Sony Magic Gate Memory Cards

Hello,

I hope you can help here. I've just started working for a new employer and one of the first projects I have is to develop and interface/reader for Sony's memory cards. CF and SD cards, IDE and SCSI I've done before and the information was pretty much readily available. In this case however, Sony's protocols for communication with their cards at an embedded level is a bit obscure and non-existant.

The MG technology I believe is an encryption algorithm applied to the cards to 'lock' them, it was originally pushed with the Playstation 2 as allowing the storage of online customer information, bank account details, passwords etc.

Is the MG function always enabled? Because if it is I'm dead in the water from the moment go without that algorithm.

Would anyone have any experience in interfacing with these cards, please any info? It'd help me out loads as I'm just sitting here doing all of the other background work surrounding the project in the meantime.

Any help very much appreciated,

Aly :-)

Reply to
techie_alison
Loading thread data ...

Hi again Aly...

Did a bit of a search, but the news doesn't look good.

formatting link

Not strictly related to your issue, but contains discussion on the encryption system, which seems to fall into the "pay as a million in licence fees, or we sue you to death" catregory.

Sorry...

Pal

Reply to
Pal

It's not in the least obscure; it's merely unavailable without signing a license. You also cannot use the Memory Stick compatibility logo without such a license.

History: In 1999 or thereabouts, when Memory Stick was emerging, Sony made 85% of the interface documentation public and a license to use Memory Stick was free for the asking. A few years later, they clamped down and made it something like SD - you had to pay a lot to get in the club. At that time they hid all the public documentation (though if you want a copy, I can give you one).

I don't have the MagicGate specification and couldn't disclose it if I did. But the specification for vanilla Memory Stick is written in such a way as to imply that the DRM feature isn't turned on until you use it. As further evidence in support of this, modern Memory Stick media work with cameras designed for pre-MagicGate sticks.

Reply to
larwe

Hey, you bullshat your way into the job, let's see if you can bullshit your way through the task at hand!

Good Luck! Rich

Reply to
Rich Grise

That was certainly uncalled for Rich.

What in the posting smells of bullshit to you?

But now that I review your posting, your signal to noise ratio is less than 10%

Is this a pot-and-kettle thing?

Rufus

Reply to
Rufus V. Smith

That you got the job and now you seem unable to perform the job you were hired for.

And I really, really wanted to use the word, "bullshat". ;-P

Good Luck! Rich

Reply to
Rich Grise

So you don't have the information I'm looking for then? You're welcome to come round to my house and go through my redundant chips drawer and get them all working without datasheets.

Actually, my company would just love to employ someone like you who can solve all of the open tickets. Send on your resume.

Reply to
techie_alison

Really.

Would you expect to hire only people experienced with the specific parts you intend to use? That would make it pretty hard to develop anything new.

Maybe that explains why 8051s are still so common.

In most of my work, I'm hired to find out what I need to know. I've even asked questions on usenet, though the answers often resemble this one.

--
	mac the naïf
Reply to
Alex Colvin

and as we know, datasheets have all the information you need to get the part working...

--
	mac the naïf
Reply to
Alex Colvin

Agreed.

A logic analyser was one of the best bits of kit I ever bought, in that usually I can solve a puzzle with an equal amount of personal energy as to requesting information via usenet/forums with a load of hassle from people who are still bemused by flashing LEDs.

The rule of thumb regarding usenet (and which a few of the engineers I work with advise on) is simply to never post, never answer questions. Simply not to get involved. The majority of our ideas and methods of doing things are thrashed out in a continuous yahoo conference which does seem to work well. As for the Sony thing, without that algorithm it aint going nowhere, I can't implement full compatiability without it. It would have generated about

10,000 units with the dies etched in China, and it's the company's decision not to sign contracts with Sony so therefore it's a management decision. ABS braking anyone???
Reply to
techie_alison

They should be thankful you haven't found a way into the algorithm. Poking about in Sony's DRM schemes without a license is a guaranteed way to a lawsuit. I don't recall for sure, but I think there are patent considerations - I am QUITE sure there are DMCA problems.

Reply to
larwe

Yep, the US side of the company had initially thrown $100k figures about for a full licence. Also while researching this it's become clear that certainly in the US it's a criminal act to try to disect something such as an algorithm.

While I haven't tried to unravel the algorithm I'd hope that it's sufficiantly secure not to be able to do so. My 3rd year thesis was about cryptography, with the main danger pointed out that sensitive highly encrypted data (military for example) could be captured today and revealed in 5 years time when processing power has evolved. So while the bad guys might not have access to the data now, in time they will do. Equally so, the Sony MG algorithm will come out in years to come.

Reply to
techie_alison

In message , larwe writes

Wow, welcome to the land of the free. There's precedent set in English law IIRC that specifically enables you to reverse engineer (Mars vending and coin mechs) a protocol, I think you'd have to be able to prove 'clean room' techniques via documentation to be safe though. Can't quite believe that you can't take a logic analyser and scope to equipment that you've paid for without risking getting your butt fried. Hopefully we'll be able to hunt lawyer instead of foxes if they ever become that much of a nuisance over here.

--
Clint Sharp
Reply to
Clint Sharp

DMCA trumps interoperability if you are building a "circumvention device". It requires a lawsuit to determine what is and is not a circumvention device. So you might win, but it will cripple you financially.

Walt Disney bought the land of the free. Soon there will be a tax on PIC12* series micros because they can be used to make PlayStation modchips.

Doesn't matter how clean your room is:

a) If you're infringing on a patent (lawsuit will determine yes or no to this) then you have to negotiate a license, no matter how you worked out how to implement it.

b) If you're delving into DMCA, be prepared for a long and agonizing battle regardless of where the merits (if I can use such a word) lie.

The best US case law on this is a case of Lexmark vs. someone who was making refurbished ink cartridges. Lexmark had some kind of crypto protocol between the printer and the cartridge to prevent people from doing this (the cart becomes useless once it is empty and the printer won't print unless it can talk to the chip in the cart). Lexmark sued someone who worked out the details and faked the chips. The defendant won. But it was not an easy victory and this particular situation (MG cards) is not the same by a long shot.

Reply to
larwe

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.