Speaking of
the 65C02, does anyone know if replacing the 6502 in a PET
 or VIC-20 is known to trip up copy protection on any apps/games?  I
 know that for well-behaved code, there should be no problems.  I'm
 concerned about WordPro or some other pirate-fearing program trying
 to be clever with undocumented op codes, and not working with a real
 65C02. 
 There's always the possibility, but CMOS versions would be much
 better candidates for such perverse cleverness than the old NMOS parts.
 At least I vaguely recall that most if not all
 unimplemented instructions for the NMOS 6502 either sent the
 CPU careening off into the weeds or were simple duplicates of
 "legal" op-codes  due to incomplete op-decoding.   
 
Yes to the first, not just simple duplicates to the second. Some instructions
actually combine the operations of two opcodes in one instruction (and I
think in the same cycle execution time, though I need to check this), and some
have totally novel behaviour.
I doubt this is a big deal on the PET, but the VIC-20 may have some of the
paranoid software that the original poster was worried about, and there are
definitely some C64 loaders that used undocumented opcodes as part of their
copy protection scheme.
--
----------------------------- personal page: 
http://www.armory.com/~spectre/ --
 Cameron Kaiser, Point Loma Nazarene University * ckaiser(a)stockholm.ptloma.edu
-- The early bird may get the worm, but the second mouse gets the cheese. -----