FTDI driver kills fake FTDI FT232??

[fcb]

I use FTDI parts in designs, and have done for years.  They have always been a good company to deal with, sales, support, etc...

Why exactly it FTDI's problem if their drivers are used outside of their T&C's???

[rsjsouza]

Meters with fake safety ratings need to die.

I agree. Since the early days of electricity/electronics, we are taught how to carefully read the specs of anything - including test gear - before using it. Since Sparkfun is trusted by newcomers and inexperienced hobbyists, in my opinion they should be extra careful with the misleading safety ratings of the equipment they sell. 

To me a similar scenario is having an infamous capacitor brand mark their products as X2 or Y2 when they are not, "just because they are cheap". Experienced folks will recognize the brand and stay away but newcomers will trust the markings blindly.

And that is the biggest peeve with all this FTDI issue: at the beginning I was siding with FTDI but, after some consideration, "newcomers" will blindly believe the functionality of the product they bought - thus I consider this a bad practice on FTDI's part. OTOH, experienced folks trying to use this for professional purposes should know better: buying the cheapest of eBay when your reputation is on the line is asking for trouble.

[EEVblog]

Why exactly it FTDI's problem if their drivers are used outside of their T&C's???

It's been endlessly debated here.
Bottom line is FTDI know what they did was wrong and it would ultimately hurt their company, and now they have backed down.

[mikeselectricstuff]

Why exactly it FTDI's problem if their drivers are used outside of their T&C's???

Because in the case of Windows Update, those T&Cs were not made clear to the user beforehand. Even on the FTDI website they were on a seperate page to the driver download page.

Because T&Cs can not override criminal law - in this case criminal damage and possibly the computer misuse act.
 

[Simon]

Why exactly it FTDI's problem if their drivers are used outside of their T&C's???

Because in the case of Windows Update, those T&Cs were not made clear to the user beforehand. Even on the FTDI website they were on a seperate page to the driver download page.

Because T&Cs can not override criminal law - in this case criminal damage and possibly the computer misuse act.

if FTDI are so concerned about fakes and misuse of its driver it begs the question as to why they are making their driver so available through windows update.......

If i was microsoft I'd refuse to "deliver" any more FTDI drivers.

[marcan]

ere is some code that will non-destructively test for clones.
https://mrcn.st/t/detect_ftdi_clone.py

Forgive me, I'm not a python user.  I get the error:

ImportError: No module named ftdi1

I assume this means I don't have the python bindings for this.  How do I install them (I'm using Linux Mint)

Look for a package named libftdi or libftdi1, and possibly something with ftdi and python in the name. On Ubuntu it seems to be libftdi1 and python-ftdi.

I just managed to get ahold of a fake FT232RL and can confirm that the script above does indeed detect it. Specifically, it's this Deek-Robot USB to TTL board that I found in a fairly reputable shop in Akihabara (Tokyo).

Edit: photo of real vs. fake (click for high-res). I think the easiest way to tell them apart is that the real one has a larger pin 1 dimple centered between pins 1 and 2, while the fake one has a smaller dimple that is offset towards pin 1.

[fcb]

Why exactly it FTDI's problem if their drivers are used outside of their T&C's???

It's been endlessly debated here.
Bottom line is FTDI know what they did was wrong and it would ultimately hurt their company, and now they have backed down.

I must have missed the posting (in the 40 odd pages here) were FTDI admit that they are deliberatly bricking non-FTDI parts....

[marcan]

I must have missed the posting (in the 40 odd pages here) were FTDI admit that they are deliberatly bricking non-FTDI parts....

They indirectly did, on Twitter, then deleted their tweets (here's a screenshot). But the code speaks for itself - it's blatantly designed to brick non-FTDI parts and is otherwise completely useless, having been carefully engineered to exploit a minor implementation difference (that even differs for other genuine non-232RL FTDI chips) to brick clones. Whether FTDI did this deliberately or not is not up for debate - they did, period.

[Simon]

Check consumer protection laws in your country. I think most of the EU would have a good case for a return/refund.

Obviously if you bought from eBay there is no warranty, buti recently returned an 18 month old battery to Amazon because the NFC bit was fake.

If you have a case for returning a fake it's not a case of standard return times and warranty periods, it's fake and should never have been sold and the seller is at fault.

[SeanB]

Edit: photo of real vs. fake (click for high-res). I think the easiest way to tell them apart is that the real one has a larger pin 1 dimple centered between pins 1 and 2, while the fake one has a smaller dimple that is offset towards pin 1.

Well, that means the 2 Freeduino boards I have do have real FTDI chips on them, they look like the ones on the left.

[Rufus]

I must have missed the posting (in the 40 odd pages here) were FTDI admit that they are deliberatly bricking non-FTDI parts....

There not being any legitimate drivers for non-FTDI parts they always were bricks under Windows. They found a way to prevent some illegal use of their drivers and exploited it. If you want to carry on illegally using older versions of their drivers you have to jump through some hoops - cry me a river. 

I have some minor sympathy for the few people who found trying to use illegal drivers under windows meant they had to jump through some minor hoops to use the device with open drivers under linux.

Sparkfun with 30 products using FTDI parts have a reasonable response to the issue

https://www.sparkfun.com/news/1629

A. They don't think they have a problem
B. If it turns out they do it is their problem

[fcb]

I must have missed the posting (in the 40 odd pages here) were FTDI admit that they are deliberatly bricking non-FTDI parts....

They indirectly did, on Twitter, then deleted their tweets (here's a screenshot). But the code speaks for itself - it's blatantly designed to brick non-FTDI parts and is otherwise completely useless, having been carefully engineered to exploit a minor implementation difference (that even differs for other genuine non-232RL FTDI chips) to brick clones. Whether FTDI did this deliberately or not is not up for debate - they did, period.

Thanks for posting that - not exactly a smoking gun though.

If the bricking of the fake is a 'happy' accident of the way the driver works, then fair enough I say. But if the driver has been deliberately written to render the part permanently useless (by re-writing the EEPROM or perhaps some other irreversible method) then FTDI would be on shaky ground, however I think it would be fair for the driver to simply not work. After all this is what ink-jets do.

Has someone got the source code for the driver or has FTDI admitted to that also?

[fcb]

I've thought about it some more (and read a chunk of the sparkfun article).

As a manufacturer of things with FTDI chips inside, I think that FTDI's behavior is acceptable.  Here goes:

If I buy FTDI parts, I expect them and want them to be genuine - I will most likely have paid good $ and have no desire to fit counterfeit IC's in my products.  If one of my products was to suddenly 'brick' due to a counterfeit part having made it's way inside, I would take it up with the supplier of that part and expect them to make me whole again (assuming it wasn't me buying a random reel from some dodgy corner of the internet).

If I was joe-public and bought some device which was suddenly rendered useless due to this issue, then I would take it up with the supplier, if they happen to be in China and can't be bothered replying to my emails then that would be my tough-shit.

[nsayer]

I must have missed the posting (in the 40 odd pages here) were FTDI admit that they are deliberatly bricking non-FTDI parts....

They indirectly did, on Twitter, then deleted their tweets (here's a screenshot). But the code speaks for itself - it's blatantly designed to brick non-FTDI parts and is otherwise completely useless, having been carefully engineered to exploit a minor implementation difference (that even differs for other genuine non-232RL FTDI chips) to brick clones. Whether FTDI did this deliberately or not is not up for debate - they did, period.

Thanks for posting that - not exactly a smoking gun though.

On what planet is it not?

If the bricking of the fake is a 'happy' accident of the way the driver works, then fair enough I say. But if the driver has been deliberately written to render the part permanently useless (by re-writing the EEPROM or perhaps some other irreversible method) then FTDI would be on shaky ground, however I think it would be fair for the driver to simply not work. After all this is what ink-jets do.

Has someone got the source code for the driver or has FTDI admitted to that also?

Assuming that the decompilation is genuine (I have not personally verified that, but if it wasn't, I'm confident it would have been exposed by now), what other conceivable purpose could that chunk of code possibly have?

Keep in mind as well that people have used USB protocol analyzers (I've seen the posts on Twitter, but cannot be bothered to go find one right now) that have demonstrated the same sequence of actions that the decompilation shows. Again - what possible purpose is served by that sequence of operations apart from deliberately bricking devices that don't ignore the given operations (like genuine FTDI devices apparently do)?

This was not an accident. This was not a mistake. This was a deliberate, unethical act. That much is quite clear.

[Simon]

Sparkfun with 30 products using FTDI parts have a reasonable response to the issue

https://www.sparkfun.com/news/1629

A. They don't think they have a problem
B. If it turns out they do it is their problem

Yess, yess you finally got it IT'S THE PEOPLE BUYING THE PARTS AND PUTTING THEM INTO THEIR PRODUCTS that need to be targeted, worked with NOT end users who are non the wiser.

FTDI's logic was thus: we want to sell more chips and we don't like people faking our chips and using our driver. Oh nice we have found a vulnerability in the fake chip. What shall we do ?

2 scenarios:

1) as they have done, brick the fakes, and by doing so punish the end user of a product that may be well out of warranty for something they had no part in. They have no idea why they device broke and i guess FTDI hope they won't buy another one from the same supplier so hopefully get a genuine one next time so FTDI hope. The supplier of the product is non the wiser and may now be using genuine FTDI chips so FTDI shoot themselves in the foot. As no one knows what the problem is people keep buying fakes anawares although by this point the problem is probably found and corrected but potentially millions of unsuspecting and innocent users are punished and financially damaged for products they bought in good faith and are already in service.

2) FTDI stop their driver from working with a fake chip so the device can't be installed as it now has no driver support (and they could have quietly found a way to distinguish so historic devices carry on but new ones won't work). If the supplier of the equipment have not figured this out yet they get loads of returns and innocent customers get their money back and no hard feeling with a supplier that may also not be at fault and if that supplier is now using genuine chips they don't feel so bad and keep using the chips. The suppliers will then seek redress from their chips suppliers and maybe involve FTDI and the source of fakes is found and people prosecuted. Everyone comes out on top and counterfeiters are hopefully caught.

What FTDI did was selfish and stupid

[nsayer]

I must have missed the posting (in the 40 odd pages here) were FTDI admit that they are deliberatly bricking non-FTDI parts....

There not being any legitimate drivers for non-FTDI parts they always were bricks under Windows. They found a way to prevent some illegal use of their drivers and exploited it. If you want to carry on illegally using older versions of their drivers you have to jump through some hoops - cry me a river. 

I have some minor sympathy for the few people who found trying to use illegal drivers under windows meant they had to jump through some minor hoops to use the device with open drivers under linux.

Sparkfun with 30 products using FTDI parts have a reasonable response to the issue

https://www.sparkfun.com/news/1629

A. They don't think they have a problem
B. If it turns out they do it is their problem

The part they left out is:

C. If they switch to CDC parts, they'll never have this problem ever.

[Simon]

Regarding: https://twitter.com/0xabad1dea/status/525849774268768256

FTDI have no right at all to brick chips, the chips could be used by linux under different drivers or someone could write a windows driver, this means no one is violating anything other than the manufacturer who badged the chip FTDI if indeed they did, it's just another USB to serial converter, this is not like buying a fake GHD hair straighter. I surely hope FTDI have to pay damages for this, to claim they have to stop counterfeiters in order to guarantee innovation, they have to be joking, if they think serial to USB converters are hi tech they really should get out of the business.

[nixfu]

What is really funny is I started buying FTDI stuff BECAUSE OF WHAT PROLIFIC did with their drivers screwing over customers a few years ago.

[fcb]

Regarding: https://twitter.com/0xabad1dea/status/525849774268768256

FTDI have no right at all to brick chips, the chips could be used by linux under different drivers or someone could write a windows driver, this means no one is violating anything other than the manufacturer who badged the chip FTDI if indeed they did, it's just another USB to serial converter, this is not like buying a fake GHD hair straighter. I surely hope FTDI have to pay damages for this, to claim they have to stop counterfeiters in order to guarantee innovation, they have to be joking, if they think serial to USB converters are hi tech they really should get out of the business.

If you use a counterfeit FTDI part with a non-FTDI written driver, then that's fine and it's upto FTDI to pursue you under the counterfeit laws.

If you use a counterfeit FTDI part with an FTDI driver, then I think it is legitimate for it not to work.

The only issue appears to be that the FTDI driver reprograms the PID and so it won't work without faffing about afterwards, not technically bricked it appears. That will be cleared up with the new device driver issue from FTDI.

Why shouldn't FTDI protect their IP and good name???

[M@rcel]

Edit: photo of real vs. fake (click for high-res). I think the easiest way to tell them apart is that the real one has a larger pin 1 dimple centered between pins 1 and 2, while the fake one has a smaller dimple that is offset towards pin 1.

I'm afraid I have to disagree with you: the chip with the laser-engraved markings on the right is the real chip. The chip with simple printed markings on the left is fake. When you view the high res image, it's visible more clearly.

Compare with the foto in this link:
http://zeptobars.ru/en/read/FTDI-FT232RL-real-vs-fake-supereal

[Simon]

If you use a counterfeit FTDI part with a non-FTDI written driver, then that's fine and it's upto FTDI to pursue you under the counterfeit laws.

If you use a counterfeit FTDI part with an FTDI driver, then I think it is legitimate for it not to work.

The only issue appears to be that the FTDI driver reprograms the PID and so it won't work without faffing about afterwards, not technically bricked it appears. That will be cleared up with the new device driver issue from FTDI.

Why shouldn't FTDI protect their IP and good name???

Good name ? they don't have one of those now. If the driver simply refuses to work with the chip then that is fine. under linux there are non ftdi drivers or maybe i am wrong if there are then ftdi are not being hard done by, not their chip and not their driver. equally someone could write a windows driver and the chip would work, no violation of ftdi's IP.

Then FTDI can use lawful and fare methods to track down counterfeiters and deal with the fake manufacturers and not cause damage to end users who have paid their money and the fake suppliers have had their share. What ftdi have done does very little to the suppliers of old fakes, it will only hit them in the future, in the mean time innocent end users have had their equipment redered useless for no good reason, the manufacturers still have their money for the fakes. The attitude has not been to resolve the situation but to be vindictive no matter who pays...... reminds me a of a certain referendum just had on FTDI land! It's the same old tune: "we have decided we have been wronged no matter what anyone thinks therefore fuck everyone else!"

[Rufus]

The part they left out is:

C. If they switch to CDC parts, they'll never have this problem ever.

Yes, but, because they are not some no-name Chinese asshole they consider buying fake chips of unknown quality and origin to be a problem regardless of what interface they use. FTDI's driver actions mean in the future they are less likely to find fake FTDI chips in the supply chain. That combined with FTDI's excellent driver support and quality is reason to continue using them.

[fcb]

If you use a counterfeit FTDI part with a non-FTDI written driver, then that's fine and it's upto FTDI to pursue you under the counterfeit laws.

If you use a counterfeit FTDI part with an FTDI driver, then I think it is legitimate for it not to work.

The only issue appears to be that the FTDI driver reprograms the PID and so it won't work without faffing about afterwards, not technically bricked it appears. That will be cleared up with the new device driver issue from FTDI.

Why shouldn't FTDI protect their IP and good name???

Good name ? they don't have one of those now. If the driver simply refuses to work with the chip then that is fine. under linux there are non ftdi drivers or maybe i am wrong if there are then ftdi are not being hard done by, not their chip and not their driver. equally someone could write a windows driver and the chip would work, no violation of ftdi's IP.

Then FTDI can use lawful and fare methods to track down counterfeiters and deal with the fake manufacturers and not cause damage to end users who have paid their money and the fake suppliers have had their share. What ftdi have done does very little to the suppliers of old fakes, it will only hit them in the future, in the mean time innocent end users have had their equipment redered useless for no good reason, the manufacturers still have their money for the fakes. The attitude has not been to resolve the situation but to be vindictive no matter who pays...... reminds me a of a certain referendum just had on FTDI land! It's the same old tune: "we have decided we have been wronged no matter what anyone thinks therefore fuck everyone else!"

Back in the real world - I can't see any manufacturer dropping FTDI because they took a stand on counterfeit parts. Also, I suspect that someone is working on an application to unbrick those 'PID=0x0000' parts.

As for FTDI not having a good name? Perhaps not amongst a small % of hackers/makers only, but I and those other manufacturers I know won't stop putting their parts in products as long as they offer the cost/benefit that they do.

[zapta]

If you use a counterfeit FTDI part with a non-FTDI written driver, then that's fine and it's upto FTDI to pursue you under the counterfeit laws.

Users did chose to use FTDI driver, it was forced on them by being shipped with Windows (practically a monopoly) and by the OS selecting it as the appropriate driver for that hardware.

FTDI's competitive advantage is having a driver in the stock Windows distribution but they abused it and people now look for alternatives. Yet to be seen if this will cause Microsoft to include a generic CDC driver and solve this USB/Serial driver issue once for all. Manufacturer can then be able to design chips for that standard and users will not have to install drivers.

As Simon said, USB/Serial is a commodity these days, just like USB mouse or keyboards.

[Simon]

If you use a counterfeit FTDI part with a non-FTDI written driver, then that's fine and it's upto FTDI to pursue you under the counterfeit laws.

If you use a counterfeit FTDI part with an FTDI driver, then I think it is legitimate for it not to work.

The only issue appears to be that the FTDI driver reprograms the PID and so it won't work without faffing about afterwards, not technically bricked it appears. That will be cleared up with the new device driver issue from FTDI.

Why shouldn't FTDI protect their IP and good name???

Good name ? they don't have one of those now. If the driver simply refuses to work with the chip then that is fine. under linux there are non ftdi drivers or maybe i am wrong if there are then ftdi are not being hard done by, not their chip and not their driver. equally someone could write a windows driver and the chip would work, no violation of ftdi's IP.

Then FTDI can use lawful and fare methods to track down counterfeiters and deal with the fake manufacturers and not cause damage to end users who have paid their money and the fake suppliers have had their share. What ftdi have done does very little to the suppliers of old fakes, it will only hit them in the future, in the mean time innocent end users have had their equipment redered useless for no good reason, the manufacturers still have their money for the fakes. The attitude has not been to resolve the situation but to be vindictive no matter who pays...... reminds me a of a certain referendum just had on FTDI land! It's the same old tune: "we have decided we have been wronged no matter what anyone thinks therefore fuck everyone else!"

Back in the real world - I can't see any manufacturer dropping FTDI because they took a stand on counterfeit parts. Also, I suspect that someone is working on an application to unbrick those 'PID=0x0000' parts.

As for FTDI not having a good name? Perhaps not amongst a small % of hackers/makers only, but I and those other manufacturers I know won't stop putting their parts in products as long as they offer the cost/benefit that they do.

So all of this over a few hobbyists ? gee what a lot of hassle, really vindictive then, as I keep saying they have not harmed the seller of existing fakes onthey innocent users that have already made the fakers rich, not very clever move