Landscape RPN calculator with alphanumeric display

[Joseph_21sv]

It has been much in my attention that one well liked thing among HP fans about HP’s Voyager series is its landscape format. However, what the community appears to really desire, if http://www.vcalc.net/hp.htm and the hypothesis of an HP 15S from this forum are anything to go by, is for HP to issue an RPN programmable calculator with an alphanumeric display and a keyboard in landscape format, just like a successor to the Voyagers, though they think it not in their self interest to do it. So, absent any presumed self interest in such a project on HP’s part, is/are any forum member(s) interested in doing it themselves?

[Dale Reed]

Yes, a few of us have played around with designs for such a creation, Walter and I and a few others. I've gotten to the point of collecting a number of components, and I have dinked around with some of the parts in a test setup.

The items I'm playing with would make such a unit cost prohibitive to sell. The display I'm using, in single unit quantities, costs me more than many people would pay for such a calculator.

I've gathered together some SO-sized surface mount chips for a LiIon charge controller (and a LiIon cell), a couple different LDO voltage regulators, a MEMS-based real time clock, SD card socket, USB mini-B socket, surface mount tact switches for the keyboard, a 4-line character display, 256k x 8 SPI FRAM, etc., etc. My "dinking" breadboard uses a Freescale MCF51JM128, but I was considering using a Silicon Labs SiM3U-series microncontroller. I have some of both chips. (These are TQFP parts, something I haven't tried hand-soldering yet...) Just haven't gotten around to laying out the PCB for all of this. Something I was hoping to do over the Christmas/etc. holiday season.

I also have a 3D printer and am hoping to print the shell and keyboard buttons for it. (I doubt I can achieve the resolution needed for recessed / double-shot legends on keys, however, even with an 0.25 mm nozzle.)

Anyway, home stuff gets in the way of toys and play time.

Search the forum for "Keyboard Challenge" -- Eddie Shore started such a thread several months ago ( http://www.hpmuseum.org/forum/thread-1243.html ) for designing a calculator with 25 keys (just in time for the 40th anniversary of the HP-25, perhaps). There are older topics with some 39- and 43-key landscape machines that Walter and I proposed, as well. I would like to see one with the double-wide ENTER^ key horizonally in the center of the bottom row of keys (to double as a space-bar for typing, perhaps) and my couple of designs reflect that.

Can't find the picture of the 43-key beast on this PC, but I will edit this post later and attach it.

Edit: Here's one that I came up with (it has since been changed to swap the display to the left and the arrow keys to the right):

   

Edit: Also, I hope Walter doesn't object -- I also attached a picture of one of his landscape designs (Walter, if you want me to remove this, just let me know.) :

   

Dale

[sa-penguin]

I started messing about with PhoneGap.
Not much to show for it so far - just a HTML table, formatted to look like a (landscape style) keypad.
Numbers and basic functions on the left, ENTER key bottom center.

  table.html (Size: 23.34 KB / Downloads: 37)

[walter b]

You made me looking into my old files again. Let's see, for example:

1. This was presented on the forum in 2008 (including a LCD available at that time).
2. This in 2008 as well, showing what a better display could do.
3. This was shown in 2009, just redefining a 15C.
   

Then the WP 34S took over. Please note all my above designs are just mockups - what we call vapourware here traditionally for reasons found in the depths of the old forum (search for OpenRPN). There were layouts of portrait shaped calculators, too. And of course also other forum members designed their dreams (I remember Jeff O. and apologize for the others I forgot).

At the bottom line, layouting keyboards is fun (and graphic output makes your readers / viewers sharing it) though the real fun and satisfaction comes with a real platform IMHO. AFAICS, such a platform is missing for a landscape RPN calculator with alphanumeric display.

d:-)

[Bill (Smithville NJ)]

(12-09-2014 11:13 AM)walter b Wrote:  At the bottom line, layouting keyboards is fun (and graphic output makes your readers / viewers sharing it) though the real fun and satisfaction comes with a real platform IMHO. AFAICS, such a platform is missing for a landscape RPN calculator with alphanumeric display.

d:-)

I have been following this project as well as others for some time.

It seems to me that the main reason most of these projects never materialize is that they are hung up on how to produce the proper platform - this being a piece of hardware with keys and screens.

I always wonder why the other approach was not taken. Start with an existing platform (namely the cell phone), create an emulator of the desired product, debug it on the emulator and move it to a "proper" hardware platform when it becomes feasible to do so in the future. It would have the benefit of having created something now as opposed to only dreaming about it.

I realize that this is an outrageous idea for many people.

But many hardware/software projects in the past have been bootstrapped in this way. I have read where some of the minicomputers were developed/debugged with software emulators long before there was any hardware for the software. That way the software was ready to implement when the hardware did become available.

In my view, a working cell phone version that I can use today will always beat out a vaporware, pie in the sky version that I might be able to use in xxx years, if ever.

Just one man's opinion.

Bill

[sa-penguin]

(12-09-2014 11:13 AM)walter b Wrote:  At the bottom line, layouting keyboards is fun (and graphic output makes your readers / viewers sharing it) though the real fun and satisfaction comes with a real platform IMHO. AFAICS, such a platform is missing for a landscape RPN calculator with alphanumeric display.

There are 49-key Bluetooth keyboards. They were inspirational in designing my HTML layout. However, even if you butchered a Bluetooth keyboard for parts, the design problems (opportunities?) quickly start to mount.

For a keypad of that size & layout, where are you are going to put the display? You need a folding "shell" case: a micro-mini laptop. If you put the "brain" CPU under the keypad, is there still room for the battery? Or should that go up top, behind the LCD? Recharging Lithium, or disposables?

I would dearly love a device in this form factor. However, I suspect the more I think about it, the more respect I will have for those designing, and actually implementing, the 43.

[sa-penguin]

Just heard of the LibreCalc.

http://hackaday.com/2014/12/09/a-calcula...-hardware/

http://www.librecalc.com/en/

[Joseph_21sv]

(12-09-2014 11:13 AM)walter b Wrote:  At the bottom line, layouting keyboards is fun (and graphic output makes your readers / viewers sharing it) though the real fun and satisfaction comes with a real platform IMHO. AFAICS, such a platform is missing for a landscape RPN calculator with alphanumeric display.

It's really missing? Maybe if you are talking about calculators purposefully designed to operate on RPN expressions. However, there are several models of a landscape algebraic programmable calculator with alphanumeric display—Viktor T. Toth has the following at http://www.rskey.org (he does not claim that they are only ones to have ever been produced): Canon: 1 with graphic LCD; Casio: 2 with graphic LCD and 1 with character LCD; Sharp: 5 with character LCD; TI: 4 with graphic LCD, 2 with character LCD and 2 with segmented LCD. For what it's worth to note it, these almost universally have a pixelled display (the more portable models are also universally discontinued, but that's beside the point here except that they are the genuine calculators—at least according to US examination boards). Of course, these are all ultimately inspired by HP's Voyagers.

(12-09-2014 02:03 PM)sa-penguin Wrote:  There are 49-key Bluetooth keyboards. They were inspirational in designing my HTML layout. However, even if you butchered a Bluetooth keyboard for parts, the design problems (opportunities?) quickly start to mount.

For a keypad of that size & layout, where are you are going to put the display? You need a folding "shell" case: a micro-mini laptop. If you put the "brain" CPU under the keypad, is there still room for the battery? Or should that go up top, behind the LCD? Recharging Lithium, or disposables?

I would dearly love a device in this form factor. However, I suspect the more I think about it, the more respect I will have for those designing, and actually implementing, the 43.

Check http://www.rskey.org/CMS/index.php/exhibit-hall/96. Sharp has managed to make one-piece calculators with over two fifths more keys and a 1*16 or 1*24 display. However, the cost is that, the way they have it implemented, almost two thirds of the keys are absurdly small. But a 49 key keyboard like you have mocked up only needs a Voyager form with an extra row of key space to be implemented—that is if you are willing to settle with only one or two lines of display.

[walter b]

(12-14-2014 10:32 PM)Joseph_21sv Wrote:  ... there are several models of a landscape algebraic programmable calculator with alphanumeric display—Viktor T. Toth has the following at http://www.rskey.org (...): Canon: 1 with graphic LCD; Casio: 2 with graphic LCD and 1 with character LCD; Sharp: 5 with character LCD; TI: 4 with graphic LCD, 2 with character LCD and 2 with segmented LCD. For what it's worth to note it, these almost universally have a pixelled display (the more portable models are also universally discontinued, but that's beside the point here except that they are the genuine calculators—at least according to US examination boards).

Please tell me the models you refer to — I doubt they are in production still. This is crucial since only a calculator being in production (or otherwise easily available) can be a platform or basis for repurposing. And I should have mentioned that when I speak of calculators I mean pocket calculators like the Voyagers were or the WP 34S is (yes, I know the latter isn't landscape format).

d:-?

[Dale Reed]

Note to OP: my post above just edited to add picture of one of my designs, and also one of Walter's. (Walter, if you want me to remove that picture, let me know. I apologize for not asking permission to post.)

Dale

[Jim Horn]

Pocket sized? Then the HP-71 doesn't count...

[walter b]

Dale, feel free to quote my old designs. I feel honoured by you remembering what I have published on this forum years ago.

d:-)

[Joseph_21sv]

(12-14-2014 11:33 PM)walter b Wrote:  Please tell me the models you refer to — I doubt they are in production still. This is crucial since only a calculator being in production (or otherwise easily available) can be a platform or basis for repurposing. And I should have mentioned that when I speak of calculators I mean pocket calculators like the Voyagers were or the WP 34S is (yes, I know the latter isn't landscape format).

d:-?

The pocket calculators are:
Casio fx-5500, fx-7500G, FC-1000
TI-62 Galaxy, TI-66, TI-67 Galaxy
I presume they are all discontinued—published data on their production runs is conspicuously spotty—but not prohibitively rare. Except for the Casio fx-5500, they have few or no awkwardly small keys on their keypads. However, they all have firmware (presumably) in one-time programmable ROM, so they need to have their microcontrollers replaced entirely in order to be repurposed if you do manage to find them in good enough working order to be repurposed. As for the WP-34S/HP 30b, since it has an alphanumeric display, would landscape formatted hardware for it be such an outrageous idea? After all, Rick Furr, who curates vcalc.net is effectively wishing for HP to landscape format the 32SII and 33S.

[MarkHaysHarris777]

(12-09-2014 11:13 AM)walter b Wrote:  At the bottom line, layouting keyboards is fun (and graphic output makes your readers / viewers sharing it) though the real fun and satisfaction comes with a real platform IMHO. AFAICS, such a platform is missing for a landscape RPN calculator with alphanumeric display.

It won't be missing for long, bubba; my commitment to myself.

I saw the pattern over and time again, when I worked with IBM; particularly among 'managers' and the like-- others who live to please them... and that is interface vs. infrastructure. [ Interface vs. Infrastructure ]

Managers and marketing reps want to see the interface... if you've got one, they're interested in additional funding maybe, if you don't have one they call the who thing vapor-ware and want to fire (sack in the U.K.) you!

The 'important' piece of the project is NOT the interface. I'll restate it for the hard of seeing--- interfaces are a dime a dozen and NONE of them are going to calculate anything!

The 'important' piece of the project is the infrastructure. The infrastructure of a calculator is more complex and more difficult (from an engineering standpoint) than most people realize, while they are making up their keyboard layouts (interface). Barry asked me what my thoughts are for key labels. Actually, I had not given key labels any thought at all... there is more than enough time for that! I want the right infrastructure, and I want it to be solid, quality, and durable.

Think about infrastructure vs. interface. Which is more important to you?

Cheers,
marcus

[sa-penguin]

(03-19-2015 06:35 PM)MarkHaysHarris777 Wrote:  The 'important' piece of the project is NOT the interface. I'll restate it for the hard of seeing--- interfaces are a dime a dozen and NONE of them are going to calculate anything!

The interface should not be ignored, or tacked on as an afterthought, either.

At one extreme, I think of Steve Urkel's original Time Machine. from memory, it had very few buttons. If you wanted the year 1998, you used the "select" button to Move a cursor to Years, then pressed the "advance" button 1998 times (cue audience lughter)

You could also, in theory, use a Morse key and an LED to act as keyboard and display. Very small, but rather impractical.

In the real world, I know of system which beep whenever settings are changed. The exact same beep indicates errors, invalid data entered, or a hardware fault. I've experienced a fault reporting system that made a phone ring and deliver a recorded warning - every fifteen seconds. intermittent faults that flood error logs. User interfaces that have the most used opotions buried in five levels of submenu, while the main menu level is full of options lmost never used.

GOOD interfaces need to be customised for each individual. Computer graphics people love to design, and redesign, the graphic style, layout, and icons on web apps. For a calculator? I'd settle for a BASIC interface, void of more obvious flaws.

The classic example from here would be: arguing which side of a keyboard has the + - * / keys, and their order. Short of having a keyboard you can open up and re-arrange the keytops on, you will upset someone. However, forget an ENTER key and the protests would much louder.

In short, don't forget the interface. It's the part customers see, and use, every day.

[Didier Lachieze]

(03-20-2015 07:52 AM)sa-penguin Wrote:  The classic example from hre would be: arguing which side of a keyboard has the + - * / keys, and their order.

[Joseph_21sv]

(03-19-2015 06:35 PM)MarkHaysHarris777 Wrote:  The 'important' piece of the project is NOT the interface. I'll restate it for the hard of seeing--- interfaces are a dime a dozen and NONE of them are going to calculate anything!

The 'important' piece of the project is the infrastructure. The infrastructure of a calculator is more complex and more difficult (from an engineering standpoint) than most people realize, while they are making up their keyboard layouts (interface). Barry asked me what my thoughts are for key labels. Actually, I had not given key labels any thought at all... there is more than enough time for that! I want the right infrastructure, and I want it to be solid, quality, and durable.

Think about infrastructure vs. interface. Which is more important to you?

Cheers,
marcus

The interface is unimportant to me, for as I have mentioned previously, we can simply cannibalize it (in whole or in part) from some calculator that already exists. But, as for the infrasturcture (I guess you mean the firmware), what is better, from the point of view of your criteria, than using firmware which already exists? After all, the first RPN calculator was built for engineers, and engineers demand accuracy like no one else; they practically live by it. So, we can simply take infrastructure that already exists and put it behind the cannibalized interface and have the calculator which we speak of here in a usable concrete form. And we would not need to be particularly ambitious with its function set either, having no competition to it.

[Claudio L.]

(03-19-2015 06:35 PM)MarkHaysHarris777 Wrote:  Think about infrastructure vs. interface. Which is more important to you?

Good Core + Good UI = Useful program
Good Core + Bad UI = Only useful for a handful of gurus willing to put up with it
Bad Core + Good UI = Fun to use but useless junk
Bad Core + Bad GUI = Useless junk

The only good combination is to have both a good core and a good interface.
The core *HAS* to be good. If the infrastructure is bad, no interface will save it.
An example of a good core with a bad ui is the 'sed' utility in **nix Operating Systems. The syntax is so cryptic that it takes 3 hours reading the manual every time you try to use it. That makes a very powerful regex engine only useful to a few people that have the patience, and makes people who actually use it a lot less productive than they could be.

[rprosperi]

(03-20-2015 04:29 PM)Claudio L. Wrote:  

(03-19-2015 06:35 PM)MarkHaysHarris777 Wrote:  Think about infrastructure vs. interface. Which is more important to you?

Good Core + Good UI = Useful program
Good Core + Bad UI = Only useful for a handful of gurus willing to put up with it
Bad Core + Good UI = Fun to use but useless junk
Bad Core + Bad GUI = Useless junk

The only good combination is to have both a good core and a good interface.
The core *HAS* to be good. If the infrastructure is bad, no interface will save it.
An example of a good core with a bad ui is the 'sed' utility in **nix Operating Systems. The syntax is so cryptic that it takes 3 hours reading the manual every time you try to use it. That makes a very powerful regex engine only useful to a few people that have the patience, and makes people who actually use it a lot less productive than they could be.

For SED, check out the "Advanced BASH Scripting Guide" here. And I agree with your analogy.

Probably the most useful single doc for learning command line Linux I've seen. Readble style and quick overviews are useful, but also includes stuff that gets waaay complex and powerful. Check it out - it made SED useful for me (about 10% of it that is). SED is very reminiscent of TECO on DEC machines. Anyone here remember TECO?