HP-41C List of "over 130 functions"?

[Dan Grelinger]

A curiosity of mine has been finding an official list of the "over 130 functions" that HP advertised in the early days of the HP-41C. Early advertising materials and the original handbook (dated January, 1979) references multiple times the 'over 130' number, and the handbook has a supposedly comprehensive list of every HP-41C function, but that list is only 129 entries long, and one of them is an error (a repeat), so that list is really only 128 long. I have not been able to find a list from HP that has more than that. Has anyone seen one? Was the >130 number created during the design of the HP-41C before coding was complete, and upon actual release, some envisioned functions had to be left out? Later advertising materials used a different number. I remember one catalog saying "more than 127 functions", which strongly suggests 128 was the specific number.

[Thomas Okken]

I think this is down to a combination of errors accumulating between the people who designed the calculators and those who wrote the manuals and promotional materials, combined with insufficient fact-checking of the final result. I remember being a bit irked by this at the time; it seemed so pointless to exaggerate the capabilities of the HP-41C when the capabilities that it did have where so impressive already.

I no longer have the prospectuses from back then, but I do know of an even greater exaggeration regarding the HP-42S. In the MoHPC page for this calculator, it is mentioned that it has "over 600 functions," which, if memory serves, is something that comes out of the original promotional materials. I remember scratching my head when I saw that the FCN catalog had only 252 entries, and the function index at the back of the manual, while longer, still didn't even have half of those mythical 600 entries.

I can see how "127" could turn into "almost 130" or "about 130" and then "130" and finally "over 130," but how does "252" become "over 600"?

In a much less dramatic, but still telling example, the HP-42S manual states that the calculator has "over 7,200 bytes of memory for storing programs and data." As any hard-core 42S user knows, right after Memory Clear, MEM reports 6977 bytes free, and when you delete REGS, it reports 7200 exactly. So that's the number you get with an empty stack, no variables, and no programs... and yet, someone felt compelled to turn this into "over 7200."

TL;DR Take such numbers with a grain of salt...

[Jake Schwartz]

(04-13-2018 01:58 PM)Dan Grelinger Wrote:  Early advertising materials and the original handbook (dated January, 1979) references multiple times the 'over 130' number,

It is curious to me that any HP41 materials would have a January '79 date, since the HP41C was introduced on July 16th of 1979. Would you be able to supply any images showing a January date?

Thanks,
Jake

[Joe Horn]

Here's one possibility. When functions are "overloaded", that is, when they perform different functions depending on which argument types they are given, each of THOSE functions are tallied separately when counting the number of functions in a calculator. For example, the + function in the HP 50g is counted as 26 different functions (real plus real; complex plus complex; string concatenation; etc). That's why the number of functions in the RPL models is reported as WAY higher than the size of its function catalog. Are any of the HP-41's functions "overloaded" so that they can be counted as more than one function?

[Thomas Okken]

(04-13-2018 05:06 PM)Joe Horn Wrote:  Are any of the HP-41's functions "overloaded" so that they can be counted as more than one function?

No.

[Zaphod]

How about the IND ‘function’

That adds to other functions does it not ?

[Thomas Okken]

Indirect addressing is an important feature, but I don't recall HP ever claiming that STO nn and STO IND nn were different functions...

[Zaphod]

I just thought that’s how they added to the tally

[AndiGer]

Maybe also register arithmetics is counted seperately? STO+ nn et al.??

[Dan Grelinger]

(04-13-2018 03:02 PM)Jake Schwartz Wrote:  

(04-13-2018 01:58 PM)Dan Grelinger Wrote:  Early advertising materials and the original handbook (dated January, 1979) references multiple times the 'over 130' number,

It is curious to me that any HP41 materials would have a January '79 date, since the HP41C was introduced on July 16th of 1979. Would you be able to supply any images showing a January date?

Thanks,
Jake

Yes, I have the first copy of the HP-41C manual. I, too was surprised by the January, 1979 date, being 6 months before the release. When I get home, I will put an image up here.

[Paul Dale]

Two different length GTO, two different length STO and XROM makes 131


Pauli

[Zaphod]

I think they added all these IND variants to the 128 listed in the function list in the manual:

Quote:Section 13
Indirect Operations
An important feature of the HP-41C is the numerous indirect operations the calculator can perform. Any storage register in the HP-41C can be used for indirect operations. This capability greatly expands the power and utility of your HP-41C. An indirect address is selected by following a function with the shift key, g , and then a register address. The function then uses the number in the specified register as an address. Indirect operations are most useful in programming.
For future reference, here is a complete listing of all HP-41C functions that can be used with indirect addresses:
STO IND nn Store.
STO + IND nn Store add (keyboard form).
STO – IND nn Store substract (keyboard form).
STO × IND nn Store multiply (keyboard form).
STO ÷ IND nn Store divide (keyboard form).
ST+ IND nn Store add (display form).
ST– IND nn Store substract (display form).
ST× IND nn Store multiply (display form).
ST÷ IND nn Store divide (display form).
ASTO IND nn ALPHA store.
RCL IND nn Recall.
ARCL IND nn ALPHA recall.
VIEW IND nn View register contents.
GTO IND nn Go to.
XEQ IND nn Execute.
FIX IND nn FIX display format.
SCI IND nn SCI display format.
ENG IND nn ENG display format.
DSE IND nn Controlled decrement loop.
ISG IND nn Controlled increment loop.
TONE IND nn Audible tone pitch.
ΣREG IND nn Define accumulation registers.
SF IND nn Set flag.
CF IND nn Clear flag. [/QUOTE}

Instead of adding IND as one more function ?.......

[Dan Grelinger]

(04-13-2018 03:02 PM)Jake Schwartz Wrote:  It is curious to me that any HP41 materials would have a January '79 date, since the HP41C was introduced on July 16th of 1979. Would you be able to supply any images showing a January date?

Thanks,
Jake

Jake, here you go! I've included photos of the function list as well.

https://flic.kr/s/aHskwdJ15K

[Dan Grelinger]

(04-13-2018 06:55 PM)AndiGer Wrote:  Maybe also register arithmetics is counted seperately? STO+ nn et al.??

In the function list of 128 (129 subtracting the duplicate entry for e^x), STO+, STO-, STO*, and STO/ are four entries, so those are already counted.

Some interesting functions:
The yellow shift key is considered a function.
The ALPHA key is considered a function, in addition to AON and AOFF.
The 'Append' character/feature is considered a function.
The back-arrow (in addition to CLx) is considered a function.
GTO. (line number) is considered a function
GTO.. (add end to current program and pack) is considered a function.
The ON key is considered a separate function from the ON and OFF commands. (So turning on the calculator is a function?)
The PGRM key is considered a function.
The R/S (Run/Stop) key is considered a function, in addition to the programmable "STOP" command.
The USER key is considered a function.

There is no 'indirect addressing' function in the list. It is possible that HP considered that a 'feature' instead of a function (as documented by Zaphod).

What I was interested in was an HP documented list showing over 130 functions. I think I agree with Thomas Okken, the marketing department was not diligent about what the definitive number of functions was. The number probably originated in a meeting where the design engineers were bringing the marketing team on board and someone threw out 'over 130 functions', and that lived on for a few years before someone actually fixed it.

[Ángel Martin]

1. The GTO function definitely has a double-duty task - doing the PACKING with the two dots.
2. The polling points (even if not accessible by the user) can be considered as functions - debatable of course.

Every entry in the "byte table" *could* be considered as a function, thus that takes the count up to 256 functions ;-)

Well, obviously that's not a relevant measure, but even if there's usually some dubious claims always made by marketing in this case there is a basis for their "rounding it up to the nearest figure".

Ah, don't forget the mysterious "W" function, to date unaccounted for - who knows what they had in mind reserving an entry for such, the "path not taken" so to speak.

[Jake Schwartz]

For info's sake, here is Keith Jarrett's HP-41C pocket hex table from 1982:


Enjoy counting,

Jake

[brickviking]

This reminds me of the "over 2900 functions" in the 9750gII. I never quite figured out how they got that figure, and when I asked for a list of those I got a list that appeared to hold far less entries (mid-hundreds, if I remember correctly). I suspect the 2100 functions might be a bit closer to the mark for the HP-50G, but I can't quite remember where I saw that referred to.

From the nice Viking, not the grumpy one

(Post 200)

[Zaphod]

There is a topical programmable Casio right now being spoken about in another thread , that apparently has 664 functions
https://www.casio-intl.com/asia/en/calc/.../fx-5800P/

[pier4r]

If one considers a function accepting different types of inputs , as in a function overload (that is surely done in the source code), I would believe that 50g, 9860 series, ti89 and others would be over 2000 functions without problems.

Only because a function has the same name it doesn't mean that automatically has to work on every possible input.

For example in rpl '+' can add integers, reals, strings, mixed, lists etc.. those are like 10+ variants for the same operator.

[Dan Grelinger]

In later manuals, the list of functions dropped from 128 to 127. I examined the list and found that sometime around 1981, the yellow 'shift' key was removed from the list of functions. Reasonable, since it did not really do anything itself.