Kat -

You can run the error check with just the "automatically fix errors" box checked to untangle any confusion/complexity in the file structure. It's still not quick, but doesn't take nearly as long as if you check the "scan and attempt to recover bad sectors."

For the full sector scan-and-recover it has to read and write in every sector on the disk - even the empty ones, while to fix errors it only has to read things, mostly only the "headers" that are a short bit of each sector, and writes only when something needs to be fixed.

I don't know for sure that the new utility does it like in the old days, but the very old "scan and attempt" routine actually read each sector, wrote it to somewhere else on the disk, wrote a test pattern to the sector being examined, read the test pattern, and if it read back right, go read the original sector content and write it back in the sector.

With common error checking in this kind of program, every "write" operation is followed by a "read" and the read is compared to the RAM from which the write was made (which means waiting one full revolution of the drive to read what it just finished writing) and every time something is written to a new location the "pointers" to the information at other places on the drive must be rewritten - and of course read back and verified.

This level of error checking may not be used in normal disk operations, and I can't confirm that it's still the standard in the new error checking; but I suspect that it is fairly close to what happens now in the error check program.

Even without all the re-reads and verifies, it's a very intensive scrub of the disk, with a lot of thrashing about, and especially with a large disk there's no way to hurry it. It's even worse with most laptop drives, partly because more of them are "low rpm" drives, and party because the heads usually are "parked" after each read/write operation to minimize damage if the computer gets moved.

John