TI certainly have certainly seen off rivals such as HP or Casio to capture the lion’s share of the calculator market. The TI-84 is a real staple, and with as many units as there are out there, hacking them is a given. However, selecting an operating system for the machine can be a hassle. TI-OS is proprietary and doesn’t really want to let you do everything you’d like to. There are alternatives, but many of them won’t let you easily use your calculator to be — well — a calculator.
[Siraben] has zkeme80 which is essentially ANS Forth (mostly) with extensions for the TI hardware. You can easily extend the system, of course, because it is Forth. You can also use the machine for its intended purpose easily.
Underneath the Forth system is a Z80 assembler and Scheme, if you want to do some heavy duty changes. The OS borrows from an earlier Z80 Forth which started out with the venerable JonesForth. It also leverages low-level code from SmileyOS which, in turn, uses code from KnightOS.
If you want to experiment with the system and you don’t have a TI-84+ handy, maybe you can find one in your browser.
We’ve seen plenty of TI-84 projects over the years. It can drive LEDs, although it might not be the most cost-effective way to do that. You can even put them on the Internet.
That means that a TI calculator is using Reverse Polish Notation! How ironic.
This makes me think JonesForth should be easy to port to the Gigatron…
JonesForth is easy. it’s written in pure x86 assembler, but *extremely* well documented. You can almost learn the language itself reading the comments.
Agreed. Because JonesForth is low-level assembler, you’ll end up re-writing it.
But the beauty of JonesForth, and Forth in general, is that it’s easy enough that you’ll be able to port it to an architecture that you know (or built yourself!) in an afternoon. Do it! (And then report back.)
That’s what Forth was invented for, after all. Being the minimal system that could be brought up on anything easily.
It seems a graphing calculator would be a good candidate for a port of GraFORTH. 3D matrix operations would be a snap — likely faster on that calculator than they were on the Apple ][.
Surprised we don’t see more of the old pocket computers from back in the day. Forth would be perfect for them.
If the TI had enough pixels to make graphing worth doing, which it doesn’t. Teaching curricula based on these is a major PITA. A great project would be a good graphing calc with the TI as a subset (or your phone) but they will be banned from exams.