Basic HP Calculator


Due to the lack of basic HP calculator (except 9-E which is a one-line algebraic calculator) should we send suggestions to build a basic calcualtor with algebraic AND RPN entry (think circa 70's HP 25)?


* LOG, LN, EXP, 10^X
* x! (for ALL real numbers), 1/x, %, POWER, ROOT, SQ, SQRT
* 3-4?, 9? Memories
* Solar and Battery Power
* RPN Stack Functions: Swap, Roll Down, Drop

Good for an entry level calcualtor?


Hello Eddie,

During the HHC2004 workshop in San Jose last September, Richard Nelson suggested to the HP folks who were there to build a cheap HP-45-like calculator. Eichard suggested that HP can use such an "affordable" machine to popularize RPN. Moreover, Richard said that HP could hand out such a calculator to teachers attending conferences that involve eductaion and calculators. This approach will help HP put its foot in the door with educators who are TI-fans. HP listened. Time will tell.



Gentlemen, in "real life" would most RPN users find a "basic machine" + some programming capability a little more practical, even high school or college students?


Eddie Shore's post is a good one.

The TI25X (now discuntinued)

has a nice form factor.

[If the keys were slighly 'clicky' without bounce and it had RPN with a large ENTER key above the numeric area I'd be eternally happy. If the keys were real plastic and not rubbery that'd be even better.]

Right now (for me at least) programmability is of less import than a few other things... my features I'd like, in no particular order:
- 10 digit display
- RPN, 4 level stack + LastX, RDn and RUp
- some memories
- storage arithmetic (STO+ 05, ST/ Z etc.)
- log, ln, exp, 10^x, y^x, xth root of Y
- circular & hyperbolic trig and P->R/R->P
- usu "stats" function, n!, nPr, nCr
- multibase operation (dec/oct/bin/hex)
- H.MS<->HR conversion
- Fix/Sci/Eng/float modes
- basic financial funcs: PV/FV/PMT/i/n

Bill Wiese
San Jose


Bill Wiese gives a good abstract.

[Right now (for me at least) programmability is of less import than a few other things... my features I'd like, in no particular order: - 10 digit display - RPN, 4 level stack + LastX, RDn and RUp - some memories - storage arithmetic (STO+ 05, ST/ Z etc.) - ...]

A few more ideas and personal opinions: I really would buy a NEW hp if it has a formfactor +/- the pioneers, GOOD keyboard (a click = an entry!), IR-printer port.
And - today - I'm shure you can have programmability, equation entry/solver and even more for a few bucks.

What about someting like a 27S?

It was THE do-everything; including the IR-printer Port.
But it lacks RPN! (good for those who comes from TI? ;-)

Or an RPN programmable scientific 32S?

Or better the revised 32SII?

Like the new 33S they have both a big minus (beeing hp calculators): why don't they have the IR-printer port? really don't know what hp thought those days...

I suppose today you have cheap chips and displays in good quality. What you still don't got, is cheap but good keyboard and cheap but ergonomic design and mix of menues and many direct keyboard functions as well! I believe that this two factors still need some special engineering effort and that's definitly not cheap. But it would be worth any penny and make again THE difference between a hp and another one... hopes...


HP 42S with 64K of RAM, PC connection, clock functions of the 41C/CX Time Module and the unit conversion of the 48G.


... don't forget a clear, sharp 2-line display - I agree with almost everything else posted already.

Well, let's sum it up:

- Basic math in direct access (no menus at this level, I'd like blue/gold/black keyboard design of the HP34C)

- Basic 2 parameter stats (including L.R.)

- 10 general purpose storage registers (plus stats registers in background)

- RPN programmability (like HP42s)

Just my 0,01 Euros


@Walter B (et al):

What do you think about this efforts here? OpenRPN eg this post here

Do you think OpenRPN project is going in the right direction we talk here about?

There are really nice drafts of P- and L-formfactors. Maybe this is the only way to got a new RPN calculator with the old goodies of hp...


Openrpn is a great project, but I don´t think it will be helpful. We need a Hardware platform rather than a complete calc. Eric Smith seems to be working on hardware prototypes!

Merry chistmas



We may be able to help somehow. Our calculators will essentially just be platforms since everything about them will be open to the public. If there is sufficient interest in producing another set of lower priced electronics I will do my best to help.



And ... don't forget the shape and dimensions identical to those of a Voyager.

Merry Christmas,


I tutored an algebra student with a TI-25 Solar, and ended up giving the calculator to the student. That was a nice little basic calculator.

Many of the entry level calcs have what we want, minus the RPN.


I have been wishing for a "basic" RPN for quite some time now. I have a 48GX but do not use it to its potential, and I do not like its rather large size. Since I couldn't buy a "basic" calculator with RPN, I decided to get my 41CV repaired. If they could bring back the 41 again and put the +-/* keys on the right hand side, and sell it for under $100 I would go out and buy one.


The way I see it, most of HP's real calculators are built with the logic: "If it is more powerful, then someone can use it to do less or more... it's up to them" Unfortunately, this is not true. Parents want to buy something cheap and easy to use (does that rule out RPN for first time users?). Furthermore, teachers don't always like overly powerful calculators. During my freshman year, a teacher had a ban on all Programmable calculators (so students couldn't use graphing calculators.) Unfortunately this also meant I couldnt use any RPN calculator. I missed more problems from making mistakes in Algebraic than anything else.

Just some thoughts


This thread has drifted away from the original subject of a "Basic HP Calculator", which meant a simple RPN calculator from HP. One with only + - * /, and maybe % and square root. People were asking for this as early as 1978 when John Ball suggested "El Cheapo" in his book "Algorithms for RPN Calculators."

Reasons why we want one include 1/ It would be really nice to be able to buy a cheap and simple RPN calculator for us (readers of Dave's site) to use. 2/ It would let us introduce friends and relatives to RPN. 3/ A simple and affordable RPN calculator could be used at introductory level in schools. I think this last particularly important, as elementary arithmetic _is_ RPN (or at least stack notation), and teachers willing to use such a calculator would actually find it easier to teach the use of calculators. Then HP would have a ready market for more advanced models.

Reasons why HP have never introduced such a model include 1/ Introducing and selling a calculator costs much the same, way over a million US dollars, whether it is a simple model or a complicated one - things like getting electrical approval from many countries or setting up a production line, packaging, and promotional material, cost nearly the same for any model. 2/ But profit margins on a low-cost model are far smaller, so HP would have to sell far more of these than of more expensive models just to cover the cost. Yet this low-cost model would take away sales from the previously least expensive model, which brings in more profit per unit sold. 3/ Therefore the bean-counters at HP will not provide the funds for the calculator group to introduce such a model unless the calculator group can _guarantee_ huge sales.

What can we do to change this? 1/ Find someone to place a firm order with HP for at least a million such low-cost calculators. Do you know anyone like that? Do you control the policy of your local math education board? No, I thought not. 2/ Keep pressing HP. After all, we've only been trying for 26 years so far. 3/ Find someone else to do it - maybe the Open RPN project can make a really simple version as well as the really advanced one.

Any ideas?


Hi Wlodek, all;

I know this is somehow controvesial, but I wonder that such "El Cheapo" (I was given John Ball's book sometime ago; thanks, G.D.!) is an easy implementation, even with a bit more of memory and functions (maybe 1/x, x^2, and a few others), by using a single, commercially available multipurpose controler (ATMEL, PIC, etc.). The three problems I see so far are:

1) housing and keyboard;
2) LCD/OLED commercial display, and
3) Ownership issues.

The third one relates to HP and RPN: does RPN as both a "name" and a "protocol", in its deepest meaning, belong to HP?

I am completely aware of some private iniciatives around developing RPN calculators outside HP "borders" (I myself encouraged some researches, and right now I'm getting myself ready to go ahead with the HW stuff), but I guess that a downloadable "Do it yourself" booklet with complete instructions on how to build your own RPN-based "El Cheapo" for some of the most common controllers available and their related core systems would be, at least, "pretty neat". I am not straightly following the OpenRPN projet (I must update myself...), but I guess that they have probably got closer, if not surpassed, the "El Cheapo" basic needs. Youngers would also find their own motivation.

Any suggestions?


Luiz (Brazil)



As to your third issue, ownership of RPN, I don't believe HP has ever claimed ownership of RPN. Further, I've got a non-HP book with a 1967 copyright date that specifically refers to reverse Polish notation. The book is Computer Programming and Computer Systems by Anthony Hassitt. The RPN discussion is on pages 95-98, and after a brief introduction to RPN, the discussion focuses on RPN as implemented on the KDF9 computer.

A Google search on the KDF9 computer eventually lead me to this article on RPN: RPN was invented in the 1950s by Australian computer scientist Charles Hamblin.


Further evidence of non-infringement is the NOVUS line of RPN calculators.

The real hope for a cheap entry calculator is the Hp33s at a street price of about $30. It is both Algebraic and RPN.

Actually Hp made the Hp20s that could have easily filled this void at a price of $30. It was a quality LCD calculator with an enter (ok INPUT) key where it was supposed to be.

If Hp had released this, it would have done well, maybe too well, because many RPN addicts would bypass the $60 Hp32s/33s line.

I would love a $30 Hp20s with RPN. It would be quality and a low price with enough features for general use. Just give the community that and many of us would SHUT UP! But that won't happen because HP would lose at least 10% of their present Hp49G+ sales and probably 30+% of their HP33s sales (never mind follow up sales, 1-3 years down the road, by new converts who start out on the low end, marketing doesn't know these potential customers exist, they can only count the existing sales base). Short Sited ^%$#!!!

Just me Ranting again.


Hi folks,

I've prob said this before awhile back - the issues and difficulties in building/selling a new calc of any type are NOT of chip selection, writing firmware, or the even cost of the actual electronics (CPU, LCD, zebra connector, PCB, batt clip). There are probably a couple of dozen folks here that can write a good RPN calculator for any of a number of eligible microcontrollers, with some fraction thereof who can prob get the floating point and transcendatals "just right". (Plan B would be to just do a Nut CPU emulator and paraphrase sections of HP Woodstock or 41 code - some of this may not be copyrighted??)

RPN has no patent protection or licensing requirements. Any such items would have passed by way more than 17 yrs ago. And even when HP was selling calcs, there were other RPN calcs on market (Nat'l Semiconductor/NOVUS, APF, Corvus?).
Even the truly HP combo of 4 level RPN with T reg copy-down and with a Last X register is free for all to use.

The real costs are setting up injection molds, dealing with vendors that expect orders in at least hundreds of thousands, and creating a 'hole' in the sales channel for your product. Making a blister pack with instruction book and shipping/boxing etc. may cost as much or more than the actual electronics!!

Products don't make it past the marketing gate unless they hit the right price point for their feature set. It costs relatively little to add "features" to a calc (other than some ROM space and key legends) once it exists, so a higher price can be charged.

The elegance of a stripped-down RPN calc - even a 20S w/RPN - while eminently logical to the folks here, doesn't pass the the multiple mfgr/distributor/dealer marketing gates who need to have a given product in a given slot. Explaining such 'elegance' to folks marketing to folks buying blister pack calcs for their kids at a drugstore will just elicit a laugh from them.

Remember, calculators are not what they used to be: high-priced, higher-margin items whose cost alone justified careful selection by knowledgable folks - some of whom were using prog calcs to avoid using minicomputer time back then. No real worry about lifespan on a $20 calc either.

Bill Wiese
San Jose


The elegance of a stripped-down RPN calc - even a 20S w/RPN - while eminently logical to the folks here, doesn't pass the the multiple mfgr/distributor/dealer marketing gates who need to have
a given product in a given slot. Explaining such 'elegance' to folks marketing to folks buying blister pack calcs for their kids at a drugstore will just elicit a laugh from them.

Well, somebody is making money selling cheap algebraics for a few dollars at the drugstore, or they wouldn't be available. Selling basic RPN models in the same places at the same prices (under what's left of the "prestige" of the HP brand name) -- or better yet, making them available in large quantities at low cost (or no cost) to schools -- would help create a market for the higher-priced RPN models later. A kid whose first calculator was an HP RPN is more likely to want an expensive RPN model in high school or college than a kid who's grown up with TI, Casio, etc.


It is apparently quite feasible to copy the functionality of classic HP calculators, since Aurora currently markets a cheap RPN clone of the HP-12C, the $25 FN-1000. You can get it from Fry's.

I suspect that Aurora would be more likely to market a cheap RPN scientific than HP.


... if they could remake their Compact Scientific SC150 calculator model with RPN instead of/in addition to Algebraic notation.

It is small, cheap, lightweight, compact and has a nice form factor IMHO


Hi, all;

as I see, many good ideas and suggestions arose, so I decided to add that I thought about a "do it yourself", homebrew custom calc with a downloadable/upgradable (RPN?) system so you could enhance it. I remember I saw a Psion when the HP41 was available. The model I saw at that time (1984 or 1986) was almost like a brick, a big, squared calculator with four ports in the upper side, below the display (a dot-matrix, two line, if I am not wrong). The owner showed me that the ROM/RAM modules could be inserted/removed when the calculator was ON or OFF, no matter at all.

I always kept that image in mind, because it inspired me to think of a custom calculator. That model had no specific characteristic for me, and it looked like it could be in hands of a Physician, or an Engineer, or a Mathematician or any one that might use it. So I considered how good it would be if that calculator could be customized for anyone else's needs.

A cheap, expandable (ROM/RAM) platform with the possibility of downloadable O.S. would satisfy many users. Yeap, I know that the HP49G/G+ are almost like this, but I had something simpler, yet powerfull, in mind.

Just an idea.

Luiz (Brazil)


Hi, Luiz-Claudio,

The "brick" you described is the Psion Organizer II. Five years ago I have obtained this calculator (in fact, it is a highly customisable computer with Basic) from my friend who was to throw it away as a useless machine.

When my wife started her new pediatric office, she needed a tool to help her coworkers count white blood cell types (leucocytes) in her paediatric medical laboratory. I wrote a program in Basic for this machine and it still works prefectly. The lady working in her lab proudly shows this tool to her colleagues from other labs.

The problem arises only when you have to exchange the battery, but has been solved by storing a copy of program in a 16 kB memory module.

Just to illustrate that very old obsolete machines can still find their everyday use.


Hi Nenad;

thanks for your (always thoughtful) words. Your wife's example fits perfectly in!

It reminds me that I have to see the Croatian Calculator's Club (is it the correct name?) page/site. I have "cleaned up" my computer's HD about three times this year, and I lost some precious info, the Croatian Club correct name and related e-address.

Best regards (same to Hrast).


Luiz (Brazil)

Edited: 30 Dec 2004, 7:20 p.m.


What you've described is very similar to what we're working on at OpenRPN. Graphical (dot matrix) LCD, flash, IrDA and sram. What's missing is the four spaces for expansion modules. Instead we've elected to have plenty of memory in the machine and a connector with enough pins to handle general communication with computers, calculators, and extra pins for other applications.

Some extra room will likely be left inside the case for small modules and a few useful traces will be run out to an edge of the PCB for easy access.

We're designing a versatile and open platform. To give you some idea, there will not be individual models. Just form factors. It can be made into any kind of calculator the end user desires. Thanks to laser etching we will be able to offer custom molded key legends. Hackers won't have to spend vast lengths of time reverse engineering their calculators to learn about how they work. We will publish schematics, algorithms, ROM source code, everything.

I've only scratched the surface here and have left out a ton of details. If there are any questions I would be happy to answer them.


Hi, Hugh; thanks for your answer.

After reading your post, I just stopped, breath a little and concluded that I'd have no "global" answer so far. Your description is actually close to what I had in mind.

You see, when OpenRPN began, I saw that I'd have nothing but some spread ideas that you, guys, surely already had. Maybe I know what I'd like to see, but I'm not sure I'd be able to go for final implementation. I developped some programming in both low (assembler) and high "user level" languages, I also designed a few hardware (PC to analog interfaces, independent digital controlers, digital data collectors and other devices) but only as spare, either personal or professional "serving" activities. This sort of "wide range" activity is vital for us, in my country, to experience some professional stability and "presence", but is not a good practice if you want to work on specific design. I have not worked, neither alone nor with a team, in a whole design, and dealing with portables (battery operated) always demands a lot of time in reducing current consumption. Any ideas not exactly related to the main subjects would surely lead outway and cause unnecessary delays. I thought I should wait and, if applicable, suggest something acceptable after the design is ready for testing.

I'm glad you reached the point you mention. For me, the heaviest part would be both developing the floating-point algorithms and joining all routines as a single, stable, self-contained operating system. This is a huge task.

Congrats. I'm looking forward to see the final product.

Best regards.

Luiz (Brazil)


One product that would "fill the niche" is a 10C without its (virtually useless) programming capabilities, which unfortunately persist in the 12C and 12C Platinum.

The 10C ran from 1982-84, when it was discontinued due to poor sales. In this Forum, I have criticized its introduction as a "programmable" $80 alternative to the excellent 11C and 15C, which sold for not much more. However, with a few missing functions (e.g., hyperbolics and delta-%) to replace the crummy programming, and sold at a lower price, it could be a fine entry-level RPN calc that would pass any scholastic or testing-board criteria.

The 10C keyboard, with only one shift key, could provide only a limited number of functions. Removal of programming capability would render unnecessary nine of them (GTO, PSE, R/S, SST, BST, P/R, x<=y, x=0, MEM), and two more (INT, FRAC) would become expendable. That's a lot of space for useful functions. (Why didn't HP do that in the first place?)

Of course, as Wlodek and Bill have stated, the real hurdle would be the production costs. Much of the "body" is still in production, but the firmware and electronic documentation might no longer be available.

Edited: 28 Dec 2004, 11:36 p.m.


As subject says, I'd prefer something in the case of a 32S,

because the ENTER bar on the 10C series is a mess IMHO.

On many of my 10C series machines, the ENTER bar has

to be pressed with much more force than the other keys,

while on the portrait layout machines,

all keys need nearly the same force.

The form factor of the 10C series machines is a bit better,

so they better fit into a shirt pocket.

but what counts in the end is usability, not only portability.

So in this respect the Pioneer series machines are better IMHO.



Karl is probably voting for the 10C because HP still makes the 12c. Hp no longer makes the Pioneer case. But HP makes a 10Bii which could be trasformed into an Hp20s with RPN rather easily.

I like the 10c form and layout a bit better, but lets face it, Hp needs to make the calculator with RPN/algebraic selection to sell to as many potential customers as possible. And that means abandoning the 10c layout. I suspect the 12c platinium's algebraic mode is considered useless to perhaps worse than useless.


Minus the programming or minimal programming. Even if HP decides to spruce up the 9S to make it a two line ALG/RPN calculator would make a good intro calculator.


We had such a calculator in the past, and it was the plain slide rule HP-35. With modern chips, a Voyager shape and dimensions, LCD display, long-life batteries, RPN logic, and the removal of unnecessary programming capabilities, we would go back to the good old days of the HP 35. Hopefully the price will be in the $25-$35 range to attract new users, and start building customer loyalty and product reputation at HP again.



Hi Luis!

Maybe we shouldn't go all the way back to the HP-35!

For the last 15 or 20 years I've used the 11C as my main calculator. I also have a 16C and 48GX but use them less.

A couple of years ago, just for fun, I bought an HP-21 on ebay. (It's roughly equivalent to the 35.) It worked fine but I found it a little too constraining for use on typical light math/science/engineering, mainly from not having enough storage registers.

I then bought an HP-25 and I've loved using it ever since! For the last year I've been using it in undergraduate science and math courses (Statistics, Physical Chemistry and Organic Chemistry) with no problem.

My classmates used algebraic calculators of similar complexity: Casios and various TI-30s and they had no trouble either. Some people tried to use their cell phones but this proved unsatisfactory.

This "from the trenches" experience convinces me that HP-45 or HP-25 or HP-29C level functionality should work well, even at college level. I would suggest a $20 calculator with algebraic/RPN, a 2-line display (would make learning RPN easier) and the same function set as other low-end calcs out there. And non-volatile memory, which is sort of a given these days. TI already does all this, except it use the "wrong" entry system.

Comments, anyone?

(And let me join in to wish everyone a Merry Christmas (or equivalent) and a happy, safe 2005.)

- Michael


Just took a look to the features page of hpmuseum. If keystroke programming is not necessary for this basic RPN calc, the features of the (old) HP27 cover almost every "basic" function you will need. A calc with less functions will hardly be sellable for 25 US$.

If we want to include keystroke programming (as an appetizer), I agree the 11C or even 10C is a good model. Program steps should show the functions instead of key codes, however, creating a user friendly interface.

Just my 0,01 Euros again.


I invite all of you to please stop by OpenRPN if you are intrigued by the contents of this thread. Our current objectives are more complex than what is being discussed here, but I am open to providing enclosures, assistance, and (if enough interest arises) producing entire units.

Best Regards,

Possibly Related Threads...
Thread Author Replies Views Last Post
  HP PRIME - Need help for basic program svp. dg1969 4 1,212 10-08-2013, 04:06 PM
Last Post: dg1969
  Easter Sunday Basic Trigs (HP-12C) Gerson W. Barbosa 29 5,893 04-04-2013, 02:19 PM
Last Post: Gerson W. Barbosa
  "The Basic HP-71" Jeff Davis 3 1,136 08-30-2012, 12:20 AM
Last Post: Howard Owen
  HP's Tired Basic Calculator Line Jeff 4 1,116 02-10-2012, 01:33 PM
Last Post: Dirk Mehldau
  After the 15c LE, better chance of a basic RPN calc from HP now? nick lidakis 80 10,269 09-18-2011, 08:15 AM
Last Post: snaggs
  BASIC BEEP Howard Owen 5 1,090 05-12-2011, 05:03 PM
Last Post: Garth Wilson
  Better Late Than Never--VAL() in HP-71B BASIC Namir 25 4,317 05-08-2011, 11:50 AM
Last Post: Howard Owen
  Mardi Gras Basic Trigs (HP-12C) Gerson W. Barbosa 10 2,131 03-09-2011, 09:49 PM
Last Post: Gerson W. Barbosa
  OT: Sharp PC-1270 BASIC Compiler V3.3 XYZT 1 832 12-18-2010, 05:02 PM
Last Post: Thomas Okken
  Hrast BASIC-4X Discussion HrastProgrammer 6 1,401 12-06-2010, 01:20 AM
Last Post: HrastProgrammer

Forum Jump: