Thought I should mention this here, since there are a lot of projects around here that are USB-based using FTDI chips and thus their host drivers:
http://hackaday.com/2014/10/22/watch-th ... ake-chips/http://beta.slashdot.org/story/208837
Could this possibly affect Flash Carts with USB ports, like the Harmony or the Everdrives?
If those devices mandate the installation of FTDI drivers on Windows for functionality, and the silicon chips used are actually counterfeit/impersonating real FTDI chips, then yes.
The FT232 is a USB-to-serial conversion chip, but what I've read seems to imply that this may be behaviour FTDI is implementing into not just the drivers for their USB-to-serial chips but others as well.
That said: I'm not so sure FTDI's approach is very pragmatic. Most consumers of products using these chips have no idea what's going on under the hood, so suddenly FTDI's drivers permanently bricking the functionality of a product seems... well, "rude" doesn't quite convey my feeling. I don't even want to begin discussing (nor will I partake in) the legalities of their choices -- utter nightmare fuel. It's safe to the driving force here is money, as in some executive idiot saying "we're losing money to these counterfeiters, how can we stop it?" (you can't, don't bother, just keep making good products + focus on what you already do well).
Yeah, people will just buy new counterfeits that aren't subject to the problem.
Sounds like it's not entirely permanent, but it still is a dickish thing to do. It should be possible to use a different OS or older drivers, and reprogram it. If/how one could stop Windows Update from changing your drivers though, I don't know. I've used several different Prolific PL2303 clones, and Prolific's driver also detects some clones but isn't so extreme about it, it just reports "unable to start" or something like that. All of them work after manually installing an older driver.
I don't know about the FT232 clone, but I know the PL2303 clones don't support everything, they ignore breaks, don't support baud rates above 115.2k, not sure what else. I think it's kinda shitty that there's no generic USB-to-Serial bridge standard driver. It's such a common application, it's no surprise that there are bunch of work-alike clones. Kind of a different problem than fake labeled chips though, but maybe it would've helped.
For Average Joe who isn't a technical expert it's practically the same as bricking the device. Of course Average Joe probably doesn't even realize a driver bricked the device and would think the device itself broke instead.
Also
somebody reverse engineered the code that causes this. Does this look familiar? (this is exactly the same kind of issue that results in homebrew working on emulators but not real hardware)
