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: Tenchusatsu who wrote (29538)	9/14/1999 10:19:00 PM
From: Jdaasoc	Respond to of 93625

tench: The only info I could find at Intel web site about VC820 ftp://download.intel.com/design/litcentr/ce_docs/desktop/boards/pentiumiii/b0059.pdf I guess they are on schedule for release in Europe john

To: Tenchusatsu who wrote (29538)	9/14/1999 10:30:00 PM
From: Jdaasoc	Respond to of 93625

tenchusatsu: I found CE document for VC820 board for release in Europe but could not find Cape Cod board CE. Highest Pentium III that had CE was 600 MHZ with 512 cache. ftp://download.intel.com/design/litcentr/ce_docs/desktop/boards/pentiumiii/b0059.pdf john

To: Tenchusatsu who wrote (29538)	9/14/1999 10:51:00 PM
From: Dan3	Read Replies (1) \| Respond to of 93625

Re: but that's absolutely wrong... Are you completely sure about that? PC memory usage is pretty random. Streaming data is generally seen only from disk or network feeds, neither of which is anywhere near to being fast enough to require anything faster than PC100. Rambus has an advantage only if there is a requirement for streaming memory. Remember that interleaving DIMMs provides the same streaming performance benefit as Rambus and that capability was discarded from chipset designs once CPU caches became standard. It was useless since it provided no real benefit. PCXXX can burst a cache line as quickly as Rambus, VC DDR more quickly, and the non rambus memory types can always be interleaved if someday, some application comes along that really does need that kind of streaming memory performance. It will be very interesting to see what happens in a few weeks when some of these systems are made available to completely independent testers. Maybe Rambus longs will be surprised, maybe it will be the shorts, either way, it'll be great to finally see some of these questions layed to rest. Dan

To: Tenchusatsu who wrote (29538)	9/14/1999 11:39:00 PM
From: grok	Read Replies (3) \| Respond to of 93625

RE: <440BX and PC100 SDRAM only contributed a 3% to 5% performance advantage at the time. PC100 cost more than PC66 at that time as well and was in short supply.> Tench, The dram industry moved from PC66 to PC100 because it was a free upgrade due to normal Moore's law scaling just like the one currently underway to PC133. Even with these simple transitions supply/demand imbalances can occur, in fact, they can occur even with no transition. For 30 years the dram industry worked to provide the lowest possible bit cost and the highest performance compatible with that lowest bit cost and it has always been enough performance for the bulk of the market. Intel/Rambus is forcing a discontinuity in this unbroken 30 year chain. It is being justified by future killer apps that need drastically more bandwidth. However, microprocessor performance has increased by a factor of 1000 during these 30 years and drams have always kept pace. There is PC133, DDR, DDR-II on the drawing boards. Can you really make any clear statement that demonstrates that the evolutionary approach could not keep up with these future killer apps if they do, in fact, actually happen?