Can this problem happen in our typical use of the nspire?
Joe
know what to make of it:
http://www.omnimaga.org/index.php?topic=7943.0
Is this bricking (what ever that means) something that may happen to
us in our normal usage of rthe nspire and we should be concerned about
it, and is it something applicable to anyone who is writing programs
for the nspire, or only those in the hacking community?
Lionel Debroux
> I've read this warning 3 times and not being a developer don't quite
> know what to make of it:
> http://www.omnimaga.org/index.php?topic=7943.0
> Is this bricking (what ever that means) something that may happen to
> us in our normal usage of the nspire
Yes. It already struck several calculators last year, and it's now
striking again.
> and we should be concerned about it,
It's _your_ call to choose transferring a newer OS version with known
bugs (cZeros, deSolve, perhaps others), which (if you didn't remove the
boot2 upgrade from TI's file using the TNOC tool) risks bricking your
calculator ;)
The problem seems to be that _something_ goes wrong with the boot2
upgrade, and when the calculator is rebooted (after e.g. changing
batteries or hibernating), it's left without a boot2, which prevents it
from communicating through the USB port. In that state, it needs fixing
through the RS232 port of the dock connector.
> and is it something applicable to anyone who is writing programs
> for the nspire, or only those in the hacking community?
*Any* Nspire user, even if they don't write any programs in any language.
It's completely independent of any "hacking", one of the best proofs
being that there was no Ndless for OS 2.0.1188 when the bug struck last
year, and there's currently no Ndless for OS 3.0.1.1753 (yet).
Lionel.
Eric Findlay
The root of it is that TI has basically put in some code that recognizes when the calculator has been tampered with, and locks it down so that it cannot do anything - basically making it as useless as a brick.
If you only use the software, or as long as you use the handheld as outlined in end-user agreement, you'll be fine.
--Eric
> --
> To post to this group, send email to tins...@googlegroups.com
> To unsubscribe send email to tinspire+u...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com.au/group/tinspire?hl=en-GB?hl=en-GB
> The tns documents shared by group members are archived at
> http://lafacroft.com/archive/nspire.php
>
Lionel Debroux
> I believe you still only use the software, which will never have
> this problem.
> The root of it is that TI has basically put in some code that
> recognizes when the calculator has been tampered with,
Which is very hard at best, probably hopeless.
> and locks it down so that it cannot do anything - basically
> making it as useless as a brick.
If TI had put such code in place, it would strike with a much higher
frequency than it does...
A number of calculators that had previously been "tampered with", as you
write, did NOT exhibit the bug. The bug is something else entirely.
> If you only use the software, or as long as you use the handheld
> as outlined in end-user agreement, you'll be fine.
No.
Lionel.
Eric Findlay
That's quite the issue. Interesting.
--Eric
Joe
there is nothing that the user can do about that with regard to
removing a battery or some sort of reset. Is that correct? Has ti
made any comment on this problem and when it strikes will they fix or
replace the calculator? What if it isn't in warrenty? Has anyone
here had this happen and how was the problem resolved?
Sean Bird
Also most Nspire Touchpad purchases where probably less than a year ago.
Ok, so as I understand it the calculator simply refuses to boot up and
Nelson Sousa
warranty being expired, I don't think TI will refuse a replacement if
this happens to a user after a OS upgrade, but just in case I would
contact the local TI rep.
Nelson
Lionel Debroux
> and there is nothing that the user can do about that with regard to
> removing a battery or some sort of reset. Is that correct?
When it has reached that state, the calculator needs external,
non-standard help. Standard USB connection methods (TINCLS, TILP II,
another Nspire) cannot put it back into a state where the user has
access to the math functionality.
> Has ti made any comment on this problem and when it strikes will they
> fix or replace the calculator?
I've not seen, or been made aware, of something like that. But I'm not
omniscient :)
> What if it isn't in warrenty?
As Nelson wrote, we can think that TI would fix it nevertheless.
If they didn't, well, third parties _could_ fix that particular state of
bricking. Not that we'd be _overjoyed_ by having to cure fellow users'
calculators from the manufacturer's FAIL, but we'd do it :)
The RS232 port has been documented for years, and reflashing the boot2
through that means has been tested on real calculators recently. There's
no need to perform any soldering. See some pictures of a real calculator
on TI-Bank (one of them cross-posted to Omnimaga):
http://ti.bank.free.fr/index.php?mod=news&ac=commentaires&id=1038
The logical picture showing Tx, Rx and GND lines is part of:
http://ti.bank.free.fr/index.php?mod=news&ac=commentaires&id=1060
> Has anyone here had this happen and how was the problem resolved?
Last year, I know that one of the calculators struck by the bug after
upgrading to OS 2.0.1188 was replaced by TI, in several weeks time.
The warning about OS 3.0.1.1753 bricking some calculators, and TNOC
providing a workaround, has now reached ticalc.org.
Let's mention that if you have a Nspire Clickpad with boot2 1.1.8916
(the oldest version for non-prototype Nspires), as a prerequisite to OS
2.1.0.631 and 3.0.1.1753 working on such a calculator, you need to
upgrade the calculator with non-TNOC'ed OS 1.6.x, 1.7.2741,
1.7.1.x/1.7.2.x or 2.0.1.60 (AFAICT, these ones have not been reported
to brick calculators, unlike 2.0.1188, for some reason), so that the
upgrade to boot2 1.4.1571 occurs.
Attempting to install OS 2.1.0.631 on a boot2 1.1.8916 calculator will
leave it in a transient unbootable state, which can be fixed by erasing
the OS in the maintenance menu and transferring a fresh 1.6-2.0.1 one
through the USB port.
While we, in the open development community, cannot recommend using OS
3.0 (especially pristine copies thereof, as outlined in this discussion)
because of the math errors, and weirdness with the rechargeable battery
handling, OS 3.0 _does_ have a bright side: some Lua programming
capabilities, discovered through TI's periodic table document :)
AFAWCS, it's a stripped-down version of Lua (no io.* or os.* functions),
but at least - and at looong last ! - it brings graphical abilities.
Lionel.
Joe
unfortunately it is way beyond my comprehension and you lost me when
you got to TINCLS, TILP II. Since I need to stick to the simple stuff
I think I'll go back to reading Rudin. By the way, is there something
wrong with cZeros in OS 3.0? You mentioned that in a previous post.
Lionel Debroux
> Lionel, I appreciate you taking the time to cover that material but
> unfortunately it is way beyond my comprehension and you lost me when
> you got to TINCLS, TILP II.
TINCLS is short for "TI-Nspire Computer Link Software" (TI's official
link transfer software), while TILP is short for "TILP Is a Linking
Program" (the main third-party transfer program) :)
Nspires without a boot2 cannot communicate with TINCLS, TILP II or
another Nspire, because the boot1 doesn't have USB support.
> Since I need to stick to the simple stuff I think I'll go back to
> reading Rudin. By the way, is there something wrong with cZeros
> in OS 3.0? You mentioned that in a previous post.
Yup, I saw a mention of that at
http://ti.bank.free.fr/index.php?mod=news&ac=commentaires&id=1056
"
Nous vous avions déjà parlé d'un bug de l'OS 3.0 avec la fonctions
cZeros (recherche des racines réelles et complexes d'une expression -
niveau Terminale S). Cette fonction renvoi désormais une erreur
lorsqu'utilisée dans une fonction avec une variable locale, ce qui est
totalement anormal pour un langage de programmation fonctionnel. Le
problème a été mis en évidence avec la bibliothèque d'algèbre linéaire
incluse dans l'OS, mais il est possible que d'autres classeurs ne
marchent plus correctement.
"
->
"
We already wrote about a bug of OS 3.0 with the cZeros function (search
for real and complex roots of an expression - <the last year before the
Baccalauréat, which is the exam at the end of high school that, roughly,
enables going to the university). This function now returns an error
when used in a function with a local variable, which is completely
abnormal for a functional programming language. The problem was shown to
occur with the linear algebra library included in the OS, but it's
possible that other documents don't work correctly anymore.
"
And there are probably other places where you can get the same
information - maybe on this very group as well, I don't read every
single post.
deSolve() can fail easily on first degree DEs, depending on how they're
written. TI-68k calculators have never failed at that in more than 10 years.
Lionel.
Joe
malfunctions. We shall shortly see.
On Apr 21, 11:58 pm, Lionel Debroux <debroux.lio...@gmail.com> wrote:
> Hi Joe,
>
> > Lionel, I appreciate you taking the time to cover that material but
> > unfortunately it is way beyond my comprehension and you lost me when
> > you got to TINCLS, TILP II.
>
> TINCLS is short for "TI-Nspire Computer Link Software" (TI's official
> link transfer software), while TILP is short for "TILP Is a Linking
> Program" (the main third-party transfer program) :)
> Nspires without a boot2 cannot communicate with TINCLS, TILP II or
> another Nspire, because the boot1 doesn't have USB support.
>
> > Since I need to stick to the simple stuff I think I'll go back to
> > reading Rudin. By the way, is there something wrong with cZeros
> > in OS 3.0? You mentioned that in a previous post.
>
Sue
for what they are worth...
I had this happen to four of my 26 non-CAS Nspires during the upgrade
to 3.0 process. I could do nothing with them. I called TI-CARES. They
talked me through the reboot/re-install OS process - to no avail. They
finally asked that I return the handhelds to TI. After shipping them
back, I got a call saying that even they could not get the handhelds
out of "brick" mode. They sent me four new handhelds.
Note: These four were the only handhelds that I had attempted to
upgrade using the new Mac Nspire TE software. Every other handheld
that was upgraded successfully was done with the Windows TI Nspire
Navigator system. I really hate to acknowledge this, being a Mac user
at heart. But, it would appear that TI does not have the bugs worked
out of the new Mac system software, or at least not as of last week
when I tried to work this all out.
TI Cares could not have been better at solving my problem. I had my
replacement handhelds in only a few days.
I hope this helps...