home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!sdd.hp.com!sgiblab!cs.uoregon.edu!ogicse!qiclab!nosun!klic!keithl
- From: keithl@klic.rain.com (Keith Lofstrom)
- Newsgroups: comp.sys.ibm.pc.hardware
- Subject: Re: What's the deal? My chip says "SX-25"; Norton says "SX-33"
- Message-ID: <1992Dec31.170344.18190@klic.rain.com>
- Date: 31 Dec 92 17:03:44 GMT
- Article-I.D.: klic.1992Dec31.170344.18190
- References: <1grur8INNkg2@savoy.cc.williams.edu> <5870140@pollux.svale.hp.com> <C01LuI.542@news.cso.uiuc.edu>
- Organization: Keith Lofstrom Integrated Circuits
- Lines: 24
-
- > misc comments on wafer sort omitted.
-
- Any reputable manufacturer of chips is going to have some way to speed test
- all parts. Imagine a clock driver on a CMOS chip that is constructed from
- two big hurking devices, strapped together with metal, to drive the whole
- chip clock at speed. Now imagine one of those metal straps has an open
- in it, resulting in a smaller effective clock driver. You now have a
- chip that will work dandy at low speed, and won't work at high speed.
- Failures like this are common, and can account for half the DC functional
- devices on a wafer, especially if the yield is low anyway.
-
- Personally, I would be a little frightened of *anything* that "didn't quite
- make it" on a wafer that did - something that's marginal can become more
- marginal over time. Chip speeds degrade because of hot carrier injection
- and other things. A reputable manufacturer will leave just a little surplus
- speed to account for this degradation. If a cheesy board manufacturer is
- eating into this surplus to save a few bucks, they are doing you no favors.
- If you run across such, please post names, so we all can avoid them.
-
- Keith
- --
- Keith Lofstrom keithl@klic.rain.com Voice (503)-520-1993
- KLIC --- Keith Lofstrom Integrated Circuits --- "Your Ideas in Silicon"
- Design Contracting in Bipolar and CMOS - Analog, Digital, and Power ICs
-