home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!olivea!spool.mu.edu!uwm.edu!daffy!uwvax!trappist.cs.wisc.edu!zakarian
- From: zakarian@trappist.cs.wisc.edu (Armand Zakarian)
- Newsgroups: comp.sys.handhelds
- Subject: Re: HP calculator questions
- Keywords: Solve, Integrate, HP34C
- Message-ID: <1992Dec18.194912.29471@cs.wisc.edu>
- Date: 18 Dec 92 19:49:12 GMT
- References: <Dec.18.00.26.45.1992.16613@aramis.rutgers.edu>
- Sender: news@cs.wisc.edu (The News)
- Organization: University of Wisconsin, Madison -- Computer Sciences Dept.
- Lines: 34
-
- In article <Dec.18.00.26.45.1992.16613@aramis.rutgers.edu> shrier@aramis.rutgers.edu (Shrier) writes:
- >
- > Back in 1980, I purchased an HP34C calculator, which has to be
- >considered a beautiful feat of design. Code density was very high; it was
- >amazing what one could do with a quarter K of RAM! The looping, addressing,
- >and other programming features struck me as more clever than earlier HP
- >calculators. The accuracy control mechanism for integration seemed very neat
- >too.
- >
- > Anyway, I have several questions for the Net:
- >
- [...]
-
- >2) I never did get around to using the R/S button to investigate the algorithms
- >behind SOLVE and INTEGRATE. The Solver uses two or three previous values,
- >so the method must be somewhat more elaborate than the secant method. What
- >is the algorithm, and how are initial guesses chosen, if the user doesn't
- >supply the pair?
- >
- > Much more interesting, how does integration work? The manual is well
- >written, but hides the details. It seems that the number of sample points
- >is doubled for each iteration, and the last three iterations are compared
- >to assess convergence. OK, but how are the points and weights chosen?
- >
- [...]
-
- It is my understanding that all HP calculators have used the same
- equation solving and integration algorithms. These were authored
- by William Kahan of Berkeley (a recent Turing award winner) and
- were described by him in a couple of articles that appeared in
- the HP Journal around 1979-80.
-
- --
- Armand Zakarian zakarian@cs.wisc.edu
-