What is wrong with the 28C/S?


Given how much the 42S and 32Sii have been selling for on E-bay, I am surprised that the 28C/S seem to be selling for relatively low prices. According to Holger Weihe's data set, these calculators have averaged $40/copy. There are some current auctions in the same range for what look like reasonable quality calculators. These calculators retailed for $235 (per the museum), almost twice as much as the 42S and much more than the 32S.

I have never used one of these before - when they were selling new I was put off by the hinge (seemed like a reliability problem and difficult to hold in one hand) - but now it seems like this may be a good time to add one to a collection.

PS: Thanks for keeping the ebay price history.


I've wondered the same thing. It's only slightly larger than the 32sii/42s, but has a larger display, and more functions and memory. It's well documented and a great way to learn user rpl before using the 48/49. On the downside, some of the 28s units have battery cover damage (what was HP thinking when they designed the 28/19B battery cover?).

If you can find a good one (I've found 3 "mints" for $60 or less) they are a keeper!



The battery cover weakness is legendary. Another is the overall case construction. I gave one to my son, who usually kept it in his too-full backpack. It was apparently flexed constantly, and most of the little plastic pins holding the thing together broke. It's now in a few dozen pieces in a ziploc bag.

But otherwise a fine device, lacking only an input port. (If I remember correctly, some folks found a way to hook up an RS-232 connector.) It's probably now the least expensive of the HP RPN calculators.


Umm, it's not a 41C/CV/CX or 42S :)

I'm not a fan of the 28/48, etc that don't have std 4-level RPN stack, etc. They offer more features than I can use, and if I had a problem severe enough (even back in late 80s) I'd use a PC with fullsized keyboard etc. and some math software.

The 42S offered some real improvements to the 41. Unfortunately they got rid of nice keyboard, no I/O for upload/download, printer, etc.

Waiting for my dream successor to the 42S:
- original 4-level RPN;
- 4 or 5 line x 20 char (approx) alpha LCD;
- has slot for SmartMedia flash memory card;
- small database mgr to manage data on flash;
- orig HP keyboard/keytops (like 11C/25C/41C etc)
- non-staked keyboard for ease of cleaning & repair
- bidirectional IR/RS232 port
- USB port so device acts as a generic bulk USB device

Hell, I wouldn't even mind if this thing had a form factor like a 41C or HP25C! Feels good in my hand.

Bill Wiese
San Jose, CA


For me, the 28's were a major improvement over what came before, especially the 28S (enough memory for subdirectories, lots of object types to use, this wonderful stack that I no longer have to worry about something falling off the top). But there was this battery cover, the hinge, and gee, if only the display were a bit bigger.

So when the 48's came out, they had everything the 28's had, and everything they didn't, and the contest was over.

It's abundantly clear from postings of "wish lists" that there are two camps here, both respectful, but both very positive. If you post here, you almost certainly like either RPN or RPL, and probably not both. The 28's don't do RPN, the 48's do RPL so much better, and thus you have 28-as-collector's-item, not 28-as-favorite-calculator.


IMHO, the demand for HP32Sii and HP42S is because these are 4-level RPN, simple programmable models. The 28C/S is the first RPL model, and for many people, me included, the operational and programming paradigms derived from the fundamentalistic reverse polish (1 10 FOR syntax), and the "infinite" stack (and things like nagging error messages when you just tried to clear a now-inexistent x register) are VERY unfriendly.

Oh, the simplicity of a "normal" CLx, and the NULL behavior of the HP41...

For those inclined to RPL, etc. (which I don't like), the 48 series is still available, with many newer features than the 28 series. And the hinge and battery door problems are gone, and the I/O is bidirectional...


What they said.

Yes, all valid points and good insight. IMHO, the 28 series represented a shift in thinking away from the classic 4 level stack RPN to RPL. If a 28 series machine was the first HP calculator you ever held in your hands and learned to use, it was great, which made the 48 series even better for you.

But, if you where born in the 40's or 50's and spent years waiting until you could afford an HP and your first machine was a classic, a Woodstock or a 41, the 28 series seemed foreign. Quick, open your 28S and give me the sine of 49.5, without having first reading the manual to find the function on the trig key. It was just, well, different. Too different for me. I didn't need a new calculator, so I just handed it back to the salesman and said "Thanks, but no thanks". Along with the 48 series machines, they were not a calculator to me, but something I didn't want or need to spend the the time to learn. I had 4 level RPN and I was VERY happy. Call me an vacuum tube guy in the solid state world, but I just didn't like it. I had a computer for that stuff. If I wanted to add 100 columns of sales commissions, give me 1-2-3. But , when I wanted to find out how fast to spin a shaft on the plant floor somewhere, give me my 42S any 'ol day.

I think it important to note that the timing of the 28 series also caught HP with their proverbial pants down in the PC market. The paradigm shift that occurred in the mid 80's through the early 90's was so huge, it caught just about every major engineering based company flat-footed. HP was no exception, at least in the PC area. They where always 6 to 9 months behind everyone else, but anything they put into the marketplace always had the proper HP feel and quality. I think that the mindset of the times led to creeping feature-ietus which crossed over to the calculator product line. After all, didn't technology demand new designs every 18 months? Somewhere along the way, the line between computer, calculator and hand-held got very blurry. I can't find an ounce of fault in what they did, or in their products. But somewhere in there is a lesson. The paradigm for a hand-held calculating device has not changed. You cannot ignore the fact that we still have the same ten fingers and two eyes with which to operate the device. Adding more keys with menus only hides the basic functions we use 98% of the time. Why make it harder to do the basics?

I still think the best thing that could ever happen to the calculator product line is for it to be given/sold to the Agilent group. I often imagine the "real" HP engineers with HP-41CX's on their desks wondering to themselves what will they do when their trusty 41 dies. Go buy another on eBay?


My opinion is slightly different, perhaps because I was born in 1966 :-)
When I want a sophisticated tool, I love the 48 with MK and CAS (Erable-Alg48). Along years it has been totally customized to fit exactly to my needs. I am surprised almost everyday by its power.
But when I want "only" a calculator (a very good one) or a shirt-pocket calculator, I use my 15C or my 32SII with their programs and solvers.
They both systems are the best (in my opinion):really I can't say if I like more either RPN or RPL Only that I prefer them instead algebraic system.
Happy new year to you all.


The first HP calculator I owned was a 28S. I liked it, wrote a few programs for it, and it helped me in college. It was an interesting machine. But as a calculator, I didn't find it as well thought-out as my previous calculator, a TI-66 (which I had in high school; in retrospect, it was maybe the best calculator TI ever made). The menu system is not a bad idea, but having to navigate through menus just to compute a sine or log is just *wrong*.

A few years later, I got an HP 32SII and lived happily ever after. Well, almost.


>The menu system is not a bad idea, but having to navigate through menus just to compute a sine or log is just *wrong*.

Actually, you don't _have_ to use menus to compute anything, since you can always type into the command line and press ENTER.

But you're right in one thing -- RPL machines are so full of seldom- or never-needed functions, that remembering how to do stuff is a major undertaking... particularly for people like me, who can't remember what they had for dinner yesterday. 8^(



I think we can all appreciate that "hiding" the trig functions in a menu was a way to reduce the number of keys on the non-alpha side of the keyboard to an almost-manageable number.

Isn't there a "Custom" command (or something like that -- it's been too long) whereby one may create a user-defined menu structure? If I remember right, I was able to bring all of the functions that I used routinely out to a custom menu, and so make access to them relatively convenient . . .

Comparing the layouts of the 28 C/S and 48G models presents a nice case study in some of the difficulties and trade-offs involved in designing feature-packed alphanumeric calculator hardware.


Hi Paul,

Yes, there's a CUSTOM feature. How you use it (I *think*, it's been a long time for me too) is to give the MENU command either a number or a list:

1 MENU ---> takes you to the ARRAY menu, I think
23 MENU ---> I'm pretty sure this is the USER menu
{ SIN COS TAN } MENU ---> but this one "loads" CUSTOM

and until a "reload" (clear, reset, whatever), when you go to CUSTOM, you'll then see the softkeys as:

[ SIN ] [ COS ] [ TAN ] [ ] [ ] [ ]

If you use { SIN { } COS { } TAN { } } MENU, you'll then get

[ SIN ] [ ] [ COS ] [ ] [ TAN ] [ ]

but I don't think that the other "tricks" of the 48 series (define a shifted function, give a different button label, _etc._) are available.


All good points. The 32sii/42s have a fair amount of computing power in a true shirt pocket size. The 28/48/49 have a lot more power, but increase the form factor. For me, the 28s' size to performance ratio still gives it a sweet spot between the 42s and a PC. I've written several small programs for the 28s that would be more difficult/impossible on the 42s, and a lot more portable than lugging a notebook PC.

I think this discussion has really got to the heart of HP's business decisions. The "high-end" calculator market has been made obsolete by the PC/notebook. The very low end can be solved with $15 plastic calculators. The few of us caught in the middle don't really matter to a multi-billion dollar corporation's bottom line. I must be getting old...


Possibly Related Threads…
Thread Author Replies Views Last Post
  [HP-Prime - Geometry App] What is wrong with this ? Jean-Michel 2 1,476 10-27-2013, 06:26 PM
Last Post: Jean-Michel
  HP Prime: What's wrong with this program? BruceTTT 6 1,983 10-16-2013, 11:14 AM
Last Post: Eddie W. Shore
  Correct me if I am wrong Denis Doyon 51 11,489 08-14-2013, 02:53 PM
Last Post: Carey
  HP-41C Transition to HP-28C book? Dan Grelinger 6 2,151 03-30-2013, 10:32 AM
Last Post: David Hayden
  [WP34s] What am I doing wrong? Dieter 10 2,751 11-18-2012, 04:52 PM
Last Post: Dieter
  I really like the HP-28C keyboard Eddie W. Shore 13 3,887 06-28-2012, 04:31 PM
Last Post: Matt Agajanian
  2K==A lot of place for (HP-28C) stuff Matt Agajanian 4 1,720 05-21-2012, 09:59 PM
Last Post: Dominic Richens
  Wrong answer when evaluating log(10^-12) Daniel Greenidge 4 1,670 01-26-2012, 05:54 PM
Last Post: Daniel Greenidge
  HP 28C - horizontal bars across display ? Kao_Tai 0 753 09-01-2010, 04:54 PM
Last Post: Kao_Tai
  HP-28C ROM checksum addresses? Christoph Giesselink 0 755 07-25-2010, 08:25 PM
Last Post: Christoph Giesselink

Forum Jump: