. Were they to seriously look for salvation in new controlling ownership, it would be very nice to have a "next best thing" of potential high value on the distant horizon. Alas, we haven't seen any "things" of this type from RMBS. From all appearances this is a decidedly "one trick" Bus. Two max, if you count the dirty tricks. Of course, they could always "assert" that when the CPU's and MPU's arrive at 10 Terrabytes/sec. the world will be forced to get on the Bus once more. But then I doubt that even Mom & Pop would believe it now. ...Even if it were true.<g> 0\|0

To: cordob who wrote (81318)	3/2/2002 8:00:07 PM
From: Bilow	Read Replies (2) \| Respond to of 93625

Hi cordob; The pricing is the average of the five lowest priced modules showing up in the 256MB PC133, PC2100 and PC800 categories on PriceWatch. Notes: (1) RDRAM is cheaper in pairs so you have to be careful here. (2) Every now and then obsolete PC600 stock shows up and you have to avoid those prices too. (3) I used to take pricing more or less daily, but it got to be way too long, and too much trouble. So now I try to get figures for after US market close on Fridays. (4) The oldest price dates to March 26, 2001: #reply-15563113 (5) The rules for calculating the prices were set forth here: #reply-15563097 I would think that to most Rambus investors, having RDRAM only 70% more expensive than SDRAM would count as "price parity". Of course to industry that ain't the case. Re Samsung's comment that RDRAM, SDRAM and DDR are at price parity, with DDR cheaper than SDRAM. It's true that DDR slipped slightly below SDRAM prices this past week, at least for chips with the same organization and density. But on average, SDRAM isn't manufactured at the same organization and density as DDR. So the figures that I use for the cheapest modules still show DDR considerably more expensive than SDRAM. Of course there is also a delay between chip pricing and module pricing, and there are also differences between the popcorn logic on the modules. The same applies to DDR vs RDRAM prices. It's not surprising that Hammer is supporting unregistered DDR. (1) As you note, it's faster. In fact, since the memory interface is built into the CPU that machine is going to be screaming fast. (2) The Hammer has a 128-bit wide bus, so that means that it can support twice as much unregistered DDR as a 64-bit wide bus machine, provided that they implemented separate address / control buses. (3) Hammer is for the future, and industry will have 512Mbit chips as mainstream then. With 512Mb x8 DDR SDRAM, it would be reasonable to get 1 to 4 rows of memory to work unregistered. With a 128-bit bus, that works out to memory sizes from 1 to 4 GByte. I'm sure they figured that that was where the sweet spot for the Hammer memory type would be. But I would still expect that they would support registered DDR. Are they not going to support registered DDR at all? If so, kindly send a note to Jdaasoc, who's been arguing with me about the significance of registered DDR for most of a year. -- Carl