Any love for the 28's?



#2

In the topic on the $600 42S, Jandro Kirkish wrote:

Quote:
...and my favorite, my 28s -- a calc that I have really grown into

I was interested by this because, having lurked here for a long time, I've seen little love expressed for the 28C and 28S.

I know for my own part, my 28's are the least used of my calculators. They were so totally eclipsed by the 48S and X that they fell into disuse. The 15C and 42S were handier to grab for a quick calculation, and for more heavy-duty work the 48's were both more capable and easier to handle. (Nowadays the 50g does most of the hard work.)

But during their brief moment at the top of the heap, '87 - '90, they were indespensible to me. I can still remember how amazed I was by their algebraic capability. The unit conversion capability, while awkward compared to the 48 implementation, was extremely useful at the time. The solver let you store an eqation and then solve for any of the variables, a most impressive feature then even if we take it for granted now.

I bought my 28C several months before taking the FE exam and there were quite a few questions where I was grateful to have it; I could solve them almost instantly. I still have a wooden stand designed to hold it in the fully open position which I bought from EduCalc. (I didn't really get the hang of RPL programming until I got the 48, and still prefer the old keystroke method.)

But at the time, the 28's offered an amazing amount of capability that no other handheld I'm aware of could compare to. Does anybody else have any kind words for the 28s?


#3

I remember taking a Math/Computer Science programming class during my Master's Degree in '88 or '89. One of the course projects was to create a program that would optimally solve a fairly nasty boundary value ODE problem. Since I had no computer access at the time I asked the prof if I could use my 28S instead. He said if I could provide printout of the program and results I could try, but he was more than positive it wasn't possible (I also had, and still have, the printer). After several days of programming and an all-nighter until 5:00AM I produced a program and output that rivaled the computer programs of the day. I still have that calculator (although the the thermal printout has faded), but must admit I feel REALLY clumsy using it today. I can't remember where 90% of the functions are. I should pull it out of the collection this summer and give it another go for posterity.

CHUCK

#4

The 28S is one of my favorites. Although the 48 is more powerful, I like the user interface of the 28S better. The dedicated alpha keyboard has always seemed much better than the alpha shift of the 48. If only they'd had a clamshell with better I/O capabilities...


#5

I second that. I bought the 28C when it came out (at the time I had a friend working at HP who made me benefit from his company rebate) - i was an EE student at that time and the 28C helped me get through a number of exams. AFter about a year, its memory was entirely stuffed with programs I had written, and I immediately jumped on the 28S that came out just at the right time. As a student, these were huge investments for me, but well worth it. Today the 28C is my office calculator (my other HP calcs are too valuable).
I bought the 48 later, but I never came to like it. The display is of lesser quality, but what is worse for me, I don't like the looks: It's ridiculously big and almost intimidating with all these buttons. The 28 has a certain elegance about it. Of course, I regret the missing input capability, and became a victim of it once when I lost 14k of valuable programs, but hey, looks matter too!

#6

Quote:
Does anybody else have any kind words for the 28s?

As Eric said, the dedicated alpha keyboard is a lot easier to use than the shifted alpha keys of the HP-49/49/50.

Quote:
The unit conversion capability, while awkward compared to the 48 implementation

For conversion of simple units (such as length), I agree. For compound units, the HP-28's capability is very useful, particularly when Imperial<->SI engineering conversions are needed. I also like the HP-28's units library with definitions.

Quote:
I bought my 28C several months before taking the FE exam

I found my HP-15C perfectly adequate for the EIT/FE exam in 1994.



Marias said,

Quote:
i was an EE student at that time and the 28C helped me get through a number of exams. AFter about a year, its memory was entirely stuffed with programs I had written,

I managed to fill the HP-28C's puny 2 kB of RAM just trying out solutions in symbolic algebra -- completely inadequate for its capabilities. I wish that the HP-32S/SII had that 2 kB, though...

Quote:
I regret the missing input capability, and became a victim of it once when I lost 14k of valuable programs

A few years ago, I made the argument that HP was wise not to release an HP-42S with the 32 kB of RAM it could support (instead of its built-in 7 kB), unless I/O were added. If users are given much RAM, they will use the calc as a repository of programs or as a data collector, then will inevitably get burned without electronic upload/download.

-- KS


#7

Karl wrote:

Quote:
If users are given much RAM, they will use the calc as a repository of programs or as a data collector, then will inevitably get burned without electronic upload/download.

I agree in all senses (and this started to thrill my brain since the HP-33S apparition).

-- Antonio

#8

I, too, have been attracted to the 28s because of the dedicated alpha and less-cluttered keyboards. My main interest is in units conversion. I even went so far as to purchase some period HP literature (a 22-page 8.5x11 booklet) that compares the 28s and 48sx. Unfortunately, it just does not answer my questions!

Quote:
The unit conversion capability, while awkward compared to the 48 implementation

For conversion of simple units (such as length), I agree. For compound units, the HP-28's capability is very useful, particularly when Imperial<->SI engineering conversions are needed. I also like the HP-28's units library with definitions.


Can someone elaborate on this?

Is the 28s units conversion capability equal to the 48sx (which I have)?

Can it append the unit to the numerical value like the 48sx?

Thanks.


#9

I have both a HP-28S and a HP-48SX, and IMHO the HP-48SX wins hands down over the HP-28S in the units department. The HP-48SX can create units objects, which associate (append) the unit type to the numerical value. You can perform arithmetic operations on two values, you can even combine two objects with different units, and it will automatically convert them to a common unit and perform the arithmetic operation, for example: 100 cm^2 ENTER 1 m^2 + yields 1.01 m^2. The HP-28S can't do any of this. Also units conversion is much simpler on the HP-48SX. Simply enter the value with its initial unit, then hit Left Shift and the final unit, and voila the conversion is made. The units are grouped by type on the HP-48SX (length, force, temperature etc), making them much easier to find than the alphabetized list on the HP-28S. Bottom line, the HP-48SX can do anything with units that the HP-28S can do, it can do more and it can do it better.

Edited: 16 Feb 2009, 10:51 p.m.


#10

As Mike said, there is no comparison between the 28 and 48/49/50 units utility.

On the 28, you would enter the number you want to convert on one level, enter the "from" units as a string on the next level, then enter the "to" units string on the final level, and then hit "CONVERT." As such:

3: 25

2: "ft * lbf / s"

1: "Btu / h"

You'd be left with:

2: 115.656071711

1: "Btu / h"

You couldn't incorporate the units into the solver then enter mixed units for the variables and have the calculator automatically check for dimensional consistency and convert units.

At the time, it was very useful (it sure beat looking up all the unit conversions and doing them one at a time). But it wasn't a breakthrough; the HP 41 Thermal and Transport Science module added this capability to the 41, and unit checking/conversion was built into all the programs on it as well. If you opted for "Units" when running the program, then after prompting for the value, it would prompt for the units which you would enter as an alpha string. It gave instructions for incorporating this capability into your own programs. As a chemical engineering major, this ROM was worth more than its weight in gold to me.


#11

Thanks Michael and Steve, this is the answer I needed.

#12

Actually, the HP-28 series was revolutionary in many respects, not just RPL. They were the first HP's (to the best of my knowledge) to replace the fixed 4-level stack (x,y,z,t) with an unlimited stack. This was important when performing RPN calculations with more than three pending (parenthesis for algebraic notation) operations. It also introduced contextual and definable soft keys. I particularly like the simple layout with only one shifted key.

But my HP-28S also has unique problems due its design. When fully opened to use its alphabetic keyboard, it has a large footprint and ceases to be a hand-held device. I find it awkward to keep shifting my eyes (and head) from the alpha keyboard on the lower left to the display on the upper right. Unless the surface it is placed upon is completely flat, the alpha keyboard (which has no rubber feet) tends rattle around as I press the keys. And finally, the lack of calculator to PC I/O is a real killer, especially with the 32K HP-32S, which is why I later bought a HP-48SX. Not only do I save my HP programs and data on my DOS PC, I find it easier to write/edit my programs using a text editor on my PC and then port them to my HP.

I had been using my HP-28S occasionally until a few years ago when the battery door broke (another design flaw of the HP-28). So now it just sits in its pretty accessory soft leather case, sans batteries.

Edited: 16 Feb 2009, 1:23 p.m.

#13

Wow it's nice to have "inspired" a thread on one of my favorite topics-the 28s! This is really the only vintage hp I own but the longer I have it the more impressed with it I seem to get. I bought my 28s back in 1990 because I had come into some cash and had been lusting after it for years and was impressed by my friends 11c and RPN in general. I was in my early 20's and going through kind of a "loser" phase: college dropout etc (perhaps some of you fellow Gen-X-ers can relate). I hadn't even taken a calculus class yet. I loved the 28s from day one, but really didn't have the math background to fully appreciate it, save for solving and graphing equations. Long story short, I was back in school around ten years later getting a math degree ... It was only then that I fully appreciated that d/dx key--wow, my hp can do that? (It was around this time that I discovered this Museum, and my obsession with hp calcs blossomed) I saw my fellow math & engineering friends with their 48g's and I thought they were pretty cool but I felt I had the advantage with the dedicated alpha keyboard, just one shift key and a remarkably uncluttered keyboard for such a powerful machine. I got my math degree but physics and engineering are my real interests, plus my amateur astronomy hobby, and the hp 28s has served me well in all of these dabblings. (And in my day job too as a scientific programmer.) So 20 years later I find myself saying less often "Wow my 28s can do that?" and more often "Of course my 28s does that!" Recently I have found the ability to define custom units is very handy in astrophysics as but one example ...

I agree with the praise as well as the criticisms raised by the other posters: I just wrestled with the battery door last night putting in new batts, the thing creaks and crackles like it going to fall apart--and it did once; thanks superglue!--but it has always been that way. And its still working. Many of these flaws are related to the clamshell design, which is paradoxically one of its strengths, as other commenters have noted - and if you can put it on a desk, it makes it a lot easier to key in programs and expressions. I am constantly hunting for functions on my 50g where most buttons have three or more function apiece, but on the 28s it is a snap. Also it fits nicely in a jacket pocket (it is in my pocket now as I write) whereas a larger RPL model would not. So Steve, you ask if there is any love for the 28s--well clearly there is plenty from my corner!

Cheers,
Jandro

#14

I like the 28 series too. Got my first at the end of my undergraduate degree and spent the vacation writing chess for it.

The dedicated keyboard wins it for me. When I enter programs or even call functions on either the 28, 48 or 49, I do so in alpha mode. I never remember where the function I want is located in the menus but I usually remember the name.


- Pauli

#15

Being a long time HP-41 addict (incurable since 27 years), some time ago I bought a mint HP-28S for my collection. It was love at first sight! What a machine! I also own a HP-48G. Comparing both units the 28S is miles ahead. Great user interface, well-arranged keyboard, excellent manuals (I own both HP-28S and HP-48G manuals), everything one expects from a good calculator, it allows almost an intuitional handling, everything on it gives me a feeling of quality . Now I'm looking for a second one for work, since I don't wanna wear the one from my collection.
For today's calculator usage its memory is sufficient, the HP-35S I recently bought doesn't offer more. I'm doing more complicated calculations via PC (Octave, Scilab, Mathematica, Excel).
In my opinion this calculator still represents everything HP stood for in the "good old days". Yes, I think it comes close to my all-time favorite HP-41.
Hopefully my enthusiastic posting doesn't start a flaming war about the best calculator, it is purely my two cents.

Rainer

P.S.: I found a HP-42S for 15€ on a flea market, with manual, two books, pouch and box, 3 months ago. It looked like new, after replacing the batteries it works. IMHO paying more than 600 bucks for such a machine is crazy ...


#16

German flea markets trading HPs? You're a lucky guy, aren't you?


#17

Marcus,

I have many of my collection items from flea markets, but I admit, it's a time consuming hunt occupying many weekends. I found the 42S among a bunch of computer crap, the seller himself announced his sales as "Computerschrott" (=computer scrap). When I took a look at his offerings I was amazed seeing the HP-42S. He was convinced it would be defect, so I dealt for the books and the supposed to be dead calc as a give-away, and it worked :-). I was absolutely happy about ...

Unfortunately opportunities like this one are very rare.

IMHO the "Bay of Pigs" - or how you ever want to name it - became one of the most expensive ways to complete a collection of HPs, bargains are rare, though sometimes possible. I ask myself who is weird enough buying heavily worn Voyagers for almost 300$, or hp-il peripherals for irrational sums from the same seller.

OK, this was a little bit off topic, sorry.

So I'll close with a "Horrido!" to all hunters of vintage HP calcs,
and beware of being ripped off.

Kind regards,

Rainer


#18

According to my age-predictive calculator model, Nearly everyone who hails the HP 28C/S was born from 1970 to 1974- a very narrow range of years. Any earlier and people tend to migrate to the Voyager, any later and they would migrate to the Pioneer or Charlemagne series.

I have both the 28s and 28c. While I find them adequate, the battery door and leaky N-cell batteries are enough to prevent their employment for all but special use, and for the size, I'd rather have a slightly smaller Pioneer (32s,32sii,42s) in an Astech Case, or a 48g series calculator.

I did not fully appreciate the RPL legacy of the 28 series until I read Chapter 1.1 of William Wickes' Book: "HP 48 insights". After learning the 3rd generation calculators started with the 28S series, I decided to add all the clamshells to an otherwise complete collection of Nut, Voyager, Pioneer, and Charlemagne Calculators. The former see the least use of any other family, for stated reasons.


#19

I'm not in the 1970 - 1974 birth date range :-)

My first HP was a 34c which died.

- Pauli


#20

you're wrong. I was born in 1965 and yet hail the 28C/S. On the other hand, I already had an HP16C when i bought the 28C.

#21

Allen, I wish I'd conform to your age-predictive calculator model, but I was born in 1962. So I guess your predictor algorithm is running on a TI :-).

#22

Quote:
I did not fully appreciate the RPL legacy of the 28 series until I read Chapter 1.1 of William Wickes' Book: "HP 48 insights". After learning the 3rd generation calculators started with the 28S series,

I was planning on quoting the same information. The HP-28C was the first calculator that could use equations as arguments (Wickes said this much more elegantly that I did). This (and other features) help justify its place in history as the first 3rd generation calculator.

#23

Could somebody shed some light on the fact and reasons, that the representation of calculating a numerical integral (and possibly other function implementations & stack diagrams too) have changed in the course of development of the newer calculators:

As an example (28 Pocket Book)
------------------------------

3: 'X^3+4*X+8'
2: {X 1 3}
1: .00001

2: 52
1: 5.1992166102E-4

If entered on my HP50G - "Integral: Too Few Arguments"

Because that might have been discussed "ad nauseam" in the past, pls. point me to the relevant postings then.

Best regards,

Peter A. Gebhardt


#24

Peter,

I do not own a HP-50g, however, I do own a HP-48SX, which I believe works the same way, and in order to perform integration you need 4 stack entries as follows:

Level 4 - Lower Limit (1)

Level 3 - Upper Limit (3)

Level 2 - Integrand ('X^3+4*X+8')

Level 1 - Variable of integration ('X')

The accuracy is specified by the display mode, i.e. in your case it would be FIX 5. To do numerical integration, as opposed to symbolic integration, you must set Numerical Results Mode (set flag -3). Also, on the HP-48SX it is much easier to do this using the EquationWriter application and I assume the HP-50g has something similar.

Hope this helps.

Michael

Edited: 20 Feb 2009, 3:55 p.m. after one or more responses were posted


#25

Michael,

Thx. for your response.

Exactly the "reversed" usage of the stack in the newer models raised my question, why the change took place (in your case with the 48SX).

Best regards,

Peter A. Gebhardt

#26

Peter A. Gebhardt stated,

Quote:
As an example (28 Pocket Book)
------------------------------
3: 'X^3+4*X+8'
2: {X 1 3}
1: .00001

2: 52
1: 5.1992166102E-4

If entered on my HP50G - "Integral: Too Few Arguments"

Because that might have been discussed "ad nauseam" in the past, pls. point me to the relevant postings then.


Michael de Estrada responded:

Quote:
I do not own a HP-50g, however, I do own a HP-48SX, which I believe works the same way, and in order to perform integration you need 4 stack entries as follows:

Level 4 - Lower Limit (1)

Level 3 - Upper Limit (3)

Level 2 - Integrand ('X^3+4*X+8')

Level 1 - Variable of integration ('X')

The accuracy is specified by the display mode, i.e. in your case it
would be FIX 5.


Yes indeed, the stack arguments for integration in the HP-28 differ from those in the HP-48/49/50.

I recall mentioning that in a thread from 2004 you might find interesting:

Quote:
"Arnaud's (HP-48) program can't even be run on a 28C because its integration accepts arguments in a different format (dummy variable and limits enclosed in a list), which I don't know how to automate. It also didn't accept value-loaded variable names for the limits."

It should be noted that the integrand-function uncertainty provided in stack level 1 in the HP-28 does not specify the same thing as the ACC parameter in the HP-42S and HP-71B Math ROM.

An excerpt from my post of 28 Nov 2005:

Quote:
28C, 28S:

(These models) specify uncertainty in a particular decimal digit as a numerical parameter to the INTEG function.

42S, 71B Math ROM:

(These models) specify a relative uncertainty as a per-unit fraction of the function-value magnitude, as a numerical parameter to the INTEG function.


Also, from a thread in 2006:

FIX for integrand-function uncertainty in the HP-48/49/50 works as in the HP-33s/35s, not as in the HP-15C, HP-34C, HP-32S/32SII, and HP-41 Advantage.

-- KS


Edited: 21 Feb 2009, 11:31 p.m.


Possibly Related Threads...
Thread Author Replies Views Last Post
  TED Talk: Adam Spencer: Why I fell in love with monster prime numbers Les Bell 3 588 09-05-2013, 12:54 PM
Last Post: Ken Shaw
  One thing I really love on HP PRIME... dg1969 0 277 07-31-2013, 06:50 AM
Last Post: dg1969
  Best Keyboard of the HP 28/48/49/50 Series Eddie W. Shore 9 947 07-12-2012, 07:21 AM
Last Post: Dave Britten
  HP-28--better design? Matt Agajanian 7 694 06-27-2012, 09:27 AM
Last Post: David Tellet
  Formatting 28/48/50 programs Matt Agajanian 9 899 06-06-2012, 06:01 PM
Last Post: Gilles Carpentier
  Matrix/Array 28/48/50 Question Matt Agajanian 9 750 06-04-2012, 10:31 PM
Last Post: Matt Agajanian
  RPN Simuator for 28/48 Matt Agajanian 8 682 06-01-2012, 07:28 PM
Last Post: Les Wright
  28/48--In perspective please Matt Agajanian 12 1,027 05-18-2012, 10:11 PM
Last Post: Dan B
  How did the 28 work? Matt Agajanian 14 1,326 03-09-2012, 08:01 AM
Last Post: David Hayden
  Nothing says I Love You like a 15C LE! Todd Garabedian 2 344 09-26-2011, 01:06 PM
Last Post: Artur-Brazil

Forum Jump: