HP-55 had potential


Hello all.

I'm trying to put the HP-55 in perspective. As a preprogrammed calculator it has quite a lot going for it. But, as a 50 step programmable calculator, there's much to be desired. So, my question (or questions): what was the goal/objective for this calculator's limited programming capabilities and unmerged (with some exceptions) keystrokes?


Simple. Offer an affordable basic programmable calculator as an alternative to the very expensive HP-65. Since it lacked continuous memory or any external storage, there was little point in having more than 50 steps in program memory. However, it is quite unique in having a built-in timer mode, any many laboratories used it to time experiments and tests.


The HP-55 at $395 was half the price of the HP-65. Of course for the price difference, it had somewhat less functionality than the HP-65 and no permanent program storage. The HP-25 on the other hand was introduced about 8 months after the HP-55 at half it's price ($195). The HP-25 and HP-55 were much more comparable to each other so the HP-25 looked like a real bargain at the time. Even though $195 was a lot of money in 1975 (for me anyway), it was much more affordable than $395 or $795.


It depends on what capabilities were important to you. For example, the HP-55 offered linear regression, whereas the HP-25 did not.


Very true. Compared to the HP-25, the HP-55 had the following additional features: A Timer, more display digits, 12 more storage registers, linear regression, Factorial, H.MS +/- math, Y^2 Summation, and 16 built-in conversions.  Compared to the HP-55, the HP-25 had the following additional features: 6 additional conditional tests, Engineering notation, fully merged program steps, storage register arithmetic, INT, FRAC, PAUSE, NOP and was smaller and lighter.  All in all, the HP-25 was roughly in the same league as the HP-55 (unless you needed the timer) at 1/2 the price.


Sounds reasonable. Interesting.


If you read the entries in the museum for the 65 and the 55, you will see that the former, introduced in 1974, was $795!!! That is a lot of money!! At that time, a new car was between what, $3000 and $10000<

The 55 was about half the cost.


At the time of the HP-65 introduction, it would have cost me one month's take home salary after paying taxes. I had to wait until the introduction of the HP-34C, before I could afford a programmable calculator.

Edited: 31 Mar 2012, 7:11 p.m.


At the time of the HP-65 introduction, it would have cost me one month's take home salary after paying taxes.

You were doing well. In 1974, as a new US Navy Ensign with a new Ga. Tech EE degree, I made $534 per month before taxes. There wasn't going to be an $800 HP-65 in my possession ever, or even a $400 HP-55! My first scientific was a SR-50. My first programmable was a TI-56 (a pretty nice machine for its era). All of these were much less costly than the HP-65, -55, and even the -25. The 1977 era brought the HP-67 (only $450), which was really an excellent buy when compared to anything programmable that HP had put out earlier. The worthy but butt-ugly competition was the $300 TI-59.

I don't know of anyone who seriously considered buying a HP-55 after the HP-25 came out. It was way too much money for way too little performance. That could also have been said of the HP-65 as well, at any time during its market life ($800 in 1973 is equivalent to $4100 in 2012, and that's just for the basic calculator).

Edited: 1 Apr 2012, 5:59 p.m.



I have an HP55 and thanks to J. Ernesto I could grab both the application manuals written for it back then. You know, I saw some original programming techniques - after building the related program algorithms - that actually amused me. At that time they had to twist and shrink an algorithm in such a way its functionality and versatility would make them desirable to have - and also the equipment to run it. Take, for example, both Numerical Integration programs - pp. 106 to 109, HP55 Mathematics Programs, Rev. B, Feb/75. The "updating-data-while-computing" technique (efficient handling of registers contents) saves program steps and enhances efficiency. One pays the price of not being able to repeat a particular step in the process, i.e., we need to take note of the results from each iteration or else the whole calculation must be started over. And that's what could be done at that time with such resources.

The programming characteristic that teases me the most is the 'jump-if-not-true' embedded address feature in both HP55 tests, x<=y and x=y. I am not sure if the HP65 also does it, but the HP55 is the only calculator I know that has an embedded jump address in a test function. I mean, the tests for x<=y and x=y have the program step to jump to already stored in their own code, as it happens with the GTO instruction code itself - a single '-' preceding the step address. It took me by surprise when I red the HP55 manual for the first time, back in 1999.

Edited: 31 Mar 2012, 7:55 p.m.


Hello Luiz,

I would have loved the HP-55! Back then ('82) I had a TI-51-III (same as TI-55). I had just learned Calculus (or at least I thought I had :-) and knew no numerical methods at all, but I tried the very limited numerical integration program below. Those were the days!



00 RCL			16 2
01 1 17 RCL
02 * 18 3
03 RCL 19 +
04 2 20 1
05 f(x) ;(*) 21 =
06 = 22 STO
07 SUM 23 3
08 0 24 +/-
09 RCL 25 +
10 2 26 RCL
11 + 27 4
12 RCL 28 =
13 1 29 1/x
14 = 30 RST
15 STO

(*) f(x) is any of the following built-in functions:
sin, cos, tan, sinh, cosh, tanh, log, ln(x), e^x, 10^x, sqrt(x) or 1/x.


n -> STO 4
(b-a)/n -> STO 1
(RCL 1)/2 + a -> STO 2
0 -> STO 0 STO 3

2nd RST 2nd R/S
when the program stops on error,
press CLR RCL 0 to get the result


Integrate f(x) = sin(x) from a = 0 rd to b = 1 rd, using n = 50 intervals, should return 0.4597054.
Actual result is 0.4596977 (1 - cos(1)).

Edited: 31 Mar 2012, 9:21 p.m.


Hi, Gerson;

I still have my TI57: 50 prgm. steps, LBL, GTO and SBR(!), compact program steps with merged codes, something like the HP25. Eg: [2nd][Prd][0] would occupy one prgm step. Mine is still working, no working batteries though. It is almost a reduced TI58, although the TI58/59 had very restricted merging capabilities in program coding.

You are correct: those were the days. I would not like 'the days' themselves to be brought back, I'd rather like to see/meet more people with the same minds (the good ones!) we used to find back then.


Luiz (Brazil)


The limited step merging capabilities of the TI-58/59 was more then offset by the 'huge' (in contemporary terms) memory and the large number of possible labels.



In 1975 my parents approved of my request to get a programmable HP calculator. There was the HP-65 with its magnetic card reader, and there was the HP-55 and the new HP-25C. After looking at cost, built-in features, memory registers, I went with the HP-55. It had a lot of features that were built in and therefore were instantly available. The timer was cool too and 20 memory register were nice.

What about the programming. Well I pushed it to the limit. I did use the programming steps to perform linearized regression, where the program steps would transform the values I entered before adding them to the summation registers. I was able to evaluate several series and polynomials.

When the HP-67 came out, I realized that this new baby brought more power with its card reader--read and write both programs and data. I did manage to buy one (despite my father's objections ... long story)


PS: It's nice to see two HP-55 emulators for the iPod/iPhone/iPad.

Edited: 31 Mar 2012, 8:33 p.m.


I have an HP-55 and for an everyday calculator, it was a nice machine: simple keyboard with common conversions, and I love the timer.

In fact, the HP-55 is the reason I implemented a timer in the WP-34s so I would be able to replace it (using it daily was to much for such an old machine).


I'm finding in my usage, the 55 has an everyday 'go-to' factor when I want an RPN methodology to solve manual calculations. In terms of programming though, I've got to think twice how to optimise my coding for a routine since the unmerged keystrokes takes away from program memory. But, it does provide a challenge.

If I start getting a Chico Marx haddock, then, I'll grab my 67, 41CV/42S or 32S-II.

Edited: 5 Apr 2012, 12:39 a.m.

Possibly Related Threads…
Thread Author Replies Views Last Post
  HP 55 Olivier De Smet 8 2,693 08-26-2013, 04:54 PM
Last Post: Matt Agajanian
  HP-25/25C Potential Matt Agajanian 14 4,200 07-17-2013, 02:57 AM
Last Post: Matt Agajanian
  Update of Emu48 to v1.55 Christoph Giesselink 4 1,938 05-26-2013, 05:28 PM
Last Post: Andrew Nikitin
  Celebrating the HP-55 my first HP Calculator Namir 6 1,982 08-25-2012, 02:00 PM
Last Post: Thomas Radtke
  OT--APF Mark 55 Question Matt Agajanian 6 2,384 07-19-2012, 10:51 PM
Last Post: Matt Agajanian
  Restoring an HP-55, advice appreciated ! Alberto Fenini 4 1,686 05-15-2012, 03:43 AM
Last Post: Alberto Fenini
  11C's potential Matt Agajanian 7 2,495 05-14-2012, 08:26 PM
Last Post: Matt Agajanian
  Got my 55 today Matt Agajanian 0 909 04-02-2012, 05:20 PM
Last Post: Matt Agajanian
  55 already registered for HHC 2011 in San Diego! Sign up today Gene Wright 7 2,403 07-26-2011, 02:04 PM
Last Post: Jim Yohe
  HP-55 odd behavior revisted Martin Pinckney 0 823 06-13-2011, 10:09 PM
Last Post: Martin Pinckney

Forum Jump: