23 Apr
2012
23 Apr
'12
12:34 p.m.
You have contradictory findings because it's your
limited and narrow
experience. Whatever or however many instances you have dealt with,
they come through a filter of circumstances before they get to you.
Sure, I don;'t, for example, see those devices whcih fialed in production
or initial testing. On the otehr hand, nobody else outside the factory
does either, so my expericne probably is relevant for what I do.
I've received more digital cameras to repair
lately than 1960's
discrete-component calculators. Does that mean modern digital cameras
are less reliable than 1960s discrete calculators?
OK, we can all be silly about this.
I can say without fear of contradiction that I've replaced more TTL than
microcontrollers in HP9800s for the simple reason that there are no
microcontrollers in HP9800s. And these machiens are actually a
counterexample to my comments in that the complex ICs -- the RAMs and
ROMs -- fail much less often than the TTL.
I've repaired more HP calculators than no-name 4-bangers. I've repaired
more Leica cameras than Kodak Instamtics. In bother cases that;s not
because HPs and Leicas are unrelaible, but rather because HPs and Leicas
are worth taking the time to repeir, 4-bangers and Instamatics often are not.
Adding another data point, I will contradict the
findings of the two
of you: I've seen more failures of [simple] SSI devices than
[complex] VLSI.
None of these 3 anecdotes (mine or those of the two of you) are
indicative of the comparitive reliability of these items.
Hmm... When I find that I rarely have to rplace a 555 (in somethign
commercal or something I've made) but that I do have to repalce VLSI
devices quite often, I know which I consider to be the more reliable.
Maybe incorrectly.
-tony