The Museum of HP Calculators

HP Forum Archive 15

 HP 27sMessage #1 Posted by ken on 1 Sept 2005, 11:26 a.m. well these things appear to be a dying breed. Is there another calculator which could solve like this one. I would first create a formula using the calculator interface and it would be stored indefinitely (key feature) P=I*E when I would press calc I could solve for any value (key feature) I could press a number and select variable p soft button and then press value for I and then solve for E just by not typing a value. it would also solve any direction. Place any 2 values and get the third.(key feature Is there a current calculator made by HP or anyone else which has a solver as flexible as the 27s. If so please do tell because I still cant find one.

 Re: HP 27sMessage #2 Posted by Mike H on 1 Sept 2005, 12:33 p.m.,in response to message #1 by ken The 27s uses an algebraic solver. It is very easy to use and quite handy. Other HP models that used the algebraic solver were the 18C, 17B, 28C, 28S, and 19B. The new 17BII+ calculator also has this feature. The 33S also has an algebraic solver that works very well, but it is not as intuitive as the 17B More information about solvers can be found here: Solver Edited: 1 Sept 2005, 1:44 p.m.

 ... and HP-17BII; -22S; -32SIIMessage #3 Posted by Karl Schneider on 1 Sept 2005, 11:02 p.m.,in response to message #2 by Mike H Quote: Other HP models that used the algebraic solver were the 18C, 17B, 28C, 28S, and 19B. The new 17BII+ calculator also has this feature. The discontinued HP-17BII carrid over the HP-17B algebraic solver, and "Let" and "Get" work properly in all respects (unlike in the 17BII+). Quote: The 33S also has an algebraic solver that works very well, but it is not as intuitive as the 17B The HP-33S algebraic solver is the one from the discontinued HP-32SII and HP-22S. The 33S also will perform a "direct solution" (instead of numerically iterated) in some cases. Unfortunately, it will always return the direct solution when possible, no matter what solution range the user provides... -- KS

 Re: ... and HP-17BII; -22S; -32SIIMessage #4 Posted by GE on 2 Sept 2005, 6:22 a.m.,in response to message #3 by Karl Schneider The solvers of the plamtop line can be used too : HP95LX, 100, 200... Maybe these are easier to get nowadays, and you get a nice DOS-capable device with a full keyboard.

 Re: ... and HP-17BII; -22S; -32SIIMessage #5 Posted by Gene on 2 Sept 2005, 7:25 a.m.,in response to message #3 by Karl Schneider I dont' believe that's true. It will return the direct solution if the equation is entered as "X^2-4", but if you put "X^2-0*X-4" as the equation, it will not directly solve it but will allow the user to give estimates that guide what the solver finds.

 33S solver: We've discussed this before.Message #6 Posted by Karl Schneider on 3 Sept 2005, 3:42 a.m.,in response to message #5 by Gene Gene stated, Quote: I dont' [sic] believe that's true (that the 33S solver will always return a direct solution when its decision rules say it can). It will return the direct solution if the equation is entered as "X^2-4", but if you put "X^2-0*X-4" as the equation, it will not directly solve it but will allow the user to give estimates that guide what the solver finds. Gene, I know that you have a vested interest in that you prepared the Learning Modules for the HP-33S, but we've covered this topic before. Ading the superfluous term "+0*X" provides a second occurrence of the solution variable in the equation, precluding the 33S by its own strategem from attempting to find a direct solution. Granted, this offers a workaround, but what rational user would include "+0*X" in an equation? It's just wrong that the calc cannot return -2.0 as a solution to "X2-4 = 0, given an initial search range of -2.1 to -1.9. I outlined quite clearly in this space how I thought the 33S' SOLVE decision tree was flawed, and I remember that you agreed. Shall I dig up the thread? -- KS

 Re: 33S solver: We've discussed this before.Message #7 Posted by Gene on 3 Sept 2005, 9:23 a.m.,in response to message #6 by Karl Schneider Hi Karl. Oh, I don't disagree about the solver. It's just that in THIS thread, no mention was made about how to force the 33S to avoid the direct solve. That's all I was attempting to do here. Sometimes the only information people might have about a model is the current thread/discussion. And, I would hope I'm not that biased in most posts. :-) Gene

 "Algebraic" 17/19/27 solver: Same problem!Message #8 Posted by Karl Schneider on 4 Sept 2005, 8:38 p.m.,in response to message #7 by Gene Gene -- The issue about the HP-33S solver, when in Equation mode, always returning the direct algebraic solution when its logic determined that it could, got me thinking about the vaunted "Algebraic" solver from the 17B/19B/27S models. Same problem! The 27S and 17BII will always return X = 2.00 as the solution to "X^2-4" or "SQ(X)-4", even when -2.00 is pre-stored in X. So, KinHPo apparently didn't put a lot of new work (of questionable value) into the SOLVE algorithm for the 33S, by adding an algebra-performing "direct solution" method in Equation mode. Instead, they just copied those algorithms -- warts and all -- from the 17BII, which then was still in production when the new-model efforts were initiated. I should have suspected as much... -- KS

 Re: ... and HP-17BII; -22S; -32SIIMessage #9 Posted by Ken on 2 Sept 2005, 8:36 a.m.,in response to message #3 by Karl Schneider Well I bought the 32S yesterday. Returned it today. I guess I am just spoiled but the interface was a far cry from the 27s. if I am going to go through all that work then I will stick to my 3 dollar TI calculator:) I will still need to lay hands on the 19BII but have not found one on the shelf yet.

 Re: ... and HP-17BII; -22S; -32SIIMessage #10 Posted by Ron Ross on 2 Sept 2005, 8:54 a.m.,in response to message #9 by Ken And you won't unless you are extremely lucky. The Hp19Bii hasn't been made for about 5 years now. Ebay is where you will find this creature. You can buy the cheaper Hp19 (w/o RPN over the newer Hp19Bii which has RPN). If you don't need trig, the Hp17Bii+ is very similiar aside from its being a business calculator w/o precidence in its algebraic mode. It has the same solver (aside from perhaps an unreliable Let and Get function) and has 32K of RAM vs 7K on your Hp27s. Or buy an older Hp17B or Hp17Bii (and get the Let and Get functions). You might also consider an Hp38G or an Hp39G+. They only hold 10 equations at a time, and only allow single letter variable names (pathetic, I know). But they have an equation solver list which is functionally similiar to your Hp27s. Actually, if you can buy the new Hp40G series (overseas model), you get an equation writer that is compariable to the Hp48/49 and a nice algebraic calculator. But it is a graphics and not a pocket calculator like the Hp27s.

 Re: HP 27sMessage #11 Posted by bill platt on 1 Sept 2005, 12:55 p.m.,in response to message #1 by ken The new 17bii+ has the same interface, but alas it has no trigonometric functions :-( I like this solver interface a lot, too! If you have a 49g+ or a 48g or any of the 48-49 models, you can build essentially the same interface quite easily, complete with soft menu keys and all. It even has more features. It does have a slightly different operating paradigm but it is not too terribly difficult to get used to and is more versatile to boot. BTW, this is the sort of thing that the 48 series has that the classic RPN machines do not--a clean, direct interface for solving equations for any unknowns.

 Re: HP 27sMessage #12 Posted by Jan (CH) on 1 Sept 2005, 3:03 p.m.,in response to message #11 by bill platt HP 27 S is truly an excellent machine. Particularly the solver is most useful, powerful and still easy to use. HP 27S are seldom and very difficult and expensive to get second hand. When my 27S was stolen, I bought a used HP 19BII. It should be much easier to find an used 19BII; may be even a new one. They are not so old. 19BII has almost the same capabilities as 27 S, even some extra goodies. It has algebraic and RPN logic, not implemented on 27S. The solver has some implementation differences to 27S, but this has no consequences in practical life. 19BII was sold as “business consultant”, which is a bit misleading. It contains all the math functions available on 27S, such as trigonometric functions. All in all 19BII is an excellent replacement for 27S Jan

 Re: HP 27sMessage #13 Posted by Ken on 1 Sept 2005, 3:27 p.m.,in response to message #12 by Jan (CH) Well thank you all for your assistance. I am now confronted by a new problem who carries a line of HP's. I visited compusa, office max, Circuit city, office depot, I want to hold one and push a button before I find a fitting replacement for my deceased friend the 27s.

 Re: HP 27sMessage #14 Posted by Charlie O. on 1 Sept 2005, 3:51 p.m.,in response to message #13 by Ken Walmart and Fry's Electronics both carry hp's. Walmart is easier to find but Fry's is more likely to have the calc you seek.

 Re: HP 27sMessage #15 Posted by bill platt on 2 Sept 2005, 9:27 a.m.,in response to message #13 by Ken Hi Ken, (Is your 27s still in existence? Many here can make good use of a "deceased" machine.) You need to buy a *used* machine--forget the retailers. The models you are really going to appreciate: 27s, 19b, 19bii ,and I think the 18c will be good, too---check the description of it in the museum or at craig finseth's hpdatabase. are no longer manufactured, so look in or post to the classified ad section of this forum, or go to ebay.... All you can get new from hp in your interest bracket (direct from HP via online interface): 17bii+ (no trig maths) 48gII (really a 49g+ with less memory and no expansion) 49g+ (current "top dog" of hp graphing) 39g+ ("high school" formula interpretive interface version of the 48gII) note that the 39/49/48gII listed above all have "issues" with the reliability of the keyboard....search comp.sys.hp48 and you'll see what I mean. You will get more for your money going used. There are no scientifics with that hp27s interface anymore. The *only* scientific in current production is the 33s which has the less than optimum solver interface, though I use it and find it quite acceptable, though I like my 27s solver more. If you need a solver with trig more than anything, then you can adjust to the 33s and because it has full RPN and Algebraic programming capability, you can in fact do more with it htan the 27s in terms of programming power; it is just that the equation interface is not so graphically cool. Do be sure to understand that there are a couple of bugs with the 33s (which one of these days I expect they will iron out) all of which can be worked aroundwithout to great trouble. The most serious generally are the rectangular to polar conversiosn bug, and hte h.ms to dec hrs conversion (for negative numbers ther eis a problem but ok in positive range). Both of these are easy to fix with simple short programmed routines. I hope this helps you. Regards, Bill Edited: 2 Sept 2005, 9:32 a.m.

 Re: HP 27sMessage #16 Posted by Ed Look on 3 Sept 2005, 2:45 a.m.,in response to message #15 by bill platt What was the solver used in the old 34C?

 Original HP solver [was: HP 27s]Message #17 Posted by Karl Schneider on 3 Sept 2005, 3:08 a.m.,in response to message #16 by Ed Look Ed said, Quote: What was the solver used in the old 34C? The original one, ported virtually without changes to the 15C and the 41C Advantage module and subsequently (with modifications to the interface) to the 32S and 42S and their successors. The solver on the models specifically mentioned work only with an RPN program defining the mathematical equation. The 34C/15C/41C-Advantage do not directly support multi-variable equations. However, more than a year ago, I posted a work-around for this limitation, using the indirect storage register. -- KS Edited: 3 Sept 2005, 3:10 a.m.

 Re: Original HP solver [was: HP 27s]Message #18 Posted by Ed Look on 19 Sept 2005, 1:12 a.m.,in response to message #17 by Karl Schneider Ah, the i register! What an ingenious idea! I wish I had the nerve to use my 34C again! It shows signs of advanced age. I'm not sure that one more power on will not kill it! But I find that the 32SII is just as pleasurable to use, except that after all these years, I still know the 34C layout by heart (you know what they say about college sweethearts)... and believe it or not, I've found the 33S to be quite a joy to use, too! I use it the most these days (to spare the 32SII extra wear), much more than the more advanced 48G, 48G+, or 49G+. I had originally guessed that the 49G+ would be my workhorse; it turned out to be the much maligned, and undeservedly so, 33S! If I have to do a similar program for the 33S, I will keep your brilliant idea in mind... if you don't mind!

 Re: Original HP solver [was: HP 27s]Message #19 Posted by bill platt on 19 Sept 2005, 10:19 a.m.,in response to message #17 by Karl Schneider Karl, Back when you originally posted your multivariable method, I printed it out and I consider it a must-have AND must-know for the 15c. I think it really should have been in the manual. The question is, had nobody at HP thought of it?!! regards, Bill

 [15/34/41] SOLVE/INTEG with selected variableMessage #20 Posted by Karl Schneider on 21 Sept 2005, 2:22 a.m.,in response to message #19 by bill platt Bill -- You as well as Raul Lion and others (several years ago) complimented the workaround I posted for using selecting a variable for SOLVE or INTEG on a 15C, 34C, or 41C (with Advantage module). My original post at was not particularly explicit. I will follow through shortly on my promise to write a better and more comprehensive one. -- KS

 SOLVE/INTEG with MISO functions on HP-15/34/41Message #21 Posted by Karl Schneider on 24 Sept 2005, 2:19 a.m.,in response to message #20 by Karl Schneider Note: This is a first draft of material developed for an MoHPC Article that I announced an intention about 30 months ago to write. Comments are welcome and encouraged. If anyone can suggest a more practical equation for illustration, please suggest. The equation should be both solvable and integratable with reasonable calculation times. ====================================================== Using SOLVE and INTEG with MISO functions on HP-15C, HP-34C, and HP-41C/CV/CX BACKGROUND OF "SOLVE" and "INTEG": Numerical rootfinding ("SOLVE") and numerical integration ("INTEG") of single variables were built-in or made available as sophisticated machine-coded functions on seven RPN-based Hewlett-Packard scientific calculator models. SOLVE and INTEG are present on the following models: HP-34C HP-15C HP-41C/CV/CX with Advantage Pac HP-32S HP-42S HP-32SII HP-33S (Excluded from this list are the RPL-based models, the algebraic-entry models, and the HP-71B with Math ROM.) The SOLVE and INTEG functions were introduced in their original paradigm on the HP-34C in 1979, and later carried over to the HP-15C in 1982. For the HP-41C/CV/CX, the same functions were adapted with only slight modification in the Advantage Pac introduced in 1985. Two fine articles regarding SOLVE and INTEG on the HP-34C appeared in HP Journal in 1979 and 1980, respectively. These are available on CD Volume 3 and the DVD-ROM from the MoHPC. This original paradigm requires that the function to be solved or integrated, be programmed as an RPN routine. The routine must be designed to accept an input value of the variable to be solved or integrated in the stack x-register, and to return the calculated value of the function to the x-register. This user program is executed repeatedly by the SOLVE or INTEG function in order to calculate the desired result. Moreover, the program may include a SOLVE or INTEG instruction, thus allowing INTEG to be executed within SOLVE, or vice versa. This capability, incidentally, was not provided on the successor Pioneer-series models HP-32S and HP-32SII, or on the present HP-33S. STORAGE OF VARIABLES; MISO AND SISO FUNCTIONS Storage registers for variables on the HP-32S, HP-32SII, and HP-33S are designated by single letters. The Pioneer-series HP-42S allowed alphanumeric identifiers for variables. SOLVE and INTEG on all four of these models require the user to specify (by letter or name) the root variable to SOLVE or the dummy variable of integration to INTEG. This feature is well-suited for multiple-input, single-output (MISO) functions. The 34C, 15C, and 41C/CV/CX, however, provided only numbered and "indirect" storage registers. The paradigm of SOLVE and INTEG on these models was designed primarily for single input, single-output (SISO) functions. It does not require that the input variable even be stored to a register. TO USE 34C/15C/41C "SOLVE" AND "INTEG" WITH MISO FUNCTIONS: Assume that a MISO function is to be programmed as an RPN routine on a HP-34C, HP-15C, or HP-41C/CV/CX. A user may wish to use SOLVE or INTEG successively for any selected "focus" variable with all others held constant, and without having to edit the RPN routine. Here is a simple procedure for this purpose: 1. Designate a numbered storage register for each variable in the function. 2. On the 41C/CV/CX, designate another numbered register as the indirect storage register. (Do not use a stack register, as it will be overwritten by SOLVE or INTEG.) In PROGRAM mode: 3. Program the function as an RPN routine that meets the basic requirements and, also, always retrieves each variable from its designated storage register the first time it is used. (On the 41C/CV/CX, the label beginning the routine must be an external label.) 4. The second instruction of the routine (after LBL) should be "store to indirectly-referenced register". This is "STO (i)" on the 34C and 15C; "STO IND nn" on the 41C/CV/CX. In RUN mode: 5. Store the desired constant values of the fixed-value variables to their storage registers. 6. Store the register number containing the floating variable to the indirect storage register. 7. Invoke SOLVE or INTEG in the usual manner. This procedure utilizes indirect storage to make the RPN program more flexible. SOLVE and INTEG feed the each value of the floating variable as input to the user program, which immediately stores the value indirectly to its designated register. Each variable in the function is then recalled for use in calculations, so the user program need not be structured to receive any particular variable from the stack. EXAMPLES: ```f(x, y, z) = 2*x - ln y + 1/z x in R1; y in R2; z in R3 indirect in R00 (HP-41 only) HP-15C/HP-34C program: HP-41C/CV/CX program: LBL A LBL "AA" STO (i) STO IND 00 RCL 1 RCL 01 2 2 * * RCL 2 RCL 02 LN LN - - RCL 3 RCL 03 1/x 1/x + + RTN RTN In RUN mode, solve for x such that f(x, y=15.1, z=5.3) = 0 HP-15C/HP-34C: HP-41C/CV/CX: 15.1 15.1 STO 2 STO 02 5.3 5.3 STO 3 STO 03 1 1 STO I STO 00 0.5 0.5 ENTER ENTER 5 5 SOLVE A "AA" (enter AA in alpha mode) XEQ "SOLVE" (Answer is 1.263007749) Next, solve for y such that f(x=0.7, y, z=3.3) = 0 HP-15C/HP-34C: HP-41C/CV/CX: 0.7 0.7 STO 1 STO 01 3.3 3.3 STO 3 STO 03 2 2 STO I STO 00 0.5 0.5 ENTER ENTER 6 6 SOLVE A "AA" (enter AA in alpha mode) XEQ "SOLVE" (Answer is 5.490560270) Finally, integrate f(x=1.7, y=4.1, z) over z = 1.0 to 6.0. Specify 5 decimal-digit absolute function accuracy: HP-15C/HP-34C: HP-41C/CV/CX: FIX 5 FIX 5 1.7 1.7 STO 1 STO 01 4.1 4.1 STO 2 STO 02 3 3 STO I STO 00 1 1 ENTER ENTER 6 6 INTEG A "AA" (enter AA in alpha mode) XEQ "INTEG" (answer is 11.73682) ``` Edited: 28 Sept 2005, 2:48 a.m.

Go back to the main exhibit hall