August 22nd, 2005, 04:36 AM
Using CHKDSK to define the bad sectors will not work. In the old PC/MSDOS, the process of identifying them is by clusters and what happens is that the whole cluster is marked bad so that the operating system avoids writing into that cluster.
Norton Disk Doctor could pinpoint the specific sectors that are no longer usable and these are marked bad so that the other "good" sectors could still be used. In a floppy, it's quite easy to deal with bad sectors, just run NDD to move the data out of the unreadable sector, extract the data by physically copying the file then just simply shoot the floppy into the garbage can; or try to reformat it using the MSDOS command FORMAT A: /U to see if it is the floppy drive head itself that is defective--or the servomechanics of the drive that have gone awry.
For HDD, it is another issue. A bad sector would mean "unstable" or unreliable data area where data recovery may be undertaken but where there are too many bad sectors, then the final option is to back-up all data then do a low-level format of the HDD (as this could attempt to minimize number of physically bad sectors). There used to be a utility program that could try and reformat the bad sectors but I can't recall the name of the product (heck, that was way back in 1987-88).
I don't know if Norton System Works still carries the program "calibrate.exe" but this is another way to attempt to eliminate the bad sectors where still feasible. What it does is reformat the HDD without losing data as it immediately rewrites the data to the sector it reformated. It takes a very long time, though.
Note that I'm not speaking for Norton in this regard... it just happened that way back then, Norton Utilities could do a lot of things that a user needed to keep data storage stable.
Since you are using XP, you won't see the physical map of the HDD as you defragments it. Try using Norton Speeddisk instead and it could show you those clusters marked bad as it defragments the drive. I'm not sure if it still works the "old way" where when you put your mouse pointer to the marked cluster, it would identify the specific sectors that are really marked bad.
- G -
Si vis pacem, para bellum!
August 22nd, 2005, 11:09 AM
You are undertaking academic research. The problem with that is you need to go to a level of detail that is of no interest to users or support engineers.
As you have discovered, the tools that are generally available are for repair and recovery, or to analyse the stability of the drive. Most people are just not interested in how badly broken something is, or where it is broken. They just want to fix it and move on.
How important is it that the tool runs under Windows XP?
My recommendation would be:
1. Get a covering letter from your college to establish your bona fides.
2. A precis of your research plan and objectives.
3. A list of your specific questions and issues.
Write to all the major HDD manufacturers with this information and see what response you get.
It is reasonable to assume that the manufacturers will have the software that you are looking for, or at least the output from it.
You might be lucky and they will supply you with the software, and even some free HDDs
Many organisations are highly supportive of academic research, particularly as they seek to recruit graduates. They might also be interested in your findings and consider it to be "free" research (grant $$$s?)
August 22nd, 2005, 09:10 PM
Thanks, I will try to do that. I will keep everyone posted of my findings.