HP 49g+


Well, I got my 49g+ this morning, and opened it up. I haven't had much time to learn how to do everything, and won't have time until saturday, but one of the things I was wondering about is the speed in comparison to the TI-89.
I am planning on having a full test completed this weekend (HP 49g+, TI-89, HP-48gx (belongs to a friend), HP 32sII (just for kicks))
This morning, I quickly tried doing some factorials on this calculator to test the speed. I decided to use 252! since that is just about the highes number the 32sii can give you. The 49g+ takes about 1.5 seconds to calculate this (giving you all the digits) The TI-89, to do the same thing takes 10 seconds. No joke! Furthermore, on teh 49g+, multiplying or doing common operations with this 498 (or so) digit number occur instantaneous (just like if u were adding 1 and 1). On the TI-89, about 13 seconds to multiply 252! by 270. The 49g+ takes less than 1 second!

I am personally amazed by the speed of the 49g+

As far as the keyboard, it is great. I havent had any problems with it yes.

I have a "Phase three- marketing sample, not for resale (dated) 03/07/30)"
Serial Number: CN33103042
It came with a USB cable, the leather case, a small manual (only a few hundred pages), and a CD with the conectivity software and the PDF manual. It is easy to tell that this is a pre-release version of the calculator because the CD is a burnt CD with the title written in sharpie:).

As far as news on the other calculators. I talked to Tony Jones (whom some of you all will be seeing at the HP calculator conference this weekend) and he said that the reason they did not give me a 33s is because they do not have a marketable version yet. I dont know what that means.

He really didnt talk as much about the new calculators as I hoped.

Well, that's it for now... I will keep you posted


Congratulations on your new 49G+!



WHERE did you purchase your HP 49g+ from? I haven't been able to purchase one here in the states, unless I am looking in the wrong location.



I actually won it in the HP Create a Calculator Contest. It is not up for purchase in the US right now. I believe they will be released here on October 6th.


I saw it for preorder at $160, minus shipping, at Samson Cables' website (I preordered!).

Oh, how did my beloved 32SII do in the factorial contest?


Hi, Ben!
Can you send your contribution to the contest to my e-mail, please?

vDOpNOTnSPAM@welho.com, remove DO NOT SPAM, VPN


see Ben's 51S


> As far as news on the other calculators. I talked to Tony Jones (whom some of you all will be seeing at the HP calculator conference this weekend) and he said that the reason they did not give me a 33s is because they do not have a marketable version yet. I dont know what that means.

It might sound a little bit strange but, personally, I find the news of a delayed HP-33S good news indeeed! Maybe, after all, HP are listening to our user reactions and are taking it back to the drawing board making it into what it really should be:

-Backwards compatible with the HP-42S

-RPN only (large ENTER key in the classical position)

-Equipped with serial I/O

Anything else would, in my opinion, be a mistake.


Erik Ehrling (Sweden)



I have to agree with Erik - the 33S as it appears now it is UGLY, with a capital UG. And what's the deal with that weird chevron keyboard? Overall, the calculator looks like something Ultraman would have used back in the '60s! I When I first heard about the 33S, I had high hopes of nabbing one to take the wear and tear off my trusty 32SII. I reckon I'll keep waiting...


Why don't you try to get hold of another 32SII? I managed to grab one for 35 euros (less than US $35) a month ago. It is a silver bezel with some "sticky keys" but now I can put my brown bezel into storage.


Good idea, Bert, and it's one I'm pursuing. A "lightly used" silver-bezel unit would be perfect for my desk drawer at work.


They're probably taking more time just to find ways to make it cheaper and increase the profit margin. They seem to think that by making it look like a cell-phone that it will be more marketable.

Tom Scott


There's growing interest in the HP33S, even among people who have never owned an HP calculator before, because of the recently-announced calculator restrictions imposed by the National Council of Examiners for Engineering and Surveyoring (NCEES). NCCES prepares professional licensing examinations which are used throughout the U.S. for engineers and surveyors.

Starting in 2004, NCEES has banned all calculators with text-editing capability, including (but not limited to) the HP48, HP49, TI83, and TI89. It looks like all currently available graphing calculators may be unacceptable under their definitions.

Programmable calculators are OK, as long as they are not graphing calculators. Unfortunately, it seems that nobody makes programmable non-graphing calculators anymore. The HP32SII may have been the last one.

NCEES says that the 32SII is acceptable, but it has several drawbacks. Since it's been discontinued, it's expensive and hard to find. It has a relatively small memory. And it's RPN only (obviously this is not a problem for everyone). The same drawbacks apply to older HP calcs.

If the 33S appears as scheduled in October, then it should resolve these issues. The 33S should be relatively inexpensive, will offer 32K of memory, and the choice of algebraic or RPN mode. Too bad it's so ugly. But there's still a lot of people who are prepared to buy one, because they need something that is both (a) programmable and (b) legal for use on professional engineering or surveying exams. The 33S may have this market to itself; I don't think TI, Casio, or Sharp offer any models that qualify.

This is still not a big win for HP, though. The new NCEES calculator policy may increase sales of the 33S, but it will also cripple the sales of the more expensive 48GXII and 49G+.


(and yes, we have power.... well, at work, anyway. At home it's all laptop)

I think that the last non graphing programmable HP was producing was actually the 20S.

There are some other programmables out there, I just haen't been able to get any of them in the US. I think 2 or maybe 3 of the citizen models have programming capability of some sort.

I think that trying to force the 33S to be a replacement for the 41CX is a losing battle- serial IO, 42S/41C code compatibility, and all.

The calculator seems obviously to be a follow on to the- here's where the detective work comes in- 32Sii. As such, it seems great. Well, aside from the freaked out keyboard.

Not every calculator *must* be another rebadged 41/48. There really is a place for a smaller, simpler, solid programmable.

I'm as much addicted to serial IO as anyone- and I desperately want it for my 42. But *not* for my 32. Even with practically endless RAM. It's just not the same machine.



You don't think a 32SII or 34C type machine could use a little non-keyboard I/O? Back when my 34C was my only calculator, I dreamt of having the ability to do stuff on it and then upload it to a PC or minicomp, and vice versa. These "low end" (not in my opinion, though, I think they're totally great) calculators are great for most things, small programs, a bit of statistics, and repetitious calculations not requiring humongous equations, which is, in my opinion, what the bulk of calculator work for many of us is.

If I could retrofit my 34C or 32SII with serial I/O... what is it that the kids say nowadays, "Woot?"



I desperately want a pioneer with serial IO. I admit it. but it has to be the 42S. (even better in my mind would be a 42 ROM with IO in a clamshell, but.....)

I just don't think the 32S is one of those- nor any upgrades. There are reasons- there will always be a testing environment where the serial IO disqualifies the calc. There will always be a price difference.

there's a place for more than one hp rpn scientific programmable, that's all :)


I'll agree with that! There DOES need to be more than one HP rpn programmable.

I read somewhere in this forum that it would so sweet if they reissued the 15C or 16C or 32S or 32SII or 42S. I hope HP monitors the stuff on this forum... nahhhh.


Yes, I would happily see A HP-15C Platinum alongside a HP-42S Platinum (with serial I/O). But if I had to choose I would pick the latter...

Best regards,
Erik Ehrling


The 33s on the Samson Cables website. ETA is 10/10. It will be interesting to see what it is like.


While I cannot deny the brilliance of your suggested changes, Erik, it is exceedingly unlikely that HP would do this based just on the reaction of the people who have seen the leaked 33S information.

What you suggest would be a major revamp, requiring significant new engineering and test. I'm not in the manufacturing business, but I would reckon the changes you asked for would take a minimum of 6 months to engineer, not to mention test.

More likely is that they've found bugs through beta testing and need to produce a new ROM image.


... it's not surprising (indeed, it's to be hoped!) that they'll be a bit more careful in testing before committing to production.

Sure it's ugly, but if it's got the features advertised, significant usable program memory and decent keyboard action, I'll no doubt find a way to overlook the keyboard's layout . . .

(My youngest thought it looked like a remote control -- I countered with a cell phone comparison, and he said he'd rather have a cell phone than a calculator . . . )


Yes! Surely it would be a major revamp, however much of the work has already been pursued in the making of the new HP49G+: The Saturn emulation on the ARM architecture is already there and the ROM of the HP-42S has been with us for a very long time. OK, it would be naive of me to say that these changes are either small or of a kind that could be implemented in a very short time. But many of the improvements of the HP49G+ over the HP49G seem to indicate that HP are actually listening to what the consumers care about - and a very large proportion of the user community are spelling out the same "mantra": Keystroke programmable, HP-42S compatible, I/O... (over and over again).

Best regards,
Erik Ehrling (Sweden)


If the 33S becomes an acceptable replacement for the 32S and 32Sii (despite keyboard layout),

If the "Saturn emulation on ARM..."is so real, and ROM images are surely available to HP...

Then, shouldn't many of us wait and hope and prey and ask for a 43S?

(But we should not expect the 33S to replace the 42S; it may be fine if it just covers the 32S model spot)


I think you make an unfortunate error in assuming the existence and availability of the 42S ROM helps very much in the building of a new calculator.

Based on the nature of the bugs found in the 12C Platinum, I would venture that it's ROM is a brand new implementation, not a modification/extension of the one in the 12C Classic.

HP has apparently lost the corporate knowledge that would make the 12C's ROM reusable -- perhaps because of all the corporate shuffle with ACO and the Compaq merger -- who knows?

I suspect the same is true of the other older calculator technologies, like the 42S. The ROM listing itself would be a good start, but you'd need to have the pre-compiled listing (commented!), at least, to make it feasible to re-use it.


I was just quoting the previous post, which states:

"The Saturn emulation on the ARM architecture is already there and the ROM of the HP-42S has been with us for a very long time."

I was just trying to suggest that,

IF these conditions were true (which I cannot prove or deny)

, THEN a newer 42S style machine COULD be POSSIBLE

BUT I think the 33S is not such machine.

Now, IF a Saturn emulation is available on ARM (the 12Plat failures are not proof on the contrary, because the 12Plat is not ARM based, as the 49G+ is supposed to be), the ROM images may be enough, even without all the knowledge.

I think this has nothing to do with the Compaq merger. Time and life of both products and people are enough to cause knowledge loss.

Examples as eV41, HP41X, HP42X, which some people successfully developed to emulate HP calculators on other platforms show that ROM images are quite useful to this end.

Just my $ 0.01, these are just some ideas as I have no access to HP internal information, or to 33S or 49G+ units. The use of IF and THEN should indicate the conditioned and conditional nature of these thoughts.

And sorry for my "unfortunate error", but I consider that elaborating upon someone else ideas is not the same than assuming them as proven; which I certainly tried to avoid most of the time.


I totally agree with you Andrés, and my personal hope is that the reason why HP has not yet released the HP-42S ROMs in the Public Domain is that they have not yet ruled out such an option...

In effect, building a HP-42S Platinum around the ARM-based emulation from the HP-49G+ should theoretically even be cheaper than designing the HP-33S from scratch.

Best regards,
Erik Ehrling


Thank you, Erik, for your comments. Let's keep a little hope on the "43S" idea...


I seemed to have given offence. My apologies, Andres. That was certainly not my intent.

By using the word "unfortunate" what I meant was that it is unfortunate that HP has lost this corporate knowledge. It would generally be a reasonable assumption that technical knowledge gained in producing one product could be used advantageously in the development of a new, similar one. My point is that, in my opinion, HP no longer has sufficient corporate knowledge to leverage past successes. I use the nature of the HP-12CP as my rationale.

Again, sorry if I offended you.



Of course, there was no offense. Sometimes I feel I was unable (perhaps due to language issues) to state clearly my ideas. That was the case with my first posting. Let me apologize if my intent of clarification suggested that I was offended, and let us not enter in an endless apologies loop.

Thank you again, and best regards.


Hi Erik,

I am an engineer in the consumer electronics industry, and unfortunately it is very unlikely that they can make radical changes to this design this late in the game (unfortunately).

Probably they are frantically pressuring their supplier to "dial-in" the production of new 49 and 48, particularly the keyboard. It takes time to get a new supplier to understand your needs and hold them to your standards. Also, it's good marketing to introduce the expensive models first.

Luckily this means that the problens we are seeing on the 49G+ are likely to be tweaked by the time the 33S rolls out.

Here's my hope: if they are still in production with the 12C, why not use the tooling to do a run of 15C's? Every engineer on the planet who graduated before 1990 would buy one.

Best regards,

Alex P


The 42S calculates 252! in about 1.5 sec, also. For even more impressive speed, try gamma(253.000000001) on the 42S or x! of 252.000000001 on the 32Sii and compare the answers.

Something must be wrong with the TI-89 to take 13 sec for that operation. I'll check with a co-worker on that one...


He was talking about the full 498-digit answer (vs the
12-digit floating-point approximation of the 42S).



Something indeed is wrong with the TI-89... It has about the slowest display of large integers found anywere! Sombody (i dont remember the reference) did a disassembly, and it was not impressive reading...


My TI-92 Plus takes about 1 sec to calculate 253! in approximate mode and 6 seconds in exact mode. It takes about a minute to scroll through all of the digits in the exact mode answer. :)


My timings for 253! on my TI Voyage200 are similar to the reported timings for the TI-92+, namely:

about 1 second in approximate mode.

about 7 seconds in exact mode.

about 50 seconds to scroll horizontally through all the digits of the exact answer.



Yes, the time I provided for the 49g+ (1.5 seconds) was for exact mode.
Some more times
252! in exact mode: 1.5 seconds
252! in scientific notation:the closest I could get with a stopwatch was .37 seconds, but it is considerabley less than that... pretty much instantaneous
To scroll through all the digits: 40 seconds

252! in scientific notation: 1.5 seconds

Now the question is, can you really test the speed of a calculator using the x! function, or does that simply test the speed of the algorithm?


It used to take a minute to calculate 1000! on the old 49G, the new plus does it (Exact, no LASTARG/Stack) in 26 seconds (2568 digits). So the speed is almost exactly four fold in this case. The smaller 252! takes only 1 second (498 digits). How about V200 with 1000! or whatever is the largest it can do? My code was 1000{!}HEAD MEM;TEVAL - VPN


Wow.... 26 seconds... that's a long time. The TI-89 does the same calculation instantaneous... of course its answer is 1000!:)


Don't fool around Ben! (: how much is Ben! ?) What is the largest it can do? 999! ? - VPN


Hey wait a minute! with the HP-48S it also took half a second to do 252! but you only have 13 significants numbers and the EE. With the TI-89, you have all the 497 numbers calculated insted of 13!!!

I don't know if the HP-49G+ calculate ALL the digit of 252! or only the usual 13 numbers and the exponent. If yes, that's an explanation of the apparent big difference of speed between the calculators...


Did you get a chance to play with it? Whaddya think?

Congrats, by the way!


Sorry it took so long for me to reply, but here we go.
I have been using my free time to figure out all of the basic features. FYI, this is the first graphing calculator that I really have learned to use, so some of my observations might be flawed. Most of what I am comparing it to is my 32sii (well, my school's) and my ti-89 (also my school's)

The first thing that i noticed is the speed in graphing functions. This is in comparison to the TI-89 (which is the one function I used on the 89). It graphs them 3-4 times faster, and many times faster than my friend's 48gx.

The feel of the keyboard is also great. It has that "click back" so to speak.

One worry is of the LCD screen. It seems to not have enough padding around it, so that when you press in the back, or front you will turn the pixels on (i dont know how else to say it)
Also, today I was using the calculator, and suddenly 1/4 of the screen on the right side went permanently on. I quickly reset the calculator (using on and f3) but that had no effect. I pressed a few buttons at random and the screen was back to normal. I think this might be a design flaw. Afterwards, i ran a test on the LCD screen and there werent any broken pixels

As for battery life, I have had it for almost 2 weeks, and the batteries that it came with are still running fine. Of course this is with less than an hour of use each day. I will alert you when the batteries do die.

As for cosmetic appearance, it looks extremely classy (who knows waht people will think about it in a few years) The gold gives it an extremely elegant look (and I know that is why I buy a calculator, for what it looks like :) )

I dont have a SD card to test it out with, and havent copied any files to the calculator yet, although it connected to the computer with no problems. As soon as i figure out how to copy files, i will give a speed test on that.

The quick start manual is excellent, although it is a bit redundant and complicated in places due to having to explain stuff in RPN and Algebraic mode. My one regret is it doesnt have an index. I have hardly used the manual on the computer, as it is cumbersome to have to go to my computer, open the pdf, and then do a search for what I want to know. Most of the pdf is pictures, so you can't search, and the pdf has no bookmarks, so if you want to go to page 13-2 you have to scroll down till you see it. I would much perfer a printed manual.

I personally perfer the interface of scientific calculators over that of graphing calculators, because it is so much easier to input data in scientific. That being said, the 49g+ is by far the best graphing calculator that I have ever touched for doing basic math (my experience includes using friends ti-83's, a casio color graphing one, the ti-89, and a few other) It might be because it is RPN, but the keyboard is laid out very well. As for the keyboard layout, my one gripe is that the space key is at the bottom next to the enter key, considering i have used the space key all of twice, i think it should be an alpha function. And actually, having the enter key down at the bottom is not nearly as nice as the wide, mid keyboard enter key, but it is not as bad as people make it out to be.

Overall, this calculator is a great graphing calculator (especially because of teh speed), and a decent scientific calculator replacement. Personally, for most of my school work, I will stick to my 32sii (even for calculus), because I am fast with it, and it fits so much better in my hands, but if i ever am lazy and need to grpah something, or solve a complex equation, or my 32sii is far away, i will reach for the 49g+ over the ti-89.

(oh and the leather case is great. I have dropped it a few times, and it just bounces to a soft fall)

So, in conclusion, if you like powerful graphing calculators, then this is most definitely the calculator for you. It is most definitely worth the extra 50-60 bucks in comparison to the ti-89 (and really and truly, it competes, and wins with the ti-92, plus the 49g+ is much smaller than the ti-92)

If you want a scientific calculator, find a nice old HP, or get this one, and get used to the weird enter key.



Nice to see you having a great impression of the HP-49G+ and gave us such a detailed report!

But even though you don't have a SD card, could you tell us from the quick start manual or CD manual what is the maximum (or minimum) card memory capacity (e.g., 64 MB or 128MB, etc.) is acceptable to the calculator?

Incidentally, I don't mean to step on toes and as an old guy, I once lusted after (and still don't have) one of those shining 41C series machines. But having the ability to use (RELATIVELY) cheap SD chips is a major plus over the 41CX or even 48GX, all other considerations aside.


Tested MMC 8MB from Nokia Communicator 9110i and a fresh 64MB SD card (Sandisk). Using *standard* DOS FAT16 gives a 2GB max card size/32KB clusters. Formatting the clusters bigger, might give a bigger card size, like 4GB/64K. That I tested on a PC HD drive, because the maximum SD on shops is currently 512MB. One may order 1GB.

How much do you need for a calculator? The 8MB MMC, limited by some Nokia programs, has enough room for all the programs that I have used intensively during past 4 years in my old 49G. So you may buy the least expensive one. If you use the card for something else, too, like in a camera, consider buying the largest possible. If you need Nokia compatibility then the best choice would be a 64MB MMC card. That's what I'm going to buy next! * VPN #


Wait a minute... the 49G+ can FORMAT a SD chip?? Or, do you do that first on a PC? Now, my ignorance will show- I thought that if you formatted using larger cluster sizes, you get effectively a smaller drive.

Anyway, I agree that a calculator should require gigabytes of memory, but I was more worried that there might be some minimum size requirement. Heck, the cheapest one it is! I'll be looking for a SanDisk/Lexar/etc. sale in local stores. 64MB should be more than enough until someone invents really bloated programs or starts to stuff high definition images and such on a calculator!


I appreciate your response. Enjoy your new toy!

Possibly Related Threads...
Thread Author Replies Views Last Post
  Re: Porting 49G Programs to 49G+/50G Les Wright 0 386 02-21-2007, 10:16 AM
Last Post: Les Wright
  Porting 49G Programs to 49G+/50G Les Wright 6 975 02-20-2007, 06:27 PM
Last Post: Tim Wessman
  TI-57 Emulator for HP-48S/SX/G/G+/GX, HP-49G and HP-49G+ is finished HrastProgrammer 9 1,227 01-17-2006, 12:57 AM
Last Post: HrastProgrammer
  HP 49G vs. HP 49G+ Prices Ed Look 5 712 10-24-2003, 12:43 AM
Last Post: Ed Look
  HP 49G Keyboard and future HP 49g's Nick Nicholas 1 421 05-31-2003, 10:26 AM
Last Post: R Lion

Forum Jump: