HP33S Manual Is Here!!!


Big news! The long awaited HP 33S manual is now on HP's web site, so the calculator can't be far behind!

It's sort of a disappointment in that it's a near exact 32SII clone with 32K bytes (that's right BYTES!) of memory. But you can only use that memory for programs and equations, not registers (not even indirectly). The extra memory doesn't even allow for nested subroutines beyond the 32SII limit of 7. In algebraic mode you get 13 levels of nested parenthesis and you can use complex numbers in them -- a nice touch! And programs can be written in either algebraic mode or RPN (looks like that translate into RPN to run them, they say that they run slower in algebraic mode).

No I/O, same flags as on the 32SII, same BASE stuff, same STAT stuff. Built-in constants look good and a much welcomed (at least to me) integer divide and remainder function are new.

It uses 2xCR2032 batteries, one backs up the other when you change them so there's little chance of memory loss (good thing with all that memory).

The manual looks very complete and so far I haven't seen any mistakes. Lots of examples!

Here's your bedtime reading....



I had just found this about 5 minutes before you posted (i was looking for a module for the 49g+ and noticed the 33s was listed in the support)
So, apparently there is a HP 33s font. I wonder if HP will release it. It would be very helpful when typing out programs
Still waiting for the other files to download... I really wish my internet was faster


I imagine we'll be able to get any fonts for calculators released. HP is very responsive these days.

The new 49G+/48GII font is available. I'm sure any other future calculator fonts HP uses will be too.

Good times appear to be coming.

The eternally pessimistic can keep using their 34c's. :-)



Gene, can you post a link to where we can obtain the new 49g+ font?



I have it, and I'll try to put it on my comcast webspace and post a link. My hope is that each new machine will have a font going forward...that would REALLY contribute to users being able to write material for it.



That ought to work, but ...


Clicking directly on the link opened the file up in my browser, but in IE I woudl think you could choose to download the link and it should come down to the PC.


And here is the HP-39G+ manual.



Thanks for the link.

But it's typical: In the German and the English version of the 39G+ manual pdf,
on page V-2 (Vorwort) resp. P-2 (Preface), secondlast paragraph, you can see that they don't even know how to write their name !

The words are: "Copyright 2003 Hewlette-Packard Development Company, L.P"

BTW, it's the page where they state that they're not responsible for any errors in the manual (LOL:-)))



not to mention the hardware errors


Who else should be? I guess want they want to say is they don't want to be held responsible.

12345 to delete


That "not responsible" thing has got to be yet another example of how in the US leagal system, if it quacks like a duck, and walks like a duck, and even swims like a duck, it is not a DUCK, and not even a goose; it is a lemon ;-)




Good work Katie!!!! Thanks for the tip. WIll give me something to read this weekend!


hey does the manual say why the keyboard is shaped like a cowcatcher ?


Norm, Yes they do explain the need for the cowcatcher design in the manual. Here's the excerpt...

Engineers at Hewlett Packard have put great effort into the overall design of this precision instrument. For example, many hours of effort went into the format of the "cowcatcher" keypad layout. Aside from making sure users were kept on their toes when entering data, the primary goal was to facilitate easy repair of your new HP33S calculator.

The repair division had been using the rail of mainline Burlington Northern railroad tracks to "fix" improperly closing cases on the earlier "Pioneer" series calculators. The same repair technique applies to this next generation of HP calculators and engineers wanted to make sure that cows would not get in the way of a speedy remedy for a malfunctioning device.

Edited: 8 Nov 2003, 11:00 p.m.


Don't try to B.S. Katie. You'll be up against a pro.


LOL! ROFLMAO! Did it really say all that ?


That's not a cowcatcher, it a car's radiator grill. And nobody, but nobody, in this country would design such a ugly thing. If someone gave out an award for a truely bad industrial design, we'd have a winner! I'm sorry, but HP behavior reminds me of FORD's advertised policy of Rollover Value !!! Have You Rolled Over A FORD Lately?

Edited: 9 Nov 2003, 1:36 a.m.


At first I was amazed that they put all the RPN stuff together, and left the algebraic for an apendix. Then i realized they did that only so they would not have to totally rewrite the manual. Many of the examples, and most of the advanced features sections were all from the 32sii manual (i reread it just a few weeks ago)
There is a bit of new stuff in there, but at least it has 31kb of memory


Katie posted:

"It's sort of a disappointment in that it's a near exact 32SII clone with 32K bytes (that's right BYTES!) of memory. But you can only use that memory for programs
and equations, not registers (not even indirectly)."

Let's see if I've understood it correctly ... are you saying that KinHPo has 'upgraded' the 32SII with an amazing 32 Kbytes of RAM, yet you can't have more than 26 or so data registers, so that you can handle matrices only up to 5x5 !?

Because most of my vintage SHARPs, even models much smaller in size than the 33S, can easily define in excess of 1,000 data registers or matrix elements (10 Kb), my incredibly small Casio FX-7500G (easily the smallest calc I own, despite its 8 lines x 16 char graphics screen) can allocate as many as 526 memories (3.9 Kb), and even my superb, 20-year old HP-15C (0.5 Kb) can handle 8x8 matrices.

If true, this does nothing but reinforce my point: there's no enjoyment nor pride in owning such a calculator, as we are painfully aware of just how little thought and effort have been invested in creating it.

Do you think the great HP engineer teams of the past, which painstakingly fine-tuned every feature of the calcs they designed to ensure utterly optimum performance for their intended professional customers, would've ever designed something like this ? 32 Kb for programs, yet no I/O and you can handle just 25-30 data elements ? Wouldn't have they felt completely ashamed if such a glaring case of bad engineering and careless design had slipped by ?

I rest my case: the former valued, respected, professional customers that we once were have yet again been absolutely neglected by KinHPo, which will only strive to maximize their profits by releasing sub-sub-optimal products, where noone has invested any time at all to try and improve neither features nor quality. It was easier to copy-paste the 32SII ROM and manual, add a little here and there, and that's all. So the user cannot use those 32 Kb for data ? Who cares ? After all, neither could the HP-12C Platinum user use those 400 bytes for extra registers. We're including a 32 Kb RAM chip because it costs next to nothing nowadays, but making firmware modifications so the user can actually use it *does* cost money, so there you are, take it or leave it.

Proud of owning such a state-of-the-art, quality product ? Enjoyment ?

Not me. This only adds insult to injury.

Best regards from V.


Valentin --

That really "hit the nail on the head" and "cut through the cr@p" (or any other similar idiomatic expression).

Based on what I've been reading here, I have no inclination to reward "KinHPo" by purchasing any of these "upgrades". Consider:

12C Platinum -- Slower TVM algorithm despite presumably faster processor; extra RAM but no practical means of using it; no improvement to limited programming capabilities; algebraic mode added without parentheses.

49G+ -- Terrible quality control and buggy ROM.

33S -- Bizarre stylized keyboard; much extra RAM but no practical means of using it; failure to address limitations of predecessor 32Sii (e.g., "pidgin" complex-number functionality and no built-in matrix functionality) that were not limitations in the long-discontinued 15C and 42S.

What's the chance of a well-done new product?

Valentin is absolutely right -- they simply took the predecessor products and tacked on a few things in a slipshod fashion, using updated hardware, but without "sweating the details".


BTW, Valentin, on an unrelated topic:

I appreciate your offer of assistance with the 71B. It may be a while before I ask for some.

I don't plan to take the time to become a real expert. Instead, I just want to learn my way around it, write and download a few programs to run, and generally see what it's all about.


-- Karl Schneider


I look at it this way.

The entire project took perhaps 12 months. In that short time period, HP appears to have said:

"Wow, since the 32SII was discontinued, we haven't had a product at that price/performance point. What can we do in the next 12 months to remedy that?...Let's make a 33S. It will be the same functionality as the 32SII. What size memory chips are available? ... If they are the same price, let's drop in the 32K chip...perhaps our users can make use of it."

The modifications you mention WOULD in fact have required massive changes to the internals of the 33s.

It would not have been done in 12 months...particularly given the large # of tasks underway (49G+, 48GII, others).

Want something new and completely different? Be patient.

The 33S will be selling for a lot less than the discontinued 32SII on ebay. People can have the functionality of the 32SII, probably faster (I dont' know) and have it for a lot less.

Oh, by the way, there's a lot more memory than the 32SII for programs.



IIRC, it took HP 9 months from the beginning of the project to production of the first HP 28C protoype. That was an entirely new calculator. So, 12 months to release an already existing calculator with minor modifications sounds not impressive to me.

I think Valentin is right: They put the least possible effort in this product, which is probably almost zero...

Just my 2 cents,


HP had a boatload of people working on the 28C, and according to the HP Journal article, it took much more than 9 months.

They do not today. A great deal is outsourced. They also have reissued/redone the 12c platinum, the 17BII+, the 49G+, the 48GII, and according to manuals found on HP's website, a 39G+ and 33S. Not to mention the rumored high-end finance machine.

That's a lot for the small amount of resources they have.

Again, half-full or rip-off, gone-to-the-dogs, Engineer-hating half-empty?



Hmmm... My memory was wrong, then. This happens once you're old enough :-( Besides, I'm not unhappy at all. I thought, like many probably did, that HP had abandoned the calculator business forever, and the relaunch of new calculators comes as a positive surprise to me. I'm especially speaking of the 49G+ which improves all that we didn't like on the 49G.

OTOH, judging from the papers, the 33S doesn't impress me so far. Not a lot of effort sems to have gone into its development. That's actually what I wanted to say. See, we're spoiled by the degree of perfection a 15C can offer. A lot of us had probably hoped that a successor of the 32Sii would be more like a 15C, or 42S. But it's, as you say, different times. The amount of resources that go into the development of calculators appears to be only a fraction of what it used to be.

Then again, judging from the 32S/Sii prices on ebay, these calcs aren't sooo bad even just the way they are...

Chers, Victor


Personally, I think that HP is not making their new products for those who already have 40 calculators. At least based on the color schemes, and addition of Algebraic mode, it seems to me that they are trying to get a foot into the student market. The 33s will fill a niche.

Go to the store today and try to find a programmable scientific calculator. Better yet, try and find an RPN calculator. How about finding a scientific calculator that can actually take numbers above 10^99. I am pretty certain that you will not find a current scientific calculator that will do it.

They are aiming this product at the youth, who don't have the all-powerful HP-15c, or 42s, or 41cv. They are aiming it at my generation, and I am going to buy one as soon as the are released. The 32sii that I call my own really belongs to my school, so as soon as I graduate (next year), I will no longer have a good scientific calculator.

I still have hope for the 33s. If you notice in the product specs released on the HP website, there is a direct reference to the types of keys.
"The Material of The Key Top : Plastic"

Furthermore, there obviously was some work on the ROM, and for that I have much thanks. On my 32sii, I very often use the curve fitting features, but not every function is linear.

It is also quite obvious that they have adapted the calculator since the first picture was released. Notice they moved the "ALG" and "RPN" key. Personally, I believe this will be a decent calculator It will not compare to the 41, or the 15, or the 42, but notice who HP has to compete with now. There are no longer companies making good sturdy calculators. If you want a scientific calculator now, you can buy a TI-30 or its like, a Casio (which is probably the best that you can buy today), a Sharp 4 function, or one of those free "pocket calculators". I believe that with the proper marketing (such as sending some models to high schools because it is teachers who have the most influence on the calculators that kids buy), that the HP33s can have the market.

I know it will have the market in Texas. My school is planning on buying 20 or so as soon as they come out for our UIL team.

-Ben Salinas


It is good to have some young people with a bit of sense on this forum. I will as well buy the 33s as soon as I can. I will have a 32sii and a 20s in one calc without fear of losing them.
I am sure lots of people who buy calcs because they need them will be impressed with it. If enough publicity is made..
I even think the cowcatcher design might me more ergonomic than the straight one.
I want my 33s NOW!



"H-P is producing several RPN models again ... "

"The new models, whatever their quirks, represent attempts to respond to customer requests ... "

"There are positives and negatives associated with everything -- just because the new models don't live up to the old company's standards in all ways, they still have their worthy attributes ... "

"I should resolve to see progress as well as shortcomings, and to balance the one with the other ... "

"I am an H-P calculator addict. I cannot cure my devotion to their old models, I can only control it."



Thanks Katie!

So, in the end, that strange icon on the display someone was wondering about is the "B"usy annunciator...



I wonder how long the "busy" indicator will be on for? If is was well engineered from a software standpoint [which is a big question] it should run pretty fast. They seemed to have traded space for faster execution time [or maybe just for bad programming]. Although the manual gives no explicit list of how much memory each instruction takes up the program examples show it now uses 3 bytes minimum per instruction with (all?) numbers taking up 12 bytes!

I'm not at all sure what they put in all those bytes, but they sure have a lot of room for more functions, etc.. Also I'm really curious about this keyboard symbol/ascii listing. Does this hint at some sort of printer interface?

Edited: 10 Nov 2003, 3:06 p.m.


I believe an hp33s font should be available somewhere to help write docs.
ASCII Ma (Taiwanese Chinese so it comes from Kimpo)on the left is the ASCII code.
I don't know what the item column is and as far as I am concerned it could be removed.
Symbol is what the font will display when you use this character. So you can show the key press in your documents.
ZiYuan on the right is the usual character associated with that ASCII code.
Now I just need the font and a 33s.
Please hp-Kimpo soon!



I poked around in an attempt to assuage my worst fear concerning the 33s: that the two-line display isn't supported or used in the manner that we (especially 42s users) might expect.

The only thing that seems to offer hope is the discussion of COMPLEX mode, and how the imaginary part, in Y, is displayed on the first line, with the real part in the second line (X). (That's still kludgy enough, given there are 14 digits to play with . . . )

That would lead me to believe that Y is generally displayed above, and X below, as I would hope. (I would also hope that two lines of program are displayed, not one, while in program mode.)

Well, I haven't actually READ the whole thing, but some keyword searches haven't turned up much else. There doesn't seem to be any explicit statement of whether Y is generally displayed above, in addition to X below. I didn't notice any extension of the programming capabilities to utilize the two displays rather than one. (Somebody, help me!)

There is, however, page 2-3, which explains that the X<>Y key is often used twice in succession, to briefly view the value in register Y and return it . . .

I can pretty much write off the cowcatcher as a bad marketing idea, and 26 registers with 32K program memory as a hasty approach to adding something without drastically changing the 32sII code base. But if the hyped two-line display is really a one-line display with another line for whatever, well that will be another enormous bummer!

I worried about this when I first saw the product sheet, and hope is fading fast . . . I may have to join the complainers with this one.


Just refer to page 2-2 of the 33S manual:

The X and Y–Registers are what you see except when a menu, a message, or a program line is being displayed.

Nothing is said, however, about two program lines being visible when reviewing a program. Much more severe is the fact that there are only 26 variables (plus the six statistics variables) and only 26 lables to fill 32kBytes of memory! I have used a lot more labels in my HP-41 programs that couldn't be nearly as large...

Cheers, Victor


Well, I guess it's not quite as bad as I feared.

I notice that immediately below the diagram with the two "displayed" registers in bold outline, it says 'The most "recent" number is in the X-register: this is the number you see in the display'. [Emphasis theirs, but I concur.] Apparently that graphic is one of the few changes made -- and the text hasn't exactly been gone over with a fine-toothed comb.

I noticed elsewhere that the VIEW command (e.g., "VIEW X" ) apparently causes "X=" to be put in the upper line, and the X value in the lower. (I wondered, if that's all they're doing with that upper line . . . )

H-P, please forgive my concern. Never should I have doubted ye! ;^)


My guess is HP said something like this:

"We're really moving the 32SII to a new package and aren't going to be making many changes to it at all. We can't sell the old one because we can't make them any more. We face a choice guys...the cost of a 2K chip is the same (within pennies) as a 32K ram chip...which do we put in?"

They went with the 32K ram. Will it be entirely useful? I don't know...I expect perhaps not.

But, it is there to try to use.

If HP reads this, perhaps next time they decide to use the 2K chip.

Lot of people here are of the "half empty" type. Try thinking "half-full" at least once a day. You'll live longer. :-)


I used to think of myself as generally having a "half-full" kind of outlook -- I'll have to ponder that . . .

Yes, I'm sure that "more" sounded better than "some" for the same price.

Bottom line: "more" than 390 bytes is all I wanted (storage-wise, at least) in a successor to the 32S/SII. No sense getting worked up about the program vs. variable space being out of balance, so long as there's more of the former than there used to be.

No doubt we'll be in-lining even lengthy subroutines to conserve labels!

Maybe the 33sII will allow for two-character labels and variable names?

-- Paul B.


Lot of people here are of the "half empty" type. Try thinking "half-full" at least once a day. You'll live longer. :-)

I'm the "half full" type in most things, but not where HP is concerned. It seems to me like they've thrown away the half-full glass altogether and replaced it with a nearly-empty, leaky paper cup, and hired a pack of lying weasels to convince us it's a brim-full crystal decanter.

If it sounds like I view HP with cynicism, distrust and contempt then you're getting the right impression. Over the past couple of years I've come to assume at the outset that anything HP does is a cheat or a swindle of some kind, and the onus is on them to prove me wrong.


I haven't checked yet, but maybe not only the calculator, but also the manual was more or less coiped from the 32Sii? That would explain these incosistencies. I mean, why should they rewrite the manual if the calc is essentially the same? And in the 32Sii, only the x register is displayed...

Just another 2 cents, Victor

Possibly Related Threads...
Thread Author Replies Views Last Post
  Is the HP33S is the only game in town? John Noble 28 1,032 11-16-2009, 10:18 AM
Last Post: John Noble
  Sudden death of HP33S batteries Dave Shaffer (Arizona) 1 123 03-12-2008, 11:54 PM
Last Post: Trent Moseley
  HP33s has more programing power than one might think Howard Boardman 6 364 08-23-2007, 10:20 AM
Last Post: Dallas Osborne
  HP33s - a true confession Les Wright 62 1,186 11-28-2006, 05:18 AM
Last Post: Valentin Albillo
  HP33s and HP32sII programmation difference David Boileau 7 294 07-13-2006, 04:05 PM
Last Post: Paul Brogger
  HP33S Check digit Derek 18 605 06-14-2006, 10:09 AM
Last Post: Ivan Nejgebauer
  HP33s Keyboard Issues John Daly 12 363 04-05-2006, 11:02 AM
Last Post: e.young
  HP33s CNA54 - improved one? Gour 2 148 02-17-2006, 02:19 PM
Last Post: Gour
  HP33s keyboard Al In FL 7 296 01-20-2006, 01:03 PM
Last Post: Paul Brogger
  hp 12cp and hp33s E. Young 39 1,129 01-13-2006, 12:06 PM
Last Post: Wayne Brown

Forum Jump: