Post updates on your projects here, or give other authors your feedback on their works in progress.

When posting about an upcoming program of yours, please include a link to any relevant information (screen shots, etc.) in the URL field.

To have an entry removed, please Contact Us.

Ive almost got this finished but I want to get some opinions on what you think I should put in it and if I should Write one for ASM. Please Reply. OR email me.

Thanks
Bear Grubb

Is there anybody out there that can make a windows/DOS program that will convert 86 basic to ASM. I have an Idea In mind. It should have a 86 BASIC programing Interface but when a basic command is entered it writes it in ASM. The program should also have the ability to draw sprites and write the code after they are drawn like assembly studio. Please reply on this and email me if you think you can do it.

I am working on a VERY VERY basic converter, the first version will have basics like displaying strings and inverting the screen, getkey routines, contrast up/down, off, and some other stuff. any suggestions, email me

WELL, YET ANOTHER BREAK THROUGH IN 86 TECNOLOGY. I WANTED AN ASM CONVERTER FOR AS LONG AS IVE BEEN PROGRAMMING AND NOBODY COULD OR WOULD MAKE ONE SO I DID. THIS IS NO JOKE, THIS HIGHLY TECNILOGICAL PROGRAM HAS BEEN WRITTEN FOR WIN. 98. I ONLY LIKE TI-86 CALCULATORS SO IVE SET THIS UP WHERE THER WILL BE NO PORTING THEM. ALSO THESE ARE NOT CHEAP THEY COST TEN BUCKS A PIECE. EMAIL ME IF YOU WANT ONE. HA HA HA HA HA HA HA HA HA

wow.. you.. idiot

I'm making a multithreading shell for the TI-86 calc. It'll allow programmers to easily make assembly programs that have more than one thread running (or appearing to run) simultaneously. I don't think there will be much practical use for this, though.

It's me again.
This multithreading "shell" is actually more like a set of functions and just a program skeleton that other programmers can build on. Any ideas of how to do this would be appreciated.

I've now completed a working version of a set of functions that allow multithreading in an assembly program. Download threads.zip in the TI-86 asm source directory. Thanks if you like it, or not. Whatever.

I'm working a lot more on this again. The upcoming version will include several more functions and be more capable than the first one.

The improvements are
* Compilable C source file (previous did not compile)
* Better source layouts overall
* Added timer functions (sleep, set_timer)
* Added a join_thread function (returns the return value of the given thread when it finishes)
* Better documentation (I hope)

I'll upload the newer version probably this or next month.

Woohoo!!!

I've gotten preemptive multithreading to work now. I'm actually the second one on ticalc to make preemptive threads, but mine are SO much more capable (no offence, Mike ;-).

Basically what this means is that each thread doesn't have to call thread_cooperate anymore, but there will now be some issues with deadlock and race conditions (yikes!).

Anyway, go download threadsp.zip from TI-86 Assembly Routines (I just uploaded it so it should be there soon). Do it now before you forget. :-P

Is it tacky to reply to myself?

Yeah, I think it's very tacky.

A few improvements:

* I've really cleaned up the source and found and fixed a few bugs and inefficiencies.

* Nice (thread priorities) is mostly finished

* Mutex locks are cleaned up (previously I put in too many just to be on the safe side).

As far as I've tested, it works perfectly. Granted, I haven't tested nice and a few other features I've "fixed" yet, but I've performed thorough eye-ball tests on them.

As for performance, it's pretty good. It gets called by default 50 times/second, and it adds about 600 (a rough estimate) clock cycles for the most typical case on top of the TI-OS interrupt handler (so I don't have to handle keypresses etc.).

Given a 6MHz clock speed, 50*600/6E+6 = 0.005, or 0.5% of overhead! In other words, adding preemptive threads to a program adds negligible overhead. Of course, it also adds about 500 bytes to program size (!), but that can be fixed by trimming it down and optimizing it a little.

By the way, this set of functions is (and has been from the beginning) licensed under the GNU GPL, so it's very friendly from a licensing standpoint.

Which features would anyone like to see added or removed from preemptive threads? I'd like to know if anyone is interested at all in (or is) using this in their own programs.

Does anyone even read this board anymore?

I want an upcoming program to be one that turns a basic program into tokens and then deletes the original. If someone wants to protect their program, then why is it stored in memory as the large and cumbersome text-like format? I think that you should just turn the thing into tokens that take up less space and can run fast the first time and every time after that. I don't know ASM so I don't know how to do this, which is why I want you people who do know to get off your lazy butts and stop making games and dumb stuff that nobody cares about. Thank you for trying, which I know you will.

Something else I would very much like to see is perhaps a command that can be used in basic that is kinda like the CP register (I know a little bit about ASM, but if i'm very wrong, please ignore my ignorace) or whatever that can be used with another command to return to that spot in the code. Maybe something like
:LblPt 1
...
:Input X
:GotoPt X
(If X is 1, it will go to the point above.)

In regular basic, you can't do this. You'd have to do
:Lbl A
:Input X
:If X==1
:Goto A
:If X==2
:Goto B
:If x==3
:Goto D
etc...

Or perhaps making a Return feature for use with Goto such as

:Goto A
:Disp "Good work"
:Stop
:Lbl A
:Input X
:If X>9
:Ret
:Disp "Too low, freak"
:Goto A

I know that there are ways to make this work as it is, but If there was some way to make what I wrote work, It could be used on a much more complex level.

I don't know if this is possible, because it would be tapping into the way TI OS runs a program, which well I don't know. However, imagine the use of something like this, where Basic could a least be another step up and competent programmers could do something more porductive in math class.
Anyway, I think that someone ought to step up to the challenge, like maybe those people who claim to be part of a company that makes programs. I will be very grateful and respectful to the person who actually makes this happen, because I was amazed to see greyscale and 3-D when I first saw it.

Ok, I'm retarded, before anybody laughs at me, just disregard the last post I made here because I now know that Programs are stored in tokens, and well, basic programs are dumb anyway. ASM is the way to go, except programming drives me crazy sometimes (like after 8 hours, it still doesn't do what you want it to). Thank you.

lol.. I hate when that happens
Usually, I scrap whatever I have if I cant get it to work right within a couple of hours, so I dont waste any more time, and start over. It usually ends up working better the second time :P

Yeah....you're right about assembly being the best.

You're also right about the debugging.

It's called TOKENIZATION, and the TI-OS does it the first time you run a BASIC program. It's only when you edit it that it's converted back to text.

For all of those c programmers out there, I am working on a c to asm conversion utility complete with asm debugging and emulation. If anyone would like to join in on this project just let me know.

Is this an on-calc compiler? If not, there's already some fairly good compilers out there.

If it is, then you could probably use the Small C compiler which is (supposedly) able to compile itself. Even so, the compiler might take up a significant chunk of RAM, leaving little for other programs.

Good luck either way.