For the past thirteen years of my life, I have been a dedicated, vehement member of the TI calculator coding community. I first started writing TI-BASIC on a school TI-82 in sixth grade, and got my very own TI-83 for Christmas in seventh grade. I developed my TI-BASIC skills, released hundreds of programs, then began to teach myself z80 ASM as I started high school. I have maintained a hobbyist website and community centered around programming TI's calculators for well over a decade, now going strong with over 1,700 members. It should be a shock for a pillar of the TI programming community to therefore endorse a calculator from the Enemy, the nefarious Casio, one of the two perennial underdogs in the world of graphing calculators. I believe that tens of thousands if not hundreds of thousands of students have been introduced to programming and technical fields by experimenting with their calculators and learning to program with these tools, and I believe TI is no longer fulfilling its social responsibility to fill this niche.

Several months ago, Casio released their new calculator, the Prizm, the first graphing calculator in over a decade to include a color LCD. It sports 16MB of Flash memory, ample RAM, and the aforementioned 384x216 pixel LCD. TI's first entry into the graphing calculator space in over five years had been the TI-Nspire, an "aircraft-carrier sized" device crippled by no programming ability and a feature set pandering to standardized tests, not to students. Fast-forward to late February 2011, when TI announced a color LCD refresh of the Nspire line, to be known as the TI-Nspire CX. Like many of my programming compatriots, I was excited. Perhaps TI had seen the light of their responsibility to inspire students to become mathematicians, computer scientists, and electrical engineers. Sadly, the Nspire CX follows the steady decline of TI's graphing calculators started with the first Nspire. If you're a TI calculator enthusiast, you should ignore the Nspire CX and get a Casio Prizm; read on for why.

The Casio Prizm is an attractive calculator, following current gadget design trends with a glossy black plastic case and chrome accents. To hands accustomed to holding a TI calculator, the build feels equally solid, and the square, smooth buttons are a pleasure to press. I got my Prizm two days ago, and I have continued to discover new features as i have explored it. The interface is initially unfamiliar to a TI-83+/84+ coder, hearkening more to the layout of the TI-89's software, with a main menu that leads to the math mode, graphing mode, programming tools, spreadsheet, document, and drawing applications, and more. The color screen is of course the most novel feature of the calculator, but the software is written to use the color screen richly and effectively rather than as a novelty. Casio in fact has many years of experience with color calculators; they released a four-color calculator that sadly saw only limited success in 1996, fifteen years ago. From a technical standpoint, it has a powerful 116MHz SH3 processor underclocked to 58MHz, 16MB of Flash ROM, and 2MB (or 61KB, depending on your accounting) of RAM. I am especially excited that it has the same pair of ports as most recent TI calculators, namely a miniUSB port and a 2.5mm stereo serial jack. I look forward to porting CALCnet 2.2, my TI calculator networking protocol, to the devices. The specs still on the same level as PDAs from ten years ago, but nonetheless respectable for a graphing calculator.

Of course, for a company with a fraction of the development budget and the reach of Texas Instruments, there are bound to be a few hiccups of a new calculator. Most notably, community members have complained that despite the Prizm's 58MHz, the BASIC interpreter is painfully slow. To the best of my research, BASIC programmers seems unable to draw in more than eight colors, and TI-BASIC coders will be disappointed to find that the Prizm has inferior drawing features to the TI-83+/84+ series. Even with all these limitations, however, the BASIC flavor on the Prizm is far, far superior to the pitiful and rudimentary capabilities of the TI-Nspire. As stated, sources have confirmed that the Nspire CX will not improve on the programming environment of the existing Nspires, making the color screen and sleeker case the only reason to purchase an Nspire CX.

And this brings me to the crux of my argument about why I, you, the rest of the community, and all your friends should buy Prizms. It's not that it's a vastly superior device, but to paraphrase a smart Cemetech member from IRC, "TI has basically become a monopoly in recent years" that cares only about its bottom line, its position as the foremost maker of calculators for standardized tests, and its relationship with educators. It no longer cares about one of the major groups that provides it with uncountable free publicity. Without the many tens of thousands of calculator programs ranging from science and math programs to shells to utilities to impressive games, many students would use their calculators far less, and would have no incentive to upgrade to TI's latest models. However, I feel that even without this monetary aspect, TI has a social responsibility to support developers and coders. The world is rapidly moving towards a phase where technology is an integral part of every facet of everyday life from waking to sleep, and this technological progress must be supported by successive generations that are enthusiastic about engineering, programming, and invention. I have talked to countless individuals in my role as a community figure over the years who cite their TI calculator, its programming abilities, and the TI programming community for getting them into technology fields, to technology-related majors, and finally to technical careers. TI is taking the unsupportable step of trying to actively prevent any sort of development on their devices, cripping the BASIC variant on the Nspires beyond recognition, taking legal action against community members who try to expand the programmability even of the tried-and-true TI-83+/SE and TI-84+/SE series that has relatively unrestricted programming capabilities, and fighting a "jailbreaking" arms race with the community in its TI-Nspire line akin to Apple's attempts to keep developers from freely using its platform. For a company that makes a livelihood off of selling programs and games, Apple's oft-critized attempts make worlds more sense than a powerful educational technology company trying to stop students taking their first steps into the world of software engineering from experimenting with and learning from these powerful yet ubiquitous tools. The bottom line is that unlike TI, "Casio isn't actively trying to development for [the Prizm]" and its other calculators, and therein lies the rub. Although it hasn't released an official SDK, sufficient tools exist to enable hardware hackers to figure out the device, and indeed many among the Casio and TI communities have already begun the effort. Programs can be loaded onto the Prizm painlessly; unlike TI calculators with their complex and terribly-bugging TI Connect software, the Prizm appears as a mass storage device. Even the OS is easier to replace; TI's devices are locked to require a mathematical task taking millenia of computer time to complete before the code that is needed to replace the OS can be cracked.

Sadly, the Casio Prizm will to a great degree of certainty see a much lower distribution and orders of magnitude fewer sales than the new Nspire CX, which is a terrible shame, and not based on the merit of the devices or the company. Many schools or school districts issue a blanket requirement to use TI calculators, which TI has taken years to support and encourage by pandering to schools and educators, itself not damnable, but at the total exclusion of any interest in the student, the curious mind, the budding technologist. I feel that this is an irresponsible corporate attitude, and though I don't expect it to change any time in the near future, I call on Texas Instruments to turn a new leaf, to take a hard look at their role in the educational community, and to add programming capabilites back into the Nspire calculators and start once again working with and holding a dialog with their loyal community of programming enthusiasts.

Very nice article, Kerm. I hope this makes TI really think twice about the CX and including programming abilities. I'm afraid I won't be able to get either of these calculators. I already have an Nspire, which has programming abilities, and pretty good ones at that (if and only if you "jailbreak" it).

I am with Merth. This was my first programmable calculator. I was hooked the moment one of my best friends got a TI 83+ Silver Edition from his sister and brought games to school. I already had desires to make games and programs from a young age. When I was somewhere between 5 and 7, I had tried to write "Pinball" on a disk with sharpie and have the game on that disk. After learning that wasn't how it was done, I was sad, but I really wanted to figure out how it was done. I then moved to trying to write Batch files, the only other way I knew to program at all. I would sit in front of the computer reading tutorials on how to program in Batch files for hours. It wasn't until I got a calculator that it all started to make sense. I was on my calculator (TI 84+SE from my cousin) from day until night. I'd bring it in restaurants so I could write my loop that counted from 0 to infinity, and would eventually count faster and faster, 10s to 100s to 1000s.

Programming has become a major part of my life now. I come here every day trying to get a little better at it. Even to the point of taking classes in school and thinking about it as a career. I don't think I'd be anywhere near the person I am now if it weren't for my calculator, and I'm am extremely saddened by TI's latest move.

As this basically amounts to an open letter to TI, I place my signature next to yours on this. I, like many others, started programming on a calculator very early, and it has certainly influenced my life so far (including getting to meet some pretty cool people and crash on their air-mattresses). It's a shame what they're doing. Though the business-minded side of me understands it...

I think some people should try to encourage their school to promote the Prizm over TI calcs.

Very well written article, Kerm, and I agree with your concern. The only complaint I got about the Prizm is that BASIC drawing commands are extremely slow (0.3 seconds to draw a pixel) and the Locate command is buggy, but otherwise it's a really nice calc and as you could see, it's much less locked down than the TI-Nspire.

I thought the Casio Prizm was clocked at 58 MHz, though? EDIT: Nvm you edited your post (It said 200 MHz max speed)

Amen! Thanks for the support, Shaun. I'm seriously considering of trying to find a way to voice my concerns to TI more directly in some form based on this.

DJ_O, thanks for the comments. I corrected the factual inaccuracies, thanks to Qwerty.55's help.

Edit: And _player, thank you so much. You guys are giving me awesome quotes to put in an open letter to TI; please continue to do so.

The main difference between the Nspire line and the Prizm is programmability, as was hinted in the article. It took three years for the first real assembly programs for the Nspire to appear as a result of Ndless. The Prizm was hacked within a week of shipping. Furthermore, while the native BASIC on the Prizm is slow and slightly buggy, it's at least functional. A program that works given a bit of patience is infinitely better than a program that doesn't work. Furthermore, the BASIC interpreter is far more easily patched on the Prizm than on the Nspire.

Yes, I'll admit that the Nspire CX has many more OS features out of the box than the Prizm and a better processor. However, I'll take a slower processor when it means that users will be able to run my programs without a difficult and obscure installation procedure like that of Ndless. Furthermore, all of the Nspire's OS functions can be done in software for the Prizm. It's simply a matter of community support. I can personally attest to the existence of at least one project that will go a long way towards rectifying this discrepancy.

In short, the Prizm, while less powerful, is far more open to development than the Nspire. I don't think it's TI's responsibility to contribute to the community, but it'd be nice if they didn't go out of their way to stop it either.

I think that one project that would/will hopefully make great strides towards painting the Prizm in a more favorable light to TI community members would be a z80 emulator for the devices; I certainly would love to attempt such a project if my schedule will allow. Also, I take issue with the claims that the Nspire CX will have more OS features out of the box; while that's probably true, I won't make up my mind until someone in the community holds one in their hands for a few hours and tells us all about it (unless that's already happened?). And as you say, even though the raw specs of the Prizm are lower, the real speed and memory impact will come down to how well the companies (and the enthusiasts) can write software for the devices.

Well, the Nspire has 3D graphing and, if you want to be technical, the ability to use Wi-Fi. The Prizm has unintentional programmability and the ability to act as a USB drive, which I find very useful. The 3D graphing and the fact that the Prizm's math functions aren't much better than the TI-83+ series are why I said it's better out of the box.

That's fair. What about the CX's math functionality? Is the implication that it offers something closer to the TI-89 then? Also, I would be more than happy to write a 3D graphing application for the Prizm.

You should send this to every teacher and college professor you know, and tell them to do the same, causing it to spread like a virus and hopefully hit TI in the pocketbook somewhere down the road.

It'd be a good idea to try and get Casio's products available at discounts in university stores and the like.

I'm serious about getting this message to professors and teachers. If you can get enough schools to bail on TI in favor of Casio, you can either force TI to support programmers again, or provide Casio with a large enough base to develop more competitive products.

I absolutely like that idea, and I will definitely start brainstorming some ways that I could get that started. Qwerty, definitely! What other ideas do you guys have towards that end?

Getting this out to the teachers would definitely be a good thing - it seems like TI cares about nothing else than making profit. They make devices that are not the most useful ones for either students or developers, but are the most widely used calcs due to TI's position... I'm just wondering how hard it would be for people to switch to Casio, and if TI is going to react in anyway.

Myself, I started programming calcs half a year ago on my TI-84+SE. I wasn't new to programming, so TI-BASIC felt familiar to me, but assembly was something completely new for me. I still haven't created anything large in it, but it's very interesting to program in a way that offers a lot of speed and access to the system.

Yeah, some people on my site planned to do that on their side too. Good initiative.

If I still went to school and the Prizm was sold in Canada, I would probably do so too, or tell 3rd year high-schoolers they should get the Prizm instead of a 83+. Granted, the 83+ is nice and has a lot of games, but I'm sure a 83+ emulator will be available for the Casio Prizm eventually anyway.

The only reason I'd use the prism instead of an 83+ is if my classes demanded more than the 83+ delivered, or if I wanted to buy a brand new calc. If I can get a used 83+ or 84+ for cheap, I'll go that route, especially since I'm on a budget. The Prism is for those who wish/need to upgrade.

I also place my signature next to yours on this open letter.
My trusty 89 HW2, bought in 2000, is the platform on which I started programming on a regular basis - starting with TI-BASIC, and switching to C/ASM in spring-summer 2001 after finding out about FLib, a C/ASM library for TI-BASIC programs.
Later, I studied CS at the university; I'm now employed as a paid programmer. I know that thousands of us, old-timers and relative newcomers alike, are following the same path.

As the current maintainer of libti*/gfm/tilp, I'll add that having to play catch-up with TI's undocumented linking protocols, and gratuitous changes in them (e.g. the Nspire USB endpoints, the change that currently makes it impossible to receive files from a Nspire running OS >= 1.7 with libti*, or the behavioural difference of 84+(SE) wrt. transfer of Pic, GDB and Str variables depending on whether they're using legacy I/O or Direct USB), or simply having to work around the bugs if possible (e.g. the 89T's USB code deviates from the standard, which requires restarting the program on the computer side if the 89T unplugged physically or by APD), is tiresome.
I'm writing that, even though I have hardly performed any reverse-engineering on TI's protocols myself - Romain Liévin, the creator and former maintainer of libti*/gfm/tilp, did.


Teacher and user education is one of the ways that we can use to put some additional pressure on TI. However, we all know that they don't care: we've been complaining about the Nspire for more than four years, and they have been increasingly locking it down; they have also been further locking down the 84+ (OS 2.55MP) instead of fixing their own bugs.
Additionally, TI's monopoly is, by now, entrenched so much that they won't move until we get a couple dozen percents of the user base to switch - which we'll never manage to do by mere education, let's be honest...

A more effective way to get TI to show some respect to their customers, and to help students' interest in computer science, could (and in my opinion, would) be replying to the all-out war they're waging.
We already know of at least one powerful potential attack angle. In April 2010, I wrote:

( http://www.unitedti.org/forum/index.php?showtopic=8191&view=findpost&p=141214 )
Yes, I know that it was easier said than done (especially for me, since I've never had the RunOS code). But is it always effective to restrain oneself, on moral grounds, against what has become an enemy ?

IOW: let's make a CAS run on the non-CAS Nspire & Nspire CX, either through "RunOS"-type methods or a TI-68k emulator, and get the regulation authorities to ban the non-CAS Nspire from the tests where, unlike the Nspire CAS, it's currently allowed precisely because it has no CAS.



Ndless 1.7/2.0 requires transferring two files to the calculator, and launching one of them. However, you're right, the installation procedure of Ndless 1.0/1.1 used to be more complicated

My view on this is that I plan to get a Prizm, although not for a while. I quite frankly do not understand how to program the nspire in C, nor do I wish to learn C until after I learn C# and plus compiling it for the nspire is a real thorn in my side. I am happy with my TI-84+SE because it's basic language is awesome, I love using it and it has thought me the basics of programming, plus I do not have to worry about playing games on causing my freaking batteries to leak and/or exploded. I forsake TI, they will never get it right, TI can go crawl in a hole, I really don''t care.... Casio is obviously better then TI in regards to everything, the CX is just a copy of the Prizm, but just the color screen, nothing else from what I see makes it any better from the nspire. All in all, TI-8* and Casio FTW, but it sucks because I bet TI is going to get rid of the TI-83+ line soon, thus closing off programming forever and being the BITCH of the education system forever

Very well-written article, Kerm. I also plan to get a Prizm, but I don't know when. However, I wouldn't ditch TI completely, after all, they made really good z80 and 68k calculators and they have not discontinued them yet.

Well the 68k series is pretty much dead to TI and I bet you that the z80 series will be discontinued at any moment because the teachers will tell them so `-`