FTDI driver kills fake FTDI FT232??

[serggio]

First of all: I would not kill my customers' products and thereby drag third parties into a conflict between me and another party. Dragging third parties into a conflict makes the outcome wildly unpredictable as FTDI must have noticed by now since their latest drivers no longer have anti counterfeit measures.

All it's clear, but what you will do to protect your products? How you will prevent to using your codes on clones?

Secondly is it common knowledge you can't milk a cow forever so you either have to lower prices and/or come up with a new product so I would have tried to stay ahead by creating new products at lower cost.

That is call real competition. FTDI made first and still make good products for USB bus interface with excellent functionality. Even Cypress with their CY7C65213 that copying FT232R pin to pin do not have UART Rx, Tx signal level inversion, that very useful for engineering design. FTDI have huge amount open documentation in form of AN and TN that useful for engineers and enthusiasts, while Cypress and other have one two documents, and that is all. 

All in all the competition coming up with alternatives for succesful products is the way the free market works and in many cases there is very little you can do about it especially if you are not a multi-billion company with thousands of lawyers.

Absolutely agree. Competition! Real competition - that is rear alternative solution with full infrastructure, documentation and support. But not stealing alien ideas and cloning their devises under stolen name.
If somebody will try to do same with Apple iPhone, make 1:1 copy with their current iOS version, believe me, next day Apple will do same as FTDI - release new update for turn fake clones to bricks.

[nctnico]

First of all: I would not kill my customers' products and thereby drag third parties into a conflict between me and another party. Dragging third parties into a conflict makes the outcome wildly unpredictable as FTDI must have noticed by now since their latest drivers no longer have anti counterfeit measures.

All it's clear, but what you will do to protect your products? How you will prevent to using your codes on clones?

Be clever but also accept that no protection method is unbreakable. At some point your are just frustrating your legitimate customers with copy protection measures (like potting a circuit so it cannot be repaired).

[MadDog]

How nice!

I have installed Altera Quartus 16.0. The included driver (FTDI) for the USB Blaster caused my PC (Windows 10 x64) to reset my PC after some minutes over and over again.
It took me some hours to find out that the connected USB Blaster clone caused this. It uses a ST32F to simulate a FTDI chip.

I switched back to the driver delivered with Quartus 14.0: Et voila, it works again!

May these FTDI guys burn...

[Monkeh]

May they burn indeed for not ensuring compatibility with someone elses hack.

[MadDog]

May they burn indeed for not ensuring compatibility with someone elses hack.

I wasn't aware that there is no genuine FTDI used in this device until I opened it (most people don't know this).
So it's ok for you if they crash the PC with all data lost?

[Monkeh]

So it's ok for you if they crash the PC with all data lost?

I'm certain they're doing it on purpose, I mean, it couldn't possibly be a bug with a third party's unsupported implementation of an undocumented protocol, that just doesn't happen. It's all FTDI's fault!!!!!

[wraper]

May they burn indeed for not ensuring compatibility with someone elses hack.

I wasn't aware that there is no genuine FTDI used in this device until I opened it (most people don't know this).
So it's ok for you if they crash the PC with all data lost?

It's your crappy device what crashes it, not FTDI.

[MadDog]

It's your crappy device what crashes it, not FTDI.

No, it doesnt. The driver software does. But whatever you say...

[wraper]

It's your crappy device what crashes it, not FTDI.

No, it doesnt. The driver software does. But whatever you say...

If the crappy device makes the driver doing something unexpected... That driver wasn't written for or tested with your hack implementation anyway. I guess your USB blaster is a crappy clone too.

[nctnico]

It's your crappy device what crashes it, not FTDI.

No, it doesnt. The driver software does. But whatever you say...

If the crappy device makes the driver doing something unexpected... That driver wasn't written for or tested with your hack implementation anyway.

A driver which doesn't handle an unexpected situation properly is written by someone who doesn't know the first thing about programming. Since the old driver works and the new one doesn't it is very likely this behaviour is on purpose.

[wraper]

It's your crappy device what crashes it, not FTDI.

No, it doesnt. The driver software does. But whatever you say...

If the crappy device makes the driver doing something unexpected... That driver wasn't written for or tested with your hack implementation anyway.

Somehow this doesn't happen even with clones of ft232. Now someone gets fake USB blaster with crappily emulated ft232, uses it with genuine software and then blames manufacturer for it behaving weird. I also sometimes use clone devices, but never blame genuine item manufacturer for software not working properly with a fake crap I have.
A driver which doesn't handle an unexpected situation properly is written by someone who doesn't know the first thing about programming. Since the old driver works and the new one doesn't it is very likely this behaviour is on purpose.

[MadDog]

In my case there is no counterfeit FTDI chip used. The STM32 just simulates the behaviour of the FTDI chip.

Their first driver with counterfeit protection (2014) erased the EEPROM connected to the chip with the PID/VID information. So the OS couldn't longer detect the device.
But this is not possible with the STM32 because of the different implementation. So they choose another way which causes the complete PC to crash.
This already happens if the device is just connected to the PC without using it.

This practice is not nice! I use several FTDI chips in devices which I have developed on job. But I will try to avoid to use any chips from FTDI in the future.

[Monkeh]

But this is not possible with the STM32 because of the different implementation. So they choose another way which causes the complete PC to crash.

Do they now? Do you have evidence, or are you just jumping to this conclusion because OMG FTDI ARE BAD?

This practice is not nice!

Nor is yours.

[wraper]

In my case there is no counterfeit FTDI chip used. The STM32 just simulates the behaviour of the FTDI chip.

Which is not a tiny bit different than knowingly using fake chips, even worse. Actually, if fake chip was used, manufacturer could claim they didn't know those were fake. In your case, manufacturer of knowingly made illegal clone of USB blaster and knowingly cloned/emulated FT232 which is illegal too. You are not legally allowed to use FTDI driver with anything else than FTDI ICs in the first place. And you knowingly bought this illegal clone. So how in the end it comes that FTDI is to be blamed for your knowingly bought crap not working properly?

[coppice]

In my case there is no counterfeit FTDI chip used. The STM32 just simulates the behaviour of the FTDI chip.

Which is not a tiny bit different than knowingly using fake chips, even worse. Actually, if fake chip was used, manufacturer could claim they didn't know those were fake. In your case, manufacturer of knowingly made illegal clone of USB blaster and knowingly cloned/emulated FT232 which is illegal too. You are not legally allowed to use FTDI driver with anything else than FTDI ICs in the first place. And you knowingly bought this illegal clone. So how in the end it comes that FTDI is to be blamed for your knowingly bought crap not working properly?

In what way is knowingly emulating a part illegal? The right to most aspects of reverse engineering and emulation for the purposes of compatibility are protected by law. Some aspects of the DMC in the US have tried to prevent such engineering when there is an element of copy protection involved, but that's about the limit of legal protection. Direct copying of someone else's software can get you into copyright trouble. Exactly copying a chip can get you into trouble with the specific copyright protection that exists for masks. Schematics aren't well protected, unless they incorporate patented techniques. Working out how something works, and emulating it without recourse to the internals of the original design, is how much of the electronics industry has always worked.

[wraper]

In what way is knowingly emulating a part illegal? The right to most aspects of reverse engineering and emulation for the purposes of compatibility are protected by law.

Not talking about legality of the hardware side of this this, it already starts to be illegal once you need to use FTDI driver with it. This is exactly the same as for example cloning SEGGER JTAG adapters or Saleae logic analyzers. Yes, software comes for free, as long as you use genuine device...

Direct copying of someone else's software

In this case inevitably illegally using someone else's software.

1.2          In this Licence a "Genuine FTDI Component" means an item of hardware that was manufactured for, and sold by, the Licensor or a member of the Licensor's group of companies. It does not include any counterfeit or fake products.

1.3          If you are a manufacturer of a device that includes a Genuine FTDI Component (each a "Device") then you may install the Software onto that device. If you are a seller or distributor of a Device then You may distribute the Software with the Device. If you are a user of a Device then you may install the Software on the Device, or onto a computer system in order to use the Device.

1.4          In each of those cases you may:

1.4.1          install and use the Software for your purposes only; and

1.4.2          only use the Software in conjunction with products based on and/or incorporating a Genuine FTDI Component.

1.5          The Software will not function properly on or with a component that is not a Genuine FTDI Component. Use of the Software as a driver for, or installation of the Software onto,  a component that is not a Genuine FTDI Component, including without limitation counterfeit components, MAY IRRETRIEVABLY DAMAGE THAT COMPONENT.  It is the Licensee's responsibility to make sure that all chips it installs the Software on, or uses the Software as a driver for, are Genuine FTDI Components. If in doubt then contact the Licensor.

[madires]

A driver which doesn't handle an unexpected situation properly is written by someone who doesn't know the first thing about programming. Since the old driver works and the new one doesn't it is very likely this behaviour is on purpose.

So MS has to retract the driver or replace it with a newer one soon again. Either because of its poor quality or because it's crashing the PC on purpose. Has FTDI any credibility left?

[Macbeth]

My couple of USB Blaster clones:



STM32 and PIC18. The STM32 doesn't work in Windows 10 and newer Quartus. It causes BSODs. Works ok in Linux. It is not bricked. The PIC18 works fine in Windows 10 and Linux.

It is really stretching it to say this one is FTDI's fault.

[nctnico]

It is really stretching it to say this one is FTDI's fault.

No it is not. It is their driver causing the crash.
Because a driver is such an integral part of a kernel it really should not crash under any circumstance because it usually takes the whole OS with it if something goes wrong. Therefore a driver is typically something you go through with a fine comb and catch whatever problem may be lurking. A driver is not the place to skimp on boundary and NULL pointer checks. Best case scenario: FTDI introduced a bug in their newer drivers in order to detect functional equivalent devices and more recently to remove that functionality again.

[wraper]

It is really stretching it to say this one is FTDI's fault.

No it is not. It is their driver causing the crash.
Because a driver is such an integral part of a kernel it really should not crash because it usually takes the whole OS with it if something goes wrong. Therefore a driver is typically something you go through with a fine comb and catch whatever problem may be lurking. A driver is not the place to skimp on boundary and NULL pointer checks. Best case scenario: FTDI introduced a bug in their newer drivers in order to detect functional equivalent devices and more recently remove that functionality again.

How should they catch a problem/bug which most likely is not even possible to occur with genuine ICs? Also I really doubt they care any tiny bit about emulated stuff like this one. This is not a competition for them, nor floods the market with fake ICs. Also, those are outright counterfeit devices anyway, so I doubt they really care about those shady manufacturers as potential customers. What I think is possible, because of authenticity checks in newer drivers, this poorly emulated crap does something weird and crashes not necessarily FTDI driver itself, it can as well crash USB host controller driver.

[Macbeth]

It is really stretching it to say this one is FTDI's fault.

No it is not. It is their driver causing the crash.
Because a driver is such an integral part of a kernel it really should not crash under any circumstance because it usually takes the whole OS with it if something goes wrong. Therefore a driver is typically something you go through with a fine comb and catch whatever problem may be lurking. A driver is not the place to skimp on boundary and NULL pointer checks. Best case scenario: FTDI introduced a bug in their newer drivers in order to detect functional equivalent devices and more recently to remove that functionality again.

Yes, if FTDI deliberately caused their driver to BSOD by detecting fakes (and these are not even fakes - they don't purport to be FTDI chips at all, merely emulating a protocol) then its back to "fuck FTDI". I maybe shouldn't give them the benefit of the doubt after their past shenanigans, but it could easily be just an incompatibility between the newer FTDI drivers and these particular emulations.

[Kjelt]

The driver is made for a special identifier that belongs to an ftdi chip.
No other manufacturer should be allowed to use that identifier, hence if you use a device that tries to emulate or pretend to be such a chip all results are your own problem, you should have used an official device. You can impossibly ask ftdi to test their driver with all illegal or cloned chips, that is ludicrous.

[nctnico]

The driver is made for a special identifier that belongs to an ftdi chip.
 You can impossibly ask ftdi to test their driver with all illegal or cloned chips, that is ludicrous.

Now imagine the scenario where the driver receives malformed data due to an USB issue. When writing a driver you are not done developing software when the driver appears to work. There is a lot more to it!

edit: typo

[Macbeth]

@Kjelt - bollocks the 4 bytes "belong" to FTDI. Or anyone else has to be "allowed" to use those bytes.

[timb]

I think you have 70 pages of reading to do before you can even suspect that you might have anything new to add! 

Amen. Dude doesn't even know who Farnell (element14) are, yet thinks he has something to add.

It's not just Farnell, either. Digi-Key, Mouser, etc. have all had counterfeit chips slip into their distribution stream at one point or another. It happens.