I know this thread is old, but I thought it was worth noting that I have a CID that seems to have a different final 12 bytes than the two templates for the different chips.
DSi U (Black) Console ID: 0820213814097105
DSi U (Black) CID: ACBECA05F030363532434D4D4E01FE00
I have never seen anyone else with a different final 12 bytes than the two templates but my nand decrypts fine with the two keys. I also have a bricked dsi xl (midnight blue) I have just recently bruteforced the console id for, and hopefully I can bruteforce the cid as well.
DSi XL U (Midnight Blue) Console ID: 0820474323122117
DSi XL U (Midnight Blue) CID: 5e5103e7953257373136354d00011500
I will update this post as soon as I find the CID and verify it works by decrypting my nand backup (CID has been found!). A user by the name of
urmum_69 has also graciously provided some more keys along with the type of consoles they're associated with.
DSi XL E (Burgundy) Console ID: 0820190119105117
DSi XL E (Burgundy) CID: 3DE92D1F83034D303046504100001500
DSi XL E (Red Mario Bros 25th Anniversary) Console ID: 0820211901122128
DSi XL E (Red Mario Bros 25th Anniversary) CID: 6D3FBD0435034D303046504100001500
That's all I have, but I'd love to know why that first CID is different than the templates.
Edit: I have found the reasoning for the CID on the first DSi being different. The reason is that the DSi that the CID came from does not use a Samsung EMMC chip. Instead, it uses an ST chip, which is around 5mb bigger than the Samsung chip.
GBATek has some pretty interesting sections about this chip, as well as the different templates that can be used for it in bfcl and twlbf.