FTDI driver kills fake FTDI FT232??

[daqq]

Wow, 1400 guests viewing this thread!

At one point yesterday it was close to 5000. Apparently it's big news, and I can understand why - this can get real nasty - consumers aside, this can affect a LOT of industrial equipment and who knows what else.

While I understand that FTDI is technically within its rights (license agreement says this can happen), there are nicer ways to inform the user, than to, say, shut down his industrial controller, DAQ thingamajig or similar. Message boxes and various other means of informing the user. It sure is the one that will gain the most attention, but the price for it CAN be pretty high.

Essentially, I see a potential backfire of gigantic proportions...

[Richard Crowley]

The response from Adafruit seems unexpectedly ignorant and dismissive.
Are they really that clueless?  Or are they trying to wave their arms through the problem?

[nctnico]

If someone has a fake, can you check the chips serial number(not the USBID number, not serial stored in external eeprom)?

Each of the chips has a unique serial number ( it can be read), it could be the fakes all have the same serial number because they were too lazy/cheap to laser cut / program it.

From my limited number of samples it seems each FT232R clone has a unique serial number.

[nctnico]

What did he use the FTDI IC for?

That's the whole point.  He used a fake chip... or a fake elephant.  Or something.  Or the real elephant came back and trampled the fake ivory because it wasn't real.

You see?

Again: the only fake thing about the 'counterfeit' chips is the FTDI logo on them. The rest is 100% different so it doesn't infringe any copyrights, patents or whatever. It's like the elephant being angry somebody invented synthetic ivory. See Intel versus AMD where AMD implemented Intel's instruction set in compatible products. This resulted in a race which has pushed both Intel and AMD to new boundaries of performance and price which in the end has been very benificial for the end user. Competition results in innovation but FTDI thinks they can just sit and wait for their chickens to keep producing golden eggs.

[hpux735]

I say good riddance to FTDI:

Their chips are over-priced.  It's cheaper to use a usb micro controller an implement a CDC device.

Their drivers SUCK.  On a mac, they haven't updated it since 2012.  If you unplug a device while the port is open it will kernel panic.

All this rhetoric about them being so innovative and "making cool chips for makers" is B.S..

Once, I actually spent a good 20 minutes trying to convince a visiting HP (nee procurve) design engineer to include a FTDI chip in their network switches rather than their RS232 console ports.  They ended up going with a custom solution (probably because FTDI chips cost too much).  I was annoyed because who wants to install YET ANOTHER F-ING DRIVER, but now I have to say that I see the wisdom in their ways.

This feels like the the gasping of a dying company.  That alone may be another great reason to stop designating FTDI parts in new designs.

[LabSpokane]

Again: the only fake thing about the 'counterfeit' chips is the FTDI logo on them. The rest is 100% different so it doesn't infringe any copyrights, patents or whatever.

No. It's an IC plus a driver. And the clone is infringing on the software license agreement.  There would be no issue if the clone ICs supplied their own driver. But they're not. They are using/stealing FTDI's driver in total violation of the licensing agreement. Buy the FTDI chip - use the driver, gratis.  Otherwise, write your own driver.

[(*steve*)]

No. It's an IC plus a driver. And the clone is infringing on the software license agreement.  There would be no issue if the clone ICs supplied their own driver. But they're not. They are using/stealing FTDI's driver in total violation of the licensing agreement. Buy the FTDI chip - use the driver, gratis.  Otherwise, write your own driver.

That's a perfect argument for the driver to detect that the chip is not genuine and then to refuse to operate with it.

And that's exactly what FDTI have done here... or not.

edit: and the history of the license text file in the driver package supports this too (they just added it in the current release)

[welterde]

No. It's an IC plus a driver. And the clone is infringing on the software license agreement.  There would be no issue if the clone ICs supplied their own driver. But they're not. They are using/stealing FTDI's driver in total violation of the licensing agreement. Buy the FTDI chip - use the driver, gratis.  Otherwise, write your own driver.

How can the hardware infringe on the software license agreement? The hardware can't know, which driver will get used. If you plug it into a linux machine for instance it doesn't use the FTDI-provided driver (or if it was written by them it doesn't have such a license agreement attached). The user can possibly infringe on the software license agreement.. but the hardware itself cannot.

[andersm]

The driver has absolutely no way of checking whether the chip says FTDI on it or not, so it cannot assume that every one that fails the "is it genuine" check (anyone here with a USB analyser and both the real and clone chips care to do some analysis on figuring out what exactly is the behavioural difference?) is illegally using the trademark and not just some generic brand.

If the chip is using FTDI's VID it's a pretty clear sign.

USB-CDC should be standard by now; surely some cheap chinese chip consortium could spend some time rolling a cheap CDC chip for the masses?

Creating a new market is much more difficult than latching on to an existing one. But for plain USB to serial converters, class-compliant devices really should be the standard. The other functionality offered by FTDI's devices (FIFOs, bit-banging etc.) is harder to replace.

[Chipguy]

Today I posted on the FTDI facebook page asking for an official statement.

If it turns out to be true using their products as become very risky. A simple screwup anywhere down the purchasing chain could lead to potential disaster. I would have to redesign a few products that add up to about 3000 FTDI chips per year, mostly FT232R and some others.

FTDI better gets an official statement out.
On some point somebody will decompile the driver and eventually find the piece of code that destroys the fakes.
This would be very bad reputation for FTDI.

[maxstone]

That's a perfect argument for the driver to detect that the chip is not genuine and then to refuse to operate with it.

And that's exactly what FDTI have done here... or not.

Perhaps FTDI support are just tired of having to support someone else's faulty hardware with their drivers and decided to solve the problem.

If I can remotely brick the MacBook air that was stolen from me, why can't FTDI brick their stolen hardware?

[firewalker]

Perhaps FTDI support are just tired of having to support someone else's faulty hardware with their drivers and decided to solve the problem.

If I can remotely brick the MacBook air that was stolen from me, why can't FTDI brick their stolen hardware?

Because the MacBook is yours.

The other would be like Apple bricking any third party MacOS compliant system.

Alexander.

[(*steve*)]

If I can remotely brick the MacBook air that was stolen from me, why can't FTDI brick their stolen hardware?

Should Vishay be able to remotely brick your MacBook air if it contains a counterfeit resistor?

[DaWaN]

If it turns out to be true using their products as become very risky. A simple screwup anywhere down the purchasing chain could lead to potential disaster. I would have to redesign a few products that add up to about 3000 FTDI chips per year, mostly FT232R and some others.

As a design engineer I have no control of my supply chain of parts, this is handled by a 3rd party.
Now if the 3rd party sources a fake chip I am screwed. As it is out of my control the consequence is simple: no more FTDI in my designs.
I have a few designs equipped with FTDI chips because the driver can be retrieved from Windows Update (which saves us quite a lot of support).

For personal and/or lab use (we consume quite a few of these cables here, test engineers always manage to blow stuff up..) I am definitely not going to buy any FTDI anymore! I cannot spend my time researching whether a device contains a legit or a fake chip.

Now I am working on a new design, but I'd like to retain a USB -> TTL serial chip with drivers available on Windows Update.
As far as I know the Prolific PL2303 does not have a driver on Windows Update, am I right? I find this chip also hard to source locally for prototypes (EU based, so I prefer Farnell).
Anyone know an alternative?

[maxstone]

Because the MacBook is yours.

The other would be like Apple bricking any third party MacOS compliant system.

Alexander.

Which Microsoft et. el. do all the time if they detect a fake serial (or Vendor ID in FTDI case.)

[firewalker]

Which Microsoft et. el. do all the time if they detect a fake serial (or Vendor ID in FTDI case.)

No they don;t. They disable the stolen goods (e.g. unauthorized Windows copy). They are not killing the hardware containing the stolen goods.

Alexander.

[maxstone]

Should Vishay be able to remotely brick your MacBook air if it contains a counterfeit resistor?

Why not? Although inconvenience, I would just get a refund from Apple. Who in that case would be in the wrong, not Vishay.

[kayvee]

Anyone know an alternative?

Silicon Labs do a range which are easy to use, and I find them cheaper than FTDI.

http://www.silabs.com/products/interface/usbtouart/Pages/default.aspx

[maxstone]

No they don;t. They disable the stolen goods (e.g. unauthorized Windows copy). They are not killing the hardware containing the stolen goods.

Alexander.

And your can reinstall a licensed vendor ID in to the clone part and it'll work again. Just not with FTDI software.

[Rasz]

If I can remotely brick the MacBook air that was stolen from me, why can't FTDI brick their stolen hardware?

Should Vishay be able to remotely brick your MacBook air if it contains a counterfeit resistor?

should you get free fluid change at Mercedes dealers if you drive up in this:

should you scream bloody murder if they _try_ to give you that free service, but fill oil pan with brake fluid due to differences between real and fake one?

As a design engineer I have no control of my supply chain of parts, this is handled by a 3rd party.

so you are not responsible for it, whats the problem?

Now if the 3rd party sources a fake chip I am screwed.

did anything change today? you could ALWAYS get screwed by a fake transistor that burns out after couple of hours

As it is out of my control the consequence is simple: no more FTDI in my designs.

no more transistors too? no more opamps, no more avrs? I suppose you stopped using pendrives and SD cards, they all get faked in china

For personal and/or lab use (we consume quite a few of these cables here, test engineers always manage to blow stuff up..) I am definitely not going to buy any FTDI anymore! I cannot spend my time researching whether a device contains a legit or a fake chip.

you dont have to, fake one simply wont work outright = no wasted time

Now I am working on a new design, but I'd like to retain a USB -> TTL serial chip with drivers available on Windows Update.
As far as I know the Prolific PL2303 does not have a driver on Windows Update, am I right? I find this chip also hard to source locally for prototypes (EU based, so I prefer Farnell).
Anyone know an alternative?

why would you need any? Farnell has original chips and there is zero risk of getting fake, where is the problem?


I simply dont get angry people from this thread - you bought a fake, deal with it. Imagine someone selling Daves uCurrent Gold, build using 5% resistors and lm358, all made to look like the genuine one, except the price is $10. This is your $2 FTDI cable with free shipping.

[(*steve*)]

Which Microsoft et. el. do all the time if they detect a fake serial (or Vendor ID in FTDI case.)

Do they?

The last time I saw anything of this nature it was when my neighbour asked if I could look at their computer.

It had a black background and a note on the screen somewhere that it was an unauthorised copy.

It still worked for most things and there was some sort of option provided to "correct" the fault.

Off topic, but in this case it was a brand name PC with a "genuine Windows" sticker on the side and it was still under original warranty.  I recommended that they take it back to where they had purchased it from.  I assume they did because I've never been asked to look at it again. 

Seemed like (a) false positive, (b) not bricked, (c) still mostly operational, (d) hardware certainly not damaged.

Should Vishay be able to remotely brick your MacBook air if it contains a counterfeit resistor?

Why not? Although inconvenience, I would just get a refund from Apple. Who in that case would be in the wrong, not Vishay.

Suuuuure.

[maxstone]

Which Microsoft et. el. do all the time if they detect a fake serial (or Vendor ID in FTDI case.)

Do they?

I'm guessing that you can yoll back the driver and use the FTDI util to change the vendor ID.

[marcan]

should you scream bloody murder if they _try_ to give you that free service, but fill oil pan with brake fluid due to differences between real and fake one?

You sure as hell should scream bloody murder if they go ahead and punch a hole in the oil pan.

This isn't "FTDI's driver accidentally damages clones". FTDI's driver contains code blatantly, explicitly, and clearly designed to damage clones.

[maxstone]

Should Vishay be able to remotely brick your MacBook air if it contains a counterfeit resistor?

Why not? Although inconvenience, I would just get a refund from Apple. Who in that case would be in the wrong, not Vishay.

Suuuuure.

It's guaranteed by trading standards in the UK.

[maxstone]

This isn't "FTDI's driver accidentally damages clones". FTDI's driver contains code blatantly, explicitly, and clearly designed to damage clones.

They removed their vendor ID from clone chips, nothing more.