HP Forums

Full Version: HP 33s unique features?
You're currently viewing a stripped down version of our content. View the full version with proper formatting.

Has anyone done a comparison of the features and capabilities between the new 33s and the old 32sII, besides the obvious differences of keyboard layout and extra RAM?

I have just gotten mine and have just given the 33s keyboard a cursory look. There appears to be some extra features or keys that I don't think I recognize from the 32sII, as "SGN" or "SEED" for starters. I guess I'll have to really read the manual. I was hoping that it was similar enough to the 32sII that I could just quickly scan over the manual... not that I would ever mind extra features! It does have a much nicer case than either of the 32s series models ever came with!

Not to start trouble, but I suppose red LEDs would have been too uneconomical for the user (battery drain)? Years ago, it was cool to me to mess with the calc in bed; I knew just about where every key and shifted key was and the red glow was just too cool!

Hi, Ed;

I did not own an HP32S/SII at the time they were available, neither I own any now. But I have an HP33S and it seems to me that, despite the obvious selectable RPN/Algebraic operating mode of the HP33S instead of the RPN-only available with the HP32SII and the (also obvious) two-line display, I guess the only extra operating characteristic is the constants library. I also remember that the HP32S/SII had a different menu selection, where the names of the "features" were writen above the arrow that indicated which key should be pressed for each feature shown. Now, the HP33S has a new menu arrangement, using the number keys instead of the first-row keys used in the HP32S/SII.

I had a look the HP32SII features (in brief) while writing this post at Finseth's Data Base, so maybe there are other new features available with the HP33S and not found in the HP32SII that I'm not aware of. I'd like to know about them as well. I recomend Finseth's database as a reference source, because many good stuff about each model is mentioned there. Great source!


Luiz (Brazil)

The 32sii did have the SEED command, but the SiGN is new, but it in itself is a silly command (it returns 1 if the number is greater than 0, and -1 if less, so you still need to use an "if statement" (x>0?) to check it, just as you would need to check if the number is greater than 0.

Offhand, there are a few other new commands that I can think of: the RPN/ALG, and the constants for starters...I dont have either calculator in front of me, so that's all that I can think of at 2:30 AM local time.


I find X^3 and Cubic Root on the keyboard
Then Integre Division and Remainder
SGN was mentioned and the ALG mode, too
More Program steps allowed.
Y-reg on the the display or the history in ALG mode
I don't much mind because the 32SII was "perfect" already.
Now I just thank for the additionla display line and the memory.
Yes, it's too bad that numerical sub-program labels up to 99 are not allowed.
I would use (i) as follows:
GTO or XEQ (i) with 1...26 would jump to A...Z
Fractional part would point to sublabel 01...99.
Zero fractional part would be the start of the main program eg. LBL 00 is not allowed.
GTO or XEQ (i) with 3.03 would thus jump to program B sub-program LBL 03.
Numerical registers from 0...999 would be great
Even if accessible only through (i) and if the sign needs to be negative.
That is (i)=1..26 = A..Z, but (i) -1...-999 would be the numerical register.
Is 27 == i ??
Then perhaps register 0 could be (j) !!!
Another index register would be very usefull in many situations.

re: "SiGN is new, but it in itself is a silly command (it returns 1 if the number is greater than 0, and -1 if less)"

It's not so silly for some folks. For astronomical calculations of position, it is quite useful. The north/south position of a source (declination) can be a bit of a nuisance to work with. There are + and - declinations, often given in degrees, arc minutes, and arc seconds, such as +12 13 14.15 or -15 16 17.18, where only the degree value has the sign attached. You almost always have to convert the position to decimal degrees (or radians) for other calculations. Getting the sign of the result from the sign of the degree value is easily done with a SIGN type function:

deg.ddd = SIGN(deg) * ( deg + arcmin/60 + arcsec/3600 } .

The real annoyance is declinations within a degree of zero. For these, you have to pick up the sign as a character - except for some old FORTRAN compilers, which knew the difference between (i.e. the sign of) +0 and -0 !!

HP 71B is the *ONLY* calculator on earth to distinguish between +0 and -0


Hi, VPN;

if we consider the 1's Complement mode representation and integer mode, the HP16C also distinguishes between 0 and (-0).


Luiz (Brazil)

Edited: 20 June 2004, 5:59 p.m.

In case you need to use a SIGN function on a calculator where it is not preprogrammed, the following should work:


You may be worried about the effects of possible rounding errors, but I think that the internal guard digits will take care of any "imperfection" on the division algorithm.


Talk to Norm.


Trent, you said, "Ed,

Talk to Norm.

tm "

No, he's really a cool guy and very interesting poster, but it'll get long. I'm sure he's reading everything and rolling his eyes.

But what's really cool is that because my kids (grade school age) are more attracted to the design of the 33s over the 32sII (they kind of smirk at the design of the 34C), they wanted to touch it and play with it. So, I whipped out both the 32sII and the 33s and got a chance to introduce the concept of RPN and the four-level stack to them! The sixth grader, who fools around with a HP39G (a huge blue algebraic brick), sort of got it and I think the third grader sort of not... he still asked for the equal sign.

Not to counter VPN's statement, but the 41 does also do so, when using NNN's. Not that it matters to many, anyway.

Hi Ed:

I believe this list to be nearly fully comprehensive.

What the 33s has which is different or additional to the 32sii:

INTeger divide (is it really modulo or something?)
SGN (sign boolean)
cube root, cube
An ALGEBRAIC calculator mode
A "Constants" Library
Internal precision is different; new algorithms.

Different, and deficient to 32sii:
More Memory,
But, it is a different memory / byte structure.
Far less efficient memory usage---it is not
32 kB of 32s eqivalent space---depending on how you
compare, it is only between 6 and 11 kB against the 32sii

Equations (both in Equation List and in programs) are
limited in length to 255 characters. (There was no limit
in the 32sii--except for the small memory--which made it
possible to only barely exceed 255 characters anyway).

BTW, the 32sii had the following which was not in the 32s:

Equation List and Equations in programs
R up command (only R down on 32s)
Conversion factors
fraction modes and fraction input
More flags (to handle fraction modes and equations)



Hi Bill,

Yes, nearly fully comprehensive, I think. I tried to contact you but the address I got from your postings to c.s.hp48 (after taking the hormel out, that is) does not seem to work. What am I doing wrong? :-)



this is works, when stack_x is equal zero:



And without ABS:


This is what is make SIGN to useful, because this feature not allowed in EQN.


Touché! Thanks, Czaba, you are right

I was just trying to produce an excuse by means of some error handling flag, so to cover my disregard for the zero case, but I must say there are no possible excuses.

On the 33S the only error that flags may help to handle are overflows. It would be nice to have an "error ignore" flag... Well, the old and faithful HP-41C indeed had it and, for that matter, it also had a SIGN function which answered 0 when applied to an ALPHA string.

- A01 => A0001
- Cursor keys and numeric shortcuts to navigate menus
(new menu structure)
- two lines display to show Y-register
(ALG mode: show history of your calculations)
very nice in program mode and EQN list, too!
- () (ALG mode only)