Re: Why the 33S is so long in the making Message #5 Posted by Bill Wiese on 5 Feb 2004, 7:29 p.m., in response to message #4 by jaredengineer
Hi...
jaredengineer wrote: Quote:
I have talked with the folks at HP technical support and from what I could get out of them, the problem was twofold. First the keyboard issue was one of concern, and as I understand it a keyboard similiar in material to the 9g was planned initially. Second there were what was described to me as "HUGE" issues with the processor on the calculator specifically with its ability to change between RPN and Algebraic modes.
Seems reasonable.
This is what happens when you outsource to a crap stamp-em-out company like Kinpo instead of looking back over your storied history...
IIRC, some of the HP98xx desktop series calcs w/algebraic logic translated expressions into RPN for evaluation/ execution. Such an approach avoids having to run in "Mode A" or "Mode B" and having to tweak lots of firmware so it can work in either mode (or at least not interfere!)
And HP probably did it back then with less code/overhead than is available on the 6502-derived microcontroller (Sunplus SPLxxxx) used by Kinpo!
My view of a hybrid calc like a 33S:
- generally, use step by step programming "language"
like regular HP programmables.
- when expressions are called for, an algebraic
expression can be stored or evaluated when
encountered
- in RPN mode, math & string expressions are worked on
operator-at-a-time on the stack
- unlikely a program needs to run in both RPN & algebraic
mode, so a user program is one or the other.
- in direct (non-program mode) expressions can be entered
algebraically & then evaluated or items entered on
RPN stack and operated upon...
This is NOT rocket science, folks.
Bill Wiese
San Jose, CA
|