SI
SI
discoversearch

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
RMBS 91.18-4.3%Nov 17 3:59 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: J_W who wrote (21809)6/7/1999 7:22:00 AM
From: unclewest  Read Replies (1) of 93625
 
yesterday, i had earnest discussions with guru and several other of my most highly qualified techie friends.

the following is a brief synopsis of all of those discussions including my interpretations. i hate to get technical, but....

samsung and toshiba are in production. we are certain that samsung can produce high volume parts now. if samsung is the only very high volume producer, parts will be more expensive than necessary initially. as nec, mu and others begin mass production in a few months and demand increases, prices will drop as they always have. don't get excited if you see high initial drdram costs and complaining or fud about it. prices will drop quickly.

600mhz parts will be much cheaper initially, so expect to see them. as discussed here once before, this could be a great marketing tool. the 800 is the automatic upgrade.

dell and cpq are way out in front and on schedule with rambus.

intel is on schedule for july/august production.

ibm is behind but will probably ship drdrams in work stations this year. even if they have to buy them.(ibm is setting up their own drdram production). work stations are the one money making sector for ibm. there is no way they will let themselves fall behind here.

a few months ago the big fud for drdram was power consumption and heat. now we know rambus actually uses less power and therefore produces less heat than sdram. expect to see drdram in laptops 2nd or 3rd q next year. the funny thing about the power thing is now the ddr 133 is the high power and high pin count memory, but we never see any fud about that. the real importance to the pin count is that it increases the die size. ddr die size also increases for other technical reasons. die size increases cause a cost penalty. ddr 133 is short-sighted but sufficient for some uses for now. intel needs rambus scaleability and for that single reason they will not choose ddr 133 stuff.

all future intel platforms are rdram based. as timna indicates the basic pc will have rdram within 2 years. by then, the price will allow the move.

the toshiba and 840 chipset stuff could be awesome news if it means drdram is going to make a fast move into servers. the best estimate i have previously seen for this is next summer. is the chipset viable? i can't imagine toshiba showing it around if it has problems. some have thought that due to the requirement that servers need to be 99.999% reliable some extra rdram field use might be desireable before putting it in servers. can intel demonstrate in their labs that drdram in servers is better than ddrdram? my guess yes!! if for just this one reason...the ceo of sony declared that ddr is unstable at high speeds a few months ago. he further stated that sony would use rambus. seems obvious that sony labs found something bad about ddr and good about rambus.
unclewest
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext