Author Topic: New TI-Nspire CX hardware revision K  (Read 25922 times)

0 Members and 1 Guest are viewing this topic.

Offline Vogtinator

  • LV9 Veteran (Next: 1337)
  • *********
  • Posts: 1193
  • Rating: +108/-5
  • Instruction counter
    • View Profile
Re: New TI-Nspire CX hardware revision K
« Reply #15 on: September 04, 2013, 11:12:26 am »
Actually, releasing an exploit in "binary" is senseless as almost every exploit is a binary. In case of nlaunch, it's (IIRC) a hand-crafted tcc-header. Obfuscation in any kind is senseless as the exploit part has to stay the same. Analysing which bug makes an exploit successful is as easy as comparing a normal binary with the exploit and maybe running the code with a breakpoint at the particular code line which loads the exploiting part. But explanations, how it works, should be hidden and maybe even removed.

Offline DJ Omnimaga

  • Clacualters are teh gr33t
  • CoT Emeritus
  • LV15 Omnimagician (Next: --)
  • *
  • Posts: 55941
  • Rating: +3154/-232
  • CodeWalrus founder & retired Omnimaga founder
    • View Profile
    • Dream of Omnimaga Music
Re: New TI-Nspire CX hardware revision K
« Reply #16 on: September 04, 2013, 11:55:18 am »
The issue with lack of source code, though, is if the Ndless developers suddenly become way too busy or lose Internet access completely, preventing them from working on it again nor giving the source code to someone else to take over, then nobody else can take over. I doubt the current team would do that, but life is unpredictable sometimes. We all saw in the past what happened to many projects (especially for the TI-83 Plus) that had no source available after the author vanished off the face of the Earth. However, I would be curious about how many connections to Ndless servers are done by TI.

Offline Lionel Debroux

  • LV11 Super Veteran (Next: 3000)
  • ***********
  • Posts: 2135
  • Rating: +290/-45
    • View Profile
    • TI-Chess Team
Re: New TI-Nspire CX hardware revision K
« Reply #17 on: September 04, 2013, 12:30:39 pm »
Lack of source code for Ndless, nLaunch and suchlike tools would indeed not be much of an impediment for TI. It makes little sense to make those even partially closed-source.

Quote
But explanations, how it works, should be hidden and maybe even removed.
That's what the author(s) of nLaunch did: the source code does not explain what any of the patches do. For adding CM OS support to nLaunchy, I reverse-engineered several of the patches, to find out what to patch (it was a filename), so as to be able to efficiently find the same data in other OS versions. Not that it was hard.
Member of the TI-Chess Team.
Co-maintainer of GCC4TI (GCC4TI online documentation), TILP and TIEmu.
Co-admin of TI-Planet.

Offline compu

  • LV5 Advanced (Next: 300)
  • *****
  • Posts: 275
  • Rating: +63/-3
    • View Profile
Re: New TI-Nspire CX hardware revision K
« Reply #18 on: September 04, 2013, 01:30:38 pm »
As far as I know, PTT hacks are currently only available on cncalc.org for registered users with several posts (else, you can't download). And of course, only if you manage to find them in chinese...
Well, PTTcopier is still available and findable with a simple google search.

Offline critor

  • Editor
  • LV11 Super Veteran (Next: 3000)
  • ***********
  • Posts: 2079
  • Rating: +439/-13
    • View Profile
    • TI-Planet
Re: New TI-Nspire CX hardware revision K
« Reply #19 on: September 04, 2013, 01:33:54 pm »
So easily?... :o
Ok, then TI is right if Ndless PTT hacking tools haven't disappeared and can be found so easily with Google - we can't complain... :(
« Last Edit: September 04, 2013, 01:38:11 pm by critor »
TI-Planet co-admin.

Offline Lionel Debroux

  • LV11 Super Veteran (Next: 3000)
  • ***********
  • Posts: 2135
  • Rating: +290/-45
    • View Profile
    • TI-Chess Team
Re: New TI-Nspire CX hardware revision K
« Reply #20 on: September 04, 2013, 01:46:18 pm »
Of course we can complain. Tools which can be directly or indirectly used to attack their business model exist only because they're consistently behaving as worsening pricks against us programmers and users (all of them were released after, and presumably as a consequence of, the announced vulnerability fix in OS 3.2). TI EdTech has no business crossing our freedom to tinker.

The PTT is indeed "laughably insecure and extremely easy to beat". On the one side, TI EdTech should be ashamed, both for making ineffective so-called protections and for daring to present such ineffective crap to standardized testing regulation authorities using "hardware secure" and similar void bullshit talk. On the other side, regulators should be ashamed of their own bloody incompetence, both for making silly exams and for letting companies lie to them so blatantly...
« Last Edit: September 04, 2013, 02:04:00 pm by Lionel Debroux »
Member of the TI-Chess Team.
Co-maintainer of GCC4TI (GCC4TI online documentation), TILP and TIEmu.
Co-admin of TI-Planet.

Offline DJ Omnimaga

  • Clacualters are teh gr33t
  • CoT Emeritus
  • LV15 Omnimagician (Next: --)
  • *
  • Posts: 55941
  • Rating: +3154/-232
  • CodeWalrus founder & retired Omnimaga founder
    • View Profile
    • Dream of Omnimaga Music
Re: Re: Re: New TI-Nspire CX hardware revision K
« Reply #21 on: September 04, 2013, 05:19:02 pm »
So easily?... :o
Ok, then TI is right if Ndless PTT hacking tools haven't disappeared and can be found so easily with Google - we can't complain... :(
in our defense, those tools were made after TI locked the Nspire down and tried to prevent downgrading Z80 models, partially as an act of revenge against TI's actions. I am fairly sure that similar tools will take much longer to arrive on the HP Prime or if the PRIZM gets a teacher mode but still unofficially allows ASM/C, simply because the community won't be as angry against them.

Offline njaddison

  • LV4 Regular (Next: 200)
  • ****
  • Posts: 176
  • Rating: +24/-27
    • View Profile
Re: New TI-Nspire CX hardware revision K
« Reply #22 on: September 04, 2013, 05:24:12 pm »
I hope we can get someone like anonymous to f*** up TI's website. But that's not what I am commenting about.
My nspire cx cas says "P-02131" on the back. What hardware revision is that?
And didn't extended mention in one of his posts (@ ndlessly.wordpress.com) that ndless may be possible through lua?
Also, as long as an nspire has a serial port, it should be hackable right? I don't know much about nspires, but couldn't a new version of ndless be made as a modified boot2 (send to your nspire through serial port)?
<a href="http://www.nerdtests.com/ft_nt2.php">
<img src="http://www.nerdtests.com/images/badge/nt2/5f42ec78e054645d.png" alt="NerdTests.com says I'm a Highly Dorky Nerd God.  Click here to take the Nerd Test, get geeky images and jokes, and talk to others on the nerd forum!">
</a>


Offline Jim Bauwens

  • Lua! Nspire! Linux!
  • Editor
  • LV10 31337 u53r (Next: 2000)
  • **********
  • Posts: 1881
  • Rating: +206/-7
  • Linux!
    • View Profile
    • nothing...
Re: New TI-Nspire CX hardware revision K
« Reply #23 on: September 04, 2013, 05:27:04 pm »
No, the boot2 is also signed by a private key. It is therefore only possible to use official boot2's. Ndless might indeed be possibly by exploiting bugs in the Lua framework, but those bugs have to be uncovered/found first before that can happen. Bugs that can be exploited and that can result into execution of custom code are very rare. That's the reason why Ndless 3.1 took such a long time to be released in the first place.
« Last Edit: September 04, 2013, 05:28:55 pm by Jim Bauwens »

Offline critor

  • Editor
  • LV11 Super Veteran (Next: 3000)
  • ***********
  • Posts: 2079
  • Rating: +439/-13
    • View Profile
    • TI-Planet
Re: New TI-Nspire CX hardware revision K
« Reply #24 on: September 04, 2013, 05:31:06 pm »
I hope we can get someone like anonymous to f*** up TI's website. But that's not what I am commenting about.
My nspire cx cas says "P-02131" on the back. What hardware revision is that?

Hi.

Your datestamp is probably "P-0213I" - you've got hardware revision I, the one before the evil hardware revision J.

It would be nice to check if you can install Ndless on it ;)

And didn't extended mention in one of his posts (@ ndlessly.wordpress.com) that ndless may be possible through lua?

Ndless can be used from Lua programs, but it has to be installed first.


Also, as long as an nspire has a serial port, it should be hackable right? I don't know much about nspires, but couldn't a new version of ndless be made as a modified boot2 (send to your nspire through serial port)?
We thought so and were wrong.
For Ndless (and Nlaunch), we need Boot2 and OS versions 3.1.
And versions 3.1 just don't run properly on hardware revision J, K and above (we don't know about hardware revision I yet).
A modified Boot2 is going to be rejected by the Boot1.
« Last Edit: September 04, 2013, 05:32:01 pm by critor »
TI-Planet co-admin.

Offline Vogtinator

  • LV9 Veteran (Next: 1337)
  • *********
  • Posts: 1193
  • Rating: +108/-5
  • Instruction counter
    • View Profile
Re: New TI-Nspire CX hardware revision K
« Reply #25 on: September 04, 2013, 05:32:11 pm »
If you find a bug which results in the nspire doing weird things (crashing, displaying crap) or something similiar, report it, please ;)
The serial port is nice to have and absolutely necessary to rescue a calculator with broken diags or boot2 which can both (?) be deleted with keyboard commands. So there's almost no chance in TI removing the uart :)
The possibility to upgrade the OS means they have to expose the binary to the public and it is always possible to decrypt it.
Hardest thing is to find bugs in the disassembled code and exploit them.

Offline njaddison

  • LV4 Regular (Next: 200)
  • ****
  • Posts: 176
  • Rating: +24/-27
    • View Profile
Re: New TI-Nspire CX hardware revision K
« Reply #26 on: September 04, 2013, 05:42:23 pm »
My normal nspire (grayscale) boots the TI-Nspire OS when I am using the TI-84+ keypad. I have to remove the keypad multiple times for it to boot correctly. But, I am running os 3.1 on my normal nspire. Some guy mentioned in a post that his friend's nspire cx screen would turn upside down.
EDIT:
Could an ndless 3.1 program be made to downgrade another calc through the mini usb cable?
« Last Edit: September 04, 2013, 05:43:39 pm by njaddison »
<a href="http://www.nerdtests.com/ft_nt2.php">
<img src="http://www.nerdtests.com/images/badge/nt2/5f42ec78e054645d.png" alt="NerdTests.com says I'm a Highly Dorky Nerd God.  Click here to take the Nerd Test, get geeky images and jokes, and talk to others on the nerd forum!">
</a>


Offline Vogtinator

  • LV9 Veteran (Next: 1337)
  • *********
  • Posts: 1193
  • Rating: +108/-5
  • Instruction counter
    • View Profile
Re: New TI-Nspire CX hardware revision K
« Reply #27 on: September 04, 2013, 05:43:58 pm »
Hardware faults, not reproducable on every calc.

Offline DJ Omnimaga

  • Clacualters are teh gr33t
  • CoT Emeritus
  • LV15 Omnimagician (Next: --)
  • *
  • Posts: 55941
  • Rating: +3154/-232
  • CodeWalrus founder & retired Omnimaga founder
    • View Profile
    • Dream of Omnimaga Music
Re: New TI-Nspire CX hardware revision K
« Reply #28 on: September 04, 2013, 05:50:58 pm »
Does OS 3.2.6 fit in the TI-Nspire CX/CX CAS memory, by the way?

Offline Vogtinator

  • LV9 Veteran (Next: 1337)
  • *********
  • Posts: 1193
  • Rating: +108/-5
  • Instruction counter
    • View Profile
Re: New TI-Nspire CX hardware revision K
« Reply #29 on: September 04, 2013, 05:53:47 pm »
If it doesn't, it wouldn't run :P
The CX has 64 MiB of SDRAM, so that's enough even for Windows 95 ;)