If you guys work this from scratch, I have a couple of requests to satisfy a couple of needs:

Make the dang thing a microkernel. Then we can throw stuff on top of it real easily. Like, say, AMS emulation layer? *Linux* *emulation* *layer*?!?!? I mean, yeah, that would really be cool.

Make it multitasking. Then I can steal your VM code when I go to do the Linux emulation. And being able to plot that 3D graph in the background while playing Phoenix -- sweet... excellent for teacher deception.

Gimme the source! Free! Yeah! Show TI that opening up is good! I'm really feeling zealotistic now...

Spot ASM. Heck, when I throw my PIII in there, it shouldn't take too much modification to make it work ;) (totally unrealistic, but... interesting pipe dream costing many hundreds of AAA batteries...) Actually, use a good optimizer (gcc -O20) (you don't need TI-GCC) for the computationally-intensive stuff, inline ASM for the real slowdowns, and plain old C for the rest of it.

You can steal code from YACAS, etc. for the CAS. Graphing will be an issue.

I'll be around this summer (except for three weeks... darnit) so subscribe kcarnold@yahoo.com to any mailing list, etc. you might set up. I'm really excited to see this thing actually turn up.

Kenneth Arnold

PS - urgh -- final projects -- midnight -- hating school -- mad.

trowing a PIII in their would require extra instruction to the OS as well as case modifications. think of the heat that will produce.

Linux? WHY THE HELL DOES EVERYONE WANT LINUX??? It's a BSOD. Black Screen Of Death. Sure, i'm not a Gates worshipper, but come on, model it after something decent. Get Be. Sure, it's got a nice CLI, but it's cryptic, and entirely too old to be any good now. If you are going to make a new one, make a new one. don't model. Invent.

Linux is Ds' BOMB!!!

exactly, it bombs your computer into a position of worthless x86 architechture. (now, i know it runs on SPARC and Alpha, but if you are smart, you don't run pansy OS's like Linux on em) It's tooo big anyways. it supports all the old harware you could possibly want, but it requires you to have a new hard drive for all the space, for that, it's better to just go out an get another compy. calculator linux... we need our OWN OS. something that we make. that we like. not a model.

Oay, say we use Linux as a model, rather than a port.

We can have an operating system with multitasking, POSIX-ish syscalls, everything-is-a-file, mountable fs'es, signals, drivers for linkport storage devices, etc... and you want to re-impelment that from scratch. Go ahead.

Microkernel (exokernel?) will make it simple to just skin Linux to the bare bones, write a couple of drivers, and recompile.

Kenneth

If you are going to do that, then go right ahead. seems like too much trouble for a *calculator* to me... If you want linux and all of that, go and get a computer.

oh yeah, read the last sentence of my comment (not a model.)

Impressive. Very Impressive. An ENTIRE OS in 68k ASM. The thought crossed my mind as well awhile ago... but I hope you're aware that this is going to take time and will be hard as @#$#@. 68k ASM, as much as I love it, can, and will, be a b@#ch at times.

Good Luck!

I have read all of these posts, and it seems we all have ideas about what this new OS should include.
So, I've decided to throw in my two cents here, if anyone cares to read it.

As I said, we all have different ideas. Most people want this thing to have most, if not all, of the TI-OS features. You've got to admit, even though TI's code isn't the most optimized, that would be hard to do and keep all of the memory that the calculator has now. And, it would take a long time to figure out how to do it, or even what to put in it.

I really hope that gaming and programming is enhanced on this new OS. Basic for the TI-89/TI-92 Plus IS a lot better than Basic for, say, the TI-81, but you do have to admit, it would be cool to have it enhanced some. For instance, a "CalcOff" command or something like that.

This is really something to think about. I mean, all the things that we have ever dreamed of doing on a TI calculator, but couldn't because of some kind of ROM limitation, could all be possible. And that is a lot to think about, so I suppose that my main point is to try and prioritize and figure out what this thing is going to have. I don't know if people have already had that kind of discussion, I don't know, but if they have, I haven't seen it or heard about it.

Anyway, back to the subject. I think it would be really neat if we could have enhancements for programming, in both Basic and Assembly. In a sense, we have an opportunity to completely rewrite the TI-Basic programming language. The possibilities are endless. Even better, making an on-calc compiler for it would be neat, too. And, of course, we could speed up 3D graphing and such, and all that other math stuff I never use.

Anyway, instead of naming off a lot of paragraphs of stuff I think would be cool, I think a list would be better, so here it is:

-Enhanced Basic programming
-On-calc compilers for Assembly
-Faster 3D graphing
-Faster math altogether
-New and more useful ROM calls to programmers
-Archive memory recovery feature
-Something of a GUI, but not like Windows or anything

And all the other stuff the TI-OS already has, or whatever else. I can't think of anything else, it's 2:18AM...

Anyway, I think the main focus on this should be to make a stable OS that has most, if not all, of everything the TI-OS has, and enhancements to programming for both languages, and of course, faster 3D graphs.

Well, that's my two cents. By the way, I have an unrelated question: If someone were to, hypothetically of course, say, jammed a link cable into a TI-86 linkport and couldn't get the little metal end out, what would they do? Because...well, I sort of...did that. If I can't fix it on my own, that means I'll have to send in a gray graph link cable, a TI-89, and a TI-86 to TI to be fixed...:(.

Right on!

i did the same thing to my 86 and was able to easily get it out by removing the case and then using a toothpick to push the metal tip back out..

now.. i don't know about the ti-86.. but the ti-89 needs star tools to open up the screws.. i couldn't find star tools small enough... any suggestions? (happened to me too)

This is the reason why I mentioned this...I don't have any tools to pry it open...

You don't actually need to open your calcs case, I found the better way to solve this prob!
Just take your broken cable,

|---------\
--> |=|=| |]]]]]
|---------/

and apply some crazy glue to its tip (but not too much cause you'll make the problem even worser) and stick it in the calcs linkport, hold it there for several seconds and remove.
Thats what i did with my 83 when it happened.

zkostik

Well, my 83 had some triangle top screws, and I was able to clip off the edge of a flathead screwdriver, so it fit on one side of the triangle, thus allowing me to open it. By edge, I mean the side.
<top>
|"|
|"|
|/ <- So it looks like that, where this is the edge of the head (side)

I use a set of screwdrivers from rat shack. they give you a BIG variaty of screwdrivers in various sizes and they work great. about 16 in the set. comes in a blue case with a clear top. big long, wide, but thin case. cheap to. about $8. that's my 2 cents. hope it works for you
-Will Stokes

Hi... Sorry my bad english... Just to say that I needed to open my 89 one day and I unscrewed the star screws with a normal screw driver, didn't had any problem on doing it. It came out quite easy.

Also, we need to add switchable RPN entry, and it would be cool to be able to link to more than just 89s and 92s.
While we're at it, add on calc C!

Hey you do know you can do a lot of stuff {clear screen change status line, disable break, contast +/-, calc off~the list goes on~} Using Exec commands or a basic library ~flib or basiclib?

Basic libraries would mean that everyone would have to have it on their calculator to run that basic program, and I think that only assembly programs should have libraries. As for the Exec command, it's good, but basic programmers would actually have to take the time to figure out the opcodes and possibly even learn a little assembly.

And thanks for all the help on the TI-86 linkport, though, I'm a little scared to stick glue in my calculator...but, I'll go for it.

And

*the ability to handle sub-folders!!!!!!!!!!