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: Apollo who wrote (23315)	6/22/1999 2:04:00 PM
From: TST	Respond to of 93625

Three years to late. No, that is why it has taken ten years to get Rambus to this point. Put it another way, why would any one spend the extra money to buy a color T.V. when the black & white is just fine. I think the thing to remember here is that mankind is just at the start of this revolution. Europe & the the rest of the world are easily years behind us & the PC revolution while well underway is very far from mature. Who better to determine the need for Rambus than Intel. Believe me, if Rambus had not come up with the memory solution, somebody else would have.

To: Apollo who wrote (23315)	6/22/1999 2:07:00 PM
From: john douglas	Read Replies (1) \| Respond to of 93625

Two thoughts on Rambus being 3 years late. First, perhaps we are temporarily in a valley where the current top-of-the-line hardware platforms deliver more performance than is needed by most users. This does not bother me. While I can't predict which particular killer application yet to be developed will drive Pentium IV chips and Rambus memory, I have no doubt it will arrive sooner than we expect. Better voice recognition, streaming IP-based movie rentals, high quality video telephony. Your guess is better than mine. The older software engineers will recall that when MSFT developed DOS for the original IBM PC, the architecture limited memory to 640K, because no-one could possibly conceive a PC would need more memory than that. So where are we now? 64 Meg RAM is standard. The IBM/MSFT forecast was off by a multiple of 100. Second, don't worry about competition in the game console market. Nintendo and Sony already sell their N64 and Playstations at a loss. What they care about is market share, so that they can make the profit on the game software sold. And with an 11-year old and a 13-year old in the house, I can tell you that game console performance does matter. Nintendo and Sony will pay whatever they must to put the best technology on the store shelf. Just my $0.02. jd

To: Apollo who wrote (23315)	6/22/1999 2:23:00 PM
From: Charles R	Read Replies (1) \| Respond to of 93625

Stan, Thanks for taking the time to post a thoughtful view. Chuck P.S.: There has been some good discussion over on the Intel thread about the same topic

To: Apollo who wrote (23315)	6/22/1999 2:27:00 PM
From: Tom Warren	Read Replies (2) \| Respond to of 93625

I would like a shot at this one. I think Rambus is early but not too early. Tony's arguments fall in two broad categories; cost and the need for performance. First the cost issues. The rambus design has two clear advantages that will become more compelling as chip density increases. First the narrow bus which permits the ram controller becoming integrated with the processor (and eliminates FSB bottleneck) and all other thing being equal fewer pins reduces cost. Second sdram gets its existing performance by implementation in gangs of 8 chips. Rambus gives full performance from a single chip. When a single chip holds 64 or 128 megabytes (not too far in the future) a cpu with integrated RAC and one chip of memory directly on the motherboard will be the lowest cost design. This is the model game manufacturers are using. Finally the costs of production equipment for testing and packaging are high but they represent the state of the art and will be required be these manufacturers for any next generation chips. They aren't unique to Rambus. In fact Chip Scale Packaging will become cheaper than TSOP before the end of this year according to EBN, and has the potential to become much less expensive if efforts to package an entire wafer pan out. Every time I see a marginal producer of dram bail out of the business I feel this is good for Rambus. Consolidation funnels dram profits to the top three or four who will lead to new manufacturing technology and I believe also to Rambus. There have been several such announcements in the past two weeks. Regarding the need for performance please see my earlier post: Message 10143850 I would love to hear anyones reaction to either of these.

To: Apollo who wrote (23315)	6/22/1999 2:29:00 PM
From: Zeev Hed	Read Replies (1) \| Respond to of 93625

Snas, sometimes, I wonder where these people are living? Even if the price wars of the then DRDRAMs cause the prices to be half what current DRAM's are, the total sales of DRAMs will still be in the $30 billion range (and according to our friends at Dataquest, of course $100 Billions in three four years <G>) and RMBS will still make 1.7% of that whether or not the DRAM makers make a penny. Price wars in the memory business and all other applications of DRDRAM's is good for RMBS, this very simply because lower prices means more sales, and RMBS get paid on total sales, not profits!!! How, excuse the expression that I rarely use, dumb, can these people be? I can understand worries about unknown technologies supplanting RDRAMs, but price wars being a problem for RMBS? Hogwash. What I would agree with is that price war will deflate the PE ratios in the whole industry, but that is not something we did not know, and if you look at MU you should ask, what PE deflation? Zeev