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.
Strategies & Market Trends : Bosco & Crossy's stock picks,talk area
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Crossy who wrote (3159)4/20/2002 5:34:11 PM
From: Crossy  Read Replies (1) of 37387
 
re: some thoughts on MEMS and Gilder's bullishness on it..

I read that Gilder feels MEMS devices & concepts are of disruptive nature and attributes a "paradigm" shift to them as he feels the same in VCSELs..

While I firmly back up his stance on VCSELs and on the need for higher speed memory-backplane interfaces for NPUs to achieve wirespeed routing in a "Tera"-era, I do not share his enthusiasm for MEMS - at least not on a long term basis. Why ?

Because MEMS have some inherent limitations. Switching speed is in the milliseconds instead of nanoseconds for solid state material. High voltage and high power is needed to drive "3D Mems" (also called 2-N MEMS) designs. Usually 100-300V current is needed for this application calling for the use of PMOS, HV-DMOS and HV-CMOS in driver applications - making big switching fabrics literal boiler plate-devices.

Now I do think MEMS will be a success but IMHO it won't be the concept to ultimately "bet the farm" on. In my view it will be used as a STOPGAP solution because in the near time there won't be any other technology available adressing the need for all-optical switching with similar insertion loss, low crosstalk and other important system parameters..

I ultimately envisage different concepts to prevail using architecture based on acousto-optics, electro-optics and maybe quantum-level-pure optical-processing. Even modifications of LCD-architectures or Agilent's bubble switching method (a refined technology sharing some characteristics with MEMS) look better than plain MEMS approaches IMHO..

All these methods do show promise and feature far higher switching speed than MEMS. The only problem with all of them is thair inability to deliver comparable performance with respect to important system parameters other than switching speed and will thus require considerable time for commercialization and refinement. Some lend themselves nicely to chip fabrication processes. First applications of those contending technologies should surface in those deemed application of MEMS where switching speed is most important. For example to provide "tunability" in tunable filters or lasers - where slow tuning speed does degrade the tunability characteristics in the time-domain with slow responsiveness.

In short - what I think is missing with MEMS is speed and an "elegance" in architecture - yes I think Gilder is calling premature victories in this category..

just my opinion as "food for thought"
CROSSY
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  

HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2025 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News