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.mac.hardware
- Subject: Speeding up system clocks
- Message-ID: <1992Dec31.172505.18294@klic.rain.com>
- Date: 31 Dec 92 17:25:05 GMT
- Article-I.D.: klic.1992Dec31.172505.18294
- Organization: KLIC
- Lines: 31
-
- Some of you folks are talking about putting faster clock oscillators into
- your systems. You should be aware that:
-
- 1) Chip speeds go down with temperature; about 0.5% per degree C.
-
- 2) Chip speeds can become worse with time, due to various wear-out mechanisms.
-
- Reputable manufacturers of chips and systems will leave some extra margin
- during speed testing, so expected degradations won't kill the system early.
- If you want to do this kind of experimentation, you should probably test the
- system at the maximum temperature it will ever reach, perhaps with a
- hair-dryer on it to stress things a bit (keep in mind, though, that hot dry
- air can generate static). Once you have determined how fast things can
- go in the worst case, back off the speed at least 10 percent. Or cook it
- 20C hotter!
-
- In the PC world, there are programs such as Check-It and QA Plus that do
- processor and memory tests. Are there similar programs for the Mac? A
- good test program will exercise all the worst case paths in the hardware.
- Running an application may not do this reliably.
-
- Personally, I need reliability more than I need a few percent extra speed.
- I find it disturbing that the IIci speedups mentioned by others resulted in
- such small improvements - this is a sign that things are close to the edge
- already!
-
- 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
-