You should probably send a note to the PhotoRec guy, but in my case the card was recognized by my system… I it was just corrupt, so I couldn’t access it further (via the normal methods). There are plenty of things you can try (such as working out a deal to send the card to someone with more experience), but if you’ve completely run out of options, you could simply do a QUICK format in the computer, then mount it on your system, then run PhotoRec or the like. A QUICK format does not touch most of the data on the card, only the table of contents (so to speak). You might also look (first) into buying a new card that comes with photo-recovery software. Some of the upper-end cards used to. Good luck! —Jeffrey

It’s not “DOS”, it’s “command line”, which perhaps feels the same, but in any case, it works very well if you can figure it out. The PhotoRec site has links to step-by-step instructions and such. Well worth giving it a try. —Jeffrey

More often than not, the block on the device that holds the allocation table is the block(s) to go bad as it’s constantly written and re-written. Especially since the whole block(s) that hold the table must be read, erased and then completely re-written. Therefore, good flash drivers will make all the changes they need to the data section of the disk, then re-write the allocation table once. This is why pulling a card out of the Windows machine when it’s not “ejected” can have problems. The data has been written, but the allocation table hasn’t been updated completely. To be fair, the Mac does this too but they at least warn you about it.

At the end of the day, most flash devices have a >100,000 write count lifespan per block, so unless you plan to use the device for years, you’re probably never going to have a problem.

You’re comments on card life are helpful, but you’re missing the point of zero-filling: it’s truly erasing whatever data might have been on the card. Out of curiosity, I ran photorec on all my cards, and found images well over a year old (on a card that sees 20,000 images a year). I zerofilled them all just to feel tidy, but in my case it was probably a waste of time and resources, but if you ever take “sensitive” pictures and want them truly removed from the card, you’d do well to zerofill the card, or perhaps blend it. —Jeffrey

Anyway, I have to say there are problems with how Windows handles ejecting. My computer has a built in card reader, and if I use the eject command with anything plugged into it the whole thing is ejected, forcing me to reboot if I want to use it again. Obviously this is no use at all, because I often have more than one card to check.

But I have to agree with Unbound about the memory card. Regardless of how or why it went wrong, trust in that particular card has taken a hit. For me, that is unacceptable, and I wouldn’t want to take pictures using that card, even if I still mostly trusted it. It comes down to this: cards are cheap, and easily replaced; pictures are often priceless, and irreplaceable.

I think Windows did it. We’ll see. By the way, “removable” doesn’t necessarily mean “can yank out at any time without ill effects”. Windows takes a tact that lowers efficiency but reduces the window for corruption if the media is disconnected while in use. OSX takes the position that you’re smart enough to know when you’re going to remove it, and so reduces the chance for corruption to zero if you can follow the simple “eject before disconnect” instructions, and as a byproduct, write efficiency is greatly enhanced. With Windows, too, you’re supposed to tell it that you’re about ready to disconnect the media, but few people do. If you don’t, there’s a chance of corruption, but Windows won’t tell you or warn you or encourage safe behavior. OSX does. That seems prudent, not retarded, to me. —Jeffrey

Kevin

Although cards are getting bigger, the answer to how many can be lost remains the same: all —Jeffrey