43S News

General discussion about calculators, SwissMicros or otherwise
vgoudreault
Posts: 13
Joined: Sat Oct 17, 2020 2:32 am

Re: 43S News

Post by vgoudreault »

Going back to the root of the WP43, it looks like it was seen as an optimisation and evolution of the HP42S, built around that calculator's physical format; and then updated to take advantage of the larger screen and keypad of the DM42.
But what if the current developers were allowed to come up with specs, a clean sheet physical form, in terms of capabilities, memory, screen, number
and size and shape and placement of keys; what would be an ideal WP43's form factor?
User avatar
Walter
Posts: 1942
Joined: Tue May 02, 2017 11:13 am
Location: Close to FRA, Germany

Re: 43S News

Post by Walter »

Yes, in principle. But the 43S was basically thought as an advanced WP34S featuring a display allowing for menus. The WP34S was meant as an advancement of the HP-42S (and HP-16C) but was limited by the hardware of the HP-20b/30b. Roots and history of the WP34S can be found elsewhere (~2008 - 2011).

When you start a project like the 43S, you seldom start from scratch. There are roots going back some 15 years in this case, where we were always talking about a pocket calculator. The WP43S layouts of 2012 show a keyboard of 43 keys incl. 6 menu keys already.* The chapter Release Notes in our manuals lets you follow the development. We were (and are) very happy about the display found by Eric Smith and late Richard Ottosen, then also used by SwissMicros. Actually, we were involved to some extent in layouting of the DM42, too. The WP43S Reference Manual, in particular pp. B-1 and B-19ff, tells a bit about our considerations.

If I could wish I'd like to get slanted keys like we had up to the Voyagers, but this is just a minor point. LCD and mechanics are close to optimum. Memory is never sufficient, of course -- but I'm no expert in electronic hardware. And everything else is 'only software'.

*More keys would qualify for the battleship class of calculators blowing our shirt pockets. Less keys disallow a reasonable balance between fixed and menu keys IMO.
DM42 SN: 00041 β
WP 43S running on this device

HP-35, HP-45, ..., HP-35S, WP 34S, WP 31S, DM16L
vgoudreault
Posts: 13
Joined: Sat Oct 17, 2020 2:32 am

Re: 43S News

Post by vgoudreault »

I was thinking along the lines of the rationale for the layout.
Let me explain.
At the risk of dating myself, in the early 1970's, Hewlett Packard introduced the HP-35, quickly followed by the HP-45, HP-55 and HP-65. They all had their basic arithmetic function buttons on the left of the numeric keypad. Texas Instrument offered its SR-50, with the arithmetic buttons on the right.
Of course, this was not the only difference between the TI and HP philosophies; and the argument that won me over to the RPN side was the demonstration, by my college math teacher, that one would run out of bracket levels way before running out of stack, and that RPN allowed to calculate the way one would manually solve, rather than how one would (clumsily) attempt to transcribe a written equation.
Operators precedence was a non issue with RPN; it was a mathematically SUPERIOR approach.
That was the beginning of the division of the students (at least those in the science programs; the humanity students did not care much about math) into two factions; beyond the RPN VS infix, there were other arguments exchanged: the TI calculator was internally computing to 13 decimals instead of 10 for HP, the SR-50 had hyperbolic functions ("When was the last time you used hyperbolic functions?" I asked someone who brought it up, the answer was 'never'... I may be wrong, but I think that directly accessible hyperbolic functions were omitted from TI successor models, perhaps effectively showing that they were not as desirable as others functions), and the build quality. There was a student who accidentally dropped his SR-50 from the height of his seat onto the carpet of the college library; it was busted. I once dropped my HP-55 from the height of a table onto the concrete floor of a lab, all it had was a little superficial sliver of plastic becoming partially detached, I put a small drop of super glue in there and one needed a magnifying glass and special instructions to locate the damage.
In the end, it was pretty apparent that the HP's were designed by engineers for engineers, and that selling for double the price of the competition meant that some extra expense went into figuring what the specifications ought to be, beyond the impression that HP calculators were sturdy enough to be used as a hammer.
But I digress.
HP was happy to have the primary arithmetic keys on the left side until they introduced the Voyager series (which were a step back form the HP-41's in terms of power and flexibility, so I never saw the need to indulge in one of those). There could have been some logic in the relocation of the '+-x÷' keys on those calculators since the left side is occupied by the bulk of the function buttons and to have the 4 arithmetic ones on the right highlighted them; but that basic arithmetic layout was retained for the HP-48. It took months for me to get used to it, having spent the previous 16 years with calculators having those buttons on the left side.
After that, all new HP calculators had their arithmetic functions on the right side.

The DM42 has them on the right, because it seems there is a desire to emulate the layout of the HP-42 as well as possible.
The DM41X has them on the left, evidently because that is where they were on the HP-41.

The WP43S has them on the left, because... what?

I am not questioning the validity of putting them there, if HP had them there are the beginning on the HP-35, I dare presume they they thought about it and decided after careful analysis; there must have been a reason to be different from the previous 4 functions calculators from other vendors. I am basically after the equivalent brilliant demonstration from my math teacher 46 years ago that proved RPN was clearly superior.
There ought to be pros and cons that could be rationally invoked to conclude that one layout is better; after all, this is not supposed to be a matter of religious faith.
And if in the end, this turns out to be just personal preference, one might wonder why the key assignements are simply not put in a user accessible configuration file, for ad-hoc customization using matching key labelling stickers.
User avatar
Walter
Posts: 1942
Joined: Tue May 02, 2017 11:13 am
Location: Close to FRA, Germany

Re: 43S News

Post by Walter »

There is a chapter 'Layouting' in the WP43S Reference Manual, starting on p. B-19. It contains a short discussion of your topic.

Regarding configuration, please see Section 6 of the WP43S Owner's Manual.

Generally, these manuals are your friends -- there's a lot of information in them. 8-)
DM42 SN: 00041 β
WP 43S running on this device

HP-35, HP-45, ..., HP-35S, WP 34S, WP 31S, DM16L
Peet
Posts: 145
Joined: Tue Sep 29, 2020 12:01 am
Location: Germany

Re: 43S News

Post by Peet »

vgoudreault wrote:
Mon Jun 14, 2021 12:23 am
HP was happy to have the primary arithmetic keys on the left side until they introduced the Voyager series
The first time I saw a Voyager (11c) was with the colleague who made me aware of the HP-41. I asked him about the buttons on the right and he said: this is normal with HP on a horizontal layout and showed me the HP67 & 97.

Strictly speaking, HP changed the base-layout only after the change to algebraic & RPN models in the same housing (e.g. 22S/32S) and the combi-models like the HP48.
My programmable calculators - former: CBM PR100, HP41CV, HP28S, HP11C - current: HP48G(+), HP35S, DM41X, DM42
vgoudreault
Posts: 13
Joined: Sat Oct 17, 2020 2:32 am

Re: 43S News

Post by vgoudreault »

Seems you people thought pretty much of everything.

Still, the reason for HP's original configuration remains a mystery.
And the result of the survey probably shows people voting for what they are familiar, rather than what could be best; in essence, better is what people are accustomed to.
And the same discussion could happen about why the digits on a calculator are set opposite to what they are on a telephone -- with the difference that one ends up typically punching a lot many more digits on a calculator than on a phone, unless getting one of those annoyingly complicated "please listen to all the menu items as some of those have been changed" automated phone attendant.

I basically went through the same questioning when I created my Android units calculator application, wondering where each function should go and not finding any rational to decide, ending up pretty much duplicating the placement of my HP-48's keys for the lack of better guidance. That was an unsatisfactory process.
Panchdara
Posts: 14
Joined: Sat May 15, 2021 9:02 am

Re: 43S News

Post by Panchdara »

Just a thought.

Most, not all, but most people are right handed and will press calc buttons with their right hand. I can only think that this is so the right hand clears the keyboard to 1) make it more visible and 2) no errant key is pressed?

Just a thought.
Peet
Posts: 145
Joined: Tue Sep 29, 2020 12:01 am
Location: Germany

Re: 43S News

Post by Peet »

vgoudreault wrote:
Mon Jun 14, 2021 3:22 am
And the same discussion could happen about why the digits on a calculator are set opposite to what they are on a telephone
The model for the pocket calculator was the cash register, for the telephone keypad the rotary dial.
My programmable calculators - former: CBM PR100, HP41CV, HP28S, HP11C - current: HP48G(+), HP35S, DM41X, DM42
vgoudreault
Posts: 13
Joined: Sat Oct 17, 2020 2:32 am

Re: 43S News

Post by vgoudreault »

Not entirely accurate. Bell Laboratories did conduct extensive human factor studies before settling on the layout for telephones.

https://en.wikipedia.org/wiki/Push-butt ... pad_layout
https://en.wikipedia.org/wiki/Telephone_keypad#Layout
User avatar
Walter
Posts: 1942
Joined: Tue May 02, 2017 11:13 am
Location: Close to FRA, Germany

Re: 43S News

Post by Walter »

Release 0.20.3 is here https://gitlab.com/Over_score/wp43s/-/releases. There were some improvements implemented (also on the keyboard) and some bugs removed. You get extended statistic support now. In the release package is everything you may want to try, test, and check the 43S on your computer or your DM42. Please look into the current manuals to see what's possible so far.
Enjoy!

And if you discover anything suboptimal, please report. There are certainly some issues - who finds one will get 25 test points (Pierre has got 50 already). Even if your attempt to install the 43S in whatever environment should fail, this information is appreciated - we will help you out. (I know the keyboard template on the penultimate page of the Reference Manual is outdated, will be updated in due time.)
DM42 SN: 00041 β
WP 43S running on this device

HP-35, HP-45, ..., HP-35S, WP 34S, WP 31S, DM16L
Post Reply