HP calculator with largest math (stats, logic, etc..) and data structure functions

[pier4r]

(could be that this topic was already discussed but I found none)

In the 50g AUR, page I-2 , there is a table showing with which hardware or firmware revision in the RPL family (28, 48, 49, 50) a function became available to the user.

Some months ago I admired the math pac of the 71B and I know that the 71B was the ancestor of all the Saturn-based systems (that are the RPL based systems!?), therefore I am asking myself the following question:
Which is the calculator, discontinued (to consider "final" works only, not ongoing projects), that received the most functions for math and data structure manipulation directly from HP?

For example the 71B has the math pac as extension, but that is directly from HP. So it counts. The 41 has surely something similar from HP as extension card. 9100 and 9800 could be included as well and whatever device marketed by HP as calculator, at least for the main focus of the device (the 80 series I am not sure, I am not sure whether they were "just" calculators).

I am also aware that quantity is not equal quality, but I have to start somewhere. Since at the moment I have little knowledge about the extension of the math library for older models, which place is better than this forum to ask such question?

Naively I would have imagined that every new functions in an hp calculator would be documented, and then implemented on the new high end calculator. So theoretically the 50g would have everything that the 49, 48, 42, 15C, 41, 71B, 75 (etc..) had. Is that so?

Does the 50g have all the neat functions that the 71B math pac exposes? Does the 71B have all the functions of the 15C ?

Specifically math and data structure (matrices, lists, structures, tuples, vectors, trees, etc..) functions. How to open a file or the like is too OS/device dependent.

In case there is no complete answer, could you provide pointers? I know that in the worst case I should go and compare the manuals of all the calculators, but maybe someone did some work on the topic already.

Counterexamples can also greatly help. Like "look the 71B can do this with matrices, in the RPL functions provided by HP, that is not possible". Or "the 15c could do that, the 42S couldn't".

If every high end model somehow has functions that other HP calculators do not have, which one is the most well equipped (by HP)?

I'd say the 50g. But actually the 48G was produced for long time and could be expanded, the 41C and 71B too, so I am not sure if hp ended putting a lot of math/data structure functions in the 50g, while doing less on previous models.

[Jlouis]

You have to consider several option classes and sizes to have an answer. Comparing all calculators together is like comparing apples and oranges, roughly saying, or comparing different types of oranges.

HP 41 can be very complete with modules and so is HP 71b, but let alone is not so.

Sizes and target market can differentiate calculators.

HP 50G is battleship class (hello Walter!!), while HP WP34C or HP 15C are pocket size, but powerful, so two different classes.

Graphics or non graphics?

I think it's good to differentiate all kinds, classes, etc.

Maybe in this weekend I will try do do that.

Cheers

JL

[pier4r]

Well I am already happy knowing something like:
- look the 71B with HP modules was the most well equipped, the rest follows.

Comparing calculators is a bonus, I am already happy with one model that is the most well equipped. Could also be that actually more than one model is at the top, but at least knowing one.

I am really unsure between 50g and 71B. I was too influenced by the math pac, so many functions.

[pier4r]

checking briefly the 9100 and 9800. The calculator themselves don't have many functions but the various pacs (documentation plus magnetic cards or tapes) have plenty already for the 9100. And that directly from HP, that is the calculator division produced the system and then worked to expand its usability with ready made programs.

The documents (only some of them or all? I don't know) are in the hpmuseum usb stick.

I didn't check other calculator manuals from the moment but if the trends is similar I would expect that the HP calculator with the largest math/data structure function library was the one that was produced for long time, was expandable and was targeted to professionals.

Using the approximative production timelines here: http://www.hpmuseum.org/forum/thread-10327.html (you are welcome to expand it) I would say that the likely candidates are the 9100, 9800, 67/97 (having a desktop calculator for "scientific" business helps), the 41 or the 48. After the 48 then more likely the community took over the effort rather than hp itself.

For example the user library programs for the 67/97, when redacted and supported by HP, for me are like HP contributions.

[Valentin Albillo]

.
Hi, Pier:

(09-21-2018 05:30 PM)pier4r Wrote:  Which is the calculator, discontinued (to consider "final" works only, not ongoing projects), that received the most functions for math and data structure manipulation directly from HP?

For example the 71B has the math pac as extension, but that is directly from HP. So it counts.

.

Good. I take it that for "math pac" you mean the Math ROM, right ? Are you also including the many Library Solutions books HP produced and released as an official product (with its Product # and all) such as the Math Users' Library Solutions ? I would guess yes.

Quote:The 41 has surely something similar from HP as extension card.

Extension card ? You mean a plug-in ROM, right ? The most similar to the 71's Math ROM would be the Advantage ROM. There's also a Math ROM, but it's mostly low-quality though it probably should count as well.

Quote:Does the 71B have all the functions of the 15C ?

All that are relevant, yes. For instance the HP-15C's MATRIX 2 and MATRIX 3 operations used to change the format of real matrices to help in implementing complex matrices operations aren't needed at all in the 71B's full complex matrix implementation.

Likewise comparing the HP-15C solver or integrate with the Math ROM FNROOT and INTEGRAL implementations would be wrong, their capabilities are vastly different.

In conclusion, merely comparing the number of functions available without regard to their specific capabilities is highly misleading.

As another example, some "Polynomial Root Solver", written and submitted to some HP users' library by some random user in RPN and made available by HP as a listing and/or a magnetic card, could hardly be compared to the HP-71B's "Polynomial Root Solver" (aka PROOT) written by the professional wizards at HP in Saturn's Assembly Language and released in a perfectly-documented ROM.

Quote:Counterexamples can also greatly help. [...] Or "the 15c could do that, the 42S couldn't".

There are some relevant things the 15C can do which the 42S can't. Just for instance, the HP-15C has a parallel 4-level complex stack so you can enter 4 complex numbers on the stack using just the stack. You can't do that in the HP42S.

Quote:For example the user library programs for the 67/97, when redacted and supported by HP, for me are like HP contributions.

If (and only If) they are redacted (i.e.: written) and supported by HP (not by random users) then yes. But If and only If.

Anyway, I'd suggest you restrict your count to HP-produced (Product # and such) math materials for each relevant model and disregard anything else (including such marvels as the JPC ROM for the 71 or the many ROMs written by Ángel Martín for the 41C models). And still the qualitative differences among similar functions would remain and would be highly relevant so something should be done about it if the comparison is going to be fair and the results meaningful.

Regards and have a nice weekend.
V.
.

[pier4r]

Thanks for the info and corrections.

Yes I think to start things and do not make the task impossible it is better to be a bit more strict, so considering HP own products rather than HP-supported collections of contributions.

[Joe Horn]

(09-22-2018 10:32 PM)Valentin Albillo Wrote:  I take it that for [HP-71B] "math pac" you mean the Math ROM, right?

HP's name for it was "Math Pac":

[JSBach]

(09-22-2018 10:32 PM)Valentin Albillo Wrote:  

Quote:Counterexamples can also greatly help. [...] Or "the 15c could do that, the 42S couldn't".

There are some relevant things the 15C can do which the 42S can't. Just for instance, the HP-15C has a parallel 4-level complex stack so you can enter 4 complex numbers on the stack using just the stack. You can't do that in the HP42S.

The HP-15 had 2 parallel stacks, one for the real parts and one for the imaginary parts.
The HP-42 had something better: Every stack member could be a real or a complex number.
(Conversion (x,y) <-> y+i*x: the "COMPLEX" function)

The HP-42S system is IMO much easier to use. (The same goes for matrix functions, solver and integrals.)

JSBach

[grsbanks]

(09-24-2018 11:48 AM)JSBach Wrote:  The HP-15 had 2 parallel stacks, one for the real parts and one for the imaginary parts.
The HP-42 had something better: Every stack member could be a real or a complex number.

Very true, but entry of a complex number from its real and imaginary constituent parts uses two stack levels. What was in T when you enter a complex number is lost. This doesn't happen with the 15C.

[Ángel Martin]

(09-24-2018 01:41 PM)grsbanks Wrote:  

(09-24-2018 11:48 AM)JSBach Wrote:  The HP-15 had 2 parallel stacks, one for the real parts and one for the imaginary parts.
The HP-42 had something better: Every stack member could be a real or a complex number.

Very true, but entry of a complex number from its real and imaginary constituent parts uses two stack levels. What was in T when you enter a complex number is lost. This doesn't happen with the 15C.

Nor does it happen with the 41Z module ;-)

For example, the following sequence:

2, ENTER^, 1, ZENTER^
4, ENTER^, 3, ZENTER^
6, ENTER^, 5, ZENTER^
8, ENTER^, 7,

places the complex numbers in the complex stack:

U: 1+2i
V: 3+4i
W: 5+6i
Z: 7+8i