Marshall,

I don't know which post you are replying to because you have not provided any context and the responses in this sub-thread are numerous enough that my browser is not showing your title with indentation to indicate which post.

I can only assume that you are replying to my post titled "I/O and parity for '51 belongs in Chat". That title has a typo but the forum software did not accept my title edit of "I/O and parity for !51 belongs in Chat" -- that is, non-8051.

Now, you said "8051 does calculate parity!". We all know that it does. The 8051's ability to calculate parity is not in question here. I said "The '51 UART I/O subsystem does not have parity capability ...". The point is that some here are under the illusion that "parity" is somehow bound to "I/O" and are arguing that as the basis for not sharing portable methods of parity calculation that might be superior (or more interesting) compared to the methods that have shared thus far.

In an 8051 parity is a function of the ALU, not an I/O subsystem. And besides, the whole concept of "parity" goes way beyond only "I/O" anyway.