The below does not apply to makers of toys, hair removers, can openers and such

unused locations can be used as scratchpad

I would consider that as an example of an utterly stupid attempt to "let the world see how smart I am" or "desperate means to save $0.17" what it really is, is grasping for straws. This is right in line with "why do I see strange things happen when I use SCON as a scratchpad".

That something is possible does not mean that it is a solution.

I have seen a galore of harebrained schemes to 'get by' with whatever chips that have been chosen without considering the scope of the project in detail before starting.

I can see "using unused locations as scratchpad" if you are a 10 day camel ride from the nearest distributor and I sympathesize with those that have supply problems, but more and more I get the suspicion that often this is just being too lazy to find the supplier. This, by the way is not the problem, the REAL problem is this "I'll use this chip" before even having a hint as to how many resources will be needed. YES, Mr. amateur, I have heard "I have this chip" but when you consider the total cost of what you are doing, you should be able to spend $5 buying the uC that is right for the job.

if the uC you have chosen can not do what you need, get another uC, do not come up with 'schemes'

[edit] one more, if you justify such harebrained actions by "this improvement need addutional resources" it is STILL not a justification, that just goes to prove that you have not learned that if rel 1.0 use more than 75% of resources ysu WILL be screwed.

Erik

for those that suspect me of a misspell of 'harebrained'
http://www.bartleby.com/64/C003/0153.html