Here are some bad news about the TI-84 Plus C Silver Edition from TechPoweredMath (thanks to Kerm for summing things up on Cemetech. There is also a calculator info page being built up there):


-Like what happens when running FX-9860G BASIC programs on a Casio PRIZM, TI-BASIC programs that uses Pixel-On/Off/Test functions will not display properly on the new calculator. This also means that programs using Line() will obviously look weird as well. We do not know what will happen with old 8xi files when sent to the new calculator. The good news, though, is that this will allow far more details in games (providing drawing commands aren't as slow as on the Casio PRIZM) and if a game only make use of Text() and Pxl() commands, they'll still look OK, even if no longer full-screen.

-Similarly to when the TI-83 Plus replaced the TI-83 calculator (now the TI-82 STATS) in 1999, all TI-83 Plus, 84 Plus and Silver Edition ASM programs and applications will not run at all on the new color models.


Although this is bad news in general, this is not the first time this happens in the calculator community, the most recent example being the redesigned FX-9860GII which now uses a PRIZM processor. On that calc, any ASM or C program that do not use the official SDK no longer run on new models. The TI-89 Titanium suffered from compatibility issues as well with older TI-89 programs. For the TI-89 Titanium, however, HW3Patch solved all those issues and almost no program needed to be ported.

In other words, when the new calculator comes out, the TI community will have a lot of porting work to do if they want old programs to run on it. That's unless, of course, a savior decides to create an emulator or shell that can run older programs in their old school resolution. One thing we should be glad, though, is that unlike the TI-Nspire, where TI has fought non-stop against Ndless, the TI-84 Plus C Silver Edition will at least have native ASM support in the first place.

That would be cool, unless they inadvertently based the OS on 2.71MP

And that means that people who have much experience in z80 ASM and Ti-83+ "in general" will not find it hard to write new games!  Sure it means new routines such as tilemappers will have to be redone, but I at least look forward to ASM games in color

Yeah true. I hope it's as easy as porting games from the FX-9860G to the PRIZM (which almost had the same hardware except the screen).

I'm hoping they will yet. There is still time for them to take these sort of things into consideration.

I'm with Hot_Dog, i'm looking forward to finally having a nice z80 color platform to play with. I just hope that it will be able to handle fullscreen color graphics quickly enough to support decent games. Now that the screen won't be divided into groups of 8 pixels, just filling the screen is going to take much longer (or will at least need to execute a lot more instructions).

A few very important programs will need to be rewritten. Remember how the GameBoy Color had the ability to play Gameboy games and even give them a little color?

If we still have most of the same RAM areas, we can do these:

MirageOS: If this is ported, we can make the LCD updating using 3x3 pixels and it can be zoomed in or out. Otherwise, sprite commands and drawing commands will need to be rewritten. Once this is done, most games should work again

DoorsCS7: A new version for the new calc can be written with updated ION/MOS/DCS7 calls.

Grammer: I can try to port this so that older Grammer games work on the new calc (with scaling up pixels, basically). This actually shouldn't be too difficult, and as well, I will be able to create new routines to work with color (I've already been trying to figure out a way to make B/W and color versions of programs work across platforms because of my Prizm)

Also, we may need to make a program to "fix" compatibility. What I mean is something like this:
We will need to have a list of RAM equates (which we have) and their sizes. We can go then find equivalent values on the new calc. Then, when MirageOS or something tries to run a program, it will instead parse the file, adjust addresses accordingly, then run the modified version (without changing the original, except for SMC). That sounds a bit daunting, but not too much so. There are a few more steps involved and it wouldn't work for all cases, but it would be useful! (We would also need to modify ports and other fun things)

Yeah, that might turn out to be a tough thing to get around :[