# HP Forums

You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3
Follow up of the discussion started here:

Informing myself on the HP-35s I saw that the form factor of the 35S is not as big as the 50g (that is also out of market) but the calculator has quite some powerful characteristics and formula/equation memory. (it is also programmable).

Therefore I opened this thread to crowdsource a list of scientific calculators that are still produced, have a not oversized form factor (more or less like a 6'' inches toucchscreen only phone), have memory (continuous) and allow formulas (also in form of a program) to be saved for later use.

Programmability is not stricly required. RPN or algebraic does not matter. HP or not.

Super small example:
The sharp el 506w from around 2002 has at least 6 variables (or registers) for single numbers and 4 registers for formulas that share a memory of 256 characters.
On the 506w I can save a formula like 'A^2 + B + 35* C' where at first I fill the registers A, B, C with the data, then recall the formula (say, saved in F1), and execute it.
I can use formulas with the solver (then I need to include the register X) or also not.

An obvious competitor for the 35S is the DM42.

Any other? If possible with enough memory for a little equation library, say 5000 characters (in algebraic form), or 100 formulas of 50 characters each.
(12-06-2017 12:18 PM)pier4r Wrote: [ -> ]An obvious competitor for the 35S is the DM42.

I'm pretty sure the DM42 does not have equation support.
(12-06-2017 12:24 PM)Thomas Okken Wrote: [ -> ]I'm pretty sure the DM42 does not have equation support.

Really? With the free42 I can write a program that is nothing else than a formula (indeed like a macro for the keystrokes), using registers to hold the variables in input.
On a 506w I can save a formula like 'A^2 + B + 35* C' where at first I fill the registers A, B, C with the data, then recall the formula (say, saved in F1), and execute it.

Is this feature removed in the dm42? Or is it me using a wrong terminology?
(12-06-2017 12:29 PM)pier4r Wrote: [ -> ]Is this feature removed in the dm42? Or is it me using a wrong terminology?

I think you're using the wrong terminology.

The DM42/HP-42S is certainly programmable and you can structure programs in such a way that they can be used with the built-in solver and integrator. But that's not all you can do with programs.

A formula-based calculator uses formulae exclusively for a solver/integrator or for you to build a table of x,y values without having to write a program at all and doesn't need to have any further programmability (if that is a word...).

The HP-35S has both features. It has a formula memory for use with the built-in solver/integrator and you can write programs for it. The DM42 does not have a formula memory. If you want to program a function to use with the integrator or solver then it has to be in the form of an RPN program.
Usually when a calculator supports formulas, i assume that it supports a solver where the formula/equation can be entered in algebraic format and without the need to use any kind of complementary programming.

The 42S has got a nice and powerful solver with integrated menus, but requires the use of custom programs to implement the customized algorithm.

Virtually all the scientific/financial calculator manufacturers have created pocket models able to do formula entry to be used in their solvers without the need to create additional programs.
The downsize is that we are limited to do what the model allows us to do.

Now, the old style 42S solver with programming is very appealing to me as it gives all the flexibility to create our own algorithms.
(12-06-2017 12:44 PM)grsbanks Wrote: [ -> ]
(12-06-2017 12:29 PM)pier4r Wrote: [ -> ]Is this feature removed in the dm42? Or is it me using a wrong terminology?

I think you're using the wrong terminology.

Ok fixed, or at least I tried to clarify.

@jebem: "Virtually all the scientific/financial calculator manufacturers have created pocket models able to do formula entry". Can you list a couple that are still produced?

I mean not only entering formulas, also save them.
The 42S solver requires a program something like this... for X+Y=5

LBL AA
MVAR X
MVAR Y
RCL X
RCL Y
+
5
-
RTN

or some such... as I recall - it has been a while. :-) Please excuse me if that's not quite right above.

You have to declare all your variables using the MVAR (memory variable) function then write an RPN expression evaluating the function to = 0.

The MVAR will create the SOLVER menu labels when you enter the solver and choose AA as the program for it. It will then display menu labels for X and Y.

So, doable, but not quite the same.
Yep. That's pretty much it.

Code:
```00 { 22-Byte Prgm } 01▸LBL "AA" 02 MVAR "X" 03 MVAR "Y" 04 RCL "X" 05 RCL+ "Y" 06 5 07 - 08 RTN```

Point is, you have to write a program for the equation in RPN. With an "equation-enabled" machine you just have to throw the equation itself at it.
(12-06-2017 01:16 PM)pier4r Wrote: [ -> ]"Virtually all the scientific/financial calculator manufacturers have created pocket models able to do formula entry". Can you list a couple that are still produced?

Hi, Pier4r,

For example, a current pocket size calculator from Texas that sports continuous memory and a solver among other nice goodies:
Texas Instruments TI-36X PRO
This one is probably the best bet and "benefit to price" ratio among the equivalent competition offering, according to the people actually using it and posting here in this forum.

Many Casio machines were known for their formula entry environment (with continuous memory), like the fx-5000F (from 1987) among many others.
They are out of production, of course, as any other calculator made in 2002 or 2010 or even 2015 for that matter.
There are notable exceptions, like the HP-35S that seems to be on sell today (Dez-2017). This makes the HP 35S 10 years old on the market now. For such a alleged flawed calculator, this should be a success study case!

But then Casio has got newer models like the fx-991 X Classwiz series that, despite limited to what the built-in algorithms can offer, also have continuous memory and a solver.

Casio also offers the current top of the line, scientific, non-graphic, programmable fx-5800P pocket calculator. Not only it is programmable as it offers a solver and formulas as well.

And of course we should not forget the current SHARP pocket offerings.
They look a bit physically fragile to me when compared to the Casio or Texas, but it is just an impression, as I do not use calculators to do my work anymore (school days are gone long time ago and I use computers at work).
I have this nice looking and feature rich Sharp EL-W506X, currently in production, that should fulfill your specifications as well.

I didn't check the physical sizes to fit your requirements, although they are smaller than the full sized graphic models (or, as an old time member of this museum used to call them: the battleship calculators).

Now, a word of warning:
Continuous memory on all these machines, 35S included, are not really continuous as it depends on good battery energy to maintain the memory contents integrity.
Adding this to lack of input/output interface and we end up with seriously limited machines.
Even a reset to restore a machine crash will result in total memory loss.
(12-06-2017 03:50 PM)jebem Wrote: [ -> ]For example, a current pocket size calculator from Texas that sports continuous memory and a solver among other nice goodies:
Texas Instruments TI-36X PRO
This one is probably the best bet and "benefit to price" ratio among the equivalent competition offering, according to the people actually using it and posting here in this forum.

Was just going to mention the 36X Pro. It doesn't allow you to store a whole library of equations, but you can enter one to solve, evaluate, or generate a table of values.
(12-06-2017 04:19 PM)Dave Britten Wrote: [ -> ]Was just going to mention the 36X Pro. It doesn't allow you to store a whole library of equations, but you can enter one to solve, evaluate, or generate a table of values.

Yes I checked the manual, and the 36X is rather limited in terms of memory. It has even less than the 506w from 2002 (on the 506w I can manage to save 4 formulas, if those together need less than 256 chars). So while it is an input, it does not quite fit. Now I check the others.

edit1.
The casio fx-991EX has quite some amount of functions, but in terms of memory is again not better than the 506w. Some variables/registers and no formulas to save. Impressive how the 506w is still "up to date", really the common needs for a calculator are quite mature.

The 506w has A,B,C,D,E,F,X,Y,M as variables/registers and F1,F2,F3,F4 as formulas (that in total can use 256 chars).

edit2.
The casio fx-5800P hits home! The size is quite close to the 35S and it has enough variables and space for formulas (aside from programs). So we have a real competitor (plus the dm 42S that requires a bit more work)

edit3:
from sharp, unless one goes for the graphing calculator (that is too big), the calculators there don't have really much memory than the 506w that I have. 4 formulas and 9 variables/registers.

edit4:
TI scientific calculators are more or less like the sharp ones, I think the sharp are better.
(12-06-2017 03:50 PM)jebem Wrote: [ -> ]Now, a word of warning:
Continuous memory on all these machines, 35S included, are not really continuous as it depends on good battery energy to maintain the memory contents integrity.
Adding this to lack of input/output interface and we end up with seriously limited machines.
Even a reset to restore a machine crash will result in total memory loss.

Of course the continuous memory is there until one messes up the change of the batteries, therefore a calculator like that cannot be mistaken for a battleship one (like the 50g). Indeed they are useful for certain use cases.

Like quick computations of a given type (so one reuses the formulas) without need of a lot of space.

Using computers/tablets is all fine but I find (and maybe I am not alone) that sometimes
(a) I want to develop a concept on paper, and having a calculator with me helps more than a bulky computer or a phone with no tactile keyboard.
(b) I don't want to use a computer, I want to sit on the sofa or in a chair and be comfy. Still having all the tools needed.
(c) less request of energy and time to get results (start the system, open the app, load the program, possibly write a program, get the results, visualize the results). It should not be underestimated.
Recap after surfing on the major producers (TI, HP, Casio, Sharp), given the limits stated above.

Hp35s. ~ 30K of ram
Casio fx5800P (the dimensions should be there, although it seems a graphing model). ~ 25K of ram.
Many sharp models have 9 memories and 4 formula memories.
DM42 (with a bit more effort to use programs as formulas).

Note that the hp35s has the two largest dimensions being: 6.22' x 3.23' in freedom units or in real units: 15.80 cm x 8.2 cm
Casio fx5800P: 6.4 x 3.2 .
Ti graphing (or HP and others), overblown with 7+ inches. One inch seems not much, but not in this case.
6.22 or 6.4 is still like a big phone.
DM42: can someone measure it?

Any other suggestion is welcomed.
(12-06-2017 10:43 PM)pier4r Wrote: [ -> ]Recap after surfing on the major producers (TI, HP, Casio, Sharp), given the limits stated above.

Hp35s. ~ 30K of ram
Casio fx5800P (the dimensions should be there, although it seems a graphing model). ~ 25K of ram.
Many sharp models have 9 memories and 4 formula memories.
DM42 (with a bit more effort to use programs as formulas).

Note that the hp35s has the two largest dimensions being: 6.22' x 3.23' in freedom units or in real units: 15.80 cm x 8.2 cm
Casio fx5800P: 6.4 x 3.2 .
Ti graphing (or HP and others), overblown with 7+ inches. One inch seems not much, but not in this case.
6.22 or 6.4 is still like a big phone.
DM42: can someone measure it?

Any other suggestion is welcomed.
Dimensions: 144mm*77mm*max13mm
(12-06-2017 10:43 PM)pier4r Wrote: [ -> ]DM42 (with a bit more effort to use programs as formulas).

If you're going to count the DM42 in there as a currently available machine with solver and integration abilities then you should also count the DM15/DM15L, which as we all know, is more or less a clone of the HP-15C but with more memory. Mine (M1B variant) reports 20 registers in use and 211 registers free after a reset.
@massimo. Only 144mm ? That's is a golden god!

@grsbanks: good point! Has the dm15 still around ~400 bytes of memory for programs (and therefore formulas)?

Recap:
Hp35s. ~ 30K of ram
Casio fx5800P (the dimensions should be there, although it seems a graphing model). ~ 25K of ram.
swissmicro DM42 (with a bit more effort to use programs as formulas).
swissmicro DM15 with registers and program memory (program as formulas) ~1500 bytes program memory, see below.
Many sharp models have 9 memories and 4 formula memories.

hp35s : 6.22' x 3.23' in freedom units or in real units: 15.80 cm x 8.2 cm
Casio fx5800P: 6.4 x 3.2 .
6.22 or 6.4 is still like a big phone.
DM42: 144mm*77mm
DM15: 88mm*59mm

Any other suggestion is welcomed.
(12-07-2017 12:59 PM)pier4r Wrote: [ -> ]@grsbanks: good point! Has the dm15 still around ~400 bytes of memory for programs (and therefore formulas)?

Much more than that in the M1B variant, which is what you get when you order one nowadays.

Total space is 231 registers (1617 bytes) with a 7-byte to 1-register trade-off. So, in its default configuration after a reset, that's 20 registers and 1477 bytes for programs.
The Casio fx5800p has the form factor of the graphical models of the same era but the display is not pixel addressable from the built in programming language. It is just a nice big multi-line text display that I am sure would be great for middle-aged eyes.

I have often thought about buying one for my collection but it would have to be a fairly expensive grey import to the UK.
For the ones interested in the Casio fx-5800P, here is an good analysis by Manolo Sobrino.
(12-07-2017 06:43 PM)jebem Wrote: [ -> ]For the ones interested in the Casio fx-5800P, here is an good analysis by Manolo Sobrino.

nice pointer!
Pages: 1 2 3
Reference URL's
• HP Forums: https://www.hpmuseum.org/forum/index.php
• :