Seagate announced the 'cuda 7200.11 1TB like ages ago and they are still nowhere to be seen for buying, at least here in Europe.
I guess that this new drive will be out for buying in Christmas if it was announced today.

Wait, so, is the actual raw data on the platter all scrambled in encrypted form? OR is it in normal form and there's just software logic combined with hardware gateway controller jazz?

http://en.wikipedia.org/wiki/Advanced_Encryption_Standard
http://csrc.nist.gov/cryptval/des.htm

Folks, this stuff isn't that hard to look up. AES is widely-studied, widely deployed, and not known to contain any backdoors. Which doesn't mean the NSA (or any other TLA) can't force it, but it does mean it should be secure against casual hacks. And since the point of this is not to protect your banal tinfoil hat secrets from the men in black, but to make it easier for corporations to avoid having hard drives walk out the door with your identity and financial information on them, that's good enough.

Provided it's well-implemented, of course. One would hope Seagate has had their implementation reviewed by crypto specialists; ideally they'd have their implementation FIPS validated through CMVP but if that had happened, or was coming soon, I'm sure Seagate would've put it in their press release. So this likely isn't (yet) a NIST-approved implementation even though AES itself is an approved algorithm, and the government wouldn't allow these drives to be used for storing Secret information without additional safeguards. And as sigher notes, Seagate hasn't specified the key length, which is critical to the robustness of the encryption. But assuming Seagate has done the right things here, this is a useful technology for securing corporate and personal information that we can hope is widely adopted (not just by customers, but by other drive mfrs).

http://csrc.nist.gov/cryptval/

Also -- The drive writes sectors and clusters just like normal: it's the data bits within them that are (transparently) encrypted. So there's no added fragmentation issues, ordinary tools like defragmenters work fine, etc, and there's no reason to expect it to have an impact on performance (AES is fast and, even by embedded processor standards, HDs are slow). Of course reconstruction of a corrupted disk with low-level tools would be pretty much impossible (that's kind of the point). And encryption should not increase the size of the data (if it did, in any kind of proportional way, it would provide a means of attack).

I'm sure it's crackable. Nothing is fools proof.

Quickly read Seagates announcements.
Could not find answers to the following questions:
1- What is the perfomance hit, if any, when using their encryption hard drives? Does it effect seek and/or transfer rates? If so, can this be quantified?
2- Is there a backdoor to their encryption method. Seagate refers to AES as "government approved". Anyone know if that includes a backdoor?

Buttressed? Here come all the butt jokes.

You said "buttressed". Uh-huh-huh-huh.

Yeah! Buttressed!Buttressed!Buttressed!Buttressed! Eh-heh-heheh

Hmmm, if second option, then all you have to do it change out the logic board and you're in...

Pretty secure, but I think that would not be enough. Likely the board also encrypts the data going to the HDD.

More secure.