Re: Yet another HP35S question Message #10 Posted by Les Wright on 8 June 2007, 11:06 a.m., in response to message #2 by Dave Johnson
Quote:
I do not believe the 35s will have many more capabilities than the 33s (maybe more storage registers and if they are nice more program labels).
Jeepers, I hope there are more labels! With all of those registers, being restricted to 26 measly labels is like being in a candy store with all of the goodies inaccessible, under lock and key, expect for a few choice morsels at a time.
Quote: There is debate about matrix ability.
I have an abundance of optimism that the the mysterious (i) and (j) key labels may actually refer to matrix element addresses under certain conditions. Even if there is no intrinsic matrix abilities, if there are adequate registers and labels it should be feasible for a geeky sort with too much spare time to port, say, the MATRIX routine from the HP41 Math Pac module to the 35S paradigm. This could be challenging if i and j are the only indirection registers available--remember on the 41 any of the 319 registers can be used for indirect addressing. But if it worked, it should be pretty fast--cut up the 33S if you will, but there is no doubt that it is somewhat faster than the 42S and much faster than the 41 or 15C, and the 35S should inherit this speed. Of course, only the most determined sorts will want to enter the 500+ steps into the 35S, but it should be a labour of love for the truly nerdy.
For the busy and practically minded, I expect computer software will just have to do for matrix work, or the 15C, 42S, or one from the 48 or 49 series for those who have them. But I do think the 35S will prove excellent for routines that iterate over a series of computations to convergence and produce a single result, as in the computation of special functions. My 33S is loaded with routines I have written (or borrowed and adapted) for the incomplete gamma functions, which in turn can be called to compute the error functions, the cumulative normal distribution, and the chi-square distribution. They are really zippy and I look forward to transferring them to the 35S.
Les
Edited: 8 June 2007, 11:07 a.m.
|