I may have discovered a new way to crash a calc(Or at least mess it up). (Someone else may have discovered it earlier, I don't know) I am on a TI-84+ SE OS v 2.55 MP. So what I did was: I went to the Y= screen, and entered these two functions: "X^1//2" and "X^2//1" (Where // is the thick fraction bar). I pushed graph, and the screen got messed up. I pushed the Y= screen to see what went wrong, and the top of the screen was shifted down four rows. I pushed down three times, and then the screen went blank. I pushed ON, and It showed a garbled Y= screen, and then it cleared the RAM.

Caveats: It does not work with MathPrint, or in a program (It just messes up the screen). You also need to be on OS v2.53 or higher (To get to the fraction bar). I do not know how it will work on the other TI-83+ family calculators, so if anyone can test this on other calculators, that would be helpful.

Which operating system are you using, 2.53 MP or 2.55 MP? Also, those operating systems cannot run on a TI 83+/SE without a modification of the OS.

This sounds like a modern-era re-incarnation of the equ>string bug.

How do you even access the lone thick fraction symbol, though? It's not listed in CATALOG for me when I press Theta/3 (I have OS 2.53 MP)

EDIT NVM I realized I had to use n/d and it changed to the thick symbol. Also I can't recreate it under OS 2.53MP.

EDIT: Tested under OS 2.55MP (both MathPrint and Classic) and it didn't mess up either.

I turned Mathprint off and did this and it graphs just fine. But I am on a TI-84 Plus 2.55MP so no SE.

I've tested it on wabbitemu with no programs, and it displays normally. From this, I conclude that it must be caused by software. If people can report what apps they have installed, it may help us eliminate the apps that do not cause the crash.

Did you have the Graph3 hooks installed when the crash happened?

I have no Idea. I have Catalog Help, DoorsCS7, Inequalz, Omnicalc, PuzzPack, and SciTools. It worked, even if the RAM had just been cleared (No hooks enabled).

It may only occur on Low-RAM models....

Strangely, my nspire's 84+ pad's keys are starting to go out (Good thing I've got another one, thanks jimbauwens!) so I can't test it. Neither down or enter work, or 0

Commentary while I crash my calc in different ways.

https://sites.google.com/site/seanga2/dropbox/Crash.zip

MathPrint is very buggy, it is probably the cause of the crash you are talking about. Crashes are very common on MP as the MP OSs are very buggy. Although that does not explain why only you seem to be able to cause the crash... I would just recommend turning MathPrint off unless you really need it

Or, you could install an older OS like 2.43 and not have to worry about MP causing crashes any more.

The crash doesn't work with MP. Duh. Have you been reading what I say at all?



Plus: it's too slow.

Yeah, that is fine, but even without MP enabled OS v2.53 is still buggy due to a lack of thinking on TI's part sorry for not reading that one part, but you do not need to be rude about it.

The topic post wasn't very long. Try to be courteous enough to read the entire thing before dispensing advice.

On the topic of 84+ crashes, has anyone ever experienced the 2 bytes token RCL glitch? I'll try to see if I have a screenshot of it to post it.

EDIT: Here is a screenshot: (It happens in newest OSes as well as some older ones)

Interesting find. Perhaps your OS didn't know what to do with that thick slash token, and executed random code as a result. That's probably something close to what happened, but I'd be interesting to see if anyone who's dug through the MP OSes would bring to the table regarding this.

Also, there's no need to be nasty like that.

Wouldn't it just err:syntax? But I guess it is TI...

There are lots of tokens that are for text-entering semantics only. Most of them aren't in the catalog, like the international symbols and all the greek lettering (Which would be much nicer to have access to as vars!) and it seems odd that this token would behave so strangely upon being unrecognized.

But that's TI for you

I've heard of a few other people having similar problems, which makes me think that this is more of a manufacturing problem or quality issue than your fault.

It might also be because I took it apart, and was doing various tests on its PCB.

Normally I'd be scared stiff to do that, but since I have a spare, I figured it was worth a shot. It's probably just some grime between the rubber and the pages, as some work occasionally, others reliably, and some not at all.

Now I can even less regretfully start modifying it even more! Maybe the nspire just sees them as GPIO, at least on a low level, which means a Bit-bashed SD card port would be fairly simple.

Not meaning to intentionally bump this post, but I also had this exact same problem. I went on a Google trek and it landed me here. (Coincidentally I had registered here a while back so I thought I'd post my solution.)
I couldn't find the exact problem, but I think an application (not program) called Calc Utils. I forgot where I got it from exactly, but it was off of Ticalc.org somewhere. I think it may have edited something inside my operating system. (If possible?)

Removing all of your programs/apps/and vars did not solve the issue. I had to re-install my calc OS to solve this issue. So if you read this post coming from Google, that was my solution.