So here is the story.
Here is what happened recently about the Nspire...

Let's do that in a "diary-style"



April the 6th :
TechPoweredMath's blogger seemed to have had an interview with Mark Fry, a TI manager, who revealed a great-looking Periodic table, that would be released with the upcoming OS 3.0.

Here's the link to the activity (download) : http://education.ti.com/calculators/downloads/US/Activities/Detail?id=16149

This looks very nice, with some graphics quality never seen before...




April the 7th :
According to several almost-official and trustable sources, the OS 3.0 should be released very soon... Some issued already have been reported (with cZeros() for example). See here.

This would be the first time an OS would break the ascending compatibility with the TNS documents.

It's then recommended to stay with the current OS (2.1) especially if you're soon going to have some exams...

For now, we have no idea if it will be possible to downgrade from 3.0 to previous versions...




April the 7th (later) :
The TI-Nspire OS 3.0 is out !
Its version is 3.0.1.1753 to be exact.

Most of the new features that come with this OS have already been discussed, as they were announced in the previously released Release Notes.
- the Vernier Data Quest App has been added
- both 3D graphs and Differential Equation graphs are now possible
- Exact values appear in the graphs of functions
- new Document type, *.tnsp (allow online publishing of TI-Nspire documents)

Quote:
More interesting, however, is that TI has released different OS files for the new Nspire CX Series, with extension *.tco for the TI-Nspire CX and extension *.tcc for the TI-Nspire CX CAS. These two file groups join the original *.tno and *.tnc for the original TI-Nspire and TI-Nspire CAS, respectively. A quick analysis comparing the *.tno and *.tco file types yields some interesting results. First, in order to save on size, the example *.tns documents are no longer included in the OS Upgrade for the *.tno file (although they are present in *.tnc). Secondly, both TI-Nspire.img and boot2.img are in fact smaller on the TI-Nspire CX OS than in the regular TI-Nspire. Finally, although the 84 Emulator has been removed from the TI-Nspire CX, the full TI-84 Plus ROM and emulator are still present on the TI-Nspire CX. Does this mean it will in fact be possible to use the Emulator, either through an actual TI feature or a later hack???

It should also be noted that a new boot 2 version is present in these files. In addition, nspire_emu will not run the OS.

As a closing note, it is probably best to hold off on upgrading the OS on an actual Nspire until further testing is done. As far as is known now, this OS very well might block downgrading/Ndless/countless other things.


WARNING: Upgrading to OS 3.0 will block your ability to downgrade to an older Operating System!

Update: According to tests performed by critor, removing the new boot 2 with TNOC will allow one to successfully upgrade and then downgrade back to an older Operating System!! Be sure to remove the boot 2 before attempting to upgrade!!!




April the 8th (at midnight, less than half a day after the OS 3.0 release) :
The Anti-Downgrade protection has been broken !
It's now possible to get back on previous OSes.



The trick was basically to removed the new "boot2" introduced in the 3.0 os files, so it doesn't get updated in the calculator, thanks to the tool TNOC.

For now, we don't know what to do with the calculators that have already been updated with the 3.0 OS with the new boot2.
It might be possible to manually remove the boot2 from the calc using the RS-232 port, but for now we don't know the protection enough to tell you... We'll test and report.


April the 8th (less than 2 hours later):
OS 3.0 fully downgradable !

What we thought was possible before was right : you can manually remove the boot2 from the calc to replace it with the older one (1.4), in order to downgrade then, from OS 3.0 to 2.1.

Here's the USB/RS-232 setup required :





April the 9th :
Updated version of TNOC available !

Here are the sizes with and without TNOC :






April the 13th :
After having analyzed (with lots of tricky 'hacking') the .tns file containing the Periodic Table, we discovered that it was in fact programmed in LUA !

GoPlat quickly released the first game third-party program : 15Puzzle.



He then explains how he did it.



April the 14th :
- Adriweb release the first LUA to TNS converter for Unix/Linux/Mac computers, written in bash shell. It requires 7za (7-zip).
Link to download (updated version)

- TI releases the online TI-Nspire Document Player. It is capable of viewing all the .tns valid files (to the exception of ndless-based ones.)
It does not however support keyboard interaction (mainly used for Lua games)

- Adriweb starts to work on dumping all the lua functions and API available.

Progress is quickly made and all the API functions for "platform", "gc", "toolpalette", and various stuff in "_G" is found.
He soon finds more about the API and almost everything in it is found, thanks to a better dumping code.
Log here : link

- A google Docs document is started in order to have an unofficial documentation of the API.




April the 16th :
-The unofficial Lua API documentation is moved to HackSpire.
- Documentation improvements




April the 18th :
- Old .tns files incompatibility detected.

-More Lua games are coming this week !
Release of Apcalc's "Block Dude" in Lua.


Release of Extended's 9sweeper in Lua :


Links to download on ticalc, omnimaga or ti-bank




April the 19th :
- CAS bugs in OS 3.0 first detected.

- Bad move from TI to block the downgrade process since it's impossible to go back to 2.1 to avoid the bugs.




April the 20th :
- Battery status issue detected on OS 3.0

- Transfer fail to start with battery at 0% (although it's not)





April the 20th (a bit later):
Big news : Updating your TI-Nspire to OS 3.0 (without having TNOC'ed it) can brick your calculator !
The calculators fails to start because of a boot2 corruption ?


Everybody's encouraged not to update to OS 3.0 except if Lua is really wanted. TNOC is the only way to be sure to avoid the boot bug.




April the 21st - 24th :
Connect 4 in Lua


nFighter in Lua


Psychedelic Tunnel in Lua


Downloads on ticalc, omnimaga, and/or Ti-Bank.




April the 26th :
First unofficial OS runs on a TI-Nspire thanks to "OSLauncher" (ndless required)


-Adriweb releases the first game in Lua to use some custom TI-Image format (see hackspire) : Ti-Basket

Download link : http://www.ticalc.org/archives/files/fileinfo/437/43726.html




April the 28th :
As reported by Critor on Omnimaga and TI-BANK, TI has removed OS 3.0.1 from their website and was replaced with OS 2.1




April the 29th :
TI sends an email to its calculator customers in order to inform them about the OS 3.0 bugs.
TI recommends not to install the OS, and for those who already have, to call the technical support.




April the 30th :
TI removes the Nspire 3.0 software form its website.

Quote:
Due to an update issue with TI-Nspire™ handheld operating system (OS) and TI-Nspire computer software, we have temporarily removed select versions of 3.0 from our website. An update to TI-Nspire version 3.0 will be available by mid-May. If you have already updated your handheld's OS or your software successfully, there is no cause for concern. It will function normally. We regret any inconvenience.

For TI-Nspire™ CAS handheld OS and computer software, additional issues may cause unexpected error messages in these specific situations:

Differential Equations
Solving certain classes of differential equations using deSolve() when v and v' are on the same side of the equation, e.g., deSolve(0.01*v'+0.0025*v=0 and v(0)=100,x,v).

Symbolic Integrals
Using the symbolic integral function to solve certain classes of definite integrals may require TI-Nspire CAS handheld to take a long time to compute, e.g., Integral((2-x^3)^2-tan(x)^2,x,0,0.902) or .

Note: The Numeric Integral command (nInt()) can be used instead to find the numeric value of an integral.

Limits
Solving certain limits of form a^x in exact mode, e.g., limit(5^(-x),x,3.1415926535898).

If you have already updated to TI-Nspire version 3.0 and you have any questions, please contact TI-CARES.





May the 2nd :
TI releases a 2.1.1 updates that pretty much doesn't do anything...




May the 6th :
the CAS OS can now be run on Non-CAS models thanks to "OsLauncher" by Lionel Debroux


More info here




--------------------
Adriweb

Thanks to Critor, Ti-Bank, Omnimaga, several authors here and there.....
Woah Adriweb, just read the whole thing. Thanks a lot! I knew about most of these stuff from Omnimaga, but still thanks.

Very good Very Happy
Thanks a lot Smile

I hope more people will be interested in all that Smile
This is an extremely informative article, Adriweb! If you don't mind, I'd like to write a summary of the excellent first post in this thread and post it as a Cemetech news article, since we haven't mentioned anything Nspire-related on the front page in a very long time. I would of course credit you and Critor with collecting this information. Also, to address the actual content, I'm impressed with the Lua programs; they seem roughly as easy as Prizm C programs to write, although some of them look a bit on the slow side. Have you seen enough of the Nspire and the Prizm to be able to judge how fast Prizm C programs run compared to Nspire Lua programs?
You know damn well the new OS 3 isn't going to have the LUA support, at least not without a ton of encryption, signing and/or other security layers to prevent homebrew app execution. You may actually be better off in the long run with an exploitable, if buggy, OS.
DShiznit wrote:
You know damn well the new OS 3 isn't going to have the LUA support, at least not without a ton of encryption, signing and/or other security layers to prevent homebrew app execution. You may actually be better off in the long run with an exploitable, if buggy, OS.
Assuming you're talking about the Prizm (?), it's not really even that buggy, just lacking a few features out-of-the-box, which we'll be soon fixing. And we don't have to exploit it to run our own programs; it happily runs unsigned add-ins made with community tools. If you were instead talking about the retracted OS 3 version for the Nspires, I'd say you made a mistake in getting an Nspire instead of a Prizm. Smile
KermMartian wrote:
This is an extremely informative article, Adriweb! If you don't mind, I'd like to write a summary of the excellent first post in this thread and post it as a Cemetech news article, since we haven't mentioned anything Nspire-related on the front page in a very long time. I would of course credit you and Critor with collecting this information. Also, to address the actual content, I'm impressed with the Lua programs; they seem roughly as easy as Prizm C programs to write, although some of them look a bit on the slow side. Have you seen enough of the Nspire and the Prizm to be able to judge how fast Prizm C programs run compared to Nspire Lua programs?


1. Sure, go ahead Very Happy

2. I sadly haven't seen anything on the Prizm yet so I can't tell you anything, sorry.
KermMartian wrote:
DShiznit wrote:
You know damn well the new OS 3 isn't going to have the LUA support, at least not without a ton of encryption, signing and/or other security layers to prevent homebrew app execution. You may actually be better off in the long run with an exploitable, if buggy, OS.
Assuming you're talking about the Prizm (?), it's not really even that buggy, just lacking a few features out-of-the-box, which we'll be soon fixing. And we don't have to exploit it to run our own programs; it happily runs unsigned add-ins made with community tools. If you were instead talking about the retracted OS 3 version for the Nspires, I'd say you made a mistake in getting an Nspire instead of a Prizm. Smile


I was referring to the Nspire. I don't have one, but I know from experience with PSP and Wii homebrew how companies deal with exploits like this one. If you want to run Hombrew apps on your Nspire, the retracted OS 3 might actually be better than the new one they're going to release, since the new one will definitely be coming with some new security layers to keep you from running your own Lua code. That said, there are alternatives like Ndless which would be even better for unlocking the hardware of the NSpire, at the cost of newer features that may be required by your school.
KermMartian wrote:
This is an extremely informative article, Adriweb! If you don't mind, I'd like to write a summary of the excellent first post in this thread and post it as a Cemetech news article, since we haven't mentioned anything Nspire-related on the front page in a very long time. I would of course credit you and Critor with collecting this information. Also, to address the actual content, I'm impressed with the Lua programs; they seem roughly as easy as Prizm C programs to write, although some of them look a bit on the slow side. Have you seen enough of the Nspire and the Prizm to be able to judge how fast Prizm C programs run compared to Nspire Lua programs?


agreed. You spent some real time into making this Smile great job! I personally already knew all of this, but it refreshed my memory some. It definitely will fill in anyone who has been living under a rock for the past 6 months.
Adriweb, is the lack of seeing Prizm stuff from not having one yourself? If not, there are several videos on Cemetech of programs in action that might interest you. DShiznit, the fact remains that having to do all kinds of illegal and/or immoral (and certainly "disallowed") stuff to use a piece of hardware you bought outright for reasonable uses like running your own software is completely foreign to me. That's why I can't understand why anyone would want to own a piece of Apple hardware, especially their iDevices. Ashbad, a rock and/or mostly visits sources other than TI-Bank and Omnimaga, because even ticalc.org had very sparse coverage of the features and problems of 3.0.
Adriweb, thanks for the update. I hadn't heard that TI has actually retracted OS 3, but it is probably good that they did so!

KermMartian wrote:
That's why I can't understand why anyone would want to own a piece of Apple hardware, especially their iDevices.

I don't like derailing threads over this all the time....but silly Apple FUD is silly. You can do whatever you want to Apple hardware if you don't have a contract with the phone company for an iPhone/iPad. If you do have one of those devices and one of those contracts, the DMCA explicitly allows for specific sorts of tinkering. In fact, they have traditionally been a hardware company, so the real trick is doing what you want with Apple software if you don't have the hardware to accompany it (Hackintoshes) - but even that is not an entirely fair accusation. F/OSS *nix types should keep in mind that every time you print something you're almost certainly using Apple-owned software (CUPS). Not to mention Webkit, Clang/LLVM, Bonjour, or Darwin OS, the 15 ton elephant in the room.
And almost everyone with an iPhone or a 3G iPad has an explicit contract with a phone company that says they can't do fun things like install non-App Store software on their devices. Anyway, let's not turn this into Apple Rox vs. Apple Sux Debate Number 9001 and keep focusing on the shortcomings of Texas Instruments instead. Razz
KermMartian wrote:
And almost everyone with an iPhone or a 3G iPad has an explicit contract with a phone company that says they can't do fun things like install non-App Store software on their devices.

This is a shortcoming of cellphone service providers, not phone makers (cf. all the locked down Droids that can only install from vendor-approved marketplaces). To keep the discussion on track I'll point out that our outrage with TI is not that they produce a locked down device, but that we've come to expect differently of them and built a community around the programming capabilities they have provided us and now they are removing those capabilities.
elfprince13 wrote:
This is a shortcoming of cellphone service providers, not phone makers (cf. all the locked down Droids that can only install from vendor-approved marketplaces).


Wrong. Any android phone can be connected in USB debug mode and have 3rd-party apps side-loaded onto it using Sideload Wonder Machine or the Android SDK. I did just this with my dad's AT&T droid phone.
http://androidandme.com/2010/03/news/att-the-most-crippled-android-experience/
Quote:
... Developers will also be unable to easily load and test their apps on the device before they release them. ...


If TI requires Lua apps to be signed, they'll be doing the same thing.
Quote:
Someone over at xda-developers already posted the hack to get around the non-market apps block. It’s is actually not a hack. Users just need to download the Android SDK and run a few console commands. Tether away.

I've yet to have a problem running 3rd-party apps on my dad's Android phone, despite what AT&T has done. Apple can fight this the same way, they just choose not to.
May 19th: TI releases OS 3.0.2.1791, which prevents third-party Lua documents from working (as reported by Goplat on Omnimaga and shrear on TI-Bank). Sigh.
Lionel Debroux wrote:
May 19th: TI releases OS 3.0.2.1791, which prevents third-party Lua documents from working (as reported by Goplat on Omnimaga and shrear on TI-Bank). Sigh.
Unless they've been loaded and then saved on 3.0.1...
Lionel Debroux wrote:
May 19th: TI releases OS 3.0.2.1791, which prevents third-party Lua documents from working (as reported by Goplat on Omnimaga and shrear on TI-Bank). Sigh.
I think I called this from the first minute I heard about the Lua support. Smile Are you convinced that you should switch to coding for Casio calculators yet? Also, did TI fix the hundreds of math glitches in OS 3 yet?
KermMartian wrote:
Lionel Debroux wrote:
May 19th: TI releases OS 3.0.2.1791, which prevents third-party Lua documents from working (as reported by Goplat on Omnimaga and shrear on TI-Bank). Sigh.
I think I called this from the first minute I heard about the Lua support. Smile


I believe I called this actually...

Me, about a week and a half ago wrote:
You know damn well the new OS 3 isn't going to have the LUA support, at least not without a ton of encryption, signing and/or other security layers to prevent homebrew app execution.
  
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 3
» 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