programming in C - Page 1

does anyone know any good sites that have tutourials on or teach how to use C? I would like to learn how to use C, so that I can make my own games. I have made some in BASIC, but...BASIC stinks compared to ASM. I wanted to make the original Chrono Trigger in basic, but I realized how stupid it would have been. Now that I saw chrono fantasy in ASM...I feel that there is hope...if I can program in C...lol

On my website ( http://www.squalenet.net/ ) you can find a C tutorial I'm currently re-writing (there are 13 chapters, now).
But it is in french, only...
the old version (more complete, but quite obsolete) is available at http://www.ti-fr.com (this site is sometimes down), in section programming or something like this... but also in french.

there are also some lessons, in english, here :
http://www.technoplaza.net/programming/
but I don't know if it's been updated recently.

Anyway, you can use TIGCC's documentation
once you have understood the basics, this doc is quite enough.

(or, you can also use a book about C programming, to learn the basics ; some functions of the book will not be available for TI, but the language is still the same)

here is a list of useful links : http://tigcc.ticalc.org/tutorials.html

chrono trigger on ti that would be great !!! But a long work to do !! but maybe that can be done.

I want to learn C programing (shouln't be so hard, i already use it with PC's. ) I just need to learn the Ti-related functions. And I also need time

edit : whoo there is less difference bettween C for PC and C for Ti than I tought. Beware TIGCC , I'm coming

If you already know the C, tigcc's documentation should be enough : it gives the functions, their prototypes, and what they do.

and if you have any questions regarding C for TI, there is a dedicated board on this site : sections/1740-c-programming

thanks...now the question is....how long until i figure out how to program in C AND make chrono trigger???(10 years...) also, I forgot to say i would like to know how to send stuff I make in C to the calc, and what things I can use for the calc...If anyone knows a site or is willing to tell me, I would greatly appreciate it.

to make something like chrono trigger, aside from technical problems, you'll have to do a HUGE work for maps/gfx/etc, so if you do this on your own it's gonna be very long

to send what you program to your calc, you can use TiLP since TIGCC gives you a .89z just like programs you download... however many people prefer to use Virtual TI because it's faster to send and to recover from crashes (and if you know assembly, it has an integrated debugger)

Pollux :
to make something like chrono trigger, aside from technical problems, you'll have to do a HUGE work for maps/gfx/etc, so if you do this on your own it's gonna be very long

sure ! it would be long. However I'm all yours to hepl as long as my shedule allows it

Pollux :
to send what you program to your calc, you can use TiLP since TIGCC gives you a .89z just like programs you download... however many people prefer to use Virtual TI because it's faster to send and to recover from crashes (and if you know assembly, it has an integrated debugger)

maybe learnig assembly might be usefull then. i'll thnik about readind the docs

thanks...if you figure out how to use assembly before me, please give me some pointers...Ya know, I now think it will take 15 years to make...lol...ti-89 titaniums will be outdated by then, but, I'm still gonna try.

One good point , i learned assemblly with a motorolla chip so it's quitte familiar to me. I still need the ti specific stuff.
But I like c better, easier.

if we want to do chrono trigger, asking the cf programers for some tips should be a good idea

edit : oh, tigcc does the asm compilation too great

is not something like Chrono Trigger a project far too big for programmers with no experience related to ti-calculators ?

yes, maybe. I'll see what I can do with it first anyway. I never it would be easy, or even possible. But then, if I renonce because it's not exactly what I have learn, I'll just go no where.

and what's the use of spending my time learing C and asm if i can't have fun with my Ti afterwards

well, i have programmed, on calc and on comp, just not in C. and I think it is possible to make chrono trigger...(If chrono fantasy is possible, then chrono trigger is...i hope)

yeah, if you start small part by small part you could do something... but be aware that you'll have 90% chances to fail (no single Zelda clone has ever come close to being complete, even though there were dozens of such projects...), and don't be afraid to rewrite code from scratch over and over again, as it is likely that all the code you write at the beginning will be complete crap (even if it works, it could be unstable, slow, or unmaintainable)
above all, don't start with the scenario, the graphics, the readme and the credits

i won't start with that stuff, so don't worry

> edit : whoo there is less difference bettween C for PC and C for Ti than I thought
Many stdio.h functions were reimplemented (hardly any is built in the system), and anyway / therefore, they should never be used for efficiency reasons.

hmm, i often use fopen()/fwrite()/... when i don't really care about size-efficiency, it's faster to write

you mean using
ST_helpMsg("Hello World !");

insted of printf("hello world") ??

right. I'll keep that in mind

the first thing I'm gonna do the readapt the progame I made for computer for my Ti. (they are realy basic and should work.)

great, now if anyone would like to help me, and know how to write in C, they can. Oh yeah, I would ask time to team for tips, but don't they only speak french except for the translator? (I don't understand French...)...If I do end up finishing a chrono trigger....I'll make sure it'll be awesome...although...I can't promise it soon...or even in a while...maybe a decade...seriously...a decade...

Before writting your CT clone, make small and tiny programs to train you.

Yes, you have to be well trained before thinking about making such a large program.

I agree with what. i need to learn how to make genlib work first.

me too! I got the sprites onto graphics studio...how do I get the sprites on the calc and display them through genlib? If anyone knows, please reply. (Making sprites with SNESconv is so easy! Just gotta figure out how to display on calc)

More than learning how to make Genlib work (I don't think it's that hard), you have to learn the method needed to manage a large game, the system APIs, algorithms, etc.

Beware that GenLib is only available in its kernel-based form starting from 1.00 (the AMS native form was already badly supported for a long time). There are two major consequences to that:
* the kernel's 30 KB overhead (kernel + libs), and its RAM consumption we've indirectly talked about a few days ago (the memory consumption of CF+PreOS is too high).
* your game will be runable only by a minority of users.

Note that if I have written CF in AMS native, its RAM consumption would be higher.
Kernel Mode / Genlib have been designed for such big projects.

Since when this kind of posts is a troll ? I'm going to make everything more explicit:

> Beware that GenLib is only available in its kernel-based form starting from 1.00
A fact.
> (the AMS native form was already badly supported for a long time).
A fact (the loader overhead, for example). And IIRC, there's something else (calling convention ?) in the Genlib header which makes Genlib in AMS native mode less efficient.
> * the kernel's 30 KB overhead (kernel + libs),
A fact.
> and its RAM consumption we've indirectly talked about a few days ago (the memory consumption of CF+PreOS is too high).
A fact.
> * your game will be runable only by a minority of users.
A fact. It's been a long time since most users and programmers have dropped kernels and kernel-based programming.

Is it forbidden to warn new users that they'd better thinking about what software they're using / forcing users to use ?

> Beware that GenLib is only available in its kernel-based form starting from 1.00
A fact.
> (the AMS native form was already badly supported for a long time).
A fact (the loader overhead, for example). And IIRC, there's something else (calling convention ?) in the Genlib header which makes Genlib in AMS native mode less efficient.

Note that the nostub version of Genlib 1.00 is not avalaible because there is NO USER who asks for it. So I think it is useless to spend my time trying to port the loader for nostub version. Also note that it is less efficient due to lazyness (once again if I spent more time for something who nobody cares - even myself).