Since I don't think TI initially created the calculator to be the base of program development outside of their control, they probably feel they have no compunction to maintain a 'supported' set of ROM entry points. As a developer, I can see their point, though as a programmer interested in programming for the TI calculators, I would like to see them 'lock down' the ROM entry points.

Just my opinion.

Im feeling a little different here(i have a ti83), but i agree with many of the main points here. The difference between the ti82, ti83 and ti83+ is really totally ASTOUNDING!
The only visible difference between the ti82 and 83 is statistics and a catalog. Seems that they could run the same assembly. Nope-completly different. Same with the ti83+, they should all be compatible. But its not. There is nothing except another 300$ sdk for the ti83+, and half the time the ti83's own programs dont work because of ROM revisions inside a class. Inisdious, huh?

On another note-whoever put "language you wouldn't hear on U.S. network television" hasnt watched NBC in the past five years.

I think those guys at TI are just wasting the energy of talentous programmers. Instead of writing new, exciting software, TI-programmers have to rewrite their old programms whenever a new ROM is released, not to mention the fact they have to struggle to get any technical information on the calc. No wonder there are so few new releases, then. But maybe, as someone stated, that's actually what those guys at TI want. (No assembly=no quality games=the calc looks "serious")

what I read from a TI reply to my email about the 8k limit was that they didn't want us running the apps on our RAM. Wording it to english, I think this statement says that they don't want us to be able to pass these apps between calcs through RAM without having the other calc pay for it. Couldn't they just upgrade the ams to check if a program being run has a special "identifier" of some sort labeling it as a flash app? please reply without flaming as I do not know much about these calculators, if I've said anything wrong. :P

Based on my experience putting games on my friend's 83+s (I have an 86), you can send something that is archived to another calc and it will be archived on the reciving calc after the tranfer. I don't know if this applies to the 89, though. (until next month when I get one).

And I agree that TI should just write an identifyer into the flash apps when they sign it so that it can only be in Flash ROM (along with an AMS upgrade that removes the 8k limitation).

If you are sending an archived file from one 89 to another, it somehow manages to become unarchived along the way. I'd prefer that it didn't, but its not much of a hinderance.

True, If I wanted a gameboy, I would much rather buy a gameboy with a color screen with better resolutoin and faster drawin that costs about half as much (hmmm, must be those $50 TI buttons) but anyway I think that basic is a very, VERY valuable tool to aid you in math. Especially if you know how and what your're doing. Which leads me to my next point. Now I may be worng, but as far as I understand BASIC is just like windows. Windows sits on top of dos telling it what to do. Basic sits on top of asm and tells IT what to do. So it would do no good to get rid of it. Why dont they just add in more memory? There. Youve had my two cents.

This is at least an utter oversimplification. Let's clarify a few points...

1. Asm, short for Assembly Language, is no more than a humanly-readable expression of machine code (what your calculator's processor understands). Hence, it is not even *possible* to "get rid" of it, but for replacing the processor by anonother one.

2. TI's AMS is a fairly large computer program written in [whatever language it is written in], and probably not in assembly. TI-Basic is part of it.

3. AMS theoretically provides third-party assembly programmers with a so-called Application Programming Interface (API), that is a set of "entry points" to certain system routines giving him greater control over the machine and greater speed.

4. The problem is that there is little documentation on the API and little support. That is why most assembly programs rely on unsupported and undocumented features and system calls, and these are absolutely not guaranteed to work in later realeses of the AMS, which is quite normal.

What we could (and should) reasonably request is a decently documented and officially supported truly powerful API/SDK. A so-called asm-ready calculator should have it.

i have two questions and one statement. first of all i am working on a TI-82 BASIC program called Quest for Truth. It is basically Oregon trail meets Zelda. I am about 1/4 of the way finished and i was wondering if anyone wants the game or want to help with the code. my second question is when i am programmming in basic how come i cant make ant lbl's with the lbl # over 9.please help and if you want a demo version of the game email me at sbcman9690@aol.com thanks

If TI is getting rid of asm compatibility... so what? Programmers will find a way to get to it, like with the 85. If they are getting rid of it completely, though, they are making a big mistake. TI knows about ticalc.org, calc.org, and other sites. They know there are games. Games help sell their calcs. They should be grateful to us for helping them!

I don't know about most people but I use my calculator for just about everything. In my advanced math classes I love having the new support allowed with new flash programs. But on the other hand I use my calculator for everything. When I unknowingly upgraded to 1.05 I couldn't run any ASM. I had just got my calc and didn't know why. Anyhow I just found a rom patch program the other day when I cam onto the site and was pleased to find that all my ASM programs worked. I use these from the calender programs to the advanced periodic table. Now that I have the wonderful new ROM version I can no longer run those. I dislike this but oh well. I also hope that the person who wrote that ROM patcher will write a new patcher for the new ROM. If anyone has any information on a ROM patcher let me know because it would be greatly appriciated.u

'I Agree'
If TI released information about the arguments and return values of the functions in exec.inc (which is kind of like a jump table), then it wouldnt metter when the address of the ROM functions changed because we could use the table to get the address. Most likely, TI will release this information in the SDK for the 92+/89. Until then we can always use trial and error to figure it out....

I say no. I'm an absolute beginner in assembly, but since I don't know of any official rom upgrades (not to say there aren't any), you'd be screwed if an important program (important in your POV, that is) were written for another rom version.

If they don't want a lot of problems why didn't they write all of the 'OS's in the same assembly language

ex.:
all of ti's calculators os in z80