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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Joe NYC who wrote (118258)	4/3/2004 8:34:53 PM
From: niceguy767	Read Replies (1) \| Respond to of 275872

Jozef: Couple of comments: 1. Flash: Probably worth re-reading this release. The potential cost savings with migration to 110nm, already underway, are huge: amd.com As previously mentioned, if contract pricing remains constant the rest of the way thru FY04 (i.e. strong handset demand in an economic recovery), AMD's flash division could easily contribute better than $1.00 per share in FY04. 2. With 90nm k8 production and MSFT software, it'll take about 1 compelling 64-bit program to ignite k8 to the level where no 1st time buyer can reasonably consider the 32-bit only option! AMD's largest problem in q3 and q4 is likely to be fulfilling demand by huge enterprise workstation and mobile k8 wins!

To: Joe NYC who wrote (118258)	4/3/2004 8:54:02 PM
From: Dan3	Read Replies (1) \| Respond to of 275872

to pravin - This was posted by MrAMD on IH to pravin - (could someone post this on SI) in the last sentence you seem to be mixing both the static and dynamic currents. from a pure dynamic standpoint, your biggest savings are from reduced voltage. bear in mind that kink/history effects happen here. also for this you need to know if FDSOI or PDSOI is being used. if FD then you need not have to worry about these effects as much. for PD becos of the dynamic vt modulation happening due to the bipolar formation normally the transistor tends to average down to lower vt implying faster switching and this allows usage of lower supply voltage to get the same performance as bulk. also helping you(as you have mentioned), is the reduced load capacitance(reduced junction caps). from a static standpoint(i am presuming this what you mean by saying amd may have the same problems - off currents), things get a little tricky. normally there are different vt devices. depending on what you need the respective devices are used. if performance critical, then low vt devices if not high vt devices which cut down the static leakage. for example, we have something like ultra high speed transistors or high speed transistors or standard transistors(high VT). this is further helped by what amd is doing. there are not gunning for clockspeed, but instead going for overall system performance improvement. this allows them to pick and choose what to speed up and what not to. this where intel appears to have hit a brick wall. example : altera vs xilinx in regards to 90nm(if remember correctly xilinx had quite a bit of problems when transitioning to 90nm). altera took the right approach of not going full blast. they enhanced performance where necessary and at other places limited leakage by using high VT devices. hope this helps.