You would need to get a dump of the firmware (provided it is not hard wired to work a specific way wrather than being a processor with firmware/data areas) before you could say. It's unlikely that such data is even dumpable without significant work (aka: microprobes, xray, die dissection, timing attacks etc.) It might be interesting to see what the leads that aren't attached to anything are actually doing, you'd never know unless you check (and have the tools available to check with) - could even be some form of jtag/boundary scan or similar interface.
Not going to say it isn't an atmel, but ATF could be anything - even just a acronym for an internal name printed on just to keep their lines sorted and give visual queues for inspectors... for example - "asic trans flash V01 August 12" could easily explain the silkscreen.
http://en.wikipedia.org/wiki/Application-s...egrated_circuit
Do realize that a custom ASIC is not the same thing as a FPGA, mainly for the fact that an FPGA is re-programmable. There is no need to run any specific known core as they are generally set up to "emulate" after a fashion complex circuitry (though they can emulate cores, like ARM implemented on an actel). In the R4's case, they are implementing two serial interfaces to an SD interface and a fairly significant (when talking about asics at any rate) chunk of embedded data memory which apparently is not updateable - and even possibly some form of crypt engine to deal with DS boot up.
Hi Cory,
Thanks for the responce, yes I am aware the differences in FPGA's over ASICs, with FPGA more like designing/programming your own silicon over the ASIC's off the shelf solutions (very simplistic analogy).
I only made the slight assertion that the chip could be a ASIC it may be something entirely different. The parts count on the R4 board is so small, it screams of a high level of intergration and of course makes it really diffictult to investigate whats under the hood.
Holes see in all these devices to be investigated are.
1) Firmware or OS updates (although this just seems to be the NDS boot loader)
2) Price (cheap somestime means low off the shelf tech).
3) The signal busses. (device to flash, device to NDS) cant see that giving much away.
4) Find a really early version of the R4 cart in the hope that the circuitry was unrefined.
Unfortunately I havent got the tools to polish away micons to expose the layers, actually or the patience,
Ohhh wouldnt a JTAG interface be nice
, guess we'll never know what the unused pins are?
I cant see why all the secrecy surrounds what is exactly in these "flash Card" devices. If you look at the last
24c3 conference you will be overwhelmed by the amount of information "Hackers/Homebrewers" put out. I suppose its all about money and competition at the end of the day, which is a shame as knowledge is power... Afterall these devices are based on the ground work of others that freely put the information out there, Dark Fader, Natrium ect....
Innovation comes from experimentation, secrecy stifles innovation.
Early mods like the Xbox/Xodus stuff gave out much more information.
Shame TeamXodus didnt stick to that mindset on the DS Xtreme
.
Still my lust for wanting to know whats under the hood remains...
R....
P.S the 24c3 hacking the xbox360 is a good youtube watch, a simple 16bit compare instruction flaw, how we laughed.
Yeah, I know Microsoft have patched it, but the reverse engineering journey was a joy to watch. Maybe I'll go to the conference next year, looks more fun the E3, GDC, AGC....
Disclaimer:
I'm not at all interested at all in the "backup" functionality of these cards, just their potential for development and a desire for knowledge...