Computers and radiation

[Lester.Slaback@nist.gov]

An observation (i.e., no added help)...
Keep in mind that the chip scale and data density on magnetic media has
constantly been moving to smaller dimensions.  Hence what might have been
true a generation ago (i.e., 1-2 years ago) might not be so now.  For chip
manufacturers a single alpha track in the wrong spot will ruin a chip. 
For low LET radiation you are basically changing the odds to much lower
levels.  But I have not seen studies with modern media.
Also keep in mind that the increasing chip size and RAM count increases
the odds, both for temporary and permanent glitches. It would be
interesting to see an effects plot as a function of chip generation (86,
286, 386, 486, ... for those of us in the PC world).

-- 
the above are the personal musing of the author,
and do not represent any past, current, or future
position of NIST, the U.S. Government, or anyone else
who might think that they are in a position of authority.
NBSR Health Physics
NIST
Gaithersburg, MD 20899
301 975-5810
-----------------------------------------------------------
Lester.Slaback@nist.gov
-----------------------------------------------------------