So my TI-84+CSE was acting weird. Every time I put a string in TI-BASIC or typed a string on the homepage, it would give the "Archived" error, like when trying to run an archived file.

I thought it was just a glitch in RAM or something, so I reset the RAM, and now it won't even boot. When I press the power button, the screen turns white, then gray, then white, then it shuts back off.

The battery was fully charged before this happened, and I still see the indicator light when I plug it into my PC and it's green, meaning it's fully charged.

But it just won't turn on. It tried pushing the reset button, but that didn't help.

This thing is pretty new. It hasn't even been a week since I took it out the box. It was working fine until I reset the RAM. I didn't know that resetting the RAM could mess it up.

The best case scenario is that the operating system is corrupted. To check for this, hold the DEL key. While holding the DEL key, press the reset button on the back of the unit. While still holding the DEL key, press the ON key. The screen should flash a bit, and in less than one second, a screen should appear asking you to install an operating system. (This procedure does not actually delete the current operating system until you actually start sending one.) If the message does appear, that suggests the hardware is good and you just need to reinstall the OS.

If this doesn't work, chances are it's a hardware issue. Call the TI-CARES number and tell them about your problem. TI is actually good when it comes to replacing defective calculators.

When turning the unit on, look in the top right corner of the screen. Does the battery indicator appear briefly? If it's red, the battery might be toast. You'll still want to send the unit back to TI for a replacement.

Okay, so I did what you said, and it showed the screen to install another OS. I installed the OS provided on Texas Instrument's site and to no avail. It transferred to 100% then said it was validating, I waited awhile, it shut off, but when I turn it back on, I get the same thing.

The battery indicator does not show at all when I turn it on, nothing does. It's just a white->gray->white screen then off.

I guess I should contact TI. :/

Try removing batteries, inserting, then holding ON+CLEAR. That will stop the OS from checking the archive on boot in the event that something in ROM is bad. If you can get the calc to boot, then clearing the rom should help.

Clearing the rom by invalidating the OS and inserting the batteries while holding STAT will put the calc into a never-ending ROM wipe test. One wipe is enough to clear the ROM and should resolve any issues related to the archive.

Rather than clearing, maybe grabbing a vat repair program and sending it to the calc's RAM and running it might be able to save things. Not sure if this is for the CSE yet.

This isn't the first time I've heard of this problem with a TI-84+CSE, so I think there may be some hardware problem that's showing up in a small but nonzero fraction of the units being produced. I'd be very interested in hearing with TI-CARES has to say about it, if anything.

I wonder if this is due to the flash chip size change. Maybe software? Hopefully it is software as those things can be patched cheaply. (Running another OS on one of these calcs, dumping the archive, and checking for errors would be interesting).

It sounds like a corruption of the OS rather than the user Archive area to me, so you'd have to figure out a way to boot said other OS that wouldn't wipe out too much of the original OS. I hope it's not due to a quality-control issue on the 4MB Flash chips, although I certainly could believe that that could be the root cause.

But resending the OS would fix such a correction, no? If corruption happens on any page other than 0, you could get a one-page OS to dump everything. Ask BrandonW about getting arbitrary code running from RAM (iirc) from a bootcode exploit.

Not if the Flash chip has cells or even pages or sectors that have gone bad in the weeks or months after production. Perhaps for the sake of curiosity amihart would be able to help us narrow things down. The fact that he re-sent the OS and the problem recurred suggests to me that my hypothesis of a failed Flash chip is correct. Good idea.

Wouldn't the flash wiping suggestion throw a fit if the flash chip doesn't clear properly?

While it might be nice to know what sectors are bad, the fact that an OS resend didn't fix the issue shows that the issue is a hardware one. The self-test mode you get from holding the STAT key on boot is worth a try, but at this point, it's time to call TI-CARES.

I agree with KermM that the issue sounds a lot like a defective flash chip, with failed sectors in the OS region. There's a chance that a few bits in the boot sector have failed, causing it crash when it tries to hand off control to the OS. Or, the issue could be in the ASIC, such that when it boots, the memory mapping unit doesn't exit boot mode. That wouldn't screw up the boot code, but obviously the OS can't function without page 0 being correctly swapped in.

That actually fixed it. I guess that means it was a software issue then? Thanks for the help.


fffffff

Oh, excellent! I hope the problem doesn't recur. I'll add that to my list of suggestions when people have this issue.

Sorry, I went with the odds around here. I should have gone for the more politically correct "[s]he" or "she or he" or "he or she".

Every single time I reset my RAM this happens. I have to press on, then when the screen turns gray, hold Clear for it to boot.

This is really annoying, because that means resetting the RAM also requires me to dump all my archived programs.

Is it like this for all TI-84+CSEs? Or do they usually reset properly?

I've never had that happen to me, so at least for mine it resets properly.

Mine also resets properly with no corruption of the Archive or OS.

Since it's not happening to any of you guys, I recorded what it looks like:

That's fascinating, thanks for making the video. I'm trying to pull together what knowledge I have of how the post-RAM-reset boot sequence works to think about what might be happening there, but I'm not getting too far. Hopefully other of our OS experts will have some ideas.

I had this happen before in Feb 2002 and it costed me an entire Illusiat RPG project. The only way to fix it was to do a full mem clear after doing the ON+Clear trick.

I think it's the archive being corrupted, but I could be wrong. Apparently this is now fixable using certain tools such as Calcsys or some programs that BrandonW or DrDnar might have made, which I wish I had or knew how to use way back in the days.

It's an interesting development that ON+CLEAR lets you boot successfully. This, as others have mentioned, suggests that the archive is corrupted. If there is something in the archive you really want retrieved, I could try throwing together a program to do that, but it'd be a bit of a pain.

The OS has a function for wiping the archive, in the form of the clear all memory option, in the same menu as the reset RAM option. This is something you definitely will need to do. Make sure you select that one, and not the one for just clearing the archive, as the latter is actually less thorough. Alternatively, you may also try ON+STAT.

If, however, you have in fact done the ON+STAT thing, that again suggests that the flash chip is defective. In particular, it suggests that a bit (or bits) in the archive is stuck in the off state, or that an internal trace in the flash chip is shorted out. Again, returning the calculator to TI is the only solution.