So, the freshmen at my school have been getting 84+Ts as mandatory for the courses. For those of you who don't know, this is basically an 84+, but completely black, without I/O port, and with an exam mode indicator LED at the top. It was a bit cheaper, which is why the school chose this instead of the regular 84+. They don't care about the exam mode.

Some knew about my "calculator knowledge" and went to me, expecting me to put some games on it. That's when I found out that the Asm(, AsmComp( and AsmPrgm tokens are just completely missing on this calculator. Freeware apps don't validate. This has probably been done to prevent Z80 freaks from hacking the LED. The problem now is, how do you get arbitrary code running on it? Has someone found the proprietary key for apps? Can you send over a hacked OS?
Yeah, like the 82A, the 84+T is crippled crap, with a different product ID (0x1B for the 84+T, 0x0B for the 82A). The 82A was recently forced open; I haven't heard about the same being done for the 84+T.
RIP

And I thought a TI-83 plus was obligatory for schools. At least you don't have TI-84 plus CE's with the latest version. Wink
But you can still run ASM on the 84+CE with 5.3.1, writing ASM on-calc is what is disallowed. It sounds like the 84+T doesn't allow running ASM at all. Also, the Asm84CEPrgm token wasn't removed, just disabled whereas the OP states that the tokens have been removed completely from the 84+T.

Question for the OP: You said it doesn't have the link port. Does it have the USB port still?
ShinyGardevoir wrote:
And I thought a TI-83 plus was obligatory for schools.

Not at my school (in Belgium).

RogerWilco wrote:
Question for the OP: You said it doesn't have the link port. Does it have the USB port still?

Yes. You can actually see the hole for the port on the picture I posted. The exam mode LED is where the link port was.

RogerWilco wrote:
It sounds like the 84+T doesn't allow running ASM at all.

I suspect that if you were able to sign an app with TI's proprietary keys, it could. The calcs have the default "ProbSim" "PlySimul" etc. apps installed. The tokens seem to be gone though, and there isn't any way I know of to put them in.

Oh, I just thought of something: maybe I can create a BASIC program with an Asm( token in it (on my 84+) and send it to an 84+T. That would work if they've just removed the token from the catalog, but given that they also removed the freeware app key, I suspect they've put a little more protection in.
fghsgh wrote:
ShinyGardevoir wrote:
And I thought a TI-83 plus was obligatory for schools.

Not at my school (in Belgium).

RogerWilco wrote:
Question for the OP: You said it doesn't have the link port. Does it have the USB port still?

Yes. You can actually see the hole for the port on the picture I posted. The exam mode LED is where the link port was.

RogerWilco wrote:
It sounds like the 84+T doesn't allow running ASM at all.

I suspect that if you were able to sign an app with TI's proprietary keys, it could. The calcs have the default "ProbSim" "PlySimul" etc. apps installed. The tokens seem to be gone though, and there isn't any way I know of to put them in.

Oh, I just thought of something: maybe I can create a BASIC program with an Asm( token in it (on my 84+) and send it to an 84+T. That would work if they've just removed the token from the catalog, but given that they also removed the freeware app key, I suspect they've put a little more protection in.


That’s a good idea. I know the disabled token on the 84+CE is still in the catalog, but your idea is worth a try, given that they don’t have to completely remove it to make it look impossible.
What you can do is install a program that puts "Asm84CEprgm" to a Y-var and recall it whenever necessary.

Also, although I never owned a TI-84 Plus T, I think that concept will not apply; AsmPrgm on that calculator would be something like
ShinyGardevoir wrote:
What you can do is install a program that puts "Asm84CEprgm" to a Y-var and recall it whenever necessary.

Also, although I never owned a TI-84 Plus T, I think that concept will not apply; AsmPrgm on that calculator would be something like


Yeah, assuming they did remove it entirely it would take more than putting it into a variable to make it possible. You'd have to do something similar to the exploit found on the TI-85/TI-82 calcs. I have no experience with ASM, but I would suspect that maybe one of the shells might be able to build in an exploit to circumvent the 84+T limitations (I'm not sure what that would look like exactly though).
RogerWilco wrote:

Yeah, assuming they did remove it entirely it would take more than putting it into a variable to make it possible. You'd have to do something similar to the exploit found on the TI-85/TI-82 calcs. I have no experience with ASM, but I would suspect that maybe one of the shells might be able to build in an exploit to circumvent the 84+T limitations (I'm not sure what that would look like exactly though).

The problem is that the calc just rejects any app which isn't signed with the proprietary key (at least I think so because it refuses to store it, maybe pulling a battery during validation may help? It was quite long ago that I last tried this so I don't really remember what happened). This is currently the only solution I see, or maybe we could send over the 84+ OS, somehow.
If someone who knows a bit about disassembling .8xu files happens to read this, here is the calc's operating system (they call it version 5.1). I could only find it on the Dutch version of TI's website, but that version is publicly available. (for some reason, I can't get the link working, so you'll have to copy & paste it) https://mega.nz/#!hVtwyAba!EiVJgEq_gWbvWv0HkA7LROyXzIJ0SrB7E3GlRRbDH9A

EDIT: Also, I found this in the manual:
Quote:
Apps are independent applications that are built-in the TI-84 Plus T. Apps cannot be deleted. Additional Apps cannot be loaded.

Somewhere else in the same document, I found:
Quote:
You can transfer the operating system (OS) and built-in Apps from one calculator to another using a USB unit-to-unit cable. You must have all factory-loaded Apps on the sending calculator prior to this transfer. If any Apps have been deleted from a TI-84 Plus T, you cannot send the OS/Apps unit-to-unit from that TI-84 Plus T. You will have to download the correct TI-84 Plus T OS (with Apps) file for the TI-84 Plus T at education.ti.com/go/download. Then use TI Connect CE to load the complete OS/Apps file to the TI-84 Plus T.

So this suggests they can be deleted.
And:
Quote:
AppVars and TI-Basic Programs should be reviewed for use after the transfer between the TI-84 Plus T family of graphing calculators. Some AppVars may not setup an App as expected. Some of the TI-Basic Programs will need to be modified due to the difference in screen resolution and new commands. TI-Basic Programs created using commands available only in the latest OS version will not transfer to graphing calculators with an earlier OS version.

About the difference in resolution: they are talking about the CE. They are treating the CE and the T as equals in this document, even supplying screenshots of the CE.
I've never heard of that model, but it sounds like it specifically is not supposed to run any third-party code at all. My guess is that the internal hardware is nearly identical to a stock TI-84 Plus. I'm curious whether the testing LED is controlled through the I/O port hardware or through the GPIO block. Either way, assuming TI couldn't be bothered to spin a new ASIC (I'd bet a few hundred dollars they didn't, but I don't imagine anyone here would take me up on that), the LED is software-controlled, hence why they can't allow any third-party native code at all: it would be trivial for such code to blink the testing LED to simulate testing mode.

Regardless of how the testing LED operates, unless TI has decided to start incorporating flash into their ASICs*, you can still solder an external programmer to the flash chip and flash the old boot code and OS to the calculator. This could be done with any microcontroller with 30-some GPIO pins, or a Raspberry Pi with two of these. (No worries---it's a 3.3 V flash chip!) It's certainly do-able if you have a steady hand---but for the cheaters out there, know that it's easier just to learn math.

fghsgh wrote:
CE. They are treating the CE and the T as equals in this document, even supplying screenshots of the CE.
Hahaha, that's a glaring copy edit fail.


*Why not, TI? I want to see a calculator using a unified memory architecture based on FRAM. You have the IP for FRAM. You could completely eliminate the external flash chip from the design, offsetting the extra cost of a large on-die FRAM bank. Do it for the TI-84 Plus CE, fix that internal bus latency, and release a TI-84 Plus CE II that's much faster and has 4 MB of memory. (But please add a hardware write protect feature for firmware/applications and read-only user variables. It'd be really nice to have an API for updating appvars in-place without having to cycle them between protected and unprotected. And consider using DRAM for the VRAM.)
DrDnar wrote:
I've never heard of that model, but it sounds like it specifically is not supposed to run any third-party code at all. My guess is that the internal hardware is nearly identical to a stock TI-84 Plus. I'm curious whether the testing LED is controlled through the I/O port hardware or through the GPIO block. Either way, assuming TI couldn't be bothered to spin a new ASIC (I'd bet a few hundred dollars they didn't, but I don't imagine anyone here would take me up on that), the LED is software-controlled, hence why they can't allow any third-party native code at all: it would be trivial for such code to blink the testing LED to simulate testing mode.

This is a Dutch model to comply with the Dutch exam rules (or something). This has no use in Belgium other than the lower price. I would say we need to inspect the OS and see if the ports are used in the same way (they probably are). Now you say it, maybe the I/O tip is a red LED and the ring is a green LED. In that case, they could use exactly the same hardware and really just add an LED. Although the nspire has advanced hardware-handled timings for the LED, where you just say "blink every X seconds" and it blinks. The TI-OS programmers may have been lazy and asked the hardware engineers

DrDnar wrote:
Regardless of how the testing LED operates, unless TI has decided to start incorporating flash into their ASICs*, you can still solder an external programmer to the flash chip and flash the old boot code and OS to the calculator. This could be done with any microcontroller with 30-some GPIO pins, or a Raspberry Pi with two of these. (No worries---it's a 3.3 V flash chip!) It's certainly do-able if you have a steady hand---but for the cheaters out there, know that it's easier just to learn math.

Okay. I'll try that if I get my hands on one. If the ports match up, of course.

DrDnar wrote:
*Why not, TI? I want to see a calculator using a unified memory architecture based on FRAM. You have the IP for FRAM. You could completely eliminate the external flash chip from the design, offsetting the extra cost of a large on-die FRAM bank. Do it for the TI-84 Plus CE, fix that internal bus latency, and release a TI-84 Plus CE II that's much faster and has 4 MB of memory. (But please add a hardware write protect feature for firmware/applications and read-only user variables. It'd be really nice to have an API for updating appvars in-place without having to cycle them between protected and unprotected. And consider using DRAM for the VRAM.)

I hope you realize TI is not reading this.
fghsgh wrote:
Now you say it, maybe the I/O tip is a red LED and the ring is a green LED. In that case, they could use exactly the same hardware and really just add an LED.
It's not quite that simple. The PCB does actually need to be changed a little to add driving transistors and loading resistors for the LED. (Much as we like to complain about TI, they're not going to try to cut corners and omit those. They don't make dumb engineering decisions, they make short-sighted engineering decisions.) Unlike altering a chip, altering a PCB is super cheap.

fghsgh wrote:
The TI-OS programmers may have been lazy and asked the hardware engineers
Unlikely. It costs at least a few thousand dollars to fab a new or revised chip. The programmer's time is much cheaper. (I suppose they could have done it with a 555 timer, but I doubt it. A few cents over tens of thousands of calculators adds up.)

fghsgh wrote:
I hope you realize TI is not reading this.
They admit to dropping by the forums from time to time. What they never, ever do for any reason is actually talk to us like real people.
Lionel Debroux wrote:
Yeah, like the 82A, the 84+T is crippled crap, with a different product ID (0x1B for the 84+T, 0x0B for the 82A). The 82A was recently forced open; I haven't heard about the same being done for the 84+T.

I'd like to start a petition to officially rename the TI-82 Advanced to cripply boiis

mr womp womp wrote:
Lionel Debroux wrote:
Yeah, like the 82A, the 84+T is crippled crap, with a different product ID (0x1B for the 84+T, 0x0B for the 82A). The 82A was recently forced open; I haven't heard about the same being done for the 84+T.

I'd like to start a petition to officially rename the TI-82 Advanced to cripply boiis



I second this Graphing Calculator
DrDnar wrote:
fghsgh wrote:
Now you say it, maybe the I/O tip is a red LED and the ring is a green LED. In that case, they could use exactly the same hardware and really just add an LED.
It's not quite that simple. The PCB does actually need to be changed a little to add driving transistors and loading resistors for the LED. (Much as we like to complain about TI, they're not going to try to cut corners and omit those. They don't make dumb engineering decisions, they make short-sighted engineering decisions.) Unlike altering a chip, altering a PCB is super cheap.

fghsgh wrote:
The TI-OS programmers may have been lazy and asked the hardware engineers
Unlikely. It costs at least a few thousand dollars to fab a new or revised chip. The programmer's time is much cheaper. (I suppose they could have done it with a 555 timer, but I doubt it. A few cents over tens of thousands of calculators adds up.)

I meant that maybe they plugged an LED into the I/O port and surrounded it with plastic. That way, they wouldn't need any new hardware or PCB redesigns. You're probably right though. It was just a theory. The other theory was that they used some premade blinking LED hardware, like on the nspire. I should probably just get an 84+T and take it apart instead of theorizing like this. It is not even $5 cheaper than a regular 84+. I am starting to think our school changed to these just to make pupils unable to play games.

DrDnar wrote:
fghsgh wrote:
I hope you realize TI is not reading this.
They admit to dropping by the forums from time to time. What they never, ever do for any reason is actually talk to us like real people.

Oh really? Good to know. Maybe we should try not to insult them too much.


RogerWilco wrote:
mr womp womp wrote:
Lionel Debroux wrote:
Yeah, like the 82A, the 84+T is crippled crap, with a different product ID (0x1B for the 84+T, 0x0B for the 82A). The 82A was recently forced open; I haven't heard about the same being done for the 84+T.

I'd like to start a petition to officially rename the TI-82 Advanced to cripply boiis


I second this Graphing Calculator

Me too. But the 84+T also needs such a name!
fghsgh wrote:
I should probably just get an 84+T and take it apart instead of theorizing like this.


All this has been done, almost 3 years ago now Razz (wow, time flies).
See pics etc. here https://tiplanet.org/forum/viewtopic.php?t=18822
Adriweb wrote:
fghsgh wrote:
I should probably just get an 84+T and take it apart instead of theorizing like this.


All this has been done, almost 3 years ago now Razz (wow, time flies).
See pics etc. here https://tiplanet.org/forum/viewtopic.php?t=18822

Nice. The French we learn at school finally has a use.

I'll translate some stuff here:
- It has a lot of apps. (17 in total, all useless)
- You can't delete the apps. The delete button only hides them from the menu.
- It has built-in language apps for quite a lot of languages.
- It has the 84+SE flash chip.
- The 82A hack consisted of sending a modified backup back to the calc. The 84+T doesn't have backup functionality, so that's why this breaks. Do not update the OS of the 82A or they might do the same thing there.
- It has two different exam modes with two different LED colors (green (very restricted) & orange (allows Conics, Inequalz, and PlySmlt2)).
- There is no trace of an I/O port left on the PCB.
Update: in the disassembly of the OS (only page 00 for now), I've found these values being OUTed to port 00:

00 (only once)
C0 (multiple times)
(contents of register C) (only once)

C0 has no meaning for the link port, so it must be used for something else on this hardware. I don't know whether that something is the LED. As for the contents of register C: it calls a subroutine and probably takes the value of C from there. I don't have the time right now to check that routine though.

EDIT: there is also one out (00) on page 76, 6 on page 77 and 1 on page 7c. Will look into those later.
  
Register to Join the Conversation
Have your own thoughts to add to this or any other topic? Want to ask a question, offer a suggestion, share your own programs and projects, upload a file to the file archives, get help with calculator and computer programming, or simply chat with like-minded coders and tech and calculator enthusiasts via the site-wide AJAX SAX widget? Registration for a free Cemetech account only takes a minute.

» Go to Registration page
Page 1 of 1
» All times are UTC - 5 Hours
 
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

 

Advertisement