Robert Lunn <robert@HUEY.RDD.NECA.NEC.COM.AU> wrote:
>   A schoolboy with a plugpak and a couple of diodes can
>   defeat the code protection on a 'C84.

>   A skilled engineer with $100k+ of specialized equipment
>   can defeat the code protection on a 'C61.

Sorry, but you're only *guessing* that a schoolboy can't defeat the protection
on a 'C61.  You may be right, but you have no proof.

I wasn't claiming that the 'C84 is as secure as anything else.  I was claiming
that there is no *hard evidence* that any particular other part is more secure.
The lack of evidence does not constitute proof that other parts are more
secure.  It is only possible to prove that a part *isn't* secure.

> To the best of my knowledge, the code protection of an 'F84
> is the same as the code protection of a 'C84.

Based on what?  I'm not aware of any evidence that suggests that the same
vulnerability exists.  Are you blindly assuming that, or have you tried the
'C84 techniques on it?

If you really want to assume that the 16F84 security is weak, go right ahead.
But I *still* stand by my original statement:

> If you pick some part other than an 'F84 solely on the basis of what you've
> heard about the 'C84, any sense of security you may have is probably
> misguided (IMNSHO).  There *are* people out there who can extract the code.

And I'll add that for *many* chips it *doesn't* cost $100K.  The 'C84 isn't
the only one that is easy to defeat.  And no, I'm not going to present any
example to back up that claim.  I find it useful to be able to disassemble
other people's code from time to time, so I'm not giving away my trade
secrets any more than Microchip is giving away theirs.

Cheers,
Eric