MQ01UBD100 USB to USB!

[Jared]

[post]8951[/post] Basically I'm referring to the type of rig the FBI uses to scrounge data straight off the platters without the drive firmware being involved at all. I have no idea what it would cost to build/buy such a rig.

What you're talking about is a fictitious machine that only exists in Hollywood movies. The FBI has no such capabilities. In fact, for a case like this, they'd be contracting it out to a commercial data recovery lab, who would have to tell them exactly what I just told you.

The FBI or other large government agency might have some pull we don't have that could perhaps have some benefit. For example, they could sopeana Toshiba to see if they retain records of the adaptive information of drives they manufacture, or if they could provide details/source code relating to how the adaptives specifically work. With that sort of information, it might become possible, and perhaps if the location of Osama Bin Laden had been on there they'd have pulled it off. But, generally speaking, they have no real capabilities we don't.

 

[Blizzard]

I'm being a little facetious. Basically I'm referring to the type of rig the FBI uses to scrounge data straight off the platters without the drive firmware being involved at all. I have no idea what it would cost to build/buy such a rig.

Where did you read about this "rig" the FBI has that can "scrounge" data from platters? $250,000 is not a lot of money for medium to large DR companies so I would expect them all to have this capability by now. Think about it, if ESS purchased a domain name for $1.7 million in 2008 they could surely afford 250K .

 

[SilverPuppy]

They have (compared to any of us) unlimited resources if the need presents itself. Theoretically it wouldn't be difficult to construct a machine which would spin up the motor, move the heads slowly across the platters, and parse any valid data which resulted. It would have to be adaptable to multiple drives to be reusable, but that's feasible enough. I have my doubts that this technology exists only in the movies.

Now the difference between "any valid data" and actual USEFUL data might be quite different when we're talking about trying to retrieve a client's "stuff" or simply trying to build a criminal case against someone where even the smallest bit of valid data might be incriminating.

At some level, though, all hard drives have some sort of tracks, sectors, and checksums. Theoretically it shouldn't be that difficult to hook the drive minus its PCB up to a machine that would bruteforce the geometry and come up with an image pretty easily. It just might take a little time.

If it does turn out to be true that no one has done it, that's a billion-dollar idea for the first person who pulls it off.

 

[jol]

Theoretically it wouldn't be difficult to construct a machine which would spin up the motor, move the heads slowly across the platters, and parse any valid data which resulted. It would have to be adaptable to multiple drives to be reusable

thats what YOU think

and let me explain
you have to deal:
1. with 1000's^ of types of density
2. various location of the SA (the SA with the data will be merged together)
3. HM
4. translator

so LETS say YOU WILL find a way deal with density issue
how long will it take make sense of the data - which its completely mixed (with the SA, with no HM and translator) - and put it together in order to be useful ?
how about brute-forcing the sectors ?!

just tell me how long does it take to brute-force SHA-256 encryption ?
and then calculate if you need to brute-force just a 500GB HDD (which is about a billion sectors (976,773,168 + some more who belongs to the SA) to brute force) and what about 1, 2 ..10TB HDD ?

so you tell me is it even possible ?

 

[SilverPuppy]

I see no reason why you should become angry over this, but you very definitely seem to be. Basically you're telling me that my understanding of how hard drives work is stuck in the 1980s (when I first started dealing with them....I can tell you in great detail how FM and MFM drives work) and that since then they have become so much more complex internally that bruteforcing the platters is infeasible. We'll leave it at that. I enjoy a vigorous debate, but when someone gets angry it's not a debate anymore, and not enjoyable.

Thanks to everyone for the information. That was truly helpful.

 

[Jared]

I don't think jol was angry, he's just trying hard to set the record straight. Guys like us are sick of people telling us that we need to just yank the platters out of the drives and throw them in a "universal platter reading machine", as if there was something like a CD-ROM drive which we could just put them in and dump the data. I've even had computer techs argue with me that such a machine does exist and that I'm "just not equipped with the best equipment", even swearing he'd seen the machine in a Gillware video.

Turns out all he saw was a glide burnish machine used to clean dirt off of platters.

But, yes, hard drives are way too advanced today to have any chance of such a device ever existing. It'd take millions of dollars and decades of research to make it work for even one model of drive. You'd have to repeat the process for every new model out there, and there's thousands of them.

 

[jol]

[post]8976[/post] I don't think jol was angry, he's just trying hard to set the record straight.

exactly ^
my initial reply was:

thats what YOU think

but then I though, that maybe someone (or you) will think that I'm angry
I decided to edit the replay, and add an explanation why its not possible

the people who know me know
thats a bit hard to make me angry