Shame on the College Board. Shame, shame, shame! Sell the minimum and charge the maximum! Shame on TI, too.

For sure my old HP50g will soon have a beautiful sister!!!

But i'm a bit disappointed about the font math noticeable everywhere in screen-shot...

- * in place of dot product,
- poor sqrt sign;
- distance between base line character and exponent;
- no different size for subscript, exponent;
- etc...

On this point TI Nspire seems to offer much better readability.

What's your opinion ? Can be hoped for something more aesthetic ?

It's part of it; it's not the entire "problem". The issue is also that teachers help maintain the monopoly by instructing students to get TI's calculators, and TI caters to teachers with everything from lesson enrichment to technical support and the T^3 conference.

Don't forget tossing money to textbook companies to put TI keystroke sequences in. That's the one I find absolutely disgusting.

Teaching teachers to teach the TI calculators is fine, IMO (although balance is needed, and Casio and HP really should invest in doing that, too - and it's best to teach the concepts that adapt to any calculator, rather than teach the TI UI only), but "teaching" kids to spam button sequences instead of actually figuring out the answer is horrible, and TI's been doing that for a long time - I recall in high school trig, my school was using a textbook printed in the mid 1990s, and it had TI-82 keystroke sequence examples. (My teacher made sure we actually learned the content, not the button sequences, being careful to use the calculator as a tool, not a magic answer generator box, but still...)

That sounds largely like hearsay to me; as you know, I've written two books about TI calculators, and I've gotten nothing from TI other than polite good wishes (and a TI-84+CSE, but that was Cemetech-related).

I think that's mostly within the purview of the textbook writer, wouldn't you say? For example, in my upcoming "Using the TI-83 Plus/TI-84 Plus", I teach readers to gradually build up their understanding of the key sequences, so that by later chapters I tell them to (for example) switch to Radian mode but don't constantly remind them where the MODE key is.

I'd argue that it's the purview of the textbook writer, school board, and teacher.

The textbook writer shouldn't write button presses to the detriment of the lesson, the school board shouldn't buy books that do that, and the teachers should, if required to use such books, structure their lessons such that the button-pressers learn how to do it right.

I think they forgot to mention that the name "Prime" also suggests that there's "more to it than meets the eye" and is probably really a transforming autonomous robot?

I agree with willwac that the keyboard arrangement is kind of strange, but I too am a TI devotee, so any departure therefrom will obviously require some adjustment.

My main issue with this calculator, however, is the same issue that I have with all modern calc lines: they're becoming too gosh-darn complicated! I like that TI-83 and 84 calcs turn on to a blank homescreen and keep all their insane capabilities "under the hood." These new devices with giant menus when you first turn them on, 15132448 different Apps that were only put on for standardized tests, and unnecessary features like touchscreen and color, are doing nothing but jacking up prices and making us lose our grasp of a once blatant truth: that calculators are meant for solving math problems! Complex geometry modeling, 3d imaging, and spreadsheets are all things that should be implemented on computer software, not on a calculator, and I don't think that anyone is really going to use them on a calculator. And I'm talking at TI here too, not just HP. The TI-89 and TI-Nspire line kind of started all this nonsense as far as I can tell, for instance, by adding a notecard app to the 89. Who would really ever use that? I know I'm getting up on my soap box, and I strongly invite any counter arguments, but it just seems to me that calc manufacturers are no longer trying to compete with just other calcs; they're trying to compete with smart phones, tablets, and laptops too! As Kerm said above, you can completely understand a TI-84 calculator; but I don't think that anyone will ever understand everything about one of these new devices, unless they WORK for the manufacturer or devote their life to the task...which just seems like a lot of effort to put into what was once a very simple machine.

The keyboard layout mostly makes sense if you're used to HP, and almost completely makes sense if you're used to the 38/39/40/39gII lines.

There are some arguments about implementation details - the D-pad, the position of Enter and the operators (but that's been a long-time argument in the HP community, ever since the Voyagers came out in 1981), and the fact that x^2 is unshifted and SQRT is shifted - but overall, it's consistent with HP's history.

Now, as far as the complexity argument goes... I'd argue that one person can definitely understand these machines, although it is quite a bit harder (but the CAS system that forms the basis of the 39gII's OS was written by one person). And, they're set up so that you don't need to understand the whole machine to accomplish the same tasks that used to require whole-machine understanding. (And, I'm under the impression that a lot of the impressive software for TI's calcs doesn't really use the OS much at all, instead just twiddling the hardware directly, and it looks like the Prime will be fairly simple in that regard - and, HP's stuff tends to make it easier to use the OS while also twiddling the hardware, from what I've seen.)

Don't get me wrong, machines that are simple enough for a single person to understand them are quite nice, but that shouldn't be an argument against adding capabilities, especially when they benefit people who will never have an interest in learning a whole machine just to do an advanced function.

Also, re: calculators having to compete against phones, tablets, and laptops... yes, they are doing that to a degree. There's a few reasons for that, some of them historical, some of them current.

Remember, the modern tablet market grew from the smartphone/PDA market. The smartphone/PDA market, at least as far as programmability goes, grew from the programmable calculator market (what you would've used a Palm device for in the late 90s or early 2000s, a lot of people used an HP-65, 67, 41, or 48 for in the 70s, 80s, and early 90s).

Also, in the educational realm, which is where calculators are the most popular, it takes a lot to compete against TI's monopoly. So, you get smartphone/tablet-like features, so that parents buy a fancy calc for their kids instead of both a TI calc and a smartphone. Then, TI does the same to act as an upsell.

And, also, there is a "legitimate" tool-based reason for some of these apps - a calc that can do these things is a much more rugged platform than a smartphone or computer, and is much lower cost, even if it's also lower performance.

I've used the CellSheet app on the 89 several times for fairly simple tasks. It's handy for taking notes and doing calculations/digesting data anywhere. Why should I be forced to sit at a computer all day?



Studycards? I actually found apps like that useful back when I was in school.

They're optional Flash Apps. It's not like they're built directly into the OS taking up space that you can't free up. I'm all for optional programs that allows one to use the calculator more like a versatile low-cost handheld general-purpose computer, if they desire.

Also, most HP calcs let you customize keys, and I would guess the Prime will let you also.

Oh ok I did not know that. I've never actually owned a TI-89 myself, so perhaps I wasn't informed enough to criticize them legitimately. Still, the fact that they would even add in a studycard app proves to me that they are just catering to students taking standardized tests, not to the everyday student who just needs a number cruncher, or the intuitive learning who wants to program (yes I know the 89 has a programming app as well, but it seemed much less user-friendly to me than TI-Basic).


Really? I've never heard of that, how does that even work?

It's been a feature of HP's high-end scientific calculators since the 41C in 1979.

http://h20331.www2.hp.com/Hpsub/downloads/50gThe_USER_keyboard.pdf is the documentation on the 50g (which will be identical to the 49-series (which is the 49G, 49g+ (identical firmware to the 50g), and 48gII), and very similar (probably identical, actually) to the 48-series).

All I can really be thinking about this now is:

OH MY GOD HELL YES.

I've had a HP 50g for a long time; and I have to say, it was my favorite calculator. Now, it looks like I'm getting an upgrade! I can only hope they've made the programming language as integrated as they have with their other models.

@iconmaster, How similar is HPs programming language to TI Basic? Like in terms of syntax and functionality?

Grammar-wise, RPL's entirely different by a long shot. First of all it works entirely in RPN, which changes how one executes logic flow entirely. PRL is also a lot different in the fact that it doesn't apply the "arbitrary grammar" TI does; each programming 'word', so to speak, is a separate and modular function, and as such, all commands (save very specific cases, such as local variable definitions) can be used both on the home screen and off with equal syntax. In the Ti8x series, this isn't the case. In addition, RPL makes use of an object system (all programs are actually manipulatable objects, for instance) and contains many more functions for system manipulation (like setting mode flags, or rebinding key handles, or changing the filesystem). RPL, once you learn its very confusing syntax, can be much more powerful than any competitors'.

Of course, I'm only waxing lyrical about RPL here; since the Prime's based on the gII, they might stick closer to it's RPN roots rather than RPL, and RPN's a whole other can of worms.

There is an "HP Basic" (algebraic) version of UserRPL available on the 50g, that looks more like a "normal" programming language. I don't personally care for it, though; there are a number of "gotchas" mentioned in the manuals about flag settings and interoperability, or something like that, which IMO just makes using it that much more confusing. It doesn't seem to be well documented, and seems to have a few limitations or weaknesses. It's essentially still stack-based RPL under the hood, with syntactic sugar to make it look like typical programming languages. It might make work well for those who don't want to learn a stack-based approach, though; I don't really have any experience with it because I just prefer to write straight RPL—more control and opportunity for performance optimization that way, plus it's much more fun.

As far as I understand, the 39/40 series use a different language altogether.

One thing I think is pretty neat is that since internally, SysRPL and UserRPL are actually one and the same thing (the only real difference being that UserRPL programs are limited to a relatively small set of commands designed to be "safe" by checking arguments, etc., and throwing appropriate errors), one can easily mix User and SysRPL code. This really lowers the barrier for entry—as I was learning my way around SysRPL, I initially could use just small subroutines of SysRPL code to do the things I wanted to do but can't in UserRPL, and write everything else in UserRPL. As I became more comfortable with SysRPL, I could gradually begin writing and including more and larger SysRPL routines.

This also means that the system doesn't discriminate between User- and SysRPL programs and libraries. UserRPL programs are treated as the same "first-class citizens" as hybrid-, SysRPL, or ASM ones.

The trick is, which programming language are you asking about? HP's had a few on their calcs.

RPL actually supports some more conventional functional programming, with local variables, too. You can pull in arguments from the stack into local variables, operate on things (which does affect the stack), pull those back in as locals, and then spit out your answer as a variable. Myself, I don't use the technique to that extent, but it's easier to do that than do crazy stack gyrations (which IMO are harder because of RPL's more flexible stack - overall, I prefer RPL, but 4-level RPN can be a bit easier to keep track of) for bringing initial variables in.

However, given that the topic is about the Prime, I'm thinking it'll be the Prime's language. That's based on the 39gII's language, which is vaguely Pascal-esque. It's a fairly simple language, someone who's used to TI BASIC and has the 39gII manual would have very little trouble picking it up (although it's a lot more functional than TI BASIC - you can actually use local variables, and return things to your main function).

One thing I find annoying, though, playing with the 39gII emulator, is how semicolons are needed as line delimiters. Why? Because they're not automatically added in - unless you do all programming in emulation, at the end of most lines, you have to hit Alpha, then you can get a semicolon. Makes on-calc programming a bit more cumbersome.

I'd say, other than that oddity, it's got quite a bit in common with the normal user interface (in algebraic mode, anyway - we'll see how RPN works on the Prime) of the calc.

This is one of THE most useful features for a calculator. On the HP50g, this can be accomplished via two methods:

1) USER MODE
Basically, you can assign just about any function or program to any key. The key can be unshifted, left-shifted, right-shifted, alpha-shifted, as well left/right/alpha-shift-hold (like the shift on a computer keyboard where you hold the shift while you press the other key). With a free 3rd party library, you can also have double click (like a mouse) and long-press (where you hold the key pressed for a specified amount of time). You can also have it behave differently depending on what you're doing such as entering values, editing a command line, or editing a program.

For example, on my 50g I have multiple assignments for the MODE key. The default behavior is:

unshifted- enters mode setup screen
ALPHA - the letter 'H'
LEFT - Custom Menu
LEFT HOLD - menu of mode commands
RIGHT - marks the end of selected region for cut/copy/paste


I found that I was using the custom menu more than the mode setup, so I swapped the behavior of those keys. I used certain mode changes frequently, so I made custom assignments for those. My MODE key assignments have the following changes:

unshifted - Custom Menu
ALPHA HOLD - toggles between RPN and Algebraic mode
LEFT - enters mode setup screen
RIGHT HOLD - toggles between STD/FIX/SCI/ENG digits modes
LONG PRESS - toggles between DEG/RAD modes
DOUBLE CLICK - toggles between rectangular and polar modes


You might think that this would be confusing but it's really not. Since you're the one who assigns the keys, you know what they do. You can recall and save the key assignments and switch between various assignments sets. If you want, you can just turn off USER mode and you're back to factory settings.

Take the common task of switching between degrees/radians mode for example. The TI-Nspire requires a minimum of 9 keystrokes to switch between degrees and radians mode. I can do it in 1.

2) CUSTOM MENUS
The top row of keys, F1-F6, are used to select menu choices (like the TI-85/86). Besides the built-in menus, you can also create your own custom menus. The TI-89 has custom menus, but you pretty much have to write a program to define the custom menu. On the 50g, it's extremely simple to create a custom menu. You can have a different custom menu for each directory (folder) that you're in. For instance, I have a different directory and custom menu for each of the four subjects I teach. These custom menus can be used to access built-in function, your own programs and functions, characters, numbers, anything including sub-menus. (If you've ever had to use SINH, COSH, and TANH on a TI-84+, you'll find that the only way to access these functions is through the catalog. Wouldn't it be nice if you could put them all in a custom menu.)

Together, these two methods of remapping keys make for extremely effecient calculating (and especially so when using them with RPN). When I first saw the image of the HP Prime, one of the first things I looked for was a USER key. YES, it's there!

-wes

@wes. Wow that [i]is[\i] awesome! See I thought customizing keys would mean like, u make the enter button be the division button, or u switch the alpha and shift keys, or u make the on button do graphing...abilities which obviously seemed kinda pointless to me. But what ur describing sounds more like assigning a macro to a keypress in Excel, am I right? [i]That[\i] I could see being [i]very[\i] useful!

On the machines with a full user mode, you can assign any function, whether it's a built-in function (like your example of assigning division to the enter key), a user program, or an add-on ROM function (at least on the 41 series and the 48SX/GX) to any key (with a few exceptions) while in user mode.

(So, that's the 41 series, the 48/49/50 series, and now the Prime, it seems. Also, the 65 and 67/97 had five keys that behaved like a user menu (the magnetic card that you loaded had the assignments printed/written on it, and you'd put it in a receptacle above those keys, and the 41 had support for that as well), the Voyager (1xC) line had those keys available for compatibility with 65/67/97/41 code, and the 28 had a true user menu.)