|
Termyn8or -> RE: When you go to get rid of your computer is taking out the hardrive good enough? (4/28/2010 9:38:42 PM)
|
Smut, if those platters exist the data can be restructed to some extent.
If you want to destroy the data you must basically use an axe. Make sure it is sharp and cuts through the housing and actually through the platter. It might not even get through, but if you bend it enough that should do it. Even then fragments can be recovered. Remember, even a sledge hammer might not do this. You MUST severly deform the platter(s). One way it to be sure is to actually break the case. An axe is really the better tool. Other ways would include taking the small tools out and removing the platter(s) and bending them in half. Then you bend them back. You will see the coating coming off. Then bend it on the opposite axis, 90 degrees and cause more to shred. At that point you are pretty damn sure, but till fragments could be recovered, but the reliability of this method even exceeds those of a DNA paternity test.
The basis is, even the programs that write all zeros can't be trusted. This is because of something called depth multiplexing. They can focus the detection field to below the normal surface and pick up even overwritten data almost no matter what. I long thought that, and it is still true, overwriting the data is better, and THEN all zeros, and THEN overwriting agian. But don't leave it all zeros, because as good as you may think such a program is, it might be the first thing they use to recover data from lower layers on the platter. It provides a queiter background level and thus less error correction. So leave them a bunch of noise to deal with. It's worse for them than all zeros.
There are even better pushbutton methods to do this, but they are very hard to do. If you could force it to LLF with a different track pitch that could be good, but just give it a try. You need to steal more information to do this than you are trying to delete. All in all, for 99% of people, the axe is best.
T
|
|
|
|
