Unbricking A Counterfeit FTDI Chip

If you haven’t been paying attention, FTDI, makers of one of the most popular USB to UART chips out there, really screwed up last October. They released a driver to Microsoft that would brick unauthorized clones of their chip by setting the USB PID pair to zero. This renders the chip unusable by any computer. That Windows driver has been fixed by now, but there’s probably still a good number of bricked FTDI chips out there. [Tony G] figured out how to fix it, and it only requires a few lines in the console of a proper OS.

The bricking Windows driver worked by setting the USB PID on fake chips to 0000. Luckily, there are ways to reprogram these chips. [Mark Lord] released a set of tools that will reset the USB PID. This unbricks the chip, fixing whatever device it’s attached to. It’s also a great reminder to either update or roll back your Windows drivers.

77 thoughts on “Unbricking A Counterfeit FTDI Chip

    1. And according to my understanding, they are not even clones, they are just compatible chips, in which case they are not illegal. What ever the case, i will not be buying any FTDI chips intentionally, genuine or compatible. Nor will i be buying any prolific chips either. This one i have (genuine) is such a pain, BSODs and communication lock ups all the time. Suckiest drivers ever, plus intentionally leaving HX/HXA chips from new drivers without any reason.

        1. Did RayBan, Nike, et. al. set a bunch of peoples feet on fire in order to counter them being made in the first place? No. But FTDI did burn a bunch of people who likely didn’t even know their chips were counterfeit. The money had already left their hands, and all it did was screw over customers.

        2. Who says chinese are making sucky counterfits? I did forget to take into account one thing, that those chinese chips are mostly marked FTDI chips, which does make those particular chips counterfeit and illegal.

          But anyway, i don’t want problems like this, so if i think i’m buying genuine stuff and it is not and FTDI fubars the chip, then no thanks, i’ll buy something else. Why should i support this, pay premium and then get bitten. No way.

          If the Raybans and NIKEs etc could be disabled, i would not buy them online, so yes, if that was possible, i would avoid them for the same reason. I don’t need the crap that comes with it. I just avoided buying a programming device, because it most likely is a counterfit and the company has disabled them in drivers, so guess what, i’m not buying any of their overpriced stuff. I have affordable options.

          I’m avoiding every company that just disables hardware without any warning or pulls other stupid stunts like what FTDI just did.

          1. If you’re buying counterfeits then you’re not supporting FTDI, you’re supporting the counterfeiters who ripped you off by charging you a premium price for an inferior product. If you want to punish the people who harmed you, stop buying counterfeits – of anything.

    2. I cannot call these clones, a 3.3v counterfeit chip i got (unknowingly off ebay) failed high and fried an FPGA on a project i was working on
      Plus i have noticed differences in rise times and jittering making me believe they are not the same internally

      1. If you would read the original stories that came out about this, you would know these are not the same internally. If they were, then their USB PID would not have been reset to zero. That’s why genuine chips – clones or otherwise – continued to work properly with the new Windows driver.

        In other words, if the Chinese had made true counterfeit clones there would have been no problem, but instead what they did was re-program existing chips to function like the FTDI chips, saving them the cost of designing clones and making new chip masks for them. The Chinese really cheaped out on these and unsuspecting consumers paid the price.

          1. Devices which weren’t marked as being FTDI were bricked though because they used the same PID to take advantage of existing drivers. If they’d changed the drivers in a way that just meant these devices wouldn’t work with them then I’d have no problem with that, but intentionally breaking them is a malicious act and potentially criminal (destruction of property)

        1. Yes there is if the clone is going to be labeled with the same brand and logo. These ARE counterfeit units because customers can’t tell the difference between these unauthorized units and the originals.

      1. I think he was trying to make the point that a clone is an identical copy, and the counterfeit is just a copy. If you buy a clone you get exactly the same product as the original, but if you buy a counterfeit you might get something of lower quality.

        1. The distinction of counterfeit is that it pretends to be something else. Clones are OK, as long as they didn’t steal the copyrighted design files. Drop in driver compatible replacements are OK too. It becomes counterfeit when it has someone else’s part number and logo on the package. That’s part of the original controversy, as far as we know all the affected chips where counterfeit, but of course you can’t detect that from a software stand point.

          1. It doesn’t matter if they were counterfeit in the sense that they had FTDI’s branding, what made them break was that they were not FTDI clones. A true clone would not have been affected. The chips that were altered (their PID changed to zero) were able to be altered because they were another chip altogether which had been re-programed to pretend to be an FTDI USB to UART chip.

    3. clone = emulates a chip from another manufacture

      counterfeit = mimics the outward appearance of a product from a different manufacturer

      AFAIK, the vast majority of the bricked chips mimicked the outward appearance, so they count as counterfeit (and clones, as they often worked fine.) They got into the supply line of plenty of legit products and then FTDI’s code caused all sorts of issues.

      FTDI created and shipped malicious code that more than likely broke the property of plenty of people who never agreed to their EULA. (What if you stuck my device into your PC and bricked it thanks to FTDI?)

      Think of the criminal liability had FTDI instead been a human being. I’m sure 5 years would have been the minimum time served.

        1. EULA has to be displayed and explicitly agreed to. When was the last time you plugged in a keyboard and it popped up an EULA for you to agree to the licencing for their drivers?

          1. Also the validity of EULAs is dependent on the region.

            And I don’t give a flying fuck if all the asshole politicians sign that goddamn surrender bill labeled TTIP or similar, I think the population agrees to NOT click yes on that one, if alone since they were never given access to it, so it’s valid for the weasels only.

      1. Why would you try to use FTDI drivers with something else? There is no way FTDI can be responsible in any way for counterfeits behaviour with their drivers, intentional or not.

        1. Exactly. It’s the fault of the counterfeit chip for using an FTDI USB ID in order to use the FTDI Windows driver. If they’d bothered to write their own Windows driver this never would have happened. The Chinese were too lazy to make clone chips that would work properly with the FTDI driver, and they were too lazy to write their own Windows driver that would work with their crap chips, and everyone blames FTDI? Nonsense.

    4. If it says “FTDI” on top and it isn’t it’s a counterfeit and not a clone. If it says something else it’s a clone. Clones are ok, because the user doesn’tthink he is using the real thing, and doesn’t expect it to work like the real thing. Also, if the user bricks it by doing something that should only be done on the original it’s the users fault. If it’s a counterfeit it’s the sellers fault for selling forgeries.

      1. But a bunch of people are blaming FTDI because their counterfeits don’t work, and vowing to never buy FTDI again. Guess what? They weren’t buying FTDI in the first place, so it’s no skin off FTDI’s nose.

        As for me, if I ever need a USB to UART chip I’m going to choose FTDI.

  1. Nothing new and nothing special. But the point is the delivered windows driver still kills the chips
    even from wires we used dranded from FTDI and ordered by an realy big retailer.

    Under linux you can get these chips with some lines of code back to working. But not with windows so far even if i kill
    the new driver from the system they are still dumped.

    and the driver from FTDi is still the 2.12.00 WHQL with the Destroy Hardware Funktion

    1. Nope, just too dumb to realize the implication of bricking peoples kit to prove a point. Yes there are fake FTDI chips out there, but bricking my gear because the manufacturer was duped into buying fake chips is like Volvo removing the engine from my car because I fitted after market brake pads.

      1. .. at the risk of restarting the debate, FTDI would have made themselves more friends if they had simply made their driver bork with a warning when it encountered the fake chip, or refuse to operate above 4800 baud or whatever. bricking the chip was a huge blunderbuss shot to the foot for FTDI.

        1. Anything they could have done would have caused a massive uproar from people that spend too much time on twitter and not enough time doing anything of worth either way. If they just put up a message on the screen (is that even possible from a windows driver?) people would have complained about them “spamming their screen” or similar.
          I’m still waiting for a report of a death related to the driver “bricking” life support machines that people kept trying to use to justify all of the drama.

          1. That is true, but that gets to what I’d call unreasonable complaining. This is a complicated issue, and I understand their dilemma, but I don’t think they chose a good solution to deal with the problem.

          2. Someone will always complain, but if you seriously think there weren’t any choices that would have resulted in less then I don’t know what to tell you.

      2. It’s more like your Volvo dealer removing inferior after market brake pads known to cause accidents the next time you bring your car in for service, then telling you that you can’t drive your car again until you install genuine Volvo brake pads. Don’t like it? Either commit to after-market and never take your car back to the dealer (don’t install FTDI drivers for your non-FTDI chips), or stay legit and buy Volvo brake pads (source from FTDI, not some Chinese web site).

    2. They should perhaps have considered that part before they outsourced production to some place where wages are low, corporate taxation lenient and, consequently, the government resources available to prosecute things like theft of IPR is scaled accordingly :-p

  2. if these chips are like the hdmi copy protection hdcp keys then even legit chips can be flagged as fake.

    i read horror stories of people not being able to play a blu ray movie for example because the chips used in the tv to handle the hdcp being stolen from the factory (some but not all of the chips or hdmi keys) then used to make some hdcp strippers.

    then when the hdmi/hdcp consortium blacklists that run of keys then every tv in transit, stores and homes that uses those keys will not be able to play the bluray movies that have that blacklist

        1. if they did that, it’d break the functionality of the chip. the counterfeited chips don’t 100% emulate the way the FTDI’s work, so FTDI uses that to determine the fakes. but since its an eeprom write it needs to be there.

  3. I guess we don’t get an actual walkthrough on how to do it with the link provided?
    It looks like we need libftdi and libusb, then run that ft232r_prog with some args to fix the chip lol.
    Was this more of a “theres a way” instead of “heres how” article :D

    1. It is easy if you think as electronics engineer and for the chip itself.
      What if this chip is fitted on a PLC or a CNC machine for instance? Still stop using it and replace the equipment?

    2. Why the hell would i do that when they pulled a stunt like this and then many people say you can’t get the chips directly or from a listed distributor anyway? I will never trust FTDI again and will not intentionally buy any genuine or clone or compatible chips again.

      Also i will not touch Profilic chips either. Their driver sucks so much. BSODs, because of trying to read with wrong settings to what the device on the other end is sending, port lock ups while uploading code etc. Unbelievable, and they also blaim “piracy”, but my chip is genuine according to them. Also they droppe the HXA/XA support for no reason.

  4. “It’s also a great reminder to either update or roll back your Windows drivers.”

    Or gee, maybe it’s a reminder to buy authentic chips instead of cheap illegal knockoffs. Instead of “unbricking” your pos chip – spend a few bucks and buy a legal one. Wow, how not-rocket science is that.

    1. “spend a few bucks and buy a legal one”:

      Prolific PL2303HX
      Microchip MCP2200, MCP2221
      Winchip CH340T, CH341A (USB to parallel)
      [please add some more – I guess that Cypress also has some cheaper and VERY cool]

      I think that the Chinese should quit this stupid FTDI business and sell the microcontrollers used for the counterfeit FTDI parts under their own brand as they are actually of such a high quality that FTDI has to damage them to prevent people using them.

      1. Prolific faced a similar problem the PL2303 was cloned and counterfeited at one point 90 percent of the PL2303 chips being sold were counterfeit. They phased out production, introduced the PL2303HX and stopped making drivers for the PL2303. A far more elegant solution than bricking peoples chips.

        1. sorry correction it was the pl-2303hx REV A, do not buy a pl-2303hx REV A it is Counterfeit. all pl-2303hx REV A still being sold are counterfeit. check out Prolific’s site.

    1. Tell the manufacturers who unknowingly bought counterfeit FTDI chips to replace the equipment they soldered the faux chips into. What’s that? Not gonna happen, ever?

      Same sad story as with the “capacitor plague” where counterfeit electrolytic capacitors with faulty electrolyte got mixed in with the genuine one actually made by the companies whose logos were on them. Apple offered warranty extensions on a few of their things, but no guaranty that the board(s) they replaced would not also have bad capacitors. If you were “lucky” the 2nd, 3rd or 4th time it went *poof* it was under the extended warranty.

      AFAIK, with the rest of the electronics industry it was essentially “Not our fault for getting suckered by ‘name brand’ components at suspiciously low prices. Chuck the thing and buy a new one from us, with no guaranty that it won’t also have flaky parts left over from the batch of junk we bought.”

  5. Well i do not think bricking devices is a great idea. i don not think being fakes is it correct to release drivers / firmware to cause harm to said fakes. and if your code it detecting fakes then your aware there fakes so … your knowingly attacking fake products … i can see this leading to a possible class action lawsuit. there are courts to resolve issues like this not damaging others peoples devices. even if there fakes. even if there basically stealing your products.

    lets think of another example …. windows detects linux install on another hard drive … it wipes the firmware to that drive rendering it useless.

  6. Or don’t buy prolific chips at all. The drivers are horrible. With my genuine (atleast prolific tool says so) prolific PL2303HXA (don’t know which rev) with newest drivers has lock up and BSOD problems. Even a clone/counterfit/compatible chip can not cause those, the driver does.

    If you have a device sending data at certain settings and you try to read it with other settings (like when device is sending data with no parity, and you try to read with parity on) it causes BSOD on win7. Then all kinds of weird lock ups where the driver just stops functioning and when you reconnect the device, it will not get a port number and you have to boot the windows, except when you do, it will not shut down correctly.

  7. The fact that a 3rd party device is lazily designed to be driver-compatible with your product does not give you the responsibility to support it, but NEITHER does it give you the right to conduct cyberattacks against it in the wild. They would be perfectly within their rights to deliberately make the drivers incompatible (it would still be a bad idea from a community relations standpoint), but attempting to damage (deprogram, really) the 3rd party hardware is just unethical.
    I hope the lawsuits cause mortal damage to the company.

  8. I for one would not use any FTDI products before this mess, and for sure not using any after. The whole EOL chips and windows 8 / OS X Yosemite driver incompatibility was enough for me.
    Upgrade your OS? Sorry you cannot use the 15 programming dongles you own any more. They are not compatible, but in reality they are, because you can use their win8 driver if you have the registry, or even there vista driver.
    To me they have just given away the market, and good riddens.

  9. This is a two wrongs don’t make a right scenario. FTDI really didn’t think through the implications of breaking products unsuspecting folks had bought. How is hurting folks who have no way to know what is fake or not any use? Far better scenario would have been for FTDI to pop-up a message telling folks it was counterfeit and recommending they rebuy. Personally I have been careful to buy both genuine prolific (who also had drivers that stopped working with counterfeit prolific devices (but didn’t brick them)) and FTDI adapters because I have found neither works 100% in all scenarios.

  10. Thanks, Alex Balcanquali, that was useful information. On top of the mess they made which makes them look like kids playing with fire, no business acumen or skilled quacks.

    TLDR for those who don’t want to wade in the FTDI fanboys one-sided, immature tirades interspersed with reasonable comments –
    FTDI messed up. Then once people told them – ‘you messed up’, they became aware that they messed up and fixed it.

    And as to what a professional and trustworthy company would have done instead of the brutish tactic, (in an alternate timeline, if you’re thinking of FTDI) – looped ASCII string “NON GENUINE DEVICE FOUND!” – from Wikipedia, https://en.wikipedia.org/wiki/FTDI

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.