Today, I use whole-disk-encryption on everything, but I still wonder if there's ever a hidden partition or unallocated space with something on it, and it's not worth worrying about it.
https://books.google.com/books?id=BjoEAAAAMBAJ&dq=hard+disk+...
> Platters are typically made using an aluminium, glass or ceramic substrate. As of 2015, laptop hard drive platters are made from glass while aluminum platters are often found in desktop computers.
> In 1990, Toshiba released the MK1122FC, the first hard drive to use a glass substrate... It was originally designed for laptops, for which the greater shock resistance of glass substrates are more suitable.
If you want or need to render the data on a drive inaccessible by any and all means, there really is no option other than physical destruction of the drive.
But my understanding, both as an industry standard and as common sense (insofar as common as computer literacy is), is that the only true way to ensure data destruction is total physical destruction of the physical medium the data resides on.
But most people don’t even bother. And with browsers saving passwords it’s probably even a worse idea now.
I do not think I have seen any credible threats that over-written data extraction works on modern drives. Maybe there are tricks where the SSD knows better and allocates blocks off to the side where it is not getting wiped, but those are supposedly addressed by TRIM (with dubious vendor support)
If someone comes across a drive that is zeroed or randomized, who is going to attempt some forensic investigation to retrieve the contents?
The NSA don't have some big magic machine that can recover data that's been overwritten. It's gone.
The idea of using some latent amount of data on the disk and somehow averaging it all out to get the old data might maybe have worked a tiny bit on old MFM drives in the 1980s.
A low-level 0 write just takes a long time on a drive of even 1TB. So I sledge. And then I can look at the drive and know it's OK to toss.
Last time around it was IDE drives.
OS encryption would prevent controller from understanding what's on the drive and managing it correctly, up to and including effectively disabling trim command as controller would never see blocks as empty.
This is back in days of ocz vertex. It wasn't like advertised / warned on the box, it would work... Just like western digital didn't warn about wd red becoming SMR drive while being advertised for Nas. But it would eventually kill performance and drive. You'd have to read Tom's hardware and anadtech etc to learn about it. I don't really follow anymore so I don't know how nicely os full drive encryption and ssds play today.