If you know where to go on the Internet, you can pick up an FTDI USB to Serial adapter for one dollar and sixty-seven cents, with free shipping worldwide. The chip on this board is an FTDI FT232RL, and costs about two dollars in quantity. This means the chips on the cheap adapters are counterfeit. While you can buy a USB to serial adapter with a legitimate chip, [Syonyk] found a cheaper solution: buy the counterfeit adapters, a few genuine chips, and rework the PCB. It’s brilliant, and an excellent display of desoldering prowess.
Why is [Syonyk] replacing non-genuine chips with the real FTDI? The best reason is FTDIgate Mk. 1, where the official FTDI driver for Windows detected non-genuine chips and set the USB PID to zero. This bricked a whole bunch of devices, and was generally regarded as a bad move. FTDIgate Mk. 2 was a variation on a theme where the FTDI driver would inject garbage data into a circuit if a non-genuine part was found. This could also brick devices. Notwithstanding driver issues, the best reason for swapping out fake chips for real ones is the performance at higher bit rates; [Syonyk] is doing work at 3 Mbps, and the fake chips just don’t work that fast.
To replace the counterfeit chip, [Syonyk] covered the pins in a nice big glob of solder, carefully heated both sides of the chip, and slid the offending chip off when everything was molten. A bit of solder braid, and the board was ready for the genuine chip.
With the new chip, the cheap USB to serial adapter board works perfectly, although anyone attempting to duplicate these efforts might want to look into replacing the USB mini port with a USB micro port.
Or just don’t bother and buy a board/cable with another UART to USB chip – like the Chinese CH340, Microchip’s MCP2200 or similar.
Reworking the counterfeit boards does not make any economical sense, IMO. If you really need the features of the FTDI chip (like the bitbang mode) then buy a board with genuine chip. If you don’t (and most people don’t), then just go for one of the above and be done with it – and you will still save money. They are equally well (or better) supported than the old FTDI chip that doesn’t use a standard CDC class driver.
That bit bang mode is interesting. The datasheet mentions that the “CBUS Bit Bang” must be enabled in eeprom.
Is that easy?
You configure that using FTDI config tool, just plug the usb-serial converter that is using an FFTDI, open the tool, config it, click save, done.
thanks
check out libftdi for controlling the chip
I used the FT232R or FT231X synchronous bit banging feature for my device Nusbio.net and can support 8 IO with SPI and I2C between 15 and 28 Kbyte/S for Windows and .NET. As mentioned above the FT232H is better and getting cheaper.
The FT4222H is also good. Though with the FT232H and FT4222H I can’t seems to transfer faster than 2.4 Mb/S in SPI despite the promise of the datasheet.
While Microchip’s normal products are just fine, the MCP2200 is a kluge. It’s not a drop-in USB device, it requires configuration and it does forget that configuration from time to time. It’s not a good plug-and-play device as they are frequently not detected properly when plugged into a USB port. My recommendation is to stay away from the 2200 and use the real FTDI part or look at Silicon Labs for an alternate.
I’m almost certain that the MCP2200 is really a PIC16 or PIC18 adapted to be a USB Serial chip, just from the way it behaves sometimes. That being said, I haven’t had many problems with it and it’s cheap. However, I built a diagnostic tool that used it, and our vendor in China couldn’t get to work at all. I would say 50% of the time it wouldn’t be detected by their PC. So good enough for us maybe, but the rest of the world no.
Sorry it was the MCP2221 that I had problems with.
The MCP2200 is a preprogrammed PIC18F14K50 and you can erase + reprogram it using a PICkit. The read protection is enabled though. They are/were sold slightly cheaper than actual PIC18F14K50 chips. The question is whether they’re fully working PIC18F14K50 chips or whether they’re chips that failed the tests but enough of it works to be a MCP2200.
The knockoff boards are much better than the shit I see sold using ftdi chips. look for 5V and 3.3V combined as well as the pinouts. Every board out there is 5$ more with a ftdi chip minimum and the chip is maybe 2$. FTDI is a scam in my opinion even with the modicrum of features the official chip has over the fakes.
Agreed, I could never justify the paying the highway robbery-level “because we can” prices for FTDI serial adapters, even if they’re reusable between projects – and anyway, the shit they pulled with their drivers made me ferociously hostile towards them.
I don’t bother with the FT232R anymore. The CP2104 is just about as good (except for crappy GPIO support, but it does have GPIOs at least).
But the FT232H is awesome. The FT2232H is almost double awesome. Why can’t people work on cloning *those* to try to drive the price down?
I recently talked my boss into letting me shoehorn an FT232R into one of our products so I didn’t have to bust out an RS-485 USB adapter.
I’m not familiar with the FT232H, and a cursory skim of the datasheet doesn’t highlight anything amazing. What makes it better?
As a serial adapter, they’re not any better – but then again, a CP2104 is also just as good and not any cheaper. So if I just need serial, I go with a CP2104.
But the fantastic thing about the FT232H are the alternative interface modes. It’s not a USB-to-serial anymore. It’s USB to lots of different things. So suppose you’ve got an SPI flash on the board which *normally* talks to the microcontroller that the FT232H is there to talk with via serial. But you want to be able to program the SPI occasionally (say, for initial programming).
In that case, you can just tie a GPIO to the reset line of the microcontroller, and have a bus transceiver enable the FT232H to take over that SPI. Then you swap the FT232H into MPSSE mode, lower the reset line, and you can program the SPI at ~megabytes per second. Basically, any kind of serial interface you can handle fine.
The FT232H is also just super-useful as a high-speed interface chip to, say, an FPGA. The synchronous FIFO mode can push ~5 MB/s with very little fuss. An FT2232H is almost twice as awesome because it gives you that high-speed interface *plus* a second interface you can configure as a JTAG interface to the FPGA. I say ‘almost twice as awesome’ because you can’t do synchronous FIFO on one interface with a second interface on an FT2232H, so you have to use async FIFO mode.
Basically, an FT232H is an FT232R + FT245 + new features. What’s the cost? Basically the cost of an external crystal and a serial EEPROM.
That sounds excellent. I’ll definitely have to order some samples ASAP.
CH340/CH341 is trash. OSX drivers are terrible or nonexistent.
But I don’t have a shitty computer so why would I care about that?
+1
That would not be an argument for me. Apple is no way to go for me. So Windows and Linux drivers would be an argument.
interesting, I still have one of those chips and a probably fake chip on a serial board, but wonder if this is necessary for my use on my mac environment..
IDK but on Linux such operation does not make any sense, everything works OK no matter if fake or genuine. I prefer CP2102 anyway.
There is fake CP2102 also, and they are worse: they ouput 4.4v even when configured for 3.3v operation. Nice when you use 3.3v only devices
When making something that uses eBay USB-serial cable I add zener clamp diode on 3.3V com lines, just to be sure. I measured well over 3.3V with some cheap interfaces.
Got a link for identifying fake CP2102s? I recently ordered one from a cheap Chinese vendor as a test and it’d be interesting to test if it were fake.
No, just found out that one I got from banggood output 4.4v instead of 3.3v, I didn’t dig further.
buy CP2104 next time, there is separate pin for Vio
CP2102s are horrible too. They require polling of modem signals via control request instead of a dedicated async end point. So the driver has to send a EP0 request every few milliseconds; and do so whether you are using HW flow control or not.
Most drivers, like the CDC spec, setup a dedicated EP for pushing out modem control status only when it changes.
That sucks I guess…for my purposes I almost never use hardware flow control so the CP2102/CP2104 work fine (more stable than the bargain FT230 chips). Good to know, though. And Silabs’ configuration software sucks A LOT.
This is what happens when the driver is written by the kernel development team. Malware from the chip developer company does not get signed by their private key and then pushed out into userland without any vetting.
If you’re having any problem programming your Baofeng with Chirp in MS/Win, it’s worth the effort to boot Ubuntu live CD and try it on that platform.
If you can deal with 1Mbps, then the CP2102 beats the FTDI in price/performance and also allows you to vote with your wallet regarding scummy driver tactics.
I have nothing but good experience with them, though your mileage may vary.
I have an FT-991 and an IC-7300 that agree with you.
Why replace a perfectly good and very rugged Mini B connector with the weak Micro B connector? I have NEVER had any problems with the Mini B but I’ve had plenty of problems with the Micro B connectors as they age and get a lot of use. The connector and cable ends get worn to the point where a power connection gets quite iffy. A brand new cable helps, but is still wobbly in an old Micro B port. Eventually the cable plug wears some and there goes the connection quality to crap again.
The heavier Mini B connectors have bigger contacts and I’ve never seen one get wobbly. The cable plugs in solidly *ka-chunk* and it’ll stay there.
Most devices, the argument for thinness of the Micro B connector makes no sense because the device is plenty thick enough for the Mini B.
The Micro B claims to be able to withstand a greatly larger number of insertions and removals, but in practice it has been the opposite, less reliable than on phones and PDAs with Mini B I used prior to 2007.
+1
Another selling point is that the mini can be hand soldered fairly easily, not so the micro.
USB Mini pins are 0.8 mm apart. USB Micro pin pitch is 0.65 mm. They appear similar enough to casually write “look into replacing the USB mini port with a USB micro port”, but anyone attempting to actually do this will be in a rude awakening when the short, non-bendable pins don’t actually line up.
Sometimes it seems like Brian just has to throw in a line like that to see if he can get a rise out of people.
Decent article, that last bit of advice aside.
I’d be tempted to go the other way and replace the mini with a full size B connector and put it all into a nice housing. Micro-USB just gives me the willies.
I’d actually love to see either a 2.5″ USB drive or just a 2.5″ USB enclosure (either 2.0 or 3.0, preferably 3.0) that used a proper full-size USB-B. The Micro-B 3.0 is just way too fragile for any kind of field use. Some of the newer models with the USB 3.0 controller and connector built directly in the HDD itself are even worse. Attempting any kind of recovery when a few hundred miles from the nearest coastline is a complete PITA.
There is a full size USB 3.0 B connector. http://www.datapro.net/images/usb_3_b.jpg
I’m well aware of that. What I haven’t seen is a 2.5″ drive enclosure that uses one of those instead of a micro-B or the rare one that uses a female A (needs an A to A cable).
You ain’t the only one! When I built my custom keyboard, I made sure to buy a controlelr with a USB Mini-B connector… I know that the Micro standard claims it’s durable for more insersions, but the stupif housing wears and becomes flimsy… You will NEVER catch me installing a USB Micro-B connector onto ANYTHING I ever build.
USB C… That looks interesting. Having seen C for the first time in person with my new Nintendo Switch… I have to say it feels robust, sturdy… And teardowns of the Switch show me the housing the use has a steel securing bracket that can be screwed down… Now THAT’s a robust looking connector!
Fact.
As somebody who repairs USB micro jacks on smart-phones this is very common.
Even more so a PITA is the non-standardization of the usb micro female socket…
Having reworked a few SMD PCBs in my time I would suggest that cutting off all the chip pins with a craft knife is the way to go. The pads can then be cleaned up one by one without risking ripping them off and destroying the board. You are not going to reuse the fake chip, in any case, so why take the care to preserve it?
Removing the connector is another kettle of fish entirely as the thermal mass makes it very difficult. I have just replaced one on a tablet and it was a nightmare!
Agree, I’ve done this with a Swiss army knife as the curve helped.
Very quick and easy. But as you said the chip is sacrificed.
What you need is a hot air rework tool. It makes this kind of project a breeze and I’ve even had reasonably good performance out of the cheap chinese versions(~$60). Well worth the investment.
+1, combo rework station (hot air+iron), like YH8786D are so cheap and very good for money, the price of a bunch some genuine brand tips
Right. There’s many very easy ways. It’s *very* much at the beginner level. Either cut the legs off, or use a large tip (a knife works great) and lots of solder (it should take a few seconds), or use any hot air gun… There’s even specialized tips for this kind of work, like the T15-1001 from Hakko.
To be fair, the pic looks like chipquik was used – that’s what I’d use anyway in the rare cases when I want to desolder something – no scratching of traces under the IC, no overheating the PCB…
I tried this some times and I often damaged the PCB by doing this. I prefer hot air or if not available or suitable (many small components in close vicinity) then the described “tin flood” method. With some care a normal “hardware store” heat gun can be used.
Solution 1: Avoid FTDI and their clones
Solution 2: Don’t “do” Windows.
Best used in combination.
Avoiding fake FTDI clones is a sensible thing, but why avoid genuine FTDI?
They make fine chips and they work very well indeed. All the fuss over the “FTDIgate” stuff was because the fake chips were affected; no real FTDI chips were affected at all.
A significant number of people had purchased boards from reputable suppliers in good faith, and were bitten when the drivers were arbitrarily changed with no warning and no opt-out, and left with bricked hardware. #1 rule for companies – don’t destroy something that doesn’t belong to you.
It would have been fine for the driver to refuse to talk to the chip – taking the extra step of effectively destroying the hardware was disgraceful.
FTDI can go screw themselves – I will not use their chips in designs, and hopefully there are enough people around to do the same in the future.
Exactly so!
I have something for you to ponder. How come it is commercially advantageous for those evil sellers to fake those chips using a much more complex, programmable chip? How come they are able to sell that cheaper than the original FTDI chips? What is the effect of a monopoly on signed Windows drivers on the price of outdated old chips?
You’re assuming a lot on the performance of the fakes. There’s a quality difference, and that has nothing at all to do with Windows drivers or their cost.
Its a mask rom 8051, if that is much more complex or “programmable”..
Avoiding genuine FTDI is sensible because they’ve weaponized their driver. There’s no guarantee that a driver bug in the future might not affect genuine chips by mistake.
The best move is to pick chips that use the class driver. My choice is the Cypress Semi CY7C65213.
That does surprise me. An actual chip that uses CDC.
I wonder why most of the USB serial chip people avoid the CDC driver. I had assumed that it was missing some features.
So far as I know, there are only two things that truly separate the Cypres chip from an FT232: 1. It doesn’t support their bitbang commands. 2: It doesn’t use a weaponized driver.
To be fair to FTDI, if the fakes weren’t a problem there wouldn’t be as many people complaining about what FTDI had done.
FTDI aren’t the only one not to use the CDC driver – and each and every one of them would appear to prefer to go through the process of getting drivers certified and signed on Windows than use it. (As far as I’m aware, FTDI is one of the few that’s actually bothered producing their own Linux driver. It’s not the one that’s in the kernel – but they did actually write one)
I have had problems with the “I2C” bit banging in the past. I put it down to not handling clock stretching at the time. (Which would be unusual if it was true)
Uhhh… Okay. Ignore that. The FTDI linux drivers are binary only – and Silicon Labs produce source (If you have an account…) for drivers for the CP210x. (The Linux drivers in the main line kernel supposedly started because silabs only produced a binary driver)
*Lots* of I2C devices/things don’t handle clock stretching right. It’s part of the reason SMBus added a timeout on how long transactions can last. Doofy I2C devices can hang the bus.
It was more that clock stretching in devices is reasonably rare these days. Still present, though.
Basically shows up any time you have a microcontroller be an I2C slave, with the exception of a rare few which have good enough I2C peripherals to handle it.
+1
“Avoiding fake FTDI clones is a sensible thing, but why avoid genuine FTDI?”
Because fake chips are labeled as genuine, so that most buyers don’t even known about the fakes and are in perfect good faith but FTDI hits them as criminals ruining their work. This is a terrible move that should be punished with the only weapon we have at hand: our wallet.
I’m guessing it was an attempt to try and make the manufacturers using the device be very sure that they didn’t have fakes. It has raised awareness at how big the problem of fakes are (Otherwise there wouldn’t be as many people complaining) but it has backfired a bit.
As a manufacturer I would try to be completely safe and make sure there is no FTDI logo on the chip I use. Then I am completely sure to not have a fake FTDI.
“why avoid genuine FTDI?” – because of “FTDIgate”, because of them deliberately making drivers which destroy (or at least disturb) stuff. I would accept, if the make new drivers which just refuse to work with clone chips. But what they did is just criminal computer sabotage. They deliberately interfered with the operation of devices which they do not own.
“The chip on this board is an FTDI FT232RL, and costs about two dollars in quantity. This means the chips on the cheap adapters are counterfeit.”
It’d probably be better to say that the chip on the board *claims* to be an FTDI FT232RL.
Also, why buy the boards with fake ones (which enriches the manufacturers and distributors of the counterfeit chips) in order to save a buck or two rather than make your own boards (if you’re interested in SMD work) using any of the free PCB layouts on OSHpark and the like, or simply buy legit FTDI breakout boards for a few bucks more if you’re lazy?
Better: use another chip so that counterfeiters get no money and FTDI get no money.
I threw away my (genuine) FTDI chips when they decided to knowingly destroy hardware they didn’t own – I will not support them in any way and hope they go bankrupt.
FUCK YOU FTDI AND FUCK THE CHIPS YOU RODE IN ON!
+3
But its fine to clone a chip/its API and use other brand drivers as a crutch to sell crap chips..
You must be rolling in money working for free AND letting anyone copy and under-cut you…
It’s not fine to attack your end customer by bricking their devices. Not all users of fake FTDI chips _knew_ they were fake. A few legitimate distributors got had by the fakes. That meant that random devices, made from distributor stocked chips got BRICKED, with no warning…
You can sue a competitor for pirating your device. You must prove they actually pirated it. Reverse engineering is a PERFECTLY LEGAL thing. Now, the chips sold _as FTDI_ chips are indeed counterfeit, as they claim to be something they are not. If those chips were sold with a different part number, or even with the same part number, but marked as being non-FTDI, then they would be perfectly permissible.
The issue is not that the chips were reverse engineered, but that the chips are sold AS FTDI PARTS.
FTDI has a legal right to go after that, not that the chip’s functionality was duplicated, so YES, it is totally alright to clone a chip, as long as you are not just copy-pasting the design (you must reverse engineer it), and as long as you are not falsely claiming it as a genuine original (that, they DO, unfortunately)…
The issue that stands, is rather than have the driver put up a pop up saying “The USB to serial chip used in this device was detected to be a counterfeit device. You must replace it with a non counterfeit FTDI chip to continue using this driver.” They could also have not completely screwed over people by giving them till a date, or X number of uses before the driver refuses to communicate with the chip, to give people time to source a new chip, or replace the device. INSTEAD… FTDI updated their driver, with no warning, to actively disable the functionality of the device, so it would cease to operate on ANY computer, ANY OS, with ANY driver!
That was wrong. They attacked end users, including people who had themselves been duped by counterfeiters, who never knew they had non original FTDI parts. Wdid was ABSOLUTELY WRONG, and as far as I’m concerned, they are a company I will NEVER trust. The features of their products mean nothing to me, as I will never buy their parts. I replaced my Arduino USB to serial adapter that I use on Pro-Minis with a CH340 based device. Screw FTDI!
Counterfeit FTDI chips, it’s a similar thing to the counterfeit electrolytic capacitors with the faulty electrolyte which have plagued the electronics industry since the 1990’s.
Except with the FTDI fakes they work just fine and would continue to do so for essentially forever. The bogus capacitors will almost all certainly fail at some point.
FTDI decided to *make* the counterfeit chips fail – punishing the end users who have no idea that a chip in their consumer product is counterfeit or even what an FTDI USB to RS232 converter chip is. All they know is that one day their thing they bought at Walmart no longer works so they complain to the company that made it – and that company very likely has no idea their purchasing department bought a batch of counterfeit chips.
Like the capacitors, there’s such a huge number of the chips out there that it is possible to purchase them in bulk surplus lots at below manufacturer suggested or direct prices. Buying low helps the manufacturer’s bottom line.
For example a company buys several hundred thousand, then alters a product or discontinues it before using up the supply. With no other products requiring those parts, the company sells them off cheap (having more than made back the purchase price of the original lot, or sells at a loss to write off on taxes) to get the storage space back.
Surplus R Us buys those leftovers and passes the savings on to others. Meanwhile Super Surplus (not actually a surplus electronics company) knowingly buys counterfeits and undercuts Surplus R Us a couple cents per part.
The manufacturer buying from either surplus company has no way to know, they just go with the best price. If their product depends on some function or performance specification only the genuine part has, then they’ll find the counterfeits in QA testing. But if the application doesn’t depend on such things, the product goes to retail – then FTDI decides to crap on the unknowing end users and their ire falls upon the manufacturers who had no clue about the problem, and no intent to produce a product with counterfeit components.
Meanwhile, Super Surplus has folded up shop and is gone so there’s nobody for the ripped off manufacturers to go after.
FTDI smugly sits back and says “You should have bought directly from us, only directly from us. It’s the only way to guarantee you get genuine chips.”
So yeah, piss on FTDI. *WE* out here don’t deserve to be pissed on by FTDI. FTDI should be pressuring the UK government to threaten China with trade sanctions if they don’t crack down on the companies producing phony FDTI chips.
1) It’s not fine to attack your end customer, but FTDI didn’t attack a customer. This doesn’t necessarily make it right, though.
2) I doubt the drivers have any access to the sort of levels of the OS that would let them pop up any sort of information on the screen – not without also installing an application. (Can driver updates that started out as drivers also install applications? Probably, I suppose)
3) What would most end users do if they got a “This device will stop working in X uses”? Probably click the thing that dismissed it as soon as it appeared. Then complain when it stopped working.
4) There may be no way for the driver to detect the device is counterfeit. The first attempt was to work around a difference between the devices to write to the EEPROM. It’s possible that the second one does something similar.
5) The devices don’t “cease to operate on ANY computer, ANY OS, with ANY driver!” – the Linux drivers have been able to talk to the “bricked” ones for a while.
Well, as they say, it’s their funeral… as far as I’m concerned, those fuckers are deader than a doornail. Just use something else that doesn’t try to cut your throat the moment you’re not looking in the name of self-defense…
It’s cheaper to replace the fake than to populate a board from OSHpark. If it’s for personal use I’d stick with the fake and use an old driver. FTDI isn’t getting another penny from me if I can help it.
What’s to like about microusb? I’ve had horribly luck with it compared to mini.
Well, like it or not, micro USB is more ubiquitous.
Not to me. I have ~10 mini USB cables lying around and struggle with swapping back and fourth of the 1 micro USB cable I happen to own.
I really don’t want to go out and buy more cables and the difference in size is ridiculously small. The components and casing are always bigger anyway so can’t really see the point in the micro except for uber thin phones.
Cool story, bro.
For me it’s the other way round. I have several micro USB things, just the Bluetooth mini keyboard I use as a remote while watching Youtube uses this damn Mini-USB connector.
There is only one worse thing: A LED working light, what is really good for its purpose, but uses 3,5mm mono headphone jack for charging. Of course there is a USB wallwart and a suitable cable, but not just use a (micro) USB connector on the device, so I could use one of the many USB cables lying around instead of needing this one special cable?
Jack plugs short out briefly on insertion, so they’re a terrible idea for a power supply. We all should’ve learned that lesson from the ZX81.
@Greenaum: I know :-( Luckily no damage up to now. Perhaps I thought of connecting the cable to the lamp before connecting it to the USB, but I am not sure :-)
Might try this on my PIC programmer which got “bricked”.
On the flip side, there is a lesson not to cheap out and buy a £70 tool and not a £122 tool when the difference is shoddy parts and defective programming. Genune ICSPs are expensive for a reason!
And that reason is greed. You should buy the £47 tool instead that has no expensive-as-fuck FTDI time bombs in it at all.
At some point we’re going to have to figure out if fake chips are a good or bad thing. So far, people (including me at times) have enjoyed the low cost of a fake chip. But on the other hand, when everything is fake, reliability and quality will go down hill fast.
There is no figuring it out – it is bad for the consumer and the real manufacturer. The only one that benefits is the counterfeiter.
The consumer can benefit on a cheaper solution.
My FT230X stock from DigiKey works every single time. (Why are people still using the FT232? Unless you need every one of the RS232 handshaking lines which seems unlikely.)
I get a few percent squirrely FT230X in production runs (stay in reset for a really long time on startup). Only fix is to replace. Just switched to CP2104.
I love the quote from THGTTG.
I vaguely remember there being a story on here of a startup (I think) who had a MCU-based platform to end all USB-to-serial converters at prices at least competitive to FTDI (originals).
I wish I could find that, but as far as I remember, it was basically just a firmware for PIC or something. Does anyone have a link?
A USB-serial firmware is great and all, but we’re actually buying USB-serial chips to avoid having to pay the USB-IF thousands of dollars and force customers to install a different driver for every USB serial device.
The title reads as though there were a way to unbrick the dead clone chips…..wouldn’t that be nice.
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/drivers/usb/serial/ftdi_sio_ids.h
Look at line 37. :)
You mean like this? http://hackaday.com/2015/02/08/unbricking-a-counterfeit-ftdi-chip/
Why wouldn’t there be a way? All the morons at FTDI did was reset a freely rewritable number in that stupid chip…
I have a few old 16-bit DOS programs that I use to troubleshoot and repair Electric Vehicle motor controllers from the mid 90’s.
The FTDI adapter was the only one I could get to work with the programs under DOS BOX, on Windows 10.
USB to Serial Adapter Testing:
http://www.youtube.com/playlist?list=PLQdu_G7xyFIQz7z3yqhWUcRPp7SVei3qV
Funny thing: The new mainboard from MSI I bought yesterday has an (internal) RS232 connector. I did not think they make them any more. I hope I still have one of the Sub-D9 slot brackets around from the 1990ies and did not throw them all out. Not that I really need the port, but if it’s there it’s nice to bring it out.
Sorry but when did we accept the fact that a major company can sabotage and destroy our things just because they throw a hissy fit about their intellectual property? What if a bold bloke from Microsoft came to your house smashing your PC with a baseball bat just because they found a counterfeit pirated piece of software on it? There are legal ways to letigate against the counterfeiters instead of destroying peoples property and if they sit in China, well thats tough luck. FTDI should have recieved way more flak for this.
No.
It’s more the equivalent of Microsoft bricking your system via an update because you have a pirated copy of Windows on it. (It has been suggested that part of the reason they haven’t done this in the past is because more pirated copies turns into more legitimate ones by making it ubiquitous)
FTDI makes their money by selling chips – and the money they make out of that pays for, amongst other things, developing the drivers. (I’m sure you’d like a percentage of your income being taken by somebody pretending to be you and damaging your reputation…?)
I’m not excusing them of their actions – it was stupid how they handled it – but a few people in here seem to want to excuse the counterfeiters. (“if they sit in China, well that’s tough luck” – a post on a previous article mentioning FTDI complained that the counterfeiters may have to pay cash for driver development if Microsoft changed their certification process (And thus, for whatever reason, FTDI should pay it for them))
Excusing the counterfeiters was not my point. The point was that a company shouldn’t be allowed to come to your house and break things which don’t belong to them. They saw that they can’t go the legal road but frankly that is not the problem of consumers or hobbyist. This is business between ftdi and the counterfeiters and not between ftdi and guys who bought a device which accidentally has a fake chip in it. Basically a company coming to your house because they see fit is something we or any form of democratic country should not tolerate.
I would be very surprised if what they did was illegal. Ignoring the big “We don’t accept responsibility for damage” in the licence agreement, (And if you didn’t get that on your driver when you installed it, you probably got it as part of using Windows in the first place) it’s not a piece of software that was designed for that piece of hardware.
Palm, for a while, had their devices pretend to be an iPhone. This was how they had their iTunes compatibility. Apple had no responsibility to work with Palm devices. They went to some effort to block it. I’ll grant you, they’re not bricking it – but they’re deliberately taking away functionality.
Let’s take another example. If I buy a stolen car, the police can come along and take that off me. I may have bought it in good faith and be entirely ignorant of it being a stolen car – but that doesn’t mean it’s legally mine.
And yes, my second example does involve stolen goods and what the counterfeiters have done is more akin to fraud.
Blocking it from working with their driver is completely OK. Damaging other’s property is not and I think as this damage was done ON PURPOSE this is hopefully not covered by any disclaimer.
there was a huge backlash against FTDI when they did this (back in September 2014) and most tinkerers avoid their chips now, so no it wasn’t accepted, but that is years ago now and the discussion was done, they are unsavory assholes, it was established and agreed upon already.
“On 24 October 2014, in response to the criticism, FTDI withdrew the driver and admitted that the measure was intended to protect its intellectual property and encourage users to purchase genuine FTDI products. The company also stated that it was working to create an updated driver which would notify users of non-genuine FTDI products in a “non-invasive” manner.”
To finish the story:
“In February 2016, it was reported that FTDI had published another driver on Windows Update with DRM components intended to block non-genuine products. This time, the driver will communicate with affected devices, but all transmitted and received data is replaced with the arbitrary, looped ASCII string “NON GENUINE DEVICE FOUND!”, which could cause irregular interactions with devices.”
What they did will never be forgotten and will never be forgiven. Hopefully their “IP” can keep them warm, they’ll never see any business from me or anybody else I have any sway over.
Yes
Nice article.
It would be even better if FTDI had some better prices when buying from their web store. The shipping costs are too high.
Since when replacing fake chip with genuine one fixes the fake?
Dont cheap out on components, counterfeit parts can really ruin your day
couldnt a hexeditor be taken to the driver to hack out the detection and the hacked driver then hosed by someone like the torrent sites or something?
Newer Windowses search for online drivers first whenever you connect anything to the computer, just to keep them updated. Also hacking driver would be piracy, which is a big no-no, especially in USA (so called “land of freedom”). Best solution is not to use anything made by FTDI, because no one can be sure he has a genuine chip or fake one, even major distributors…
Hacking drivers isn’t piracy – but distributing them (Modified or not) would be a Copyright violation. And you’d have to give people instructions for disabling the drivers signature check on Windows 10 machines.
64-bit Windows requires drivers to be digitally signed by Microsoft so they cannot be modified. And Microsoft charges the manufacturers to sign their drivers, thus why some of them get pissy about third parties skipping the messy process of getting it done and just cloning their API.
There are ways around the digital signature – but it does involve flicking the right switch in the options.
Good luck getting an end user to follow them. (One page I found had 2 methods. One of them involved disabling secure boot)
It is enough to use the older driver without the malicious code in it. The “update” was not more than introducing the software-bomb.
After such anti-competitive measures, how could I have sympathy if FTDI goes under.
It was a bit stupid but it’s not anti-competitive. Other USB to serial chips are available. And, if a company has the capabilities to produce silicon that emulates the hardware, it’s surely not beyond their capabilities to also produce their own command set and their own drivers.
How could it not be anti-competitive, if its sole purpose to stop or break a competitor’s chip?
If they held a patent, they should have created an embargo or restriction on the other chips, what that did was taking the law into their own hands.
It is not a competitors´s chip, with its own band and drivers. It is a fake, wrongly using their brand and drivers. And their driver worked ok with the right chips. If people were using their driver with the wrong kind of chips, it is like putting diesel in a gas car and complaining the diesel seller broke your car.
No, it’s like selling gasohol without warning after years of normal gasoline, well knowing, that the own built engines can tolerate it and the others getting their seals and gaskets eaten away.
[Some complaint about how you are not fixing the fake FTDI, but just replacing it. ]
I get this measure on a small scale, you have a board and for the sake of it, wanting a real thing, it is worth paying for a real deal chip to replace the fake one.
But… if you really want a few more, just grab your favourite cad program and draw the freaking ~5 component board and be done with it. I made quite a few like this with the olderr FTDI using a crystal.
If you want to be smart, chances are whatever micro you are connecting to the FTDI comes in a similar flavour with USB, for a smaller price than the 2 together.
Worst idea ever! Not only is the desoldering just a MESS, no, you acutally pay the cheating chinese selling copy-chips and FTDI for the original in the same project. Just don’t do it!
This is pure sin. Dont use FTDI or god hate you! You lost ability to get childrens and when you die you go straigh to hell. Yes, rhar so true. Do not trust fake media.
I am just going to jump into the dogpile and say “Just say no to FTDI”. We can admire the (admittedly pointless) hack that this article highlights, but …… There are plenty of USB to serial chip alternatives that are dirt cheap and work just fine without any extra effort. What is the point here? The cheap gadgets from China I currently use have a CP2102 chip. Work great.