I am a black link success story. I have sent ams 2.04 to my 89 using a black link. my misteak, but it can be done. my batteries were farily fresh, i think.

I got brave and tested the AMS 2.04. Not very many ASM progs work on it becuase of the 24k barrier. Some didn't even run. BUT YES IT IS POSSIBLE TO SEND IT ON THE BLACK LINK CABLE. I still have 2.03 on my PC so I went back to it, cause I know it works. I agree with everyone else don't use anything higher than 2.03 cause TI doesn't want us to write ASM program or else they wouldn't put the barrier, that and compatablity and program useage is very poor right now.

What will TI do next?
Kaizer_911

Hey pplz- When i try to send ams 2.04 - i have ams 2.03, HW 2, black cable; ti-graphlink crashes ~ Does ne1 know whats up? *Thanks*

I just heard something terrible: it seems that even the internal organization of TIOS VAT table is changed in AMS 2.04. If this is true, this means that even SYM_ENTRY structure defined in TIGCCLIB is no more valid, so any programs which access to VAT will not work, regardless of legality of its usage. The same is true for my file routines from stdio.h: fopen, fput, etc.

Disasterous!!!

I wrote a program that hides a varible (for use in some BASIC programs I write) with tigcc lib 2.0 and it seems to still work under AMS 2.04

If it works, I will be very happy. This means that the VAT is still the same, and that the information I received is wrong :-)

Thank's to God. I received the confirmation that VAT is still same on AMS 2.04. So, you may expect that programs written with TIGCC will work on AMS 2.04 :-)

Yippee! Maybe SOMEONE at TI has some sympathy for us poor programmers. Although it's strange how they hang onto their stupid program size limit even though they know we can break it. Maybe they have a hardware version 3.00 up their sleeves... (ominous music here) Why do they want such a stupid limit anyway? Does anyone know? Will their SDK allow people to ignore the limit for non-flash programs, thus forcing people who want to write large non-flash programs to buy it?

I actually have no idea if this will happen, but I think there will be "limitations removed when programming is done with TI-SDK" things in the next version, if not this one.

- Jeff

According to someone in TI's "TI89/92+" discussion group, TI added some 43 new functions into AMS 2.04 bringing it up to 1506 functions (as compared to 1463 in AMS 2.03)

Are there any new functions in the catalog?

i don't think there are any more in the catalog, these are just function built into the rom, that can't be accessed through the catalog window.

I say that we all just keep AMS 2.03 on our calculators and never again download a new AMS version from TI, unless is is completely compatible with the older versions. Then, after everyone has AMS 2.03 on their calculators, someone decompile ti's AMS and make our own, that would be completely compatible with all other versions. This way TI would never be able to limit our ASM programs, Hardware limitation, and anything like that. If anyone has an idea of how we can stop TI from screwing up everthing, please e-mail me.

Maybe TI released the new version of AMS because they're probably about to release the SDK and probably had to add some new stuff to it in order to get the programs to work properly the way they want to.

But pretty much, if TI keeps adding these incompatiblities between rom versions then I say we make our own or do that Linux idea (it'd be nice to have a computer and a calc w/ the same OS, only have to learn how to program for one then).

if linux is ported or based on it still won't have any of the functionality of the real linux since how are you going to run x windows on a cpu that is not even 100 mhz or has color.

amen. i don't get why evrybdy thinks that linux on a calc will/could be cool... yes, linux rules, but on a calc? the hardware just wouldn't do it justice

"i stood in front of her face when the first bullet was shot"

Anyone else think TI should put a hold on adding new things to their AMS until they optimize the speed? I mean, nothing aside from the collapsable folders added much since 1.00, and the cost in asm functionality has been huge. I'd consider upgrading if they could make things faster. I think it would be possible to do so judging that with a 68k chip the 3D graphing is still slower than HP's 3D graphing. They must be using really bad routines.

they aren't using bad routines when there OS is written in C.they have added a ton of features since 1.00, better archive support, flash apps support, more mem. also 1500 some functions.

I know C is very powerful, but 68k assembly is a lot more. Graphing 3D graphs could be very fast.Just they need to optimize some codes.

The 89's 3D graphing is slower because it is at a higher resolution then HP's. The default xgrid and ygrid settings are 14 for the 89, and I read somewhere that they default to 7 on the HP. Change the xgrid and ygrid settings to 7 and the speed will greatly improve.

That also causes the graph to be smaller, which doesn't help in some situations, but you do have a point there.