>Kalle Pihlajasaari said:
>
>>I must say I PICs in the field fail very rarely, perhaps 4 out of
>>1000 that I have had to do with.  All done with a PICstart
>>programmer at standard voltage.
>
>If 0.4% of hobby-progg'd chips fail, that's a very effective advert for
>production-quality programmers! :-) Still, that raises a question which I (as
>a newcomer) would love to know the answer to. How reliable are the PIC chips,
>when programmed to production standard?.
>
>Andy W.

We make a circuit board which is used in a medical device. It uses a 16C54
to sequence some valves. In the past several years, we have made tens of
thousands of these assemblies. We have the 16C54's gang-programmed by the
distributor. I won't say we've NEVER had one fail because of the PIC part,
but offhand I'd estimate the failure rate is probably less than one in ten
thousand. Maybe way less.

Reg Neale