This could be confused for the augmented reality cards. I think they are called cartridges.--T3a 18:08, 8 April 2011 (CEST)
^ should at least be a redirect ( "Game Cartridges, Game Cart Hardware, Game Carts" )
- 90 00 00 00 00 00 00 00 ... => 45F80090 for SSFIV, C2FF0090 on Rayman3d.
- Is it just me or do these byte orders keep changing. 9000000000000000 corresponds with "C2FE0090" not "9000FEC2".
- "header command has some initial dummy bytes, and eventually responds with a 0x200 byte header.". Dummy bytes, Where? The redundant A500 cmds? --Jl12
- You need to endian swap the data. 9000xxxx is the correct form. The upper bits specify information about the intended platform for the gamecard, and the lower bits specify the media size of the card.
- There are dummy bytes in the response for each 16-byte command. In the wiki these dummy bytes have already been stripped. --Neimod 17:41, 3 June 2011 (CEST)
Could you post the size of the original header reply with the dummy bytes. I'm trying to inject logic ( send raw commands ) with a 3DS card. It works well up to the 3E 00 00 00 00 00 00 00 command, I get 'data' returned from the header packet as well but only 00 bytes. --Jl12
- The size depends on what speed you are clocking out. Just keep reading until you get something that is non-zero. Also, what device are you using for reading out the data? --Neimod 09:09, 4 June 2011 (CEST)
http://pastebin.com/y22zZcZN - That's for 'Street Fighter IV 3D'. There's an NCCH header and cartridge ID. The same method returned x00 00... with a different cart, though. Not sure why there's a difference. I actually have it hooked up to a laptop, where a device over USB reads the pins and the 3DS cart is connected to it there. ( ie. the laptop acts as the 3DS. )
http://pastebin.com/38QxNGqn - That's the cart that was returning an empty header. ( "rayman 3D" ). Seems to work, now.
How did you actually decrypt the encrypted commands, Neimod?--Lazymarek9614 21:06, 1 November 2011 (CET)
- Sorry, it is a little too soon to reveal that. --Neimod 02:00, 2 November 2011 (CET)
- I think the 3dbrew members are one team, why do you want to hide it?--Lazymarek9614 21:23, 2 November 2011 (CET)
- I see the two sides of the problem here Neimod and Lazymarek9614. This forum like many public homebrew forum,s are there to try to find a way to load Homebrew. Letting everyone know your progress on what you are doing lets people understand how the scene is performing and people who have the knowhow can possibly help. But the downside to this is if Nintendo gets the drift on what you trying to do, they have the advantage, they know how the 3DS better than anyone, so releasing data on some exploits and WIP could lead to Nintendo patching it before it is finalised. Thats the decision the homebrew devs have to make.
- It would cause more harm than good at this point. Showing that progress is being made keeps the community active, but public understanding of the encryption of the gamecard protocol is not vital for the pursuit of homebrew. It would only advance the pursuit of piracy. --Neimod 02:47, 3 November 2011 (CET)
- Responsible disclosure. Try not to devalue the 3DS. Nintendo succeeding is also in our best interest, longevity/software wise.
- Don't bolt doors shut before you open them for yourselves. Remember N can patch or change or obfuscate anything posted here.
- Despite what I said. lol. If you can disclose the method without any specific technical information that would be nice. Privately if necessary.
- About the discussion further up on this page, I was able to inject logic. With a laptop and hcs microcontroller. ( microcontroller -> usb -> CTR cartridge ). I can obviously only get as far as what I can do without encryption, as I don't know how to encrypt or decrypt for now. Jl12
The protocol encryption is RC4 seeded by SNOW 2.0, not AES CTR. The main page is uneditable due to it thinking there is emails in the page.