Cool... so are these "Exclusives" just for people who want to submit tutorials? We already have a few good ones. For instance, I used karma.ticalc.org to learn TI-82 ASM.

Oh, now I remember. The "exclusives" were tutorials you could write that you'd send in to Morgan. However, I learned everything from the tutorials that were already written, so I don't think I could write anything new, interesting, or worth reading. But maybe it will be different for other people.

Wait...so if I want a tutorial made as part of this, I email morgan? Cool.
This looks like a great idea.

No, the point is to make a tutorial on a language and suggest it for posting.

I write up and essentially port it to the ticalc.org source, but no the writing and style of the tutorial is done by the Author and I just add it to the site and make link it from all the appropriate basic calculator pages.

If you made a tutorial, I sincerely doubt anyone would want to read it. Not trying to be offensive, but Doors whatever-acronym-you're stealing-from-windows-now is slow and useless, and so is all of your BASIC crap (with one or two possible exceptions, and only to the useless part; all BASIC is slow). There are far too many BASIC tutorials, most of which are bad, and only one or two good ASM tutorials.

Thanks a lot for your helpful and insightful contribution as well as your endless suggestions as to how to fix the problems you point out.

It might be time to dust off my 68K BASIC string manipulation tutorial.

Let me advertise an excellent fresh project here: the Unification FAQ. See link. All contributions are welcome.

I learned most of my TI-BASIC from messing around with the calculator and looking at other people's programs, so to me, a BASIC tutorial wouldn't really help. But the assembly tutorial definitely helped me understand assembly language a lot better. It also helps (for BASIC or ASM) if you have some familiarity with basic functions in basic, C, etc.

Basic is so under-rated these days on TI's calcs. Sure, assembly language has it's perks and all but when I want to type up a program to perform tasks in math faster (yes, believe it or not Basic can be fast) I'm restricted to using Basic. No, this is not because I don't know assembly... Just the fact that assembly language requires "assembly" on a computer is a big deterent. ;-P

/Added/ Just a few tips for Basic programmers who want to make programs for performance of "well-useless txtness":
Try to cut down on the code wherever possible. This could be by leaving off ")" for "output(1,1,A" etc
Also, if your typing up your programs with TI-Graphlink NEVER USE LOWERCASE TEXT!!! When you transmit it to your calculator or save it the text will change from "Hello" to "HeLlO". This does not look good in your program. Make them on your calc if you really want lowercase.
When requesting a user to input values, tell them what var it will store to. The format of the entered value IS DIFFERENT when storing to a "A" "Matrix" or "List"
*just a few things I noticed whilst browsing through a few thousand basic programs.

In AMS, you can't leave off ending parentheses.

Mostly, the last suggestion just confuses users.

See link for more (the "Good BASIC" posts)

Yeah, that's something that always annoys me about programming on my TI-89. You have to include all the parenthesis and quotes at the end.

You can use AutoClbr.

AutoClBr doesn't work in the tokenizer.

Oh. I've never actually used it. I just assumed it would.

bah, if they can't understand a Prompt, then they don't /deserve/ your program ;p Actually, you can usually tell what data type you have to enter, even without it telling you, by the type/description of the program... like AASTAT (link). Of course you would enter a list because who's going to find the mean, median, and mode of a number or a matrix? And you can always outline how to do the input in the readme. That's what readmes are for, people!

Why use Prompt when you can use Input?

Prompt saves bytes, if you're a really picky person about it. But I think using Input is a lot better...

:Input "X:", X
which will show
X:_

is a lot better than
:Prompt X
which shows X?_

or maybe its just me.

I agree... there's really no point to Prompt, except for that it can input to some variables that Input can't, IIRC. (But it's still not worth it, so use Input)

True, but on a z80 based calc it can be very difficult to know what variable to enter and in what context as the program doesn't hint as to what var is being used. One useful thing when the average programmer is not feeling overly paranoid about the safety of their source is that when the calc prompts an error you can goto the problem in the Basic program if it's not protected. I don't advise protecting your programs where some variables are unknown. That plus testing your program would be useful. Oh, just a friendly reminder SUBMIT YOUR PROGRAM TO THE APPROPRIATE DIRECTORY! You'd be surprised as to how many 83 programs are in the 83+ dir and vise versus. (a few hundred);-P

Yeah, I know, that's annoying about people submitting their stuff in the wrong category, but I don't think that happens as much now, since Morgan is more careful :)

As for not knowing what type of variable to enter into a Prompt/Input... I still say you'd have to be some sort of moron not to know what to put in there. If the program is called ADD2NUMS, for instance, I doubt you'd be enetering a list or a matrix. And if the programmer isn't being ignorant, they'll give you a Disp or text in the Input (Input "text",A) to let you know what you're supposed to be enetering in. I don't see how there could be confusion. And as said above, if you don't know what to do, read the Readme file (which should have been done in the first place).

Ending parentheses and unneeded variables fill up quite a lot of space on the calc, and if the program doesn't "clean up", so to say, deleting the files it created, it makes even more clutter in your memory. I've shaved a 1500 byte file down to about a 700 just by eliminating ending parantheses, deleting unneeded text, and changing some variable names around (such as making :a + b -> a instead of :a + b -> c. RAM is a big issue with all of the Z80 calcs except for the 86, which has a bit more, so that couple hundred bytes can make a difference.

AMS

It depends on which calculator you are programming for.

Whoops, sorry, didn't read it right before I responded...

I've never had that problem with Graph-Link before. What calculator are you using?

The problem will occur when you create a program in the latest TI-83+ Graphlink Software and press "Send To Ram" (or archive for that matter)
I assume this is because Graphlink doesn't parse lowercase properly.

I agree. You learn much faster through trial and error, and the BASIC functions are very easy to figure out. On the z80 calcs, there's only a few programming commands, and it shouldn't take you more than a day to learn and be familiar with all of them. For the 68k, they give you the syntax of how the command is used, so with knowing what the code is called and the syntax, it's usually pretty obvious what the command does.

As for assembly... people like me need all the help they can get before they completely grasp the whole concept of how it works, so ASM tutorials will be good. :D In fact, it might be even easier just to write a tutorial on a specific portion of programming in ASM. Maybe sprites... or more efficient loops. That way, you can just go and read a more detailed tutorial on the stuff you want to learn about, assuming you already know a bit about the language.

Anyway, I'm throwing random ideas out... you guys have fun with them.

ehh, i don't think ASM is necessarily hard to grasp but our precious trial-and-error is unusable :( You can't just type in a one-line program and instantly run it to see what a certain command does or how it works. With making BASIC, you just 2nd-quit enter to test your program, with HTML you just ctrl+s and flip over to IE (html testing, people!) hit refresh, and even with C++, just hit the compile and run button to see your work in action. No so with asm :( I learned BASIC by simply stealing my friends calc for, oh, six periods a day and just experimenting. For PHP, i just read the first few chapters, realized it was just a oversized JS/Perl clone, and programmed a little CMS for my PHP experiments. Experimentation is a big deal. So I was thinking... a frameset page: on the left, a big textarea and a run button (the whole thing being a java applet); on the right an ASM tutorial. So when you read something you don't quite understand or you want to learn better, you just type the code in the box, hit run, and watch the output of the program spill out. It wouldn't be realtime (like games) but it would teach ASM verry well. And in the tutorial text, say some of it was "Now we're going to learn the loop". Then right after the work loop, there would be a <:a href="#?1" target="leftframe"> [+] <:/a> (ignore colons and add superscript tags). So it would refresh the left frame (assuming of course that the frameset defined it as "leftframe") with the query of 1, which the java app should be able to fetch and see that there's a query so it does a switch (or its java equivalent) with case 1 incidentally being the example for a loop. The app would write the code to the box and simulate it. good idea?

a combination of ^LINK^ and a web-based vti was what i had in mind

Wow, that WOULD be helpfull.

I tried that on-line assembler, and it didn't seem to work. It produced nothing in the output.

Whoa, I can't beleive somebody was able to make that online... I don't know ASM for the TI-83+ (it's too much different from my TI-82 ASM), so I couldn't test it.

I still say the whole thing would be easier/faster if it wasn't web based... java can be... well, slow.

What about JavaScript?

JavaScript is slower than Java. However, it could be possible to do it in JavaScript, I just don't know if there would be a point vis-a-vis speed.

What!? No, it wouldn't! Are you kidding!? There is NO WAY that you could ever program a calc emulator in JAVASCRIPT. LMAO!

As for the compiler... it technically is possible, but it would be horrendously slow and not worth it. Imagine trying to write TASM in JS... lol

You can't do file i/o in JS either...

For the emulator, you'd need to control each pixel as a separate image. That's 64*96=6144 separate 1x1 images to control. JS would take FOREVER just to swap all of them from "off" to "on." So I'm sure you can imagine how well playing Mario would work, or perhaps just seeing the home screen and typing in keys...

For the compiler... all the code would have to be in that one page (unless you kinda import javascripts) and you'd have to go through each line, interpret it, and convert it to hex... and you'd have to make sure there's no errors, like JRs being out of range and you'd have to get your addresses for the JPs (I believe that's how it works). And... if you run onto an error, you have to give some kind of error message... there's just too much. But, yes, you could "compile" a program in JS, but I don't think it would be worth it. It would take even longer than using TASM and it would be much, MUCH slower. JS is slow. Cool language... but very limited... and of course, SLOW.

That's what I was saying.

That would be amazing... unfortunately, I don't know how well that would work, for a Java app, anyway. I bet if somebody was really ambitious, they could do it by taking and mixing the sources of TASM and VTI. Either way, if there were something like this, I'd probably go right back to z80 ASM programming more. The thing that drove me away from it so much was the fact that it took me about a minute to save, compile, send, and test. And if I have an error that I can't figure out how to fix, debugging can be a real pain because you have to keep doing the process OVER and OVER and OVER...

But really, that is a great idea.

I don't know any asm, but it looks like you guys need an Integrated Developing Enviroment (IDE). See the link for an example (it doesn't do asm, though).

Yeah, that's what we were talking about, except with VTI mixed in so you could immediately test your compiled programs without any hassle. I know the TIGCC IDE can sort of do this... it would be nice to be able to do it with ASM too.

I have an alpha version of a java z80 assembler that supports all opcodes and labels. It doesn't support expressions so ld a,2+3 doesn't work yet. It's also pretty close to supporting equates. Thus if somebody is interested I can send you the code or if there's enough interest I can make a simple applet or app with a notepad like editor and the assembler.

I am also working on an eclipse plugin for z80 assembly(see link). That way you have a complete IDE available to work with.