Hi Jasen,

 > OTOH I've never found signed chars more convenient
 > than unsigned.

    I'd go a bit further. The signed char default can be a thorough
nuisance because you have to remember to switch it off at compile time if
you want to use unsigned chars (say for lookup table indices).

    However, negative small integers are sometimes necessary and there are
whole chapters in the textbooks about which notation is most effective for
which purpose. Generally twos complement is used, with the high bit
reserved as a sign bit because it suits the math and COPS chips and is
logically less unsound than other systems.

    This is a personal opinion, and should not be taken as gospel.

Best Wishes,
Bill.

---