> On the other hand, it probably doesn't make sense for Microchip to
> make 4 MHz parts in the first place.  I suspect (and have no data to
> support) that the 4 MHz parts are simply 20 MHz parts that didn't
> make it.  Most likely most of them would run at 19 MHz, so 6 should
> be a piece of cake

They would have been intended to be 20MHz parts, but without
being privy to Microchip's testing results, there's absolutely no way
to know how far short of 20MHz these -04 parts failed. Maybe a
close-but-no-cigar batch that failed at 19MHz, or could equally have
failed at 5MHz. Could be a whole mixed bag. If you test half, does
that indicate in any way how the other half perform ?

> so even a great deal on 6 MHz crystals can't save too much.  Just
> one return would eat up the savings on a lot of units.

Exactly - why bother making work for yourself. Or would a pissed
off customer just toss it in the bin as another bit of pointless landfill ?

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist