Cool. I take it a hard reset is when you pull ALL the batteries, including the cr1616?

Basically. There's shades of failure there too; sometimes you can reset it via the Memory menu, sometimes not.

epic 3-month bump.
Are there any command-line things I can use with TI-Connect(or any other linking software) that would allow me to write a simple batch file to run all the tests I need(memory test, screencapture, etc.)?

Also, the flash on my own calc seems to be degrading. Programs don't last more than a day or two before getting corrupted. This results in syntax errors in TI-Basic programs and all kinds of funky stuff in ASM ones. Would it be simple enough to swap it out with a flash chip from one of the busted calcs?

TiLP has some command line options you can use.

I'll take a look, thanks.

Seconded, TiLP would be your best bet for something like what you're describing.

I'll take a look. What about that flash chip? Is there any chance I can replace it, or could there be something else causing the corruption?

The flash chip is likely to be at fault, although I'm very interested to hear of one in which corruption has started to occur with such regularity. What OS version is that running? You _could_ replace the flash chip, but it's a delicate process.

I'm running Doors CS 7 of course, but my cousin had MirageOS installed before I got it, and I've heard bad things about that particular shell...

It does more archive churning than Doors CS, and of course I consider it inferior, but on its own it's not going to kill your archive that much faster. Why don't you just swap your whole calculator with one of the "broken" ones that's not actually broken?

[true-story] I actually received my TI-84+ SE in that condition (having the blank LCD issue); my mom bought it from a friend who had never used it. Oddly enough, enough physical bashing actually got it to work! [/true-story]

Probably a loose LCD ribbon calbe then, by the sound of it.

Ah, percussive maintenance...

Anyway, will 22 gauge wire work for calc repairs, or is that too big?

It won't work for ribbon cable repairs; even 24 gauge won't work for most things. 30 awg will.

BUMP

Did my first set of diagnoses:
http://www.mediafire.com/?zsnyoq77kto53py

And the following screenshots go along with it:

Great job, and nicely-organized in that spreadsheet. Those screenshots are more or less symptomatic of ribbon cable failure; I explained that previously, right?

Does ribbon cable failure affect what TI-Connect reads from the link port as well? Because that is what those screenshots are from.

DShiznit, I have a few 83+'s with that exact same issue, I'd bet Kerm is right with the ribbon cable on this one as the calc reads directly from the display for screenshots over the link cable. I haven't gotten a chance to test the ribbon cables on mine yet, my Multimeter went MIA recently so I need to solve that to test.

Ok then, I'll note that and see what I can do tomorrow when I have the tools to work on these.

Yup, exactly; I've fixed enough calculators to be positive on this. The TI-83+s have a non-memory-mapped LCD, which means that the contents of the LCD buffer in memory might not be the same as what is actually on the screen. Therefore, TI's screenshot software reads directly from the LCD's hardware, which of course is just as buggy as writing to the display when done through a degrading ribbon cable.