That just sounds to cool! I would love to see that!

Check out v2os.v2.nl (I think that URL is right). It is an entirely ASM OS for PC that is < meg. Why not start from there?

>Why not start from there?

Because it is written in x86 asm so there's no way you can *start* from there. It's just not portable. You could design something similar though, but that'd be pointless considering what we're going to use it for.

This sounds like a great chance to have a better, open source OS for our favorite calc. You can count on my support.

Damn! Very impressive indeed. One thing they should make sure to do is include RPN when they get to the mathematical stuff.... ability to switch from RPN to Algebraic mode like in the RPN program by Lars would be great.

totally agree. RPN !
I would like to see a Very Fast 3D graphing with a real 3D engine...that's what I received form TI-cares:

Amalfi,

Thank you for your e-mail.

Texas Instruments has no announced plans to develop a new AMS that will specifically make 3D graphs graph faster.

If you have any further questions or comments, please feel free to contact us at your convenience. Please copy this message in your response and send directly to ti-cares@ti.com for faster service on replies.

Kind Regards,

John Onufreiczuk

Hehe, morons at TI. It seems as if they just have a program that picks a couple words from your message and generates a auto-reply from that :) Either that or that guy just always sounds like a robot ;) TI is too lazy to rewrite their ROM to do something useful.

I have an idea. Why not send lots of random messages to them containing key words and see if the reply is automated? We could try "3D Graphs" in various stupid contexts, maybe "AMS 2.01" or whatever it was...

Ciaran

No, actually, Once I recieved a reply about 2.01 saying that it is a 3rd party AMS, and you can install it at your own risk. It contained almost no words from my origional email.

Yeah, TI is funny. I sent them a two-three page rant on January 1st this year about how they are trying to screw assembly programers and how their SDK is outrageous, and other similar info, and one month later (February 3), this comes back to me, with comments of mine put in:

<BEGIN TI LETTER>

Thank you for your recent correspondence.

Your e-mail has been passed to our SDK Development Team. We at TI want to hear from our customers and will work to product the best products. We will
respond to your questions as soon as we can.
<--Three and a half months later, no response-->

I hope you find this information helpful. If you have further questions or comments please feel free to contact me at your convenience. Please copy this message in your response and send directly to ti-cares@ti.com.

Sincerely,

******* ******

---------------------------------------- --------------------------------
Educational & Productivity Solutions
<--lol, they think they are productive-->
Customer Support Center E-mail: ti-cares@ti.com
<--sure TI, you must really care, that was a very thoughtful email-->
Texas Instruments
PO Box 650311 M/S 3962
Dallas, TX 75265 (972)917-8324 (Technical Help)
(800) 842-2737 (General Info)
(972)917-0747 (Fax)
---------------------------------------- --------------------------------
www.ti.com/calc

</BEGIN TI EMAIL>

Funny, huh? That same little part about including relevant parts of the message..they must really think I'm stupid.

Lol, the email I sent them was really funny. A few paragraphs of ranting and then a few questioned numbers about the sdk (no wonder they didn't really reply). Here are a few excerpts.

Hello TI. About your sdk for the 89 (and the 92 and
92 Plus)- First of all, I think that it is entirely
unfair to charge people to make FlashApps ...
Most people who program for the TI are amateurs, and
by charging them money to program FlashApps you put a
burden on them to try to sell to break even, while
most people wouldn't be able to sell (m)any copies of
programs for a simple calculator.
<--Info about calc's use in general--> The asm programs crash on it somewhat often and are hard to program well because of the lack of documentation.

... The idea of paying $100 for only three (programs) is absurd. Just look at the computer programming world- for free, you can get great C programming kits and compilers that can be as effective and are arguably better than commercial ones such as Visual C++. There is no was that I am going to spend my time building *commercial* applications for a calculator, of all things, and I believe that paying so much money to develop freeware for a calculator is a terrible idea.
All that the TI community wants from you is the ability to build stable applications such as games and science programs and to not have to change the programs with each upcoming ams, and you have consistantly hindered this ability. By not giving <--lists some specific documentation ideas--> you have prevented them from building programs that are stable and that are compatible with different ams's. You have wasted countless hours of TI programmers with this "let 'em hack" attitude (as said by someone else) that they must use to program even minimally stable programs. While the availibility of these third party programs doesn't necessarily sell as many calculators as the math software for them, they do add a significant consumer base and I know people who have bought calculators from you for playing games and for developing games.
The point is, programmers are a significant part of
your community and it is time that you recognize them.
You have angered many people with the terrible
support of your otherwise great calculators ... The general attitude in the TI community is that you consistantly hinder the efforts of programs, whether this is on purpose or just due to bad support.
... If you don't want people to sell applications without your approval then charge for sdk's that allow for people to program commercial applications while letting amateur programmers program freeware for free.
[Think not of your customers] as simple consumers who should be taken advantage of and squeezed for every last cent. I am sure that TI doesn't need to embrace this Microsoft/AOL type of business strategy that has angered so many consumers in order to prosper. You have become famous through good calculators and quality products, not through bullying programmers and discouraging the community from making good, free programs. I believe that TI can care, as your email address suggests.

[stuff concerning how the 8k limit is impractical, some other stuff, and so on...]

Hmm, what would creating some kind of open source petition to send TI sound like to everyone? After it's all edited, we could encourage everyone to consistantly send it to them and if the dissent is large enough, they might make a few reforms.

TI has one of the worst Calculator "help" lines... I once had a connection problem between the the link and my calculator, and they said "we're gathering information to help you". sounds pretty stupid, huh? well, it's been err... a year, and nothing's happened... I eventually fixed the issue myself, and I sent a email telling them of my solution (just in case), and they said they knew of the solution as a common problem WHEN THE TI-86 WAS RELEASED. at least they responded to my solution...

Since i live only 4 miles away from the texas instruments plant in sherman,Tx.(45 miniuts away from the one in dallas) i got conections. MY friends brother is geting ready to join there development team. Mabe we can negotiate. Oterwise im shure it listed in there mainframes under (crapy os 89).

Hey anyone notice how the person is usually John Onufreiczuk ? I sent 4 messages, and all of them were returned by John Onufreiczuk.

Here is the most recent one...
------------TI message -----------------------------
Thank you for your e-mail.

The TI-89 local variable are only stored during the evaluation of programs
or functions. Once the execution is completed, the variable is deleted.
This saves memory on the unit.

If you have any further questions or comments, please feel free to contact
us at your convenience. Please copy this message in your response and send
directly to ti-cares@ti.com for faster service on replies.

Kind Regards,

John Onufreiczuk

{address info}


========================= ORIGINAL MESSAGE ==========================

[ AVI2687@aol.com on 6/15/00 4:38:48 PM ]

How does the ti-os handle the ti-89's local variables?
(I know it makes it like the localed variable is only used for that function/program, and deleted when once the function/program is run)

----END MESSAGE -------------------------------
Is it just me or did the (robot/person whoever it/he is) just repeat what i said???

PS Why dont we make a hotmail/mail.com email address
(Ti_doesnt_give_a_crap@hotmail.com - tisucks@hotmail.com - TiHater@cheerful.com...)
and send a really BAD email to them -Flaming them left & right-? they cant throw a suit or anything (they'd be the kind who would)
cuz we'll make it just for that email? - As long as they dont trace it ~ send it from a library/school or sumthin....

Ill email them and ask if they're robots.
Wonder if they'd respond...

HACK AMS 2.01 OR 2.02!

Enough w/ the hacking AMS 2.01 and 2.02!!!! I think that a asm OS would be a heck of a lot better than a C OS. And anyway, when I had a TI 83+ and the link broke, I had to call TI's Warranty "Service" (There wasn't any good service) 4 times before someone actually told me that it was under warranty and I could send it in...... I just wish TI would leave all us programmers alone and make their OS easy to use and customizable...... and if we can't make any money off the programs the make for their calcs, why don't they just give us the source?

Peace,
Gtb

Does anyone know when that Triple Triad Game is supposed to come out?

I don't speak french, so i'm not posting on ti-fr... but.. if they were to do a little research on the hp48/49... and replicate that os... we would have one darn nice os... i think in the 49.. rpn/alg is switchable. also, they have support for different stacks (user-made). the programs kick butt, and the os has built-in support for not allowing asm programs to crash (freeze or whatever).. goto http://www.hpcalc.org for research on programs... i would get the hp49.. but well.. i don't need to spen an extra couple hundred bucks... i think if you more or less replicated that.. we would have it pretty good... any replies are welcome...

Once upon a time I felt in that "I would get the HP49G". Just doubting about buying a TI92+ or a HP49G. But everytime I use the HP49G emulator... I feel happy of having a TI89 :) . You know, at first, I wanted a HP49, but my mother(who traveled to the US) told me....no HP avaible in the stores, so, I bought you a TI89...is it the same?
At first, I was deceptioned..."what the H... is this...Texas Instruments??nobody knows that calc!"
Then, surprised...!
Anyway, I still wants that HP...but it has no future,it can't be better than that. A TI89 has the advantage of having a powerfull hardware..someday this people would release "an Ams that takes full advantage of the processor" that was an eternal phrase...

actually... i've used an emulator with a hp48gx rom... i happen to like each calc for certin thing.. hp is much more complex (different ways to load different date.. like libs are different from data is different from matrix.. etc).. anyways.. i think the best of both world would make the ultimate machine (duh!).. but then again.. who cares what anyone thinks?

HP49 suck's, my frind have an HP49 and it chrashes 3 times in an exam.
First i have a TI86 and i think it is better than Hp48 (not HP48gx),ti86 it's easier to use and powerfull.
now i have TI89, the best calculator(with Ti92+) in the world, i compare my ti(HW1) with Hp49. Ti89 is faster, stable, easier, powerfull,...than Hp49.
Hp imitates Texas instruments Calculators and remember the quality: TI calc's have 2 years warranty and HP products only have 6 months warranty.
Excuses my bad english, I'm from spain.

About the math stuff... How powerfull should the OS it be? It would be very hard (next to impossible) to make an OS similar to the TIOS in the math department. We should make an OS that is oriented more for gaming than math. Besides, the TIOS is good enough for math (I mean with add-on flash apps and user progs, it can't be beaten). Gaming however, it can be greatly impoved.
We could start out by modifying an existing rom, using some of the basic math routines for a start. Or, if there are copyright resrictions, we could make a prog that greatly modifies a user's existing rom to a rom more suited for gaming (the prog could be either on calc or on computer)... That way it's the user's choice, sorta like Napster. Btw, why would we make a rom like this anyway? Time could be better spent elsewhere. Tell me what you think. :-)

I will never turn my TI89 to a gameboy... but some Rom modifications would help to make things BETTER in many areas, gaming included.

Uhh, I don't know about what you're saying...

>>About the math stuff... How powerfull should the OS it be? It would be very hard (next to impossible) to make an OS similar to the TIOS in the math department. We should make an OS that is oriented more for gaming than math.

I doubt that many people would pay around $150 for a calculator that only plays games. Although they can be cool, TI games currently don't come close enough to game systems like NEOGEO, which are half the price (and color). If people are really going to switch OS's, they would want to preserve the math.

>>Besides, the TIOS is good enough for math (I mean with add-on flash apps and user progs, it can't be beaten).

It sounds to me like this new OS would actually *replace* TI's...so it really wouldn't be good enough for math unless you coded it from start.

>>Gaming however, it can be greatly impoved.
We could start out by modifying an existing rom, using some of the basic math routines for a start.

Is this physically possible? TI's OS is closed source and aside from h4x0r a while back (in the Gooey89/GIDE fiasco), who was lying, no one has claimed that they had opened up their OS.

>>Or, if there are copyright resrictions, we could make a prog that greatly modifies a user's existing rom to a rom more suited for gaming (the prog could be either on calc or on computer)... That way it's the user's choice, sorta like Napster. Btw, why would we make a rom like this anyway?

This would greatly open the doors to allowing people to modify things within the OS. If people were actually able to match the normal 89 OS in math, they could greatly surpass it in other areas. It would be open source and documented and so it might be simpler for assembly programmers to develop for it. They would know more about how it handled the memory and would be able to avoid address errors for the most part when programming.

On the other hand, TI's OS is extremely stable when you do not use asm programs. This OS might crash in the middle and so losing information might be unavoidable (unless they did something like AMS 2.03 by preserving archives after crashes) if it does, whereas normally you are fine as long as you don't run asm (I haven't lately because I haven't had quite as much time to test stuff but DoorsOS .96 is very unstable..I'll probably try TE-OS).

Personally, I think that this could be a great project. If stable and as functional as the normal OS it would provide a great alternative. Also, does anyone know how complex it would be to restore TI's OS? Isn't there some FLASH memory that can't be re-written (or can it only with an AMS update..I thought that there was a backup in case the AMS update failed)? If it would be completely safe to install this new OS and you could go back to TI-OS, this would be a great idea. It might take until the second generation OS to make it a completely feasible alternative, but if this works out, with time, it would be great...

I think they understand this.
If the OS is used for games only, and it skimps out on _any_ math-related feature, I won't use it, most of the people here won't use it, and TI will do _everything_ in its power to keep its users from doing this.

You know it, I know it, the developers of this OS know it, and TI knows it.

--BlueCalx

I once programmed my own EOS (not CAS, though). It works like the one on the TI-83, only a little better. It's in C++, but it should be easy to port. It includes converting a string to an expression and back.

Then why do they try to screw all the third party developers all the time? If they made it easier for them, more people would buy their calculators because they knew that there would be tons of programs (math, games, and whatever)to put on it. They definitely have something up their butts.

- KH

I disagree, i think TI would be more upset if the os based on math because that competes with their softwart and the supplemental income from their apps. I do agree, however, that there does need to be math functions in the os, b/c too many people use their calculator for math.

All I was saying is that we should keep it simple (hense my nickname jestbsemple). I wanted to give a few ideas that would help actually get this thing off the ground. We should start small. After that, we can do anything we dream of.

Do we "simply"know where to start from ?

build it with both a cas and an rpn for math and make them hot-switchable, with a converter between them for the math side of the os. for games and programming, we should have a true directory structure and integrated on-calc basic, c/c++, and asm programming, with c/c++ and asm compilers
we could also have integrated telnet-type chatting, and possibly a way to dial into the internet using a cellphone, although this last part would be quite difficult, both from a hardware and a software standpoint

p.s. it should also have a cli unless someone figures out how to make a gui run as fast as a cli, or where the lagtime of the gui behind the cli is infinitessimal

Geez, should it make coffee and walk the dog too?! We have to be realistic about what we can do. TI is a giant company with paid employees who work 8 hours a day, and they were able to program the AMS in C in 3-5 years. Now we think that a bunch of rather disorganized hobbyists who work in their free time can program something better than TI, in assembly language no less?! It would take CENTURIES. I think we need to be realistic here. We have shown that we can make good programs, but I don't think we can match TI's AMS. We should concentrate on what we do best.

be carefull, this could be subject to the Digital Milenium Copyright Act... reverse engeneering and all that. --damn the DCMA!

eggo

Darn DMCA! That thing ticks me off. At least we have free geocities accounts and such that they can't track and we can link to them anonymously if they threaten legal action. Oh wait, how about that pending lawsuit against someone for linking to their site because of DMCA and the slashdot.org discussion that Micro$oft wanted to censor because of the link...hrmm...that thing ticks me off. I bet you someone patents links within the next 6 months.

That's easy. Do the reverse engineering in a country that doesn't have any such laws. Like, for example, most places outside the US.

Oh, I reckon TI are breaking the law by selling their calculators in France. Doesn't the ti89 hw2 include some encryption algorithms / checksums that are more powerful than allowed by French law?

Ciaran

Please don't even get me started on the DMCA!

No it couldn't. They aren't planning to use any code from the TI-OS, hence the impossibility of reverse engineering.

--BlueCalx

They still have to crack the 512bit checksum encryption key, though. Wouldn't that imply reverse engineering?

Yeah, besides the legal issues, how do they plan on cracking the 89's encryption? Have they been hiding their secret Cray supercomputer laboratory from us?

don't we already each own a TI 89 calculator?
Why can't we cut functions out of a rom that we already paid for, and use them untill we make a better one?

I mean, we know the entry points for the functions... chop chop.

Also, anybody use that pay software for TI calcs? I won't buy it till I've tried it... and I don't think anybody will buy it... so I aint going to try it... and I aint going to buy it!

I wouldn't mind hacking some of that software to try it.. but that would be against the @#$^@@ law... methinks. I am not a judge, or even a lawyer.

>>don't we already each own a TI 89 calculator?
Why can't we cut functions out of a rom that we already paid for, and use them untill we make a better one? <<

and hence the reason that the DCMA is an un-just law

The TI-89 is a powerhouse, and it shouldn't be for games. It's somewhat good at games, but if you want a gaming machine, go buy a gameboy. Not many people will even switch to the new OS, I believe, if it were optomized for gaming. This calculator got me through calculus easily, and I'm not giving up its existing functionality just to play some Mario or Tetris. It should be written in assembly, and should be optomized to handle complicated expressions and 3D graphs much better than TIOS. Complicated integrals, for example, take forever, and I'm sure, longer than they need to be. I don't have anything against games on the calculator, but personally, I wouldn't consider switching out of TIOS unless the math capabiliies were better.

I think TIOS has a sucky user interface.
I want to replace it with something more 86ish

I choke up every time I look at the 89 and realize that TI wasted a huge portion of the screen to frilly menus that don't look like they are associated to the buttons.

I want a pretty print input mode.

I want the stack to be cleaned up a bit.

I want the keys remaped so I can delete characters properly... without risk of killing a whole line of complicated entry

I want UNDO.

If we could add these things to the 89 I would be very happy.

I and other people arround my house find the 89 to be almost as hard to use as an hp48g. Lets see if we can patch that stuff before we totaly redo something

<I want a pretty print input mode>

Just stop and think about how this would be accomplished... do you realize how hard it would be to try to type in a function if it automatically tried to pretty print what you're typing AS you type it???

Have you ever used the equation editor that comes with Microsoft Word? It's cool! That's what we need.

Yeah, something like that...

***Tucked to be close to the 1st reply)
The main point was:

MORE 86ish (More like a TI 86 OS)
**Stop wasting screen space!
And I need the UNDO feature, and delete key maped properly, cause I hate typing stuff in, and having it wiped out! Pretty Print Input would help us make less errors in our math.

I don't care if we change much else with the TIOS, it seems OK... but it is hard to use, I'll try to help make a new OS.

>How hard it is.. blah blah blah

I've almost done it with a TI-86 in Basic!
It is not releasble, and it is hella slow...

If you can do it systematically on paper, then you should be able to do it on a computer... unless you cant do it on paper...

I also believe the HP49 has pretty-print (called textbook style by HP, I think) while you're inputting.

My 7 year old sharp graphing calc that has less then a kb of memory for variables had it.. so get yur nose out of your but and be careful what your talking about.. if sharp could do it 7 years ago, we can do it now twice as easily

Ummm, if you read carefully, I never said it couldn't be done, I was merely voicing an opinion that editing an equation while it's in pretty print would be strange... I do grant that it would definitely look better and might make it easier to keep up with parentheses and whatnot. It would also have it's downsides though, because it seems it would be difficult to edit enormous complex fractions. But since I've never used anything that auto pretty-prints, so I'm probably wrong. Oh well.

Try MathCAD. It has an auto pretty print input system that is quite intuitive. I wish TIOS had the capability to format text (superscript/subscript) that would rock.